FOOD AND DRUG ADMINISTRATION
DEPARTMENT OF HEALTH AND HUMAN SERVICES
This transcript has not been edited or corrected, but appears as received from the commercial transcribing service. Accordingly, the Food and Drug Administration makes no representation as to its accuracy.
85th Meeting of:
November 3, 2005
10201 Lee Highway, Suite 180
Fairfax, Virginia 22030
TABLE OF CONTENTS
Welcome, Statement of Conflict of Interest, 1
- West Nile Virus Update - Hira Nakhasi 10
Theresa Smith 17
- Draft Guidance on NAT for HIV-1 and HCV - Paul Mied 27
- Summary of the TSEAC Meeting of October 31, 2005 37
- David Asher
- Summary of DHHS Advisory Committee on Blood Safety and 50
Availability - Jerry Holmberg
- Re-entry of Donors Deferred Based on anti-HBc Test
Results - Gerardo Kaplan 61
- Susan Stramer 64
Approaches to Over-the-Counter Home-Use HIV Test Kids
- Introduction and Questions to the Committee 77
- Elliott Cowan
- Proposal for an OTC Home-Use HIV Test Kit 101
- Sue Sutton-Jones
- Changes in HIV Test Practices and Counseling 139
Recommendations - Bernard Branson
- Role of Quality Systems for Diagnostic Tests 172
- Devery Howerton
- Psychological/Social Issues Associated with HIV 187
Testing and OTC Home-Use HIV Tests - Joseph Inungo
- Human Factors in OTC Testing - Arleen Pinkos 205
- Open Public Hearing 224
- Elliott Millenson 225
- Wesley Rodriguez 232
- Patrick Keenan 238
- Freya Spielberg 242
- Waheed Khan 252
- Patricia Charache 255
- Tracy Powell 261
- Ernest Hopkins 266
- Tom Donahue 271
- Duralba Munoz 276
- Richard Cizik 285
- Tom Myers 288
- Shawn Fay 291
- Damen Dozier 298
- James Sykes 303
- Bill Parra 306
- Neeraj Vats 308
- Terry Anderson 311
- Questions to the Committee and Committee Discussion 315
JAMES ALLEN, MD, MPH, Chair. President and CEO, American Social Health Administration, Research Triangle Park, NC
DONNA M. DI MICHELE, MD, Chair (Topic I only). Associate Professor of Pediatrics and Public Health, Weill Medical College and Graduate School of Medical Science, Cornell University, New York, New York
MATTHEW KUEHNERT, MD, CDR, U.S. Public Health Service, Assistant Director for Blood Safety, Division of Viral and Rickettsial Diseases, CDC, Atlanta, Georgia
CATHERINE S. MANNO, MD, Professor of Pediatrics, The Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
KEITH QUIROLO, MD, Hemoglobinopathy Pediatrician, Clinician Director, Apheresis, Transfusion Medical Director, Sibling Donor Cord Blood Program, Department of Hematology, Children's Hospital and Research Center, Oakland, California
GEORGE C. SCHREIBER, ScD, Vice President, Health Studies, Westat, Rockville, Maryland
DONNA S. WHITTAKER, PhD, Director, Robertson Blood Center, Fort Hood, Texas
JUDITH BAKER, MHSA, Regional Coordinator, Federal Hemophilia Treatment Centers, Children Hospital, Los Angeles, CA
NON-VOTING INDUSTRY REPRESENTATIVE
LOUIS KATZ, MD, Executive Vice President, Medical Affairs, Mississippi Valley Regional Blood Center, Davenport, Iowa
ACTING NON-VOTING INDUSTRY REPRESENTATIVE
WILLIAM H. DUFFELL, PhD, Director of Government Affairs, Regulatory Affairs Quality Systems, Gambro BCT Lakewood, CO
TEMPORARY VOTING MEMBERS:
HENRY M. CRYER, III, MD, PhD, Chief, Trauma and Critical Care, Division of General Surgery, University of California, Los Angeles, California
ADRIAN M. DI BISCEGLIE, MR, Professor of Medicine, Chief of Hepatology, St. Louis University School of Medicine, St. Louis, Missouri
SAMUEL H. DOPPELT, MD, Chief, Department of Orthopedic Surgery, The Cambridge Hospital, Cambridge, Massachusetts
HARVEY KLEIN, MD, Chief, Department of Transfusion Medicine, National Institutes of Health, Warren G. Magnuson Clinical Center, Bethesda, Maryland
ROSHNI KULKARNI, MD, Professor and Chief, Pediatric and Adolescent Hematology/Oncology, Michigan State University, East Lansing, Michigan
SAMAN LAAL, PhD, Assistant Professor, Department of Pathology, New York University School of Medicine, New York, New York
KENRAD NELSON, MD, Professor, Department of Epidemiology, Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland
THOMAS QUINN, MD, Professor of Medicine and Deputy Director, Infectious Disease Division, The Johns Hopkins University, Baltimore, Maryland
FREDERICK SIEGAL, MD, Medical Director, Comprehensive HIV Center, St. Vincent's Catholic Medical Center, St. Vincent's Manhattan, New York, New York
GORDON SNIDER, MD, Towson, Maryland
IRMA O.V. SZYMANSKI, MD, Professor of Pathology, Emeritus, University of Massachusetts Medical Center, Department of Pathology, Worcester, Massachusetts
DONALD JEHN, Executive Secretary, Blood Products Advisory Committee, Division of Scientific Advisors and Consultants, CBER, FDA
COMMITTEE MANAGEMENT SPECIALIST:
PEARLINE MUCKELVENE, Division of Scientific Advisors and Consultants, CBER, FDA
SUSAN ZULLO, PhD, Acting Associate Director for Policy, Office for Blood Research and Review, CBER, FDA
RHONDA DAWSON, Policy Analyst, Office for Blood Research and Review, CBER, FDA
P R O C E E D I N G S (8:20 a.m.)
Agenda Item: Welcome, Statement of Conflict of Interest, Announcements.
MR. JEHN: I am going to go ahead and begin here. Mr. Chairperson, members of the committee, invited guests, consultants and public participants, I would like to welcome you to the 85th meeting of the Blood Products Advisory Committee. I am Donald Jehn, the executive secretary for this meeting.
The entire meeting is open to the public today. At this time, I would like to introduce the individuals seated at the head table for today's first session.
Starting over to my right here, and going counterclockwise around the table, we have Dr. Kenrad Nelson, professor of epidemiology from Johns Hopkins. Dr. Quinn will be seated next to him when he arrives. He is also from Johns Hopkins.
Dr. Samuel Doppelt is the chief of the department of orthopedic surgery, Cambridge Hospital. Dr. Henry Cryer III, chief of trauma and critical care, UCLA.
Dr. Schreiber, vice president of health studies, Westat. Dr. Roshni Kulkarni, professor and director of pediatric and adolescent hematology oncology, Michigan State University. Dr. William Duffell, Jr., director of government affairs, regulatory affairs, quality systems, Gambro.
Coming over to this side now, we have Judith Baker, consumer representative, regional coordinator, federal hemophilia treatment center, region IX, Children's Hospital, Los Angeles.
Next to her, Dr. Adrian Di Bisceglie, professor of medicine, chief of hepatology at St. Louis University School of Medicine.
Dr. Suman Laal, assistant professor, department of pathology, New York University School of Medicine, Dr. Frederick Siegal, medical director, comprehensive HIV center, St. Vincent's Hospital, Manhattan.
Dr. Catherine Manno, professor of pediatrics, Children's Hospital, Philadelphia. Dr. Matthew Kuehnert, assistant director for blood safety, division of viral and rickettsial diseases, CDC.
Dr. Donna Whittaker, director of Robertson Blood Center, Fort Hood, Texas. Dr. Irma Szymanski, here to my immediate right, professor of pathology emerita, University of Massachusetts Medical Center. Dr. Harvey Klein, chief of the department of transfusion medicine, NIH.
Dr. Donna DiMichele, she will be acting chair for topic I discussion. She is the assistant professor of pediatrics and public health, Weill Medical College and Graduate School of Medical Science, Cornell.
Our regular BPAC chair, Dr. James Allen, president and CEO, American Social Health Administration, Research Triangle.
There are two committee members not present for the meeting, Dr. Quirolo and Dr. Katz. I would like to thank the members and consultants for attending this meeting.
Before we begin, Dr. Epstein, can I have you forward? There are four retiring members of the committee that we are going to recognize these individuals.
Could we have Dr. DiMichele, Dr. Doppelt, Dr. Klein and Dr. Laal to step forward? Thank you.
DR. EPSTEIN: Let me just remark that it is a bittersweet honor to give these awards to our outgoing committee members.
We deeply appreciate the effort that each of you have provided on our behalf, and in the interests of the public health.
We know that this participation represents a certain degree of personal sacrifice, reading long documents, sitting in a chair long hours, and scratching your brain for all of us. So, these are small tokens of our appreciation, which I am very pleased to hand to you. So, I guess one at a time.
First, to Dr. Harvey Klein, with our deep appreciation. [Award given, photograph taken, applause.]
This is to Dr. Donna DiMichele, again, with our great thanks. [Award given, photograph taken, applause.]
Dr. Samuel Doppelt, again, thank you very much. [Award given, photograph taken, applause.]
Dr. Laal will receive hers later. Thank you.
MR. JEHN: Before I turn the meeting over to the chair, I have a conflict of interest statement to read. Bear with me. It is a little lengthy.
The Food and Drug Administration, FDA, is convening today's meeting of the Blood Products Advisory Committee under the authority of the Federal Advisory Committee Act of 1972.
With the exception of the industry representative, all members and consultants of the committee are special government employees or regular federal employees from other agencies, and are subject to federal conflict of interest laws and regulations.
The following information on the status of this advisory committee's compliance with federal ethics and conflict of interest laws, including, but not limited to, 18 USC Section 201, and 21 USC Section 355(n)(4) has been provided to participants in today's meeting, and to the public.
FDA has determined that members of this advisory committee and consultants to the committee are in compliance with the federal ethics and conflict of interest laws, including but not limited to, 18 USC Section 208, and 21 USC, Section 355(n)(4).
Under 18 USC Section 208, applicable to all government agencies, and 21 USC Section 355(n)(4) applicable to certain FDA committees, congress has authorized FDA to grant waivers to special government employees who have financial conflicts, when it is determined that the agency's need for a particular individual's services outweighs his or her potential financial conflict of interest, section 208, and where participation is necessary to afford special expertise, section 355.
Members and consultants of the committee who are special government employees at today's meeting, including special government employees appointed at temporary voting members, have been screened for potential financial conflicts of interest of their own, as well as those imputed to them, including those of their employer, spouse or minor child related to discussions of approaches to over-the-counter home use HIV test kits, and the discussions of alpha-1 protease inhibitor products.
These interests may include investments, consulting, expert witness testimony, contracts, grants, CRADAs, teaching, speaking, writing, patents and royalties, and primary employment.
Today's agenda for topic I includes review and discussion of the approaches of over-the-counter home use HIV test kits. For topic II, the committee will review and discuss alpha-1 protease inhibitor products.
In accordance with 18 USC Section 208(b)(3), waivers have been granted to the following special government employees: Dr. Donna DiMichele, topics one and two, Dr. Catherine Manno, topic two.
In addition, Dr. James Allen has recused himself from the discussion of topic one related to HIV home test kits.
A copy of the written waiver statement may be obtained by submitting a written request to the agency's freedom of information office, Room 12-A-30 of the Parklawn Building.
With regard to FDA's guest speakers, the agency has determined that the information provided by these speakers is essential.
The following information is being made public to allow the public to objectively evaluate any presentation and/or comments made by the speakers.
Dr. Mark Brantly is professor of medicine, molecular genetics and microbiology and alpha-1 research professor, University of Florida, Gainesville.
Dr. Bernard Branson is associate director for laboratory diagnostics, division of HIV/AIDS prevention, CDC, Atlanta.
Dr. Devery Ann Howerton is chief, laboratory practice, evaluation and genomics branch, coordinating center for health information and services, CDC, Atlanta.
Dr. Joseph Inungo is professor, school of health sciences, Central Michigan University.
Dr. Hans Peter Schwarz is associate professor of medicine, vice president, global preclinical research and development, Baxter Bioscience, Vienna, Austria.
Dr. Sue Sutton-Jones is senior vice president, regulatory affairs and quality assurance, OraSure Technologies.
Any guest speakers will not participate in the committee deliberations, nor will they vote. In addition, there may be regulated industry and other outside organization speakers making presentations.
These speakers may have financial interests associated with their employer and with other regulated firms.
The FDA asks, in the interests of fairness, that they address any current or previous financial involvement with any firm whose product they may wish to comment upon. These individuals were not screened by FDA for conflicts of interest.
Dr. William Duffell, Jr., is serving as the industry representative, acting on behalf of all related industry, and is employed by Gambro BCT. Industry representatives are not special government employees and do not vote.
This conflict of interest statement will be available for review at the registration table. We would like to remind members and consultants that, if the discussions involve any other products or firms not already on the agenda, for which an FDA participant has a personal or imputed financial interest, the participants need to exclude themselves from such involvement, and their exclusion will be noted for the record.
FDA encourages all other participants to advise the committee of any financial relationships that you may have with any sponsor, products, direct competitors and firms, that could be affected by the discussions. Thank you. Dr. Allen, I turn it over to you.
DR. ALLEN: Thank you, Mr. Jehn. I would like to add my welcome to all the committee members, and my thanks to the retiring members. It has been wonderful working with you, and we will miss you.
We have a very full agenda today. We are going to do one more retirement here.
DR. EPSTEIN: I would like to call Dr. Suman Laal up at this time and, once again, please step forward, to thank you personally for your service to this committee and, again, this is an expression of appreciation on behalf of all of us and the public health. [Award given, photograph taken, applause.]
DR. ALLEN: Jay, I think we will have to come up with another name for retirement, because there are no retiring members on this panel.
We do have a very busy schedule today. During topic one, from which I will recuse myself, we have got a number of open hearing speakers, as well as our regularly scheduled speakers.
I would just like to remind everyone today, please keep to your allotted time. It is essential that we have adequate time for the committee to have full discussion and to ask questions, and to seek the information that we need in order to provide the proper advice to the Food and Drug Administration.
For the updates this morning, rather than running through each of the updates in sequence and then having time for discussion, I am going to allow time for discussion immediately after each one, because each topic is so different.
So, we will move straight into the first discussion item, which is a west nile virus update by Dr. Hira Nakhasi of the FDA and Dr. Theresa Smith of the CDC.
Agenda Item: Committee Updates: West Nile Virus Update.
DR. NAKHASI: Good morning, everybody. The first speaker always has to be on time, I guess. So, I will try my best to be on time, and also just to give you a quick update on the epidemic of 2005, the west nile virus epidemic of 2005, and our efforts to see how things are moving. Also, Dr. Theresa Smith will talk from the CDC perspective.
Now, the background here, if you were not here for the last three years, then I think you should be looking at this slide.
However, if you were here for the last three years, at every BPAC, I have been talking the same thing. I just want to briefly say that West Nile is a single stranded RNA virus, a flavivirus.
It is mosquito borne. Eighty percent of infection is asymptomatic, 20 percent develop mild fever, and approximately one in 150 infections result in meningitis or encephalitis.
Old people like me are at risk for neurological diseases for a period two weeks prior to symptoms, and can last more than a month, which was a new revelation during the course of this epidemic.
In 2002, when the first outbreak -- not the first outbreak, the first transmission cases -- occurred, and in that time we realized that west nile can be transmitted through blood and transplantation and other sources. However, the magnitude of risk is still unknown through the transmission. Virus titers of this blood is much lower than the other known viruses, like HIV, HCV, and IgM can persist, that antibody can persist, up to two years. However, there is no chronic stage.
This is the progression of the epidemic, west nile epidemic, in the Americas. You can see the United States started in 1999, Canada 2000, spread to Mexico in 2002, and Central American in 2003, and then to the Caribbean.
This just gives you the overall human cases over the period of time. What you see is here, that the epidemic started in 1999 and reached a peak in 2003, and the last two years it has sort of plateaued here.
You will hear more about the epidemic from Theresa. This is year, 2005, as of now, most of the states in the lower 48 states, they are human cases, but still, Washington, northeastern and some places in Virginia you still have only avian activity but not the human activity.
This is to give you an idea of how many cases of the blood donors this years. As of November 374 cases in blood donors that were detected by testing. Again, you can see it is spread out all over the place.
This just gives you an idea how we progressed since we started testing the blood as of July 1, 2003, using Genprobe and Roche tests.
The first year, in 2003, we found 880 donors who had reported to the CDC arbonet and, at that time, there were six confirmed transfusion transmitted cases. However, out of six, four out of six had very low viremia.
In 2004, we saw a dramatic decrease in the number of cases from approximately 880 to 224. Again, I want to emphasize that this is mini-pool NAT and, during this season, we started to also, in certain areas, where the incidence and prevalence of these cases was much higher, the industry introduced the individual NAT testing.
Last year we found one reported case of transfusion transmitted cases. However, that was only detected by ID NAT and it happened in a situation right before the ID Nat was instituted in that area.
This area, as of now, October 1, 2005, we saw a little bit of an increase in the number of cases, 374 so far, and it has already petered out. In the last couple of weeks we see one or two cases every week. Again, the peak was between late August, early September.
Again, this year, mini-pool NAT and ID NAT was instituted in mild cases, where the high cases, where the prevalence was much higher, the ID NAT was started right away.
Fortunately, so far we have no reported transfusion transmitted cases this year. However, there were three cases of transmission through transplantation.
As you know, we have issued several guidances over a period of time in 2002 and 2003, and now this year we issued a revised guidance, basically to update what is happening with the epidemic, and to keep on changing the donor deferral periods.
This year we show, under the new revised guidance of 2005, basically says the epidemic season starts between May 1 and November 30.
The reason I am saying May 1, for the last several years we have seen it coming earlier and earlier, except this year, where it started a little bit late, and last year it started toward the middle of April, end of April kind of a thing.
So, we said, if the cases are during that epidemic season, donors of suspected west nile infection, or diagnosed with west nile infection, should be deferred for 120 days.
What we found out is that the studies showed last year that, in some of the donors during the epidemic period, could be up to 104. Mind you, it was only one case, but there were some cases that were beyond 28 days, and the earlier guidance said 28 days, because that was based on the studies in the 1950s.
We found out that in some cases -- again, mind you, this is low level of virus in the presence of antibodies. So, we do not know whether that is infectious or not.
So, we decided that it would be 120 days deferral, would be fine. Donors would be deferred on the basis of investigational tests, and blood establishments, at their own discretion, may enter such donors after 120 days after the reactive donation, i.e., that they may not have to test them again before entering.
Earlier we had suggested that, before re-entering, they should be tested. However, the FDA recommends that we still would like to be tested those who are re-entered, be tested by ID NAT on a follow up sample during the 120 days, which will provide useful additional scientific information on the duration of west nile viremia.
Still it is evolving. The first year we thought it was only 28 days, the next year we found out 49, 50 days. Last year we found that one of the cases was 104 days.
So, it will be important to find out how long the viremic period is, because this is an evolving situation. If such a follow up sample is reactive for west nile virus, the FDA recommends that the donor be deferred for an additional 120 days from the day the sample was taken.
So, we still are continuing working closely with the test kit manufacturers to expedite the licensure, and we still are having the meetings bi-weekly during the season with the AABB task force, and I am really thankful for them, in collaboration with CDC and NIH. So, we monitor the epidemic and then make a course correction if that is needed.
So, in having this story told so far, there are still gaps in knowledge, which we in the FDA are working on. A couple of things, one is the genetic variation in west nile strains.
Even though there is limited data, I will show you some data that there may be some variation, but the question is, why are we studying that? Is it important because it may have impact on the west nile assays?
Then also, an important issue still remains whether there is any residual risk of west nile infection in the presence of antibodies. That means, those people who have low titer virus by mini-pool or ID NAT, and are having antibodies, are they infectious.
I think those studies, we want to ask those questions. So, Maria Rios' lab in FDA has collected several of these isolates, 25 in total, from various periods of the epidemic, and sequenced the structural gene, and five of them have completely sequenced the complete genome.
What she found out, and others found out, was that there was a genetic shift between 1999 and 2001 and 2002 in the sequences, at least in the structural regions. So, there was a major shift there, that 55 percent of the oral isolates in 2002 had changes, whereas 85 percent had a different group of isolates in 2003.
However, between 2005 and 2002, the majority of the nucleotide changes were in U to C and A to G, and a small number of mutations also have so far not affected the testing. So, thank God for that.
She also did in vitro infective study samples, which were mini-pool NAT positive or ID NAT positive and antibody positive or mini-pool NAT negative or ID NAT positive.
What she found out, 10 out of 16 samples, when she infected these blood samples onto viral or macrophage cells, she could see the virus replication.
That doesn't mean that virus is infectious, and it doesn't mean that it can be transmitted, but it tells you that it is infectious.
Although in vitro infective does not imply infective in vivo, it demonstrates the presence of live virus and, therefore, raises some concerns about a potential risk for transfusion transmission.
I should hasten to say that, so far, we have not seen any transfusion transmitted cases from samples, if they were antibody positive samples. So far, in real life, we have not seen that.
There remains a potential for that, even though it is low. So, what we are trying to do now is doing it in non-human primate studies as well as small animals, to see whether these samples are infectious.
With that, I think I will acknowledge my colleagues at CDC, the AABB task force, other organizations -- ABC, ARC, BSL, Roche, GenProbe, Department of Defense and FDA colleagues. Thank you very much.
We are doing a tag team, and Theresa can talk and if you have any questions, then we can come together.
Agenda Item: West Nile Virus Update.
DR. SMITH: Thank you for letting me tell you about the west nile virus epidemiology and surveillance update for 2005.
I will skip virology this year, since Dr. Nakhasi covered it so well but, like last year, I will go ahead and discuss the epidemiology of west nile in the United States quickly from 1999 and then, more fully, a brief review of last year and how things have gone this year. Then we will discus what we have seen in blood donations. As marked on this slide, all data, including 2005, was in Arbonet as of 11-1-05.
The data that we are going to be looking at is from Arbonet, a national arbovirus surveillance system. It is a web based passive system begun in 2000 in response to the new epidemic in the United States.
It includes 57 area health departments that report over the internet to the division of vector borne infectious diseases.
It includes both animal and environmental data in the form of mosquito, bird, horse and other animal surveillance, and human cases.
Human cases have a great deal of information collected, including age, sex, race, ethnicity, residence, clinical illness, onset date and outcome, blood or organ donation or receipt.
You may have seen some of these maps over time. This gives a sense of when we first saw this virus, and when we have seen activity in environment versus activity in humans.
You will notice we have something like areas in our country where we only see transmission in the environment and not yet any human cases, and the rest of the country has experienced human cases.
In 2004, we saw a great deal of activity down in the southwest. While it had been there before, it was the first time that we saw a lot of human cases.
During this year, we see that area continues to have quite a few human cases, and it is becoming a little denser in terms of the counties that become affected by west nile.
This is a quick reminder that all the data that you are looking at revolves around reports that come in, and that is always going to lag behind the actual onset.
So, in orange you see the onset of cases. In green you see the reporting of cases. So, what you see today is not going to look the same when we finalize this year's data.
Here you can again see that we have started with relatively few cases in 1999. Our peak year was in 2003. The last two years have been fairly level.
So far this year, we have had 2,581 case reports in 42 states. Of those, 1053 have been neuroinvasive disease.
West nile fever has accounted for 1,368. There have been 83 deaths. Of those 83 deaths, we have information on the age of those people in 78 cases. The median age was 78 with an age range of 36 to 98.
There were six states with cases that were greater than 100 in number. California has 824 so far, Arizona has 102. These together account for nearly two thirds of all the cases in the United States.
We have been collecting data on screened blood since 2003, when 6.2 million units were screened. Eight hundred eighteen presumptive viremic donors were recognized.
Of those, six ultimately developed west nile non-invasive disease, 137 developed west nile fever, and six were related to transfusion associated transmission of west nile.
In 2004, 8.2 million units were screened, 224 presumptive viremic donors were recognized, of which four developed west nile neuroinvasive disease, 66 developed west nile fever, and there was one transfusion associated transmission.
In this year, we of course don't know how many will be screened by the end of the year. So far, we have seen 375 presumptive viremic donors. Three have developed west nile neuroinvasive disease, 82 have developed west nile fever, and there have been no transfusion associated transmissions this year.
Here is the map from 2003 showing that the predominance of the presumptive viremic donors were in the midwestern and high plains states.
Last year, where we see a little more scattering of where the presumptive viremic donors were found, although some of the highest density areas were where the newest activity was in the southwest.
This year, you again see some scattering, but there also are some little clusters. It looks like the Missouri valley, Texas and California would be the areas that we have seen the most activity in.
In 2004, again, 224 cases were reported in 29 states -- California, Arizona, Colorado and Texas being the most common places to find presumptive viremic donors last year.
This year, we have had more presumptive viremic donors -- 374 -- in 30 states, with California, Texas, Nebraska and Louisiana being the most common places to find presumptive viremic donors this year.
During this year we have had six investigations. As of last night, four are now negative and two are pending plasma return from Switzerland.
We did have one organ transplantation transmission. In the past, organ transplantation transmission has also been associated with transfusion associated transmission. This year it was likely to be a mosquito borne infection in the donor.
There was potentially a febrile illness in that person, but that was a history that was not obtained prior to their donation.
Later, again, after there was recognized to be a problem, it was found that the day before donation, this person was IGM positive for west nile, although PCR negative.
Three of the four organ recipients became PCR positive, and two became ill. Here is a quite time line of what occurred.
On the top you see information that was available at these dates. So, on August 16 there were known to be west nile positive mosquitoes in a park in the area where this person lived.
On the 23rd, this person was injured and required surgery. On the 26th, they were declared brain dead, and on the 28th organ recovery and transplantation occurred.
On the bottom you see information that was available only after an investigation occurred. It was found that there was a possibility of a febrile illness before this gentleman's injury, and blood that was recovered from the 27th of August was found to be PCR negative and IGG positive for west nile.
In summary, west nile activity has continued over most of the continental United States. It has continued its westward expansion insofar as it has increased its presence in many of the counties in the southwest.
The full season of transmission in California occurred for the first time in this last year. Now human cases have been reported from all states except Alaska, Hawaii, Maine and Washington.
The risk of west nile virus transfusion associated transmission remains unknown. We continue to investigate possible transfusion associated transmissions, as we will next year. We have only two pending this year.
As we have seen from this last occurrence, the transfusion associated transmission is less likely to cause organ transplant related west nile virus transmissions than it was before we started our work in looking at the blood.
I would also like to point out that there are some areas of the United States that appear to be remaining important spots for transmission.
Here the Missouri Valley appears to be a consistent area since it first became involved in west nile. This may be something that we need to continue watching.
Each year we learn more about where this organism wants to live and transmit the most. Thank you. Any questions?
DR. ALLEN: Questions or comments for either Dr. Nakhasi or Dr. Smith?
MS. BAKER: A question. Is there any west nile transmission surveillance activity in the U.S. Pacific jurisdictions?
DR. SMITH: I am sorry, I don't think I understood that question.
MS. BAKER: We have the territory of Guam, and there are two other Pacific jurisdictions that are affiliated with the United States, and three others that are less affiliated. I wondered if they were involved in surveillance.
DR. SMITH: Puerto Rico is. The others have not been very active, simply because there is little evidence that it is going to get across that ocean again very quickly.
Over time, we are going to need to be able to make sure that they have the capacity for surveillance, but right now, no.
DR. CRYER: Is there a clinical difference in the course of the disease in the transplant patient that is immunosuppressed?
DR. SMITH: It appears so, but it is a little hard to tell because we have had so few to look at. There is no reason to believe that all four people didn't receive an organ that was capable of transmitting west nile virus, yet only three actually seemed to have any evidence that the transmission occurred.
Of those three, only two people got sick, but one of the people that did not get sick had a rather prolonged course of having at least PCR positive blood, and that may not be something that is due to their transplant, per se.
Just as the blood donor had long-term PCR positive blood, it is hard to tell whether or not that had to do with transplant or just some people are capable of having a little longer term viremia. I wish I could tell you more than that. Those are the things we are looking at, though.
DR. ALLEN: Are organ donors, at this point, routinely tested for west nile virus in most circumstances or not?
DR. SMITH: No, the test that we have been using on blood has only been given an IND for that purpose at this point, is my understanding.
DR. DUFFELL: I am curious about your thoughts about the drop in numbers from 2003. Is this an indication that this was almost like a medical fad at the moment, that people were looking for this, noticing it and reporting it, and then the interest in it has dropped off? Has the reporting system changed in some way that might account for the dramatic drop in numbers?
DR. SMITH: It is difficult to tell, but there does appear to be a point at which west nile becomes entrenched in an area in a way that it has covered a certain geography, it is capable of being in the mosquitoes, in the birds, in the humans, but the majority of people have never seen it. The majority of birds have never seen it. So, all are susceptible.
Why that decreases, whether it is a change in the mosquitoes, the birds or the humans is a little difficult. There is certainly evidence that, once west nile virus is introduced into an area, that people are much more careful about using mosquito sprays when they go out to garden, and wearing long sleeves and long pants.
It is also possible that birds become resistant to the transmission, making it impossible, then, for the mosquitoes to continue the cycle as efficiently each year.
I don't think it is merely a surveillance artifact. I think this is a matter of the rather complex work that goes into supporting a transmission cycle like this in nature.
DR. NAKHASI: I just wanted to make another point here. I don't think it is that interest in the surveillance has dropped.
I think as you see in the blood area, the testing has been going on year round, in most of the blood centers. You can see, when you did the number of cases in 2003 were more than 800, and last year were 224, and this year a little bit more.
Obviously, it is fluctuating, but what was happening, I think what Theresa said, that was the first time it was a full blown epidemic going through mosquitoes, a new population of mosquitoes were being infected and, once it established, and birds were infected, birds were dying.
It is known in the literature that, once the birds and other people get infected, they may develop resistance, and therefore you may see a drop in the human cases.
DR. ALLEN: Other questions? If not, we do need to move on. Thank you very much. Our next presentation, by Dr. Paul Mied, Food and Drug Administration Draft Guidance on NAT testing for HIV-1 and hepatitis C viruses, testing, product disposition, donor deferral and reentry.
Agenda Item: Draft Guidance on NAT for HIV-1 and HCV: Testing, Product Disposition and donor Deferral and Re-entry.
DR. MIED: Thank you, Dr. Allen. This morning, I would like to very briefly provide an update on comments FDA has received to the docket for our draft guidance on NAT for HIV-1 and HCV, testing, product disposition, and donor deferral and reentry.
This draft NAT guidance for industry was published on July 27, 2005, for comment purposes only. The 90-day comment period closed on October 25, 2005.
This guidance provides recommendations to blood and plasma establishments that have implemented, or are implementing, a licensed HIV-1 and HCV NAT method for source plasma or whole blood.
This guidance contains generalized testing algorithms, to be used when NAT reactive results are obtained on a pool of samples, or on individual samples of source plasma or plasma from whole blood donations.
Now, these testing recommendations deal with a NAT reactive result only, and they are independent of those for serology testing that were in the 1992 HIV memo, and the 1993 HCV memo to blood establishments.
Currently approved tests on master pools of donor samples, or on individual donor samples for HIV-1 RNA and HCV RNA may be either multiplex NAT for the simultaneous detection of HIV-1 RNA, and/or HCV RNA, or separate NAT tests for the RNA of the two viruses.
Now, the draft guidance includes six different testing algorithms that cover all of these testing situations.
The recommendations on testing, product disposition and donor deferral address the actions to be taken when an individual donor sample is reactive on a multiplex HIV-1 HCV NAT, after a negative antibody screening test, and when a master pool is reactive on a multiplex HIV-1 HCV NAT, and that reactive master pool is then resolved down to the level of the reactive individual donation by testing sub-pools and/or by testing individual donor samples.
The recommendations also address the actions to be taken when an individual donor sample is reactive on a separate HIV-1 or HCV NAT after a negative antibody screening test, and when a master pool is reactive on a separate HIV-1 or HCV NAT, and that reactive master pool is then resolved down to the level of the reactive individual donation by testing sub-pools, and/or by testing individual donor samples.
Now, in the guidance, we refer to an individual NAT for the separate viruses HIV-1 and HCV, but one of the comments that we received was to change this terminology to separate NAT, and I found it helpful to do that, at least for this presentation.
Some of the highlights of the recommendations in the draft guidance on testing and donor and unit management are, we recommend actions to be taken regarding the unit and the donor and for look back, product retrieval and recipient notification, at each point in the testing algorithm.
When an individual donor sample is reactive on a multiplex HIV-1 HCV NAT, we are recommending deferral of the donor whether or not one of the discriminatory NAT tests is reactive.
In accordance with a previous recommendation by the blood products advisory committee, we recommend that a non-reactive NAT on subpools or on individual donations be definitive for release on all units in those non-reactive subpools, or for release of those non-reactive individual donations.
For look back, the recommendations indicate, for product retrieval only, or recipient notification also, should be done.
Now, here is a generalized algorithm for using the multiplex NAT, that combines all the testing algorithms that are in the draft guidance into one algorithm.
It refers to a situation in which, whether you are resolving a reactive master pool down to the reactive subpool, and then down to the individual donation level, or you are screening individual donations in the first place, and you have a reactive individual donation on the multiplex NAT.
Units in all non-reactive subpools and all non-reactive individual donations may be released, and we recommend running discriminatory NAT tests on the individual donation, and discarding the unit, and deferring the donor when one or both of those discriminatory NAT tests is reactive.
When both discriminatory NAT tests are non-reactive -- that is, when there is a NDR or a non-discriminated reactive result -- we are offered the option of performing an additional NAT at this point, and discarding the unit and deferring the donor, regardless of the results of that additional NAT.
The result does indicate whether you should perform product retrieval and also initiate transfusion recipient notification when the result is reactive, or just do product retrieval when the result is non-reactive.
Now, some of the major comments to the docket that we have received that address the testing and unit and donor management recommendations in the guidance include a question about the need to investigate each occurrence of a NAT reactive unresolved master pool, that is, when all of the subpools or the individual donations subsequently test non-reactive.
Also, a question about the need to defer donors with NAT reactive results of the individual donation, when both of the discriminatory NAT tests are non-reactive.
These are the donors with an NDR or a non-discriminated active result. The argument being made here is that the vast majority of these NDRs are false positive. There is also a request for us to define the time frame for look back for HIV and HCV in this document.
Now, this draft guidance also contains algorithms for donor reentry for HIV and HCV that combine NAT and serologic test results obtained on the donor.
This is the recommended reentry algorithm for the three classes of donors deferred because of HIV-1 test results.
Here is the first class of donors who were individual donation NAT reactive but serology negative, the second class of donors who were not non-reactive, HIV-1, 2, combi-EIA repeatedly reactive, western blot or IFA indeterminate or negative, and the third class of donors, who were negative on the other tests, but were repeatedly reactive on the HIV-1 P24 EIA, with a neutralization test that was positive or indeterminate.
We are recommending that, after eight weeks, you conduct non-donation testing on a follow up sample from the donor, using an HIV-1 NAT and an anti-HIV-1, 2 EIA.
If the HIV-1 NAT is non-reactive and the anti-HIV-1, 2 EIA is negative, the donor may be reentered. That is, the donor is eligible to donate in the future, provided, of course, the donor meets all donor eligibility criteria.
Some of the major comments to the docket that we have received that address the recommendations for donor reentry for HIV in the guidance include the suggestion that there be an additional waiting period for a donor who tests NAT reactive during that eight week deferral period. We had recommended that that donor be permanently deferred.
That we clarify the conditions for reentry of donors who were deferred because of their HIV-2 test results.
That we clarify the requirements for Group O sensitivity for the individual sample NAT, and for the EIA on the follow up sample, and that we permit reentry for donors who initially had an invalid or an unreadable western blot.
This is the recommended reentry algorithm for the two classes of donors deferred because of HCV test results. Here is the first class of donors, who were NAT reactive, serology negative, and the second class of donors, who were NAT non-reactive, HCV EIA repeatedly reactive, RIBA indeterminant or negative.
We are recommending that, after six months, you conduct non-donation testing on a follow up sample from the donor using an HCV NAT and anti-HCV EIA.
If the HCV NAT is non-reactive, and the anti-HCV EIA is negative, the donor may be reentered. That is, the donor is eligible to donate in the future, provided all other criteria are met.
Lastly, some of the major comments to the docket that we have received that address the recommendations for donor reentry for HCV in the guidance include the suggestion that there be an additional waiting period for a donor who tests NAT reactive during that six month deferral period.
That we clarify the requisite sensitivity for the HCV EIA used to test that follow up sample, and that an additional six-month waiting period be permitted for donors who are anti-HCV EIA repeatedly reactive on the follow up sample, and for whom the RIBA test is persistently indeterminant.
We will carefully consider and discuss these and other comments submitted to the docket, and it is FDA's intention to revise the draft guidance and to issue final guidance as soon as possible. I think I will stop there and take any questions you might have.
DR. CRYER: How many are we talking about? How many donors in the donor pool would this affect?
DR. MIED: How many would be re-enterable? The estimate I have is about three years old, and it comes from one of the blood organizations.
We are talking about, assuming eight million different donors per year, possibly re-entering with these HIV and HCV algorithms, on the order of 14,000.
For NAT false positivity, it is at about one to 20,000. So, we are talking about maybe reentering 400 false positive NAT donors out of those eight million. Again, those are three or four year old estimates.
DR. ALLEN: My comment, as I have followed this, I think you have done a superb job. It is a very logical sequence.
One can argue about the parameters somewhat, which I think are reflected in the comments that have come into the docket.
My concern is, have you received comments from people at the working level within blood collection centers, who are going to have to apply these, because it is an extraordinarily complex algorithm.
As you go through the process of finalization, I would hope that would be a consideration that you would seek from the blood collection industry.
Certainly, one would hope also that the software manufacturers would try to write this algorithm into the computer based software that is used in the blood collection centers, that would help simplify this. I am just concerned with the complexity of the algorithm.
DR. MIED: Yes, Dr. Allen, most of the major comments that I showed this morning are from the blood organizations themselves.
DR. DI MICHELE: I just have one question. I am just looking at the algorithm for the multiplex screening that is reactive.
Then we go on to do discriminatory NATS. If they are non-reactive, it just seems that, if they are reactive, we go on to destroy, relabel units, notify donors and do look back.
Then, if they are non-reactive for both HIV and HCV, you either destroy and relabel the units anyway, or perform another sample. Then, whether the sample is reactive or not, you go on to do the same thing anyway? Is that correct?
DR. MIED: The difference there, on the additional NAT, the difference -- the additional NAT, which they have the option of performing, if that is reactive, they are performing the full look back, whereas, if they are non-reactive, we only recommend retrieval of prior collections, and not recipient notification. The difference there also is that a donor who is reactive on the additional NAT is no eligible for reentry.
DR. DI MICHELE: So, whether or not they are non-reactive on discriminatory tests, they are still removed.
DR. MIED: Yes, that is correct. This is a point of discussion, as I mentioned. The argument is that the vast majority of these are false positives.
However, we have in the guidance the conservative approach to defer all of those donors, but they are re-enterable, with the exception of those who have a reactive additional NAT.
DR. ALLEN: Other questions or comments? Well, we look forward to hearing how this continues, and to having another update in the future.
Our third update topic will be presented by Dr. David Asher, the FDA summary of the transmissible spongiform encephalopathy advisory committee, which met on Halloween.
Agenda Item: Summary of the TSEAC Meeting held on October 31, 2005.
DR. ASHER: And a frightening experience it was. I was just asked to do this talk on Tuesday, and I can't summarize eight hours of action-packed committee meeting in 10 minutes.
What I tried to do was put as much of the salient information as possible into a handout that was provided to the committee.
Unfortunately, I only made 75 copies and I was so tired I couldn't figure out how to reload the electric stapler, but this will be -- the other thing I have to warn you is that this is not a cleared presentation. So, it really shouldn't be cited as authoritative information. A cleared version will be appearing on the web site.
We had the 18th meeting of the TSE advisory committee on Monday. There were four issues covered, two interesting informational issues that time doesn't permit discussing, and two decisional issues.
One is asking advice on further development of a risk assessment for variant CJD and plasma derived factor 8, an extension of an assessment that was done for factor 11 in the year.
The second was a discussion of validation criteria and possible label claims for devices purported to remove TSE infectivity from blood components.
What I propose to do is to concentrate on the first few slides, which may need some explanation, whereas the last slide summarizing the advice of the committee should be reasonably self explanatory from the handout.
What prompts both of these issues, of course, is the striking observation in the United Kingdom that variant CJD has almost certainly been transmitted by transfusions of non-leukoreduced red blood cell concentrates to two of a very small number of recipients, a difference that is quite striking between this and sporadic CJD, which has never been convincingly documented as transmitted by blood, although the fear of that exists.
I do want to point out right at the outset that there is no case of variant CJD in the United Kingdom or anywhere else that has been attributed to use of a plasma derivative, human plasma derivative, regardless of the amounts of product used.
There are now 185 cases of variant CJD, the vast majority of them in the United Kingdom. The concern is that five or six of those cases have occurred in people in other countries who acquired the infection in the United Kingdom, one of whom was only there for about 26 or 27 days, and three days in France.
In addition, there are other BSE countries where exposure might occur, particularly France, where there have now been 15 cases, several of whom have been blood donors.
These travelers, of course, carry their infections with them to their home countries, where they pose some risk of transmitting disease to recipients of their blood.
As you know, there are deferral policies in place to reduce that risk for people who spent considerable periods of time in the United Kingdom and longer periods of time in France and other countries.
In addition, we know that we have had a very small BSE risk in our own country. There has only been one native borne cow, over 500,000 sick cows screened.
So, we think the risk of endogenous BSE in this country must be very small indeed. The main risk is from people who were exposed in those countries where there has been a lot of BSE.
To evaluate these risks, Steve Anderson, our risk assessor, with Hong Yang, has set up a risk assessment breaking down the elements of risk into four modules.
Two of them involve the source of plasma, the second, the ability of processing to reduce the infectivity, and the third exposure from the use of the product.
It turns out that each of these elements is surrounded by considerable uncertainty and is very difficult to model, and it was to improve the information to be used in the assumptions for this model that we asked for advice from the committee.
The prevalence of variant CJD in U.S donors is largely controlled by the prevalence of CJD in the United Kingdom.
There are problems in evaluating both that prevalence and our own prevalence. That is, however, a critical driver of risk.
A second driver of risk is reduction by processing. Plasma pool size does not appear to be a highly important factor, but reduction by the manufacturing process and, to a somewhat lesser extent, the amount of infectivity potentially present in the plasma are important.
In addition, it has been very difficult to estimate -- surprisingly difficult to estimate -- the amount of plasma derivative that an individual patient has used.
There are two sources of risk in the donor population in the United States. One is those people who should have been deferred but, for some reason, were not. The second is those people who are suitable blood donors, but nonetheless might have been infected.
We know that, if you defer for people who are in the United Kingdom for three months, that it is possible to be infected in considerably less than three months. Yet, we can't defer everybody who has ever visited the United Kingdom, because we would be deferring perhaps 25 percent of blood donors, and even more in east coast areas.
So, there is an analysis of prevalence in the United States relative to the United Kingdom. Estimating prevalence in the United Kingdom turns out to be surprisingly difficult.
There are two methods that have been used to estimate it, and they give quite disparate estimates. One is predictive models based on the number of UK cases actually observed, and the other is an estimate based on a survey of appendices removed from patients in the United Kingdom.
The models of the first sort have estimated a maximum, in more recent estimates, an estimate of, oh, 2,500 cases in the entire United Kingdom, tops.
Studies looking for the accumulation of abnormal prion protein in lymphoid tissue of the appendix have predicted about 13,000 cases incubating variant Creutzfeld Jakob's disease in the United Kingdom.
Those are possibly low estimates, because we know that earlier in infection the appendix is likely to be negative.
We know that blood of these patients contains the agent as long as three years before onset of signs, clinical signs, but we don't know how much longer, and we don't know the exact incubation periods for most people. Nine to 12 years seems to be a reasonable number. We don't know the maximum.
We don't know how many people with less susceptible genotypes might be infected, but we think that some of them surely are infected, and their blood probably contains the infectious agent as well.
The advice from the committee was to use empirical prevalence values based on the tissue survey, allowing for preclinical and subclinical infectious status in donors of all prion protein encoding genotypes.
Our second source of uncertainty has to do with screening out unsuitable donors. Estimates from analogous donor screening situations, like the HIV risk donors, suggested that questionnaires might be as high as 99 percent sensitive, perhaps 90 percent more conservative, but the committee was quite uneasy about that, and recommended considering only 85 percent sensitivity of the screening.
They were concerned, as we are, that we have no direct data about travel histories of source plasma donors. We have been using extrapolations from a travel survey for doors of whole blood.
Donors of source plasma are younger, give more plasma, are thought to be poorer, and it would be desirable to have specific data.
Another area of uncertainty is in how much infectivity to consider potentially present in the blood of donors.
All the information for that comes from animal models. Our proposal was to consider a minimum of .1 infectious dose per ml -- that is one dose per 10 mls -- more likely 10 doses per ml, with a maximum of 310, because that is the highest value in the published literature.
The published literature shows models all over, and to assume that the blood was infectious from the time that the donor became infected until the donor died.
Estimates of the actual amount of infectivity in the blood of rodents, however, tend to cluster together somewhere between two and 30 infectious doses per ml.
This study was from mice, this study from hamsters, that showed a mean of 10 infectivity appeared about halfway through the incubation period.
The committee agreed that, based on rodent data, 10 doses per ml of blood was most likely, and a minimum of 1.1 prudent. They offered mixed advice on upper levels of infectivity.
Pool size, bimodal with a peak at 20,000, a peak at 60,000, the committee agreed, and agreed that we use three different ranges for the reduction in infectivity by different classes of products, anywhere from a minimum of two log reduction to a maximum of nine log reduction, although that would be overkill for removing infectivity.
They agreed that the CDC data bases for factor eight use are the best available data, but they encourage, especially for von Willobraun's disease, collecting better data, and FDA has a plan for doing that.
They shared our concern that some experimental data suggests that cumulative, repeated exposures to small amounts of infectivity may increase the risk of infection, and of course they could be considered a model for those users of plasma derivatives who use them repeatedly. They agreed with our plan to estimate a risk per year rather than a risk per dose.
A special concern to the committee as well as to us is risk communication. The risk of transmission by plasma derivatives, of course, is entirely theoretical.
It is very difficult to communicate small and highly uncertain risks of this sort, and a concern that the notification itself might cause patient anxiety and health care provider anxiety.
It would be terrible if dentists started refusing to take care of people with a history of the use of plasma derivatives, simply because we had told them they were at some small, uncertain risk of this disease, and it was felt that participation of the patient groups in setting plans for risk communication would be helpful.
There won't be time to go through the plasma filtration. We had a very interesting presentation from UK authorities who are going to do independent evaluation of any European, Council of Europe, marked plasma component filtration device that is going to be marketed there with a claim for reduction of TSE infectivity.
They are going to require at least three log reduction in a spiking study, evidence that at least red cells maintain their functionality.
Three manufacturers, one with a Council of Europe marked filer, and two with interesting matrices, presented promising results.
The committee suggested a few modifications to our criteria, greater than three log reduction of spiked infectivity, removal of all detectable infectivity from endogenously infected animal blood, at least two animal models using two strains of agents.
One of those should be derived from either a cow with BSE or a human with variant CJD rodent adapted, and functionality of filtered components should be demonstrated.
They also noted that it would be desirable to demonstrate that any device would remove all endogenous infectivity from whole units of blood from large animals.
At the moment, the sheep is the only large animal demonstrated to have infected blood, although I think primates will eventually become available, since bioassays in sheep are generally not feasible.
They felt that, if sensitivity of transgenic mice to sheep infectivity were demonstrated, they would be an acceptable, although a huge number of mice would be required, and the expense would be considerable. I think that I can stop there. Thank you.
DR. ALLEN: Thank you, Dr. Allen. Questions or comments on this topic? For anyone who is not following it, it is a very complex and confusing area, but one in which a lot of new data continues to emerge rapidly.
Just one comment quickly. With the order of presentations to the advisory committee, it should be noted that, already with the presentation of data on efficacy of filtration to remove the prions from units of blood or plasma, the work that is being done there, we may have a device being presented that would claim to reduce risk, before there is any test to identify risk in the potential units of blood or plasma. So, this presents an interesting conundrum in terms of regulatory issues.
DR. SZYMANSKI: A questions about travelers to the United Kingdom who are going to be refused as donors. If that traveler is a vegetarian, is that considered at all?
DR. ASHER: Yes, the question is, is a question of vegetarianism considered in suitability of blood donors. First, let me say that travel to the United Kingdom after 1996 is no longer considered a deferral factor, because the agency is convinced that a full range of food chain protections put into place then should be effective in protecting against significant exposure to BSE agent.
The answer about vegetarianism is no, it is not considered. The reason for that is that dietary histories, particularly long after the fact, have to be considered relatively unreliable.
Also, people have different understandings of what it means to be a vegetarian. Many people who are self proclaimed vegetarians have, in the past, consumed beef.
One of the patients in North America came from a family that, I believe, were practicing vegetarians. So, vegetarianism is a sometimes thing, and it just can't be evaluated critically in the setting of a blood program.
DR. NELSON: How many cases were in the United Kingdom in the last year of the 158?
DR. ASHER: In the last year, just a handful. The cases are dropping in the United Kingdom.
DR. NELSON: So, the numbers are still going down, and it seems that the estimate of 12,000 is perhaps too high.
DR. ASHER: It is reassuring as far as it goes. The problem is that all the cases in the United Kingdom, all the clinical cases, have been in people who were homozygous for methionine at codon 129 of the prion protein encoding gene.
Now, that genotype is known in other forms of CJD to increase susceptibility in shortened incubation period. The problem is that people with other genotypes are susceptible to other types of CJD.
The second transfusion transmitted case, who died -- well, it was a woman who died at, I think, 81 of a ruptured aortic aneurism, had evidence of preclinical CJD.
The most likely scenario is, had she lived longer, she would have come down with symptomatic disease, which is what happens with other forms of CJD in people with those genotypes.
I understand that the last case of kuru was seen in New Guinea in 1999, and then suddenly they start reporting cases in people of the less susceptible genotype almost 15 years after the exposure.
How long the blood is infectious in variant CJD in those people during that period of time is not clear, just another example of the extreme uncertainties surrounding this.
DR. NELSON: About 40 percent of the population has a susceptible genotype, as I understand it.
DR. ASHER: Has the most susceptible genotype. The advice of the committee and of the UK authorities, is to consider all persons potentially susceptible to infection. Hopefully, the attack rate will be lower and the incubation period will be longer, but we still have the problem of what to do with infectious blood from people who are clinically normal. So, it is a mixed picture.
DR. ALLEN: Okay, any other questions or comments? We will move on then. Dr. Jerry Holmberg, executive secretary of the advisory committee on blood safety and availability, will present a summary of the department's recent meetings.
Agenda Item: Summary of the DHHS Advisory Committee on Blood Safety and Availability.
DR. HOLMBERG: As Dr. Asher mentioned, it is difficult to cram a two days presentation into 10 minutes. So, I will try to move as fast as possible.
Just a point of clarification, Mr. Chair. I understand you wanted a little update on the hurricane response. Is that still desirable, or should I move forward?
DR. ALLEN: Go ahead and move forward.
DR. HOLMBERG: Some of the issues that were discussed at the most recent Advisory Committee for Blood Safety and Availability were issues relating to the varicella zoster immune globulin, immune globulin intravenous, and a strategic plan for improving the blood safety against known and unknown transfusion transmitted complications in the 21st Century.
As far as the issue with the VZIG, there is a potential shortfall. There is a use rate of about 200 vials per month, which should probably take us through the end of January.
Also, one of the things that most recently the Advisory Committee on Immune Practices has recommended, recommendations may lead to the use of IGIV as an alternate product until there is an availability of VZIG.
That leads me to the next issue that was discussed, and that is the IGIV issue. I hope you like my slide here. I thought this front coming in here was a good analogy to what is happening. Is this a perfect storm, or is this just a market adjustment to the MMA, the Medicare Modernization Act.
I would like to dwell a lot of time on the waterfall here. This was a picture taken during my trip to Puerto Rico this summer, but I just put that in there as an analogy of what is really happening in the market place.
That is one thing, that the BPAC usually doesn't get involved, or isn't permitted to get involved with the issue of economics, and basically your charge is to look at the science attributed to the various products that are put before you.
The Secretary's advisory committee can look at different aspects, including the ethical use and the economics.
One of the things that we have found -- and I like to put this as a waterfall or a cascade, is that we have the manufacturers contributing their products.
We have seen, over the last couple of years, a consolidation of the manufacturers and also, with the manufacturers, they have also consolidated the number of distributors.
So, you also have primary and secondary distributors and then, at the hospital level, you have the distribution of that, to either the hospital, which is under the reimbursement as the DRGs, the physicians offices, which are under the part B, and then the hospital outpatient.
Well, if we look at how reimbursement is handled, what we have is, we have the manufacturers reporting their wholesale and their sales prices to CMS, and this is on a regular basis, that they are reporting to CMS for a recalculation.
IGIV is one of the very few drugs that actually is monitored on a regular basis for price, and it is adjusted on a quarterly basis. Unfortunately, that adjustment lags behind the quarter.
Most recently, in the medicare modernization act that took place in January of this year, the calculation was changed to average sales price plus six percent. As of this quarter, that is going for a -- liquid is going for $56 a gram.
However, right at the present time, hospitals, outpatients, are being reimbursed at average wholesale price at 83 percent, which is about $80 a gram.
So, you can see that there is a disparity there and, what happens, there is going to be a shift in service. So, we have seen a shift in service.
One of the things also that we identify is that the distributor in the calculation that is mandated by congress -- and this is a statutory requirement -- is that the MMA look at the price coming from the manufacturer and not the distributor.
This is where the problem exists, in that the distributors are not part of the calculation for reimbursement.
As a consequence, you have the primary distributor for the manufacturer who is selling under contract and the secondary and gray market that is selling in a spot market scenario.
If I can put it in a different way, we have the manufacturers who have been very diligent at increasing their inventories.
In the past, the manufacturers have had months and months of inventory. In an economic move, they cut back and, for a while, we were under a yellow light system, meaning that there was about a five week supply of IGIV.
However, in the last two months, since August, we have actually had a green light situation, meaning that there is a greater than five week supply of IGIV.
We also have the reimbursement issue which, as of October 28, the lyophilized was about $45.57, and the liquid was at $56.30. This is for doctor office reimbursement.
So, as I mentioned before, in the last slide, there is a shift in service. So, we have seen a shift in service from the physician offices to the outpatient setting.
Then, on the other side, we have the demand issue. We have the labeled use of the product. We have the evidence based use, that CMS has about 30 different evidence based clinical entities that they will reimburse for, and then you have the non-evidence based use of the product.
We have quite a problem here, trying to address all those issues. What we have seen, just to quickly go through this -- I have talked about all of this -- is that the industry has consolidated, there has been a change in business practice, there has been a market correction, reduction in inventory, a smaller number of distributors.
What we have found is that there are sufficient supplies of IGIV for patients who need treatment. I was on the phone quite a bit this week tracking the various complaints of availability.
Every place that I called said, yes, there is product available. It is just that when you get it from the secondary or the gray market, some of the physicians' offices are paying up to $160 a gram.
What we also found was is that it suggests that, under the allocation system, that the physicians might best serve the patients by communicating their supply needs directly to the manufacturer, and also ensure that IVIG treatment is prioritized toward FDA labeled use, and those diseases are clinical conditions that have been shown to benefit from the use of IVIG based on evidence of safety and efficacy.
What we have done within the department is we have looked, or we actually have posted -- this is my web site if you would like to go to the advisory committee for blood safety and availability.
We do have a statement there on the IGIV. We also have posted the CBER number and a place where you can report shortages to CBER.
Also, if there is denial of medicare coverage, then that can be reported directly to medicare. What we also have posted there is a page from PPTA that lists all five manufacturers with their 800 numbers, and that will -- each one of the manufacturers has agreed to have an emergency supply inventory of IVIG.
So, if there is a need, the physician can call directly to the medical director at the manufacturer, and discuss the case, and request the product.
As I just mentioned, hot lines have been established. The department has also looked at the possibility of doing an evidence based medicine study, and that is still under investigation.
However, we are looking at the CMS reimbursement. We are monitoring the cost. Once again, I have to say that this is an issue that is statutory.
Congress has said, these are the calculations that you will use to determine the rate of reimbursement. So, there is very little that we can do as far as the reimbursement, and right now we are just constantly monitoring what is the situation.
However, there is an IG assistance in looking at the problem and trying to identify areas where we can step in.
Well, the committee looked at all that and, after new input from patients, medical professionals, distributors and manufacturers, the committee remained highly concerned that persistent disruption in access to IGIV, which includes a progressive shift to treatment in a hospital, continues to compromise quality of care to many patients.
In particular, the committee believes the transfer to a hospital may impair continuity of care by their usual medical provider and may add otherwise unnecessary cost, logistical complexity, and nosocomial infectious risk.
The committee further is concerned that a change to the hospital outpatient reimbursement to ASP plus eight, effective January 2006, will further aggravate an already difficult situation, and that this shift will not be sustainable.
They recommended to the secretary that an increased reimbursement for non-hospital IGIV therapy, to a level consistent with current marketing practice.
Consider and reclassify IGIV as a biological response modifier, consider declaring a public health emergency to address short term problems.
Modify the current plan to change hospital outpatient reimbursement to ASP plus eight in January 2006, in such a way as to prevent any sudden or large decrease in reimbursement.
Reexamine whether the current IGIV supplies are meeting patient needs, and work with congress to establish a long term stable and sustainable structure.
We are continuing to investigate this. Areas where we can have impact we are looking at. However, there are some areas, such as what congress is willing to do, and what congress has already put in the MMA. Whether the MMA is going to be opened up again for discussion this year is debatable.
The other thing that we looked at was a strategic plan for the 21st Century. The committee recognized that blood is a critical element of modern medical care, and ensuring an adequate supply of safe blood is a national responsibility.
Although there have been dramatic improvements in blood safety and availability in the United States in the last two decades, the committee finds that there are compelling needs for improvement in some areas:
Minimizing destruction of the supply, and the supply of access to blood products and their analogs;
Meeting the product development needs for patients with rare disorders;
Timely funding to ensure appropriate utilization of new technologies;
Integrating presently fragmented systems for monitoring blood safety and availability;
Aligning reimbursement and funding policies with product approvals and other decisions intended to optimize blood safety and availability;
Modifying reimbursement policies, as needed, to sustain access to blood products and their analogs for all patient groups, e.g., IGIV;
Reassessing policies and their related interventions based on the evaluation of their impacts;
Intensifying efforts -- the committee recommended doing these things to intensify efforts to influence clinical practices related to blood transfusion and alternate therapies based on scientific evidence.
Accelerating responses to threats -- patient specimen/unit misidentification -- for which there are available interventions.
Utilizing formal risk communication strategies targeted to blood donors, patients and care providers, to enhance scientific comprehension and public trust.
Also, pursuing opportunities to enhance public health in the management of blood donors.
Promoting comprehensive disaster planning,including sustaining the inventories necessary for an effective crisis response.
Establishing a proactive, prioritized and goal-oriented research agenda, utilizing formal assessment tools more routinely in policy development and decision making.
Further clarifying the respective roles of government agencies and the private sector in management and oversight of the blood system.
The plan should include and encompass: structured process for policy and decision making; integration of blood systems within the public health infrastructure; surveillance of adverse events related to blood donations and transfusions; risk communication; error prevention in blood collection centers, transfusion services and in clinical transfusion settings; donor recruitment and retention; clinical practice standards for transfusion; strategic research agenda; disaster planning; stable and sustainable reimbursement; funding for promising new technology.
I believe I will call it there, but these recommendations have been forwarded to the secretary for consideration.
If these are approved by the secretary, we will be working within the department to put together a strategic plan to encompass the recommendations that the secretary approves. Are there any questions?
DR. ALLEN: Thank you, Dr. Holmberg. Questions or comments? It is an ambitious plan. I assume it is unfunded.
DR. HOLMBERG: Unfunded and, right now, we have 94 years to accomplish that.
DR. ALLEN: Thank you very much for the update. Our final update for this morning will be reentry of donors deferred based on anti-hepatitis B cor test results. Dr. Gerardo Kaplan from the FDA will start, and Dr. Susan Stramer from the American Red Cross will provide a brief statement also.
Agenda Item: Re-entry of Donors Deferred Based on anti-HBc Test Results.
DR. KAPLAN: Good morning. I will give you an update on reentry of donors deferred on anti-hepatitis B cor test results.
The issue here is that the FDA would like to update the committee on the proposed algorithms that will allow reentry of donors deferred for testing repeatedly reactive for antibodies to hepatitis B cor antigen on more than one occasion.
As a historic perspective, the 1991 guidance indicated that donations for transfusions should be tested for anti-hepatitis B cor antigen. Only negative units should be transfused.
The donor should be indefinitely deferred when they tested reactive more than once, and that reentry algorithms were not recommended at that time, because there was no supplemental test for anti-hepatitis B cor.
One of the main consequences of the screening was that anti-hepatitis B donors contribute to the safety of the blood supply.
However, many donors have been indefinitely deferred because of potential false positive anti-hepatitis B results.
The considerations for reentry is that reentry will be permitted only on the premise that historical tests for anti-hepatitis B cor antigens were false positives, and that there is no evidence for past or present hepatitis B virus infection.
In a 1999 BPAC presentation, the FDA and AABB presented to BPAC similar reentry algorithms, based on negative test results for surface, anti-cor and anti-hepatitis B surface antigens.
The committee did not recommend reentry because some of the American Red Cross data showed that there were some hepatitis B surface antigen and hepatitis B cor negative samples with hepatitis B DNA reactives, using an experimental NAT.
There was another BPAC presentation in a 2004 meeting, where FDA presented to BPAC the following reentry algorithm, based on test results for surface, anti-cor, and hepatitis B NAT.
The committee did not vote on the proposed reentry because there was a lack of data to evaluate its use. This algorithm basically said that, although the donor has been indefinitely deferred because of having tested repeated reactive for anti-cor on more than one occasion, there may be reentry if, after a minimum of eight weeks subsequent to the lest repeat reactive anti-cor test, a new sample is collected from the donor and these samples test negative for the three markers, basically surface antigen, anti-cor, and hepatitis B NAT in FDA licensed assays.
When a donor presents at the blood center to donate, subsequent to the negative test for, again, the three markers -- surface antigen, anti-cor and hepatitis B NAT, and all donor suitability criteria for donors of whole blood and components are fulfilled.
The good news is that companies have developed two key tests, and the FDA has approved them. There was an approval in April of 2005 of the Roche Cobas AmpliScreen hepatitis B NAT for screening donations in mini-pool and individual donation format.
Roche also presented some data to BPAC indicating that it is possible to enhance the sensitivity of the ID test.
FDA also, in October 2005, the Abbott PRISM anti-cor test for screening donations was approved. This test uses a different technology than other currently licensed assays, and could be used for the reentry algorithm.
So, the FDA is looking forward to reviewing that data to validate the proposed reentry algorithm for anti-hepatitis cor repeat reactive donors, and the good news is that today we are going to hear a talk by Dr. Susan Stramer from American Red Cross, and she will provide some of this data. This concludes my presentation.
DR. ALLEN: Very quickly, any questions for Dr. Kaplan at this point? Okay.
Agenda Item: Re-entry of Donors Deferred Based on anti-HBc Test Results.
MS. STRAMER: Good morning. While my handouts are being handed out and we are loading my presentation, I will thank the committee for this opportunity to present an update of what I presented at the October 21, 2004 BPAC, and also to declare my conflicts of interest.
I am a current employee of the American Red Cross, and I am a former employee of Abbott Laboratories and, as such, I still have Abbott stocks, but that shouldn't influence my judgement of scientific data.
I would like to first acknowledge my collaborators at National Genetics Institute, NGI, and Abbott Laboratories. As I mentioned, this is an update from the October 21 meeting of last year, about a year ago.
So, the questions -- so, relative to what I presented in October, the historical and current repeat reactive rates for FDA licensed anti-cor tests range from .4 to 1.6 percent. This is basically a historical number.
False positivity, as already mentioned, is high on these tests, ranging from 25 to 87 percent reported, varying depending on the specificity of the test used.
As mentioned, there is no confirmatory or supplemental test available, and as such, we have a two time allowance for donors to donate and be repeat reactive before they are permanently deferred.
I presented last time the impact of a repeat reactive notification on first time donors, and whether this prevents them from returning for their second donation.
If you look at what the Red Cross has in our data base for the numbers of anti-cor deferred donors, who lack other markers for deferral, from the period of time that we have formal tracking, it is greater than 200,000.
If you go back to the numbers since February 1987, and try to estimate a nationwide number, we know there is probably greater than a half million.
So, the question is, can an algorithm be validated to reenter deferred donors who are falsely anti-cor reactive. If so, what would be the yield of such a reentry algorithm.
BPAC, last October, did provide unanimous support for the validation of the reentry algorithm as provided by FDA and outlined by Dr. Kaplan.
We will test a follow up sample greater than or equal to 56 days by a highly sensitive HBV NAT, or DAN assay, having less than or equal to 10 copies per ml.
We also would retest the sample by a second or more specific anti-cor assay. We would, of course, also do an FDA licensed, HBSAG assay, and all tests must be non-reactive and the donor have no other causes for deferral, and the donor would then be eligible for reentry.
What I showed at the last BPAC, relative to how successful reentry would be based on the same test, what I used was a model saying how many donors were one time reactive, and then came back a second time and were reactive again.
So, looking at a data base from the year 2000 that included 6.5 million donations from about four million donors, the overall repeat reactive rate at that time was 6.4 percent.
The important point here is that 75 percent of donors who were reactive one time because two times reactive on their next donation, with another 13 percent accrual over the next three years.
That doesn't bode very well for the efficacy of the two times reentry algorithm based on the same test. If you look at the split between first time and repeat donors, one would assume that, in these first time donors, most of these represent true HBV infection.
These represent probably an overall mixture of false positivity as I mentioned, and true positive HBV infection.
So, what we -- the AABB and the industry -- had recommended was that an anti-cor reentry algorithm would be projected to have a higher yield if a different test was introduced.
The blood system would then have to convert to that new test, rather than testing these donors over and over again with the same tests, and finding the same false positives.
This slide shows you the repeat reactive rate over about the last 10 years at the Red Cross. We are running now at about a .4 percent repeat reactive rate. That is for all donors.
Over time, the repeat reactive rates have come down significantly. So, we have culled out the true positive donors and, since most of our donors are repeat donors, that is possible.
Now, one interesting piece of information is, if you switch tests, how much of your false positivity will be reduced.
So, this slide was provided to me by Hema-Quebec, courtesy of Dr. Gile Delage. It shows you here that, when they introduced HBC screening in Quebec, they introduced it on the Ortho platform.
Then, in April, they converted to the Abbott PRISM platform What you can see, if you take an overall mean, after a first culling period and then a stabilized repeat reactive rate of just under one percent, they convert it to PRISM and then they stabilized it just under .6 percent for .4 or 40 percent decrease.
You can see that there was quite an impact by changing tests, and this is actually the test, the European version of PRISM that does not contain the reducing agent that eliminates false positivity.
So, what we did, in 2001, to look at the possibility and to follow up, do anti-cor donors really have DNA associated with them, we took 3,000 anti-cor repeat reactive unlinked donations, and we tested them for HBV DNA and NGI.
These were surplus NAT samples. So, they were handled under proper conditions, and to limit contamination. The criteria for inclusion was non-reactive by all other test methods.
There was no pre-selection for first time or two time anti-cor repeat reactives. We just took the tubes out of our NAT laboratories.
From projections of numbers of donors who were two times reactive, of this 3,000 -- which is why I chose a large number -- it was to enrich the population so we would have enough donors in here who represented truly deferred donors, and we would project 708 of these 3,000 were actually two time cor-deferred donors.
So, when we tested these samples on the NGI net, ultraqual assay, the eight reaction test -- that is, the sample is extracted and processed eight times.
We used a .2 ml input and, if there is reactivity in any of the eight tests, the test is interpreted as positive.
The sensitivity was 9 IUs per ml, which converts to 31 copies per ml. We saw 19 of 3,000, or just under one percent, samples reactive.
The samples had low reactivity for DNA. Eleven of the 19 had less than 100 copies per ml. With an average of the eight reactions, how many were positive? An average of just under two.
Eight of the 19 samples had viral loads between 100 and 500 copies per ml, with about a 300 copy per ml mean, and the average number of reactions of the eight reactive was just about five.
So, in red, this slide is shown frequently. It shows you how this data, the .3 percent anti-cor positive rate, compares with other published studies, either the Roche clinical trial or a rat study showing what the frequency of DNA positivity is in isolated anti-cor reactives.
So, the purpose of me standing here today is to tell you the updated results when we tested these 3,000 samples which were characterized by DNA testing.
When we tested them by the Abbott PRISM anti-cor assay, using the licensed lots -- that is, post-licensure, which Gerardo said just happened, so these data are hot off the press, I just got the data last night.
The study was initiated following the qualification of the samples and PPTs, which hadn't been qualified on the PRISM system previously, and availability, as I mentioned, of licensed lots.
The assumption going into the testing is that the 19 HPV DNA positives that I mentioned would all be PRISM anti-cor reactive.
Also, what we wanted to know was, how many eligible donors we would yield for reentry after testing the anti-cor non-reactive. That is, how many samples would be anti-cor nonreactive by the PRISM and were DNA negative.
This testing, as I mentioned, is completed. We are doing further studies to investigate anti-cor IGM reactivity of the cor reactives, and to look at anti-surface for scientific interest, as sample volume allows.
So, these are the results. If you look at the NGI sample tests, we tested 3,000 samples. Actually, it came out to 2,999. I am off by one. Seven were QNS for the PRISM testing, leaving a total in this table of 2,992.
In red are the NGI DNA positive results, and all 19 were positive on PRISM cor. All 19 were strongly positive for anti-cor antibodies on the PRISM assay. The mean S to CO was a .14.
I should mention, in the PRISM test, which is a competitive inhibition assay, sample S to COs less than or equal to one are reactive. Sample S to COs greater than one are negative. So, here you look for a low signal to cut off ratio.
So, a .14 is a very strong mean, and the range of these 19 samples, their PRISM reactivity ranged from .02 to .49.
Of the samples that tested DNA negative, what were the PRISM results? Eighty percent of the total actually retained reactivity by PRISM.
So, using the Ortho test, as we do, in combination with the PRISM test, we only have 21 percent of donors who were deferred due to anti-cor reactivity, who would be predicted to be eligible for re-entry if they passed the follow up testing.
These are the 19 samples. Although you might not be able to see this, I highlighted in red -- these are the triplicate signal to cut off ratios, initial reactives are repeated in duplicated.
The values are very consistent, and the values above .14, which was the mean, are highlighted in blue, and there were only five such samples.
So, in summary and to conclude -- and then I just have one more slide after this -- preliminary data indicate the feasibility of an anti-cor reentry algorithm.
That is, all 19 HBV DNA positive samples were detected as reactive by PRISM anti-cor assay. All 19 of these samples were strongly anti-cor reactive, and I already highlighted their reactivities.
The caveat of this study is that the yield of reentry is probably dependent on the prior assay. I showed you, for our 1X, 2X anti-cor deferral study, that we only yielded 25 percent for going from Ortho to Ortho.
In this pilot study, I only showed you a 21 percent yield. Then, from the Hema Quebec study, I showed you a 40 percent yield going from Ortho to PRISM.
I don't have any data, if you were using the prior Abbott EIA called Corzon, what that yield would be going forward, going to PRISM.
I would like to just end with subsequent studies other than completing the study I just described. Currently we are doing a study of anti-cor repeat reactive samples with the National Genetics Institute, that is testing them all for HBV DNA.
Of approximately 5,000 deferred donors that we had over the last year -- that is, 2X cor reactive with anti-cor as their only deferral criterion -- we have invited these 5,000 donors to provide a follow up sample under NGI's HBV IND.
This was before the FDA had the proposed algorithm. So, we tested these by HBV DNA, HB SAG anti-cor, Ortho, anti-cor, and the PRISM anti-cor results are pending, as well as anti-HBS.
A study similar to this we will convert to the Roche's IND using the more sensitive method that Gerardo presented.
Of these study results to date, only 10 percent -- again, relatively low yield -- of these 5,000 deferred donors have provided a follow up sample.
Sixty-six percent, or numbers that are consistent with those I have shown, do have evidence of past HBV infection, leaving 44 percent eligible for reentry, assuming they don't show reactivity on the PRISM test.
Due to low participation rates, if you multiply out all the numbers, our yield of these 5,000 donors has only been a reentry rate of 4.4 percent. That is all the comments I have. Thank you.
DR. ALLEN: Thank you, Dr. Stramer. Questions or comments either for Dr. Kaplan or Dr. Stramer? This is certainly good to have this data, and I think at least I am a little surprised by the results that you just presented.
DR. STRAMER: The relatively low yield?
DR. ALLEN: Yes. As you said, the science is the science. So, that is very good, and we continue to learn more about hepatitis B infection.
DR. NELSON: Are donors now being screened for cor using PRISM?
DR. STRAMER: No, the PRISM cor test was just licensed on October 13.
DR. NELSON: So, presumably they will be.
DR. STRAMER: Yes, some blood centers are gearing up. The first date of implementation that I heard was December 5, but you know, validation, et cetera, et cetera, is required.
DR. ALLEN: Other questions on this topic?
DR. SZYMANSKI: I just wanted to point out that anti-surface hepatitis B surface antibody is not a reason for deferral, whereas the cor antibody is, I guess, because then you have more infectivity present if you have anti-cor antibody.
I just have noticed that sometimes our donors who have been recently vaccinated for hepatitis B also converted to anti-cor positivity.
DR. STRAMER: That is probably because they were vaccinated having had prior HBV infection, and this is a secondary or amanastic response to the vaccine. That has been seen before.
DR. SCHREIBER: Another elegant presentation, thank you, Susan. What the time between them being invited back for the re-testing and their initial tests?
DR. STRAMER: For the 5,000 in the last slide I showed?
DR. SCHREIBER: Yes.
DR. STRAMER: Immediately. Certainly, when we are inviting them back for a study, the message we can give them is not as strong as saying, if you come back for the study you may be re-enterable.
So, moving forward, hopefully after we do the validation studies and we are doing this real time, hopefully our yields will improve. The regions, to do a study, aren't that excited, as they would be to get their donors back.
DR. NELSON: I wonder if anybody has ever done a study where they do repeated DNA testing on a cor? In other words, the levels may be fairly low, but I wondered if they bounce around, those that are PRISM positive.
DR. STREAMER: Most likely they do, but in the reentry algorithm, they would have to be nonreactive, by the anti-cor.
DR. ALLEN: That is probably a study that is more likely to be done in the future than to have been accomplished in the past, and perhaps could be done on the reds data or other sources.
DR. NELSON: There may be some PRISM false positives as well.
DR. STRAMER: Of course there will be. There is no test without false positivity.
DR. ALLEN: An important object lesson to take on. Okay, we will bring this committee update session to a close. We are, at this point, running almost a half an hour behind. Good discussion.
I am going to turn the gavel over to Dr. DiMichele for the next session after a 10-minute biological break. Please keep your discussions to a minimum. Back in here ready to start in 10 minutes.
DR. DI MICHELE: Good morning, everyone. My name is Donna DiMichele. I am a committee member and serving as the chair for this particular session.
In that capacity and without further ado, I would like to ask Dr. Elliott Cowan to provide the introduction and the questions to the committee on our topic for this morning, which is approaches to over-the-counter home use HIV test kits.
Agenda Item: Approaches to Over-the-Counter Home-Use HIV Test Kits. Introduction and Questions to the Committee.
DR. COWAN: Thank you very much. I think I am stating the obvious when I say there has been considerable expectation surrounding this meeting, and for good reason.
Knowledge of one's HIV status is one of the most important weapons we have against the epidemic, and HIV tests and testing are at its core.
As we begin what promises to be an extraordinary session, I wanted to make clear from the outset why we are here.
Within the past several months, a sponsor approached FDA to market its HIV test kit for home use. In response, FDA is taking a carefully measured approach.
As a first step, the agency felt that input on what information should be provided to validate a home use HIV test kit is critical to our decision making process, and that input is necessary before we consider over-the-counter claims for HIV home use test kits.
For that reason, we are bringing those issues before the BPAC in this public forum. Let me be clear, however, that consideration of HIV test kits for home use is a multi-step process.
Contrary to what you may have seen or heard recently, we are not going to be evaluating an HIV test for approval today.
An evaluation will come in the future only after we have established the criteria by which to evaluate these tests for their intended use, and when a company decides to apply for over-the-counter status.
At that point, the FDA will determine if the test kit meets the statutory and regulatory requirements for approval, determining that the device is safe and effective for its intended use. That analysis will be a determination of whether the benefits outweigh the risks.
Let me turn, then, to the matters at hand. The purpose of this session is that FDA is seeking advice regarding the conditions necessary to support the approval of a home use HIV test kit.
In particular, we are asking the committee to consider what studies are needed to validate test accuracy, test interpretation and medical follow up based on the provision of informational material in place of a trained test operator and counselor.
My role is to provide an introduction to this session to orient you, the committee. To do this, I am going to give you information on prior public discussions of home use HIV tests, a history of point-of-care testing for HIV, concentrating on rapid HIV testing, issues to be addressed for home use HIV test kits, an overview of this session, and finally the questions that you will be considering at the end.
Let me start with some definitions. Home use tests are tests that are used at home by untrained persons without the help of a health care professional.
There are two types of home use tests. The first is a home use collection kit, in which you take your own sample, mail it to a laboratory and get your result over the phone.
There are also home use test kits, and these are those in which you take your own sample, test the sample, and read your own test result.
There are some currently approved home collection kits, and those include tests for detection of hepatitis C infection as well as HIV. I will come back to the HIV in just a couple of minutes.
There are home use test kits which are for fecal occult blood, glucose, cholesterol, pregnancy, prothrombin time, just as a few examples.
At the same time, I need to let you know that there are no previously approved home use test kits for infectious diseases.
Again, home use test kits are different from home use HIV collection kits. For the home use collection kit, the specimen is collected by the test subject, the test is performed and interpreted by a trained operator in a certified laboratory, and the individual asking for testing receives live counseling.
In contrast, with a home use test kit, the specimen is collected by the test subject and, in this case, the test is performed and interpreted by the test subject.
So, there is a lack of trained operator, there is lack of live pre-test counseling and post-test counseling at the time that the test result is provided, and there is a lack of medical referral.
Let me turn now to some of the previous discussions that we have had in considerations of home use tests.
Starting in 1986 -- you can see there is some history here -- companies first expressed interest in developing and marketing home use blood collection kits for HIV testing.
At that time, FDA and AIDS advocacy groups raised public health concerns about test accuracy in the hands of untrained individuals, and the way users would be notified of their test results, particularly centering around areas of patient confidentiality, and adequacy of telephone counseling.
Counseling is a central issue. It is deemed critical to ensure that HIV infected people understand what HIV infection means, and receive important information about recommended treatment and coping methods.
In March of 1988, FDA notified, by letter, manufacturers and other interested parties, of requirements for the approval of these test systems, and this was based on consultation with U.S. public health agencies, our sister agencies, and public sector advisory groups.
This led, in February 1989, February 17, 1989, to a Federal Register Notice, in which FDA published its criteria for HIV specimen collection systems, and stated that these systems would be available only for professional use, and that results of testing should be reported to public health care providers for reporting an interpretation of the test result to the person requesting the test, as well as counseling the individual. In other words, the test result was not to be reported directly to the individual. There were other requirements as well, such as use of licensed HIV tests and some other requirements.
In this Federal Register notice, there was also an announcement of a public meeting to discuss the collection and shipping of blood specimens by lay persons, return of results directly to the person from whom the sample was collecting, counseling outside of the medical health care environment, availability of blood collection systems over the counter, as well as kits for collection and home testing of blood for evidence of HIV infection.
The meeting was held on April 6, 1989. At this session, invited speakers spoke to the issues of regulatory issues.
We heard from CDHR, the Center for Devices and Radiologic Health, their experience in home testing, speakers on counseling, ethical issues, as well as experience among the states.
There was extensive discussion on risks and benefits of blood collection kits and home use test kits, which I will get to in more detail a little bit later.
This led, in turn, to a Federal Register notice on July 30, 1990, in which FDA had stated that they had considered the data and comments from the April 1989 meeting and, on the basis of that, decided that HIV specimen collection kits should remain for professional use only.
At the same time, we stated our willingness to work with manufacturers on requirements for premarket approval application to review data for home collection kits, opening a door.
That same month, BPAC met to consider the approval of a premarket approval application from a sponsor for its home collection system.
The committee recommended against the approval of that particular application, citing the fact that the application lacked sufficient data, and the questions remained regarding possible problems with a variety of issues, including confirmatory testing, the counseling issues, as well as compliance with state requirements in maintaining confidentiality.
Between 1990 and 1994, FDA discussed over-the-counter specimen collection kits extensively with other public health service agencies, and also with product sponsors.
During that time, there was a change in circumstances. We saw advances in technology which translated into a potential for improved accuracy of these tests.
We saw a change in treatment methods. We saw the availability of therapy for asymptomatic individuals, so that people who had a reactive test result have the ability to be treated and to maintain a life with infection.
We also saw the public's increasing desire for greater involvement in personal health care decisions. In June of 1994, BPAC met again to reexamine home use specimen collection systems, and there was an agreement that there was a benefit to having alternative means of reaching previously unreachable populations for HIV testing, and that that outweighed the potential risks of such a system.
At the same time, concerns were expressed about accessibility of a home use specimen collection kit for target groups, adequacy of counseling, while maintaining confidentiality. and effectiveness of education and follow up.
There was a recommendation from BPAC At the time for pilot studies to evaluate these studies. One more Federal Register notice.
On February 23, 1995, FDA notified the public that it was revising its guidance for specimen collection kits labeled for HIV antibody testing set forth in the February 1989 notice, that over-the-counter specimen collection kit systems may be approvable, and listed specific kinds of data that sponsors would need to submit for the review of the safety and effectiveness of these products.
However, it did not address kits for home testing, home test kits, in other words, for evidence of HIV infection.
This notice led eventually to the approval of two home specimen collection kits in 1996, one of which was the original test that BPAC had considered in 1990.
So, what has changed since 1995? One is the technology. We now have tests that have an extremely low risk of an incorrect result, and tests that are unaffected by changes in operating conditions or conditions that could affect the integrity of the specimen.
These tests are simple to use. They don't require special storage conditions. The results are available within 20 minutes and, in one case, uses an oral fluid specimen, which would eliminate concerns that were previously expressed concerning biohazardous conditions. In other words, there is no blood to be collected and no sharps. We also have more experience with these tests in non-traditional testing settings.
There is the concept that early detection translates into better outcomes. Again, with the increased availability of treatment regimens, early identification before the onset of symptoms can translate into better outcomes. Finally, there are changes in social awareness of HIV infection.
Let me turn now to some nuts and bolts of HIV testing. Traditional HIV testing requires two visits to a clinic or health care provider, one to provide the sample and the other to receive the test result, normally about a week later. It could be a little bit less, or it could be up to two weeks.
CDC has estimated that each year approximately 8,000 positive individuals do not return to receive their test results.
Contrast this with point of care testing, in which the test provides a test result in a relatively short time, so that only a single visit is required.
These tests come under the category of what we refer to as rapid HIV tests, again, tests that provide results within 20 minutes, require few steps to perform, have a visual readout, no special storage conditions or instrumentation is required. They detect antibodies to HIV, and these tests are used for diagnostic purposes only, not for blood donor screening. At this point in time, FDA has approved four tests in this category.
The performance standards for rapid HIV tests were discussed at yet another BPAC meeting on June 15, 2000. Performance standards, and the clinical trial and non-clinical requirements were discussed and concurred upon by BPAC at that time.
In terms of performance, it was determined that the sensitivity of these tests should be 98 percent, as well as a specificity of 98 percent.
The important consideration here is that this is a statistical number. Ninety-eight percent is the lower bound of the 95 percent confidence interval. It is not a point estimate.
This translates into tests that are very accurate. The next two slides are going to give you some examples regarding the four tests that are available now.
So, the four tests that are available are OraQuick from OraSure, Reveal from Admira, Unigold from Trinity Biotech, and Multispot from Biorad.
Looking at the numbers here, with the different types of specimens that could be analyzed using these tests -- whole blood, plasma, serum, oral fluid -- you can see greater than 99 percent sensitivity, and these figures are point estimates from clinical trial data that were submitted in support of the product approval.
Likewise, for specificity, the numbers are exceptionally good, specificity being the ability of the test to tell someone he or she is not infected when that is truly the case.
Regarding the interpretation of these tests, a non-reactive rapid test result is considered to be a negative result or, as a reactive result, is considered to be preliminary positive.
That is, all reactive results should be confirmed using an appropriate supplemental test, and this is consistent with the concept that FDA has abided by over the years that, although screening test results are highly accurate, the test results should be confirmed by supplemental testing.
In discussions on the role of rapid HIV tests in facilitating HIV testing, it was recognized that, even though these tests were simple to use, it is critical that a quality management system be in place to assure that testing was being performed properly, and you will be hearing more about this later from one of our speakers.
With an eye to this, and with concerns that rapid HIV tests could be used improperly, FDA approved rapid HIV tests with a series of sales and use restrictions.
The first is that sale is restricted to clinical laboratories that have an adequate quality assurance program, and where there is assurance that operators will receive and use the instructional materials.
Secondly, the test is approved for use only by an agent of the clinical laboratory, that is, they are not approved for self testing.
Test subjects must receive a subject information pamphlet and pretest counseling prior to specimen collection, and appropriate counseling when test results are provided.
These tests are not approved to screen blood or tissue donors, and a customer letter is included with all kits that are purchased which states, by purchasing this device, you are doing so as an agent of a clinical laboratory and agree that you or any of your consignees will abide by the restrictions on the sale, distribution and use of the device.
Access to rapid HIV test has been an issue that has been discussed also over the years, and the central question here is who is permitted to use these rapid HIV tests.
This is governed by something referred to as CLIA, the Clinical Laboratory Improvement Amendments of 1988. What CLIA does is, it categorizes tests on the basis of their complexity.
So, tests can fall into one of three categories, high complexity, moderate or waived. Waive tests are free of many of the requirements under CLIA for oversight, when these tests are run.
To perform CLIA waived tests, the entity must do one of only a few things. One is, enroll in the CLIA program, obtain a certificate of waiver, pay a biennial fee, follow the manufacturer's instructions, and meet state requirements.
Contrast that with moderate and high complexity tests in which there are requirements for training for each of the individuals who are running the tests, as well as a number of other requirements, including inspections of the facility.
The sponsor must apply for CLIA waiver. It is not granted automatically. An application is made to FDA after initial approval, and results of studies must be supplied to demonstrate that the device is simple and accurate in the hands of intended users.
By the way, CMS, the Center for Medicare and Medicaid Services, is also involved in the CLIA and is involved in the inspections and much of the program as well.
As far as the HIV testing goes, what is the impact of CLIA waiver? What it has the potential to do is that, again, subjects can be notified of their test result without the need to be recontacted or the need for a second visit, and fewer laboratory restrictions permit potentially wider use.
BPAC discussed this issue on June 14, 2001. There are now two tests that have received CLIA waiver, each of the two tests for two different types of specimens.
OraQuick has received a CLIA waiver for its use with whole blood, which was granted on January 31, 2003. Subsequently, for use with oral fluid, on June 25, 2004.
The Trinity Uni-Gold test was granted a CLIA waiver for use with venepuncture and finger stick specimens as well. On your handouts, there is a bit of a typo. The second Uni-Gold reference should read finger stick, instead of another venepuncture.
I should also make mention of the fact that the sales and use restrictions, which I was discussing before, also apply to the CLIA waive tests.
Now, turning to some of the recurring themes that we have heard over the years -- and you will be hearing more about this, I am sure, over the course of the day.
There are a number of benefits. Number one is, anonymous testing potentially leads to more people knowing their HIV status.
Secondly, earlier diagnosis can translate into earlier intervention and better outcomes for the patient. The home use test kit would empower consumers in making their own health are decisions.
There is the potential impact on behavior and public health. In other words, knowledge of HIV status can lead to change in behaviors that would have resulted in infections, thereby limiting the spread of the virus and having a potentially significant impact on public health.
There are also, of course, a number of risks that have been repeatedly coming up over the years regarding home use HIV test kits.
Inappropriate use of the test or test result. It is absolutely critical to understand the limits of HIV tests.
In tests that we are discussing here, these are tests that detect the presence of antibodies to HIV. These antibodies are a response to the infection by the immune system of the individual, and this response can take about two months.
Therefore, an individual concerned about a possible HIV exposure within a week could very well be infected with HIV, but would test negative.
Actually, there is no approved or licensed HIV test that has the ability to detect infection within a week after exposure.
Continued high risk behavior with a false sense of security of the negative test result would then result in transmission of the virus to others.
There is a potential for adverse outcomes obtaining a test result without live counseling. Concerns have been expressed over the years about the psychological effects of receiving a positive HIV test result without the benefit of live counseling. The biggest issue that has come up repeatedly is suicidal tendencies.
Follow up. Testing in the home leaves the decision to seek confirmatory testing and medical care up to the individual, rather than facilitate it through a counselor. Also critical is the need to notify partners so that they can be tested as well.
There is the issue of coercive testing. Concerns have been expressed about testing people against their will, for example, by insurance companies or employers. Also, forced testing by abusive spouses or family members could potentially lead to violent back lashes.
Another issue is testing by minors. This is not testing of minors, but testing by minors, the ability of a minor to go out and purchase a test, and the question is whether a minor is able to handle the test result, or properly interpret the test result when it is achieved.
A few additional issues, obtaining a test result without a supplemental test. The false positive rate is very significant in low prevalence populations or when there is little risk present.
The availability for those who need the test the most, and a potential conflict with state and/or federal health reporting requirements.
As I close my introductory remarks, I would like to set the stage for you, so that you will know what you are about to hear.
What we have done is arrange for speakers to address issues that FDA feels are key to considering HIV test kits for home use.
You will first hear a proposal by OraSure Technologies for an over-the-counter claim for its OraQuick Advance Rapid HIV-1, 2 Antibody Test, for use with oral fluid specimens, including proposed studies to validate adequate performance in the hands of intended users.
What will also be addresses is the populations that would be studied, and will they be reflecting intended users of the tests, as well as the ability of informational materials to provide counseling and other information in a comprehensible manner by intended users, addressing such issues as accuracy of testing, correct test interpretation, management of psychological and social issues, and medical referral.
You will be hearing a discussion of changes in HIV testing practices and counseling recommendations from Dr. Bernard Branson from CDC.
You will be hearing from Dr. Devery Howerton, also from CDC, who will speak to the role of quality systems for diagnostic tests.
You will be hearing from Dr. Joseph Inungu from Central Michigan University, who will address psychological and social issues associated with HIV testing and HIV home use tests.
You will be hearing an overview of the over-the-counter review process and human factors consideration by Arleen Pinkos from our sister center, the Center for Devices and Radiological Health, and the Office of In Vitro Diagnostics.
Finally, you will be hearing from the public. You will, no doubt, hear passionate arguments on all sides of these issues.
I ask that, while considering all of the information that you will be presented with today, you do so keeping in mind the need to address the following questions:
Number one: Are FDA's previously established criteria for sensitivity and specificity for rapid HIV tests also appropriate to support OTC use for home use HIV test kits.
Number two: Please comment on the design of clinical studies necessary to validate the safety and effectiveness of an over-the-counter home use HIV test kit.
Finally, number three: Please comment on the proposed content of the informational materials and the steps that should be taken to validate the adequacy of the informational materials to communicate or provide pathways to adequately address issues including: accuracy of testing; correct test interpretation; the importance of supplemental testing for confirmation of positive results; management of psychological and social issues; the availability of counseling; and medical referral.
I am going to be returning to these questions, of course, later -- it may be much later -- to assist you in addressing these questions, which the heart of these questions is, what is needed for the validation of these systems. Thank you very much in advance for your input.
DR. DI MICHELE: Thank you, Dr. Cowan. Do the committee members have questions? I actually have one. In reading the Federal Register document from 1995, regarding the licensing of collection HIV test kits for home use, there was a recommendation that the manufacturers follow up with post-marketing surveillance on the psychological and social impact of home collection kits.
Is that some of what we will be hearing later? I guess, did that ever happen, and is that some of what we will be hearing later?
DR. COWAN: I am not aware of any formal studies that have been done. However, the speaker who will address the psychological social issues will cite various references in the literature, at which these sorts of things were examined.
As far as formalized studies to address the psychological and social issues, in a phase IV type study, I am not aware of any results that came out of those.
DR. QUINN: Although it is very different, there are parallels to what the FDA probably went through with OTC pregnancy testing being used by young people, the counseling that is inherent with knowing one is pregnant or not, and so forth.
Will we hear, or would it be appropriate, to hear the steps that the FDA had to go through in those considerations and how they were dealt with, and the type of information and so forth? I don't know how far back that actually goes, as to when that was approved.
There is a big long track record and, if there are some parallels, at least -- different, but parallels -- we might be able to learn from some of those experiences.
DR. COWAN: I believe that Arleen Pinkos will be addressing some of those issues in her talk, and it is her center which was actually involved in those studies, and there are other folks here from CDRH who could help address some of those, if questions remain, after her talk as well.
DR. SCHREIBER: One of the things that we continue to see are these sensitivity specificity issues. We all know that, in blood testing anyway, the EIA is not particularly accurate in terms of the number of repeat reactives that confirm.
In fact, the rate is probably somewhere around 10 percent or so, so that 10 percent of the repeat reactives confirm positive.
Whatever we look at in sensitivity and specificity is a function of the population that it was tested on. If you look at only the high risk populations, you are probably going to get a much higher sensitivity and specificity.
The sensitivity is only a function of the number of positives that are picked up, true positives, not a function of the false positives.
If you look through the information that we have, it appears that there is still the chance for a fairly substantial false positive rate, probably in the literature that we saw, somewhere on the order of 10 percent, and we have a letter here that indicates a 50 percent false positive rate.
I think that, when we review these presentations, I think we really have to keep in -- at least I have to keep in mind how many people are tested repeat reactive, that would never confirm, and those people probably wouldn't go back and seek a confirmatory test, I am afraid.
DR. COWAN: I think it is also important to keep in mind -- to separate the ideas of sensitivity and specificity from positive predictive value and negative predictive value.
These are concepts that I alluded to a bit, where I referred to the fact that someone who has few risk factors -- a low risk person, a low risk group of people -- who are going to be taking this test are most likely going to have a false positive result, again, in the absence of any risk factors.
These are statistical epidemiological considerations that need to be taken into account. Sensitivity and specificity are more absolute numbers, whereas positive predictive value and negative predictive value take into account the group of people that are being tested at the time.
MS. BAKER: Thank you for your presentation. In the course of today's hearings, will we learn about the experience of other countries with infectious disease testing, over the counter?
DR. COWAN: That is not on the agenda at this time. We had considered that briefly, but -- I take that back. You will be hearing some data, I believe, from Dr. Inungu, on some studies that were done abroad as far as psychological testing goes, and there is a variety of experience around the world with over-the-counter testing.
At the same time, there are cultural differences from country to country, what our society is willing to accept versus what other societies are willing to accept.
We felt that it was most relevant to restrict our discussions to what we would expect to find in the American public.
DR. KULKARNI: I just also wanted to know if there is any data available on the impact of these home use test kits that are currently available on adolescents and minors. That is one of the issues that was raised with this particular kit.
DR. COWAN: Dr. Branson may be able to better address issues like that, and I believe you will be hearing something to that effect from him when he speaks.
DR. DUFFELL: I noted that your second point that you want the committee to consider deals with the design of clinical studies.
Can you tell me right now whether or not there are any such studies that have already been undertaken by a manufacturer and, if the answer is yes, are they going to speak to us about the design that they chose and the logic behind it?
DR. COWAN: I will speak to you as an FDA person and say that any questions regarding what a manufacturer is doing should be addressed to the manufacturer. The information is considered confidential and proprietary.
I am not being cagey. I believe that the OraSure people will be able to address issues like that, and you should certainly feel free to question them on any concerns that you have regarding what they are presenting as far as the clinical trial data goes, or as far as anything goes. That is really the point of our questions.
DR. DI MICHELE: Are there any further questions? Okay, I would then like to thank you, Dr. Cowan, for a very, very clear presentation. I think we understand what we are about to hear and the decisions we will be faced with.
I would like to invite Sue Sutton-Jones to the podium. She will speak on the proposal for an OTC home use HIV test kit, on behalf of OraSure Technologies.
Agenda Item: Proposal for an OTC Home-Use HIV Test Kit.
MS. SUTTON-JONES: Hi, good morning. On behalf of OraSure Technologies, I want to express our appreciation just to be here.
We have been very pleased with the adoption of our product in the market since 2002 by public health and other agencies, and we are very encouraged by the results and the feedback we have gotten from that.
To be able to be here and participate in the meeting, listen and learn from all the discussions today, is just a great opportunity for us. So, we are really excited to be here with you today. Hopefully, we will answer all your questions.
Briefly, I just want to go over the agenda. During the talk, what I will do is just elaborate on some of the key points that we feel are important, the intended use of the device as an OTC.
I just want to show you the device very briefly. We have a slide that shows you how easy the sample is to collect, we will talk about the test system itself as it exists now, and how it supports the validation of an over-the-counter application.
Then, what OraSure would propose is needed for studies to further validate an over-the-counter application for this product, our consumer labeling, messaging and packaging, and then a very brief conclusion and recap at the end.
I won't go over all the information on the slides. So, if you have any questions about a specific slide, feel free to stop me. Otherwise, I will just keep moving through them. There is a lot of information on them.
Our proposed intended use statement is very straightforward. It focuses a target or an intended user on the following, which is, it is a single-use device. It is a qualitative test.
The user won't have to worry about calculations or number ranges, as you do with some other tests. It is also a test that will detect antibodies to both HIV-1 and 2, which is important, and it is an oral fluid test.
So, there is no finger sticking, no blood collection that is going to be required by the user in their own home.
Now, this is the collection device. As you see in the picture here, it is very easy to use. You simply go around the gum line once, around the region of the gum, and then it is going to be inserted into a developer bottle, and that is it.
What it does here is, this is the flat pad, and there are no parts or anything that can harm someone with that. Then the results are simply ready to read in 20 minutes.
Now, materials in our test kit for a consumer will be especially configured for use by a consumer and not a laboratory person.
So, we are going to have pre-test and post-test counseling information in the package. We will also have a risk assessment pamphlet. We will have other printed materials in there as well.
We are also going to have pictorial based collection testing and interpretation materials in there. So, even if they don't read everything, they will be able to see it graphically.
Then they will encounter the single use device. They are going to see the developer vial, and then the test stand, which we propose will be built into the package itself.
It is also important to note that there are no hazardous components in this kit. So, disposal is simply household waste or trash. There is no special handling that will be required either with this test for the user to deal with.
Now, the device contains an internal control. It is easy to interpret. What it will do is let the consumer know that the test is functioning properly.
They will know if the sample has flowed through -- well, they will know if the sample has been collected, and then they will know whether it has flowed through the device after they have put it into the developer vial, because of this C triangle, where there is a line that appears here. That will indicate to the user that it is a valid test. So, they will see that.
The technology behind that is basically the antibody conjugate complex is captured on an immobilized protein line that is on that pad. All the consumer will have to know is simply to look for the C triangle here.
Using the same view, once the user has actually performed the test, what they will do is interpret it after 20 minutes.
For a negative, there is only going to be that single line. So, a line will appear here, they will know it is a valid test, at that C or control triangle.
If it is a preliminary positive, what the user will see is two lines. They will see, again, the control line that is going to tell them that it is valid, and then they will see a line here at the T.
This may indicate the presence of antibodies to HIV. The negative result indicates that there are no antibodies to HIV-1 or 2 in the sample tested. So, it is just a very simple two step process that an individual would go through.
Next, I would like to briefly describe some of the clinical studies that have already been performed with oral fluid.
There is a question raised today about the specificity and sensitivity, lower confidence boundaries of 98 percent, and whether that is appropriate or not but, as you saw, we do meet and exceed, with greater than 99 percent, the recommendations by FDA for rapid tests at this point.
So, our sensitivity, specificity and ease of use have been demonstrated through many of these CLIA studies that we have done here.
These studies were also conducted with sufficient statistical power and followed the recommendations of FDA for the population demographics and size.
So, we generated our claims for sensitivity and specificity based on the negative, positive and high risk populations that you see here on the slide.
We based the specificity of the test on all the negative samples that were determined from all three population groups and, of course, their sensitivity was calculated using positives we found within our high risk population, as well as known positive populations that we sourced from known infected individuals.
Now, here in this study, all of these samples were collected, self collected by the subject, and the test performed by them.
Here, again, it shows, one, it is able to be used very easily, and the results of all these studies are well within, again, the sensitivity and specificity claims so far in place for rapid HIV tests.
This is our CLIA waiver study. It was granted based on demographics in a population from a lot of different areas in our target users.
This slide here demonstrates that we had male and female, we had varying ages, as well as a fairly diverse population.
In the next slide, we also considered their educational backgrounds and experience, as well as their professional experience.
You will notice, down here on the bottom of the slide, actually none of them -- 99 percent, not none -- but 99 percent of them had no experience with our device or with rapid tests prior to participating in the study.
In this study, users were actually given a panel to use. There were negative, low positive, and high positive samples that they were to test and interpret.
This study validated the safety and efficacy of our device in those untrained user hands, as well as the accuracy of that test interpretation by an untrained user.
Now, here we also performed studies to just determine the impact of common household substances on performance of our test.
What we found was, there was actually no influence there at all, no effect. We also realized that this is going to be performed under varying environmental conditions, and so, to evaluate that, what we did was focus on these changing environments, perform the testing there, and we found no effect on performance of the test as well.
All of these results, as here on the bottom of this slide, were accurate, and they were concordant with the true serostatus of the specimen that was tested.
Now, in these slides, what we have shown you are studies that have already been done. So, we feel that the technical performance of the test actually does validate this application as an OTC.
However, what we would like to do now is go through slides where we are going to propose some of the studies that we would like to do in order to fully validate it as an OTC application.
This is just a summary slide here. What we have proposed to do is, we want to demonstrate that, with the instructions for use, the messaging, the labeling of the device, that an individual that is untrained is able to perform the test correctly, and that they are also able to interpret the test correctly and accurately.
We want to validate that the labeling and printed materials provided communicate effectively the importance of linkage to care and counseling to the user, and also communicate the options that are out there and available to them for those.
Lastly, we see a need to perform post-market surveillance studies, which were mentioned earlier, to estimate the number of individuals that are taking the test, and then also which options did those individuals actually choose from those available to them.
So, the first study here, the objective is simply, the study will be designed to test the ability of the untrained user to actually collect the sample correctly which, again, is just very simple, one time around the gum line, and then into the developer vial.
They will do this after they read the instructions for themselves, and then also interpret the results. Again, can they do it accurately based on the labeling and instructions for use that we provide to them.
Now, as expected in any study, the population will reflect the demographics of the expected or intended users. We will have untrained users, again, as I said, collect these samples themselves, read the instructions, and perform the test.
The devices will be interpreted by them. They will look for that C or control triangle which, to us, will indicate that it was a valid test and performed properly.
We will develop design or acceptance criteria prior to the study, in order to assure verification of the efficacy of the collection of the sample by the untrained user.
The size of the study population will be sufficient enough to supply statistical verification of these results.
We propose to perform another study, to validate the ability of an untrained or lay person to interpret the results accurately after reading the instructions and performing the test.
It is very similar to the previous study that I just talked about. The difference here is, these individuals will be given a positive and negative specimen panel that they will use to test and interpret.
We will have parallel interpretation of the devices by a trained reader, so that we will have a comparison between trained and untrained users, and then the size and the statistical validity of the study will be established, just as it was in the previous studies, and the ones we have done in the past.
Now, understanding the importance of counseling referral to care, or show or validate the ability of the labeling and printed material to ensure that important linkage to care.
The same rationale will apply for sample size, population demographics, and for statistical validity, as it did in the other studies.
The key objectives that I want to point out in this study is that, the ability of a user to understand will be evaluated.
What we want to know is, they understand the options for pre and post-counseling based on the labeling and instructions for use in the product.
We also want to look at key messaging. That would be messaging such as risk factors, risk behaviors, the potential for false positive and negative results, and then also the important window period where the disease may not be detected if they are tested too soon.
Now, we have not developed fully our labeling that we would use within these clinical studies. That is one of the reasons we are here today, is to hear from the audience, as well as from the advisory committee, what we should include in there.
So, this is just an example of the information or the type of information we would expect to include in our labeling and packaging material. So, it is not intended to be all inclusive, but it is one of the approaches and one aspect of it.
The linkage to care and counseling is very important. With this in mind, we are keenly aware that the clarity of the printed materials, as well as the messaging, is essential to this process.
Because of the criticality of this information, though, we are very eager to partner with other organizations that have experience within this area.
We also will solicit input and recommendations from FDA and Public Health Service during our development process as well.
Now, continuing on with labeling, our general approach again would be to provide pre and post-counseling options that are very clearly defined, so that the consumer can understand them very easily, as well as act upon them.
By establishing a comprehensive support network, the consumer will be enabled to make decisions they are comfortable with and, therefore, reduce their risk of HIV.
Our emphasis is going to be on the numerous choices an individual may make in order to reduce the risk of HIV in the general population.
Now, the post-marketing studies, the last piece of the keystone. Everyone recognizes that an over-the-counter HIV self test kit can play a very vital role in testing where both positive and negative individuals are actually linked to appropriate care, and also to risk reduction counseling.
We intend to partner with public health authorities in order to disseminate and capture some of this information.
By collecting and sharing this information, we can assess which options are the most successful in linking an individual to the multiple resources that are out there now for their use.
In this context, our packaging is a key and critical component to it. So, we have talked a little bit about the labeling and the concepts that we have about what belongs in that.
We actually want to go into now the physical package itself. We have come up with a prototype design based on the following.
We recognized early on that the packaging is an important opportunity for us to communicate key concepts and understanding to the user of the product.
So, we focused on the items here. We want to take the user through a step by step, very methodical procedure on how to use the kit, as well as introduce them to the packaging and printed material within, and the labeling and instructions for use.
We also wanted to enable multiple visual cues. So, again, if they don't read everything, they will see how to collect the sample and how to process it.
Important, it is to minimize missteps, as well as go through procedure sequencing by the user. So, the following is just one concept of how that might be accomplished.
There is our package. It is a prototype. So, it is not a fancy one here. What you are seeing is that the package will be opened by the user.
The lid will come off and then multiple layers, where we have an opportunity to message how to use the kit, message about risk factors, pre and post counseling options with it.
So, as the layers come out, it is intended to be provision of educational materials. There will be graphical representations, as we just spoke about. They are pictorial based, and will provide an individual with visual cues as to how to perform the test and interpret it.
So, now it is revealed here. What you are seeing is that you are now going to see the package. The smaller one is the developer vial.
What they will do is, they will take it out. They will open it up, and they will place it in the test stand. They will open it up. So, that is the developer vial opened.
This is actually this collection device. They will open it, once around, and then insert it immediately into that developer vial, and that is it.
So, it is just a two step process, very simple, and in 20 minutes they will be able to read their results as well.
So, in conclusion, the technical performance of our approach for an over-the-counter HIV-1, 2, oral fluid antibody test, we believe, has already been proven through studies that we have done in the past and have submitted.
The sensitivity and specificity are well above 99 percent, which is well within the FDA guidelines that have been established already for rapid tests, and are under consideration today.
Then, the validation of the test system itself, which includes the labeling and the packaging, for use by a lay consumer, will be further studied, through other studies that will go on, and that will validate that.
Then lastly, we propose to demonstrate the effectiveness of that very important link to counseling and care through post-market surveillance studies that we want to partner with public health to perform.
So, at this time, I would be glad to answer any questions you have. I do have other individuals from OraSure here, that can help out with questions as well.
DR. DI MICHELE: Thank you very much. Perfect timing. Congratulations. The committee has questions. We will start with Dr. Nelson.
DR. NELSON: I wonder if you have done any studies to document the timing of the control spot, or band.
MS. SUTTON-JONES: The line?
DR. NELSON: Yes. In a person who is positive, does that precede, or is it in parallel with somebody who is positive for the HIV band?
In other words, if somebody at the home, instead of letting the device stay in the fluid for 20 minutes, they allowed it to stay 10 minutes or five minutes or eight-and-a-half minutes, is there any possibility that they would have a control band showing that it was a valid test, but that the HIV band had not yet developed?
I wondered if you had done any studies on HIV positives to look at this sequence. We have had some experience with injection drug users using bleach to disinfect injection equipment.
What we find is that, when the drug is there, ready to use, and it requires one minute to disinfect the syringe, oftentimes they do it for 15 minutes, and we found that this disinfection is not very effective in practice.
There is a difference between what a lab would do and what a person would do. I just wondered if this has been studied in any detail.
MS. SUTTON-JONES: It has been looked at, and that is how we established the window for read. The window actually goes from 20 minutes to no longer than 40 minutes to read for the test.
There are studies that have been done. I don't know that any of them have been performed specifically -- unless one of the members of my team do -- that actually addressed an HIV positive individual, and how fast that line developed versus the control line.
It all flows together. It is a lateral flow device. So, in theory, they both come out at the same time and develop.
DR. NELSON: Are any of your studies on seroconverters, that is, within people who have been known to be infected, let's say, a month or two months, as opposed to established infection.
MS. SUTTON-JONES: Yes, there are seroconversion studies that we have done. This is an antibody test and, as long as there are antibodies to HIV, I think we have -- our claims are 400 copies are great. Then, it will most certainly detect it. We will address the window period in the labeling and following the CDC guidelines, of course, for re-test.
DR. DOPPELT: You mentioned that you put the developer in this stand. I assume it has to stay upright during the full 20 minutes. What happens if the cat knocks it over and you find it and stick it back up.
MS. SUTTON-JONES: Actually, it doesn't sit straight up. It is at a slight angle, once it is in there. What the packaging is intended to be -- and that is why we propose to build into the package -- is so that, once it is inserted, it is securely seated there.
So, you would have to turn over the whole test stand to actually turn over the developer vial. So, it is not going to be set on a flat surface.
DR. DOPPELT: If it does fall over, is the test still valid or not valid, and you put it back up.
MS. SUTTON-JONES: We haven't really considered that.
DR. DOPPELT: Does it alter the flow. I mean, it falls over, you put it back up. This is not done in a laboratory.
MS. SUTTON-JONES: That would be one of the -- we call it bash testing, but we would do some consumer testing to actually look at how they perform the test and what possible mistakes they could make with it, and label for it.
DR. KULKARNI: Even though I know that this is for oral fluid only, have you tested other body fluids? I am sure, if I know my patients, they will probably dip it in every little thing that they can think about. Would you label it that way? Do you have data on other body fluids?
MS. SUTTON-JONES: We have data on blood and plasma, whole blood and plasma.
DR. KULKARNI: I am talking about breast milk, vaginal fluid, seminal fluid, things like that.
MS. SUTTON-JONES: We have done testing where it is just common substances, and that is one of the questions, obviously, that we have to address.
Again, this is an OTC application and previously we were working, as you said, with a laboratory. So, it is all part of the consumer testing that we will have to go back and look at, those common substances.
DR. QUINN: On your slide about the untrained user study validation, I mean you did negative samples, low positives, high positives and so forth, and you give sensitivities, how well they did compared, but I wasn't quite sure, what is the gold standard.
Is that a negative sample on OraSure, compared to a western blot? In other words, how well did the untrained people do compared to, say, a trained technology on OraSure? That would be one way of looking at the data, and your potential studies, I hope, will address it.
The other is, compared to a gold standard of EIA western blot, how well do these untrained use individuals compare to that golden standard?
MS. SUTTON-JONES: That is typically the characterization we do of our specimens. Before we actually use them, we know, in general -- it depends on the study, but we will do western blot as well as EIAs.
DR. QUINN: So, for these, the low positives, high positives, that is like western blot confirmed, and that is how well they compared to that.
MS. SUTTON-JONES: Right.
DR. QUINN: Another study will be, how well does an untrained individual compare to a technician. I think lots and lots of data on that would be useful.
The follow up question is, although you have addressed many of the issues that have been raised, the two that I don't know if your future studies are going to get into, one if these reporting requirements by state. The other is no follow up, or partner notification issues. That is a little bit beyond your control. I didn't know if you had any thoughts about that, for future studies, how you might want to look at that down the road.
MS. SUTTON-JONES: Certainly, through post-market surveillance, we would want to look at it, and you bring up a good point about reporting.
We obviously are not going to distribute or work within an area where it is not allowed. So, if a state, for instance, has specific reporting regulations, then we would be precluded from selling over the counter there.
We are going to work, though, with public health services to do that, and establish that, and find out what the impact might be, if any, on them.
DR. SZYMANSKI: I wonder, to which kind of population you are intending this test. Is your purpose to meet some need that is not met right now?
MS. SUTTON-JONES: Right now what we have talked about is just the general population. What we need to do is a market study, where actually we look at who would be the most typical consumer of this product.
Right now we are guessing general population, because of the venue it would be sold in. So, we do need to do further market studies to actually pinpoint who would do it.
The unmet need is actually -- much of what was up there is the benefits for an HIV home use test. That need is not met, because there are individuals that don't get tested, or won't come back for results to a health clinic.
DR. KUEHNERT: I saw that the sensitivity and specificity is very high, but you do have some false positives and false negatives.
I was just wondering if we are going to hear today about the reasons for those false positives and false negatives, both in terms of those that are thought to be operator dependent and those operator independent, along the line of what Dr. Quinn was saying.
That would give us some insight into what the pitfalls might be. So, I didn't know if you were going to be presenting that data today. That would be very useful.
MS. SUTTON-JONES: No, but basically it is just based on the specificity and sensitivity of the assay. We have a very nice balance where they are very close to each other. That is the reason, obviously, that they occur.
There are also certain preexisting disease states, for instance, that would cause a false negative to occur. Certainly, if you are on any kind of immunosuppressive therapy, for instance, for AIDS, your antibody level plummets on some of those, and we wouldn't pick that up, but no one would.
So, it is just that has an antibody based test. So, that primarily is why you are going to see a false negative.
The other is, it is going to based on obviously also the predictive positive calculation. Therefore, we are going to have to identify exactly who might use it or, if it is general population, what the demographics of the United States are, as far as prevalence of AIDS infection.
DR. KUEHNERT: So, for the false positives, that also, you said, was based on some underlying condition and not based on, as Dr. Doppelt said, something of the cat knocking over the test.
So, it basically was some underlying condition in the patient. Is that what was thought to account for false positives?
MS. SUTTON-JONES: That, and there are just false positives that occur. It picks up something within the specimen.
I can't say that they are all related to disease states, no. There are false positives, and that is in the literature.
DR. KUEHNERT: Just one more thing. On a false positive or a false negative, there will also be data on whether, on a repeat test, whether that happened again? Will there be any data on that, whether it was associated with the test episode or, again, just with the patient being repeatedly positive on the test? Will that be sort of looked at?
MR. SUTTON-JONES: That is part of the clinical studies we do now, yes. We will look at that.
DR. DI MICHELE: Just relating to Dr. Kuehnert's question, if you read the brochure and your untrained individual for the OraSure validation studies you did for CLIA, it appeared that the errors really did segregate to one or two individuals performing the test. Is that not correct? I mean, that is what the brochure says.
MS. SUTTON-JONES: It does. I am not sure if they were the same individuals in each instance, though. Do you guys know?
DR. DI MICHELE: We can come back to that.
DR. MANNO: I have two questions about the package information. The first is, how likely do you think it is that people will go to a web site to find a local provider for care for HIV infection? Do we know how often that works?
MS. SUTTON-JONES: Well, there are generalized research studies. We would have to establish that ourselves.
DR. MANNO: It just concerns me that some of the people at risk might not be the sort of part of our population that uses on line resources.
MS. SUTTON-JONES: No, but there is a toll free number, and there will be 24/7 access for individuals to call, if they want to call and talk to someone rather than use a web site.
DR. MANNO: Would you recommend other local public health recommendations for care?
MS. SUTTON-JONES: Oh, yes. Both will be there. We will have geographic specific.
DR. MANNO: The other question I have is about packaging, and I am sure your packaging people gave you recommendations about how people open packages.
It seems to me that someone would want to see the information about the requirement for counseling as well as the testing standards prior to purchasing the product.
It might be a good idea, rather than depending on them opening these two pieces of cardboard, which I can assure you they would toss, that some of this information be on the label.
MS. SUTTON-JONES: Actually, those pieces that flip up would have printing and figures on them.
DR. MANNO: Once they get the product, I think they are going to want to use it.
MS. SUTTON-JONES: Okay, we will do that.
DR. LAAL: I wonder what the sensitivity of your test is in non-clade B infections, in the other clades of HIV.
MS. SUTTON-JONES: Actually, I don't know. Keith, do you know? Keith is from OraSure.
MR. CARDOS: For the non-clade B, we did run the worldwide panel, and that is reported within the package insert. So, we did look at the kind of full diversity. I can't quote that right now without a slide, but it is in the labeling.
The other question that was brought up on the user study, the reason it was delineated to the two users, when that test was done, it was a batch mode of randomized samples.
So, in any case where there could be a potential sample mix, that would count against us, but that was one thing in that study that we would correct in a following study, is that samples would not be given together. You would be giving them together. So, there is no chance of that.
DR. SIEGAL: A couple of questions. The negative sample is obviously prescreened for HIV negativity and, therefore, screens out false positives.
What about the characteristics of your low positive sample? Is the N actually just 200, and what were the demographic characteristics of that sample?
MS. SUTTON-JONES: Actually, the negatives are any negative, even if it comes from the high risk population. So, negatives weren't screened prior to -- we did include those in specificity. Now, to your next question, I think Keith is coming back up to actually answer that one.
DR. SIEGAL: I just wanted to know a little bit more about what you guys considered a low positive sample. What was the N. Where did they come from. Where was the geography of that population and so on?
MR. CARDOS: The low positive -- for the CLIA study, we actually had a panel that was evaluated in a larger user study, and we picked samples on the low end from really the color of the line.
You saw in the presentation there was a dark line. There can be a somewhat fainter line, and we made sure we covered both ends of the spectrum to make sure we accounted for that in the study. So, the low positive really is the reaction within the test, is the way we have delineated it.
DR. SIEGAL: My question really has to do, in part, with whether you are looking at populations of people from low risk areas, or low prevalence areas for HIV, to find out what the false positive rate is there.
MR. CARDOS: Within the population that we have, we did have the low risk population, which was a substantial set of the clinical data, and also the negatives that came from the high risk. So, there was a good mix of both of those populations in the study.
DR. SIEGAL: And what was the N?
MR. CARDOS: The N I couldn't quote off hand. I would have to look that up in the insert.
DR. SIEGAL: But it is larger than 200?
MR. CARDOS: It was larger than 200, yes.
MS. SUTTON-JONES: It is thousands.
MS. BAKER: I noticed, in your CLIA waiver studies, you did not include teenagers. The education levels were rather high compared to what I understand to be the high risk population.
I was curious if you could address primary language, education level, literacy level, the use of teenagers, and what you would term as the demographics of expected users in slide 17 and 18, when you are discussing your studies that will help to validate the efficacy of the interpretation of untrained users.
MS. SUTTON-JONES: Now, the CLIA waiver studies, the demographics that included the educational experience and professional experience, though, are based on that user. So, it is an untrained individual who is not going to be trained or prepped prior to reading the instructions and running that test, that we hope to get CLIA waived.
It really focuses on that intended targeted user group. What we are proposing to do is repeat those studies, the CLIA studies, in somewhat the same vein, but do it in our target audience.
We have not established what age, for instance, would be the lower boundary or cut off yet. In our current package insert it is 12 years of age and above, but that is something that clearly needs to be looked at and addressed, that we have not made a determination on yet.
DR. KLEIN: My question is similar, and perhaps a follow up. You obviously are looking at English and Spanish and I agree, from what I saw in your original data, the education level if pretty high.
Do you have data from other manufacturers' home use test on other non-English, native English speaking populations in the United States, of which there are many?
Of the high level of illiteracy, actually, whether it is a native English speaker or a non-native English speaker, how do you deal with that, or do you have any data at all from other manufacturers?
MS. SUTTON-JONES: Right, now, again, this is part of the development process that we are going through, and there is one bullet on one of the slides that actually has multilingual.
So, what we have now within the current product is English and Spanish and we would propose that, minimum, we would have to do that.
We are going to have to go back, though, and look at the demographics of our target user and, if it is the general population, then there will have to be multiple languages, as well as our call center will be set up with the most prevalent languages as well.
DR. DI BISCEGLIE: I guess I am still not getting a clear picture of this question of the expected user. It is not defined, and I also -- you are sort of trying to build a case that there is a an unmet medical among medically underserved individuals, but it sounds like you want to sell the kit to the general population with dollars in their pocket.
So, you kind of need to do your studies, I think, where the need is, I think, and more clearly define the expected user.
With that in mind, I had a specific question. The specificity was very good, but the place where the rubber meets the road is in the low risk individuals, the negative individuals.
You showed data for specificity with everybody all lumped together. I would like to know the specificity in the low risk group alone.
MS. SUTTON-JONES: My recollection is it essentially the same, is it not?
MR. CARDOS: That would be an issue we would want to follow up and actually recalculate the data with just that in mind and get back to you.
DR. DI BISCEGLIE: One follow up if I may, Madame Chair. The other kind of research you might want to consider -- because now you are marketing to the public directly -- is qualitative research.
You know, we are all used to quantitative research, but focus groups. Get a group of consumers together of various education levels, give them packages, and ask them what they think of it, not put in the packet what you think should be in it.
Then the very final question is, the studies, how do you intend to do them? Do you intend to send the individuals home with a test, do it at home and come back and tell you the result, or put them in a back room somewhere and do it? I suspect this needs to be tested with the person doing the test at home.
MS. SUTTON-JONES: Again, it goes back to, we wanted input about the design. You know, obviously we do the focus groups and generally do them in a room with individuals.
That is something we will have to consider, then, is actually sending the person home and getting the results back from them.
DR. NAKHASI: Following on the same line of questioning, one of the criteria that Dr. Elliott Cowan had elucidated in his presentation, was the supplemental test. Maybe i missed it in the slides.
Did you mention anything about how that will be done and how that will be conveyed to the person, and what would be the methodologies by which they would do the supplemental testing?
MS. SUTTON-JONES: That will be conveyed to the user through the packaging and labeling. One of the slides addressed that it is a preliminary positive.
However, the follow up test, the confirmatory test, is going to be done through referral to care, medical care, and counseling.
So, the follow up, we would hope to reflex them, for instance, to public health, to their own physician, and that labeling would convey the importance of doing that to them.
DR. SZYMANSKI: The pamphlet that we received was intended for a different purpose, I understand. The new pamphlet that you are going to make for this particular test, is that going to be similar, and how can you comment on that? I didn't think this was really valid for the home test using the oral fluids, that we received.
MS. SUTTON-JONES: When you say package you received, the hand out. Well, what we are saying is valid is the technical performance of the kit, because we have done thousands of clinical specimens and samples. Oh, you actually have a package insert.
DR. SZYMANSKI: Yes, all the information for the person doing the --
MS. SUTTON-JONES: It would be completely redone. It would be very different for the one that we hand out for personal use. That is part of the development process.
DR. CRYER: I just wanted to know -- maybe I missed it in your presentation -- but how often is the test invalid, and what would you propose to do for the customer if it turned out that way?
MS. SUTTON-JONES: Actually, it is very seldom the test would be invalid. We don't have to date -- you know, knock on wood -- complaints that say the test is invalid, or reports of that, from users that we have, so far.
What we do for the user, I think that is more of a marketing decision than anything else. They can obviously credit them, replace the product, help them walk through the testing prior to doing it again, if they did something mechanically wrong that gave them a funny result. So, that is something that needs to be decided.
DR. SCHREIBER: In one of our handouts, on table six, there is some data on the positive predictive value for HIV-2.
It appeared that there were actually three true positives, but the test picked up 32. It would seem that -- this is in a high risk population. The known positives, it was 100 percent, but on the other group it seemed like you had almost three times as many positives detected as were true positives.
That led me a little bit to worry about whether, in fact, the labeling should be as an HIV-1, 2 kit. I know that the 2s were taken from an African population, and maybe there is a strain difference or something, but can you comment on that?
MS. SUTTON-JONES: No, those slides weren't from the ones I presented, were they?
DR. SCHREIBER: No, but it was in our packet, the company propaganda.
MS. SUTTON-JONES: I actually don't even have a copy of that. Can someone else address that?
MR. CARDOS: If I could ask to see the table? [Peruses document.] Through the labeling for HIV-2, a lot of samples that get into an area that reported HIV-2 could end up to be determined to be HIV-1, and we had a lot of that through the insert.
You notice a lot of that labeling, that we had to have a minimum number of true HIV-2 only positives, and that was part -- as you look through that labeling -- part of that delineation in the insert.
DR. SCHREIBER: The way I read the material in here, it said that those that were true HIV-1 versus 2 were eliminated from the table.
MR. CARDOS: Correct, in what we called the total HIV-2 positives out of that population versus what we called positive.
DR. SCHREIBER: It still seems to me, at least when you read this, and maybe it really needs more detail to go into it, but it really looked to me that you have already culled out what you called the 2s, so that these are now testing as positive on your test, but in fact, there are very few of them that were truly positive.
MR. CARDOS: I think we can work to further clarify the insert.
DR. HOLMBERG: You mentioned that you would only market to those states that did not require reporting the HIV results. Do you know how many states do not require reporting?
MS. SUTTON-JONES: If I said that, I misspoke. What I meant was, we are going to have to work through that, and we are going to work with the public health services.
Again, it goes back to that collection and gathering of information that is a two-way street. We would actually share information back with public service, so that we can get a better idea of the number of individuals tested, as well as the options. So, we would work with each state.
Now, if it came down to, in a certain locality, for instance, that through our negotiations we just weren't able to work something out with them, and they did not want it as an OTC, we obviously wouldn't sell it, and would try to deal with that, but that is something that really has to be worked out.
We are hoping that that will definitely not be the case, but there are some states and localities that the reporting of the number is key and essential. So, we need to get back to those and work with them.
DR. HOLMBERG: I have another question concerning your language on the package insert. You know, we do have quite a few African immigrants who speak French, and i would just bring that to your attention.
MS. SUTTON-JONES: Thank you.
DR. QUINN: Just in terms of this state reporting, again, it comes back to how you are going to position the confirmatory testing, the supplemental testing.
You call it preliminary. It could be probable, you could use a number of different adjectives to educate the individual that this is not a definite answer, here, it is done, you don't need to do any more, you are positive, you are negative.
You are going to need to get them into a referral setting where they get that confirmatory testing. That place should be licensed to report to the state, if it is a state, and there are about 29 states -- probably Bernie Branson knows this -- but about 29 states that are not reportable, or something like that, and the others it is a reportable mandate. You could comment on that.
DR. BRANSON: There have been, up until now, 33 states that require named reporting of HIV, and AIDS, all the states require reporting of AIDS.
The number of states that require reporting is increasing, but they require reporting of confirmed positive results, as you point out.
This test, as it is already approved, we don't do any reporting of preliminary positive results from individuals.
Similarly, many states allow anonymous testing where there is no possibility of reporting in those circumstances. So, in terms of the committee's questions with respect to reporting, we are already dealing with this phenomenon with the preliminary positive, and the reporting occurs after the confirmatory testing is done.
DR. DI MICHELE: Let me ask one question, and then I will go to maybe the last question for this session, because I guess we need to move on or forego lunch, I am not sure which.
You know, obviously there is going to be a tremendous amount of anxiety in the false positive. In addressing that, I know that you have looked at interfering substances and unrelated medical conditions -- some of which you have documented -- can go on to cause false positivity.
I was just wondering to what extent individuals with autoimmune disease -- for instance, recently vaccinated, such as influenza vaccinated individuals, will be looked at with respect to the potential for false positivity, and whether you will indicate in your patient-related materials, or consumer related materials, what conditions might give false positivity, in order to allay anxiety.
MS. SUTTON-JONES: Well, we would. We are going to have to establish some of that, because we have not done the testing on an individual that has just been recently vaccinated, for instance, to see if there is any interference. So, those are the types of studies that we are going to have to do with our audience.
MS. BAKER: Are you familiar with the experience of other countries with home use HIV tests, and how will you use that information to inform your further work, or is this the first ever?
MS. SUTTON-JONES: I don't think it is the first ever. If you look on the internet, you can buy tests right now, and have them shipped to you, where you can do your own test at home. It is not legal, but it is done.
What we will do, though, is do market research, as well as liaise with different individuals who oversee different groups with our notified body, for instance, and get information from them about the performance of this test over the counter elsewhere.
Again, we are focusing on the domestic market, and there are very different cultures, and there is different acceptance, even from European country to European country, for HIV testing in general.
I think the most valuable information is probably here for us, given that this is going to be our target, initially anyway.
DR. DI MICHELE: Thank you, Ms. Sutton-Jones. I think you finally earned your right to sit down. Finally, Dr. Branson, you can come up.
Dr. Branson, from the CDC, will actually speak on changes in HIV testing practices and counseling recommendations.
Agenda Item: Changes in HIV Testing Practices and Counseling Recommendations.
DR. BRANSON: My name is Barnard Branson. I am the associate director for laboratory diagnostics in the division of HIV/AIDS prevention at the CDC.
The FDA has asked us to address some issues with respect to our experience, first of all, with rapid HIV tests, in particular the OraQuick test, and our post-marketing surveillance, as well as some of the other information that we have from the earlier experiences with home sample collection.
So, the outline of my presentation is, I will first describe what CDC sees as the role of rapid HIV tests in the initiative announced in April 2003, advancing HIV prevention.
Second, to just describe and present data that I think will answer some of the questions the committee has raised on post-marketing surveillance for rapid HIV tests and for the home sample collection HIV test kits.
I will present some information on the revisions in counseling recommendations since the committee last considered the home sample collection kit, because I think there have been a number of those, and I will then discuss some of the CDC's planned revisions of testing practices, and then a few words on the anticipated value of an OTC differentiated from a CLIA waived test, and then a comment on validating HIV tests for home use potential studies.
Advancing HIV prevention was a strategy announced in April 2003 that changed somewhat the focus of CDC's prevention activities toward more of a focus on identifying HIV positive individuals.
There were four key strategies: making voluntary HIV testing a routine part of medical care; implementing new models for diagnosing HIV outside medical settings to prevent new infections; working with persons diagnosed with HIV and their partners; and to further decrease perinatal HIV transmission.
So, in three of the four strategies of the new initiative, testing -- and in particular rapid testing -- plays a significant role.
The reason for the change in this strategy is that it is currently estimated, as of a revision this year, that there are between one and 1.2 million people with HIV infection in the United States.
Approximately one quarter of them are unaware that they are HIV infected, and approximately 65 percent of new infections are attributed to this quarter of the individuals who are infected, and there continue to be 40,000 new HIV infections annually in the country.
The role for rapid HIV tests in this initiative, and in CDC's strategy is, first and foremost, to help increase the receipt of test results.
Elliott Cowan presented some of this data but, in 2000, 31 percent of the people tested in publicly funded sites by CDC, with conventional tests and who tested positive, did not return to receive their results.
The second role, we feel, is to increase the feasibility of testing in acute care settings, by providing same day results.
Many acute care settings, such as emergency departments, number one, have very low rates of return, if they test individuals with conventional tests and, number two, are often unwilling to test individuals with conventional tests because of their inability to provide follow up in order to locate people who don't return who test positive, and therefore incur liability.
The third is to increase the number of venues where testing can be offered to high risk persons, and this in particular relates to waived rapid HIV testing.
Then the fourth role for rapid test is to increase identification of HIV-infected pregnant women, who have an undocumented HIV status at the time of labor, so that they can be rapid tested and then be offered effective prophylaxis if found to be positive.
Some of the results of our experience with rapid HIV testing, we conducted a number of studies in different acute care settings, using rapid HIV tests, and others have had similar experiences in the past several years.
Here you see that the prevalence in most of the emergency department studies were considerably higher than the prevalence found in the CDC HIV testing sites with targeted high risk persons.
A large proportion of these individuals, usually about half in each of these studies, had not been tested before, and usually approximately half of these individuals do not have identified risk factors, so that risk based screening would not necessarily help identify these individuals.
CDC initiated a series of demonstration projects looking at rapid HIV screening in medical care settings, not necessarily emergency departments.
As you see, in New York City, there were two clinics and an emergency department with rapid testing that identified prevalence of two percent among those individuals.
Similarly, in Los Angeles, with clinics and emergency departments, an emergency department in Alameda County Hospital in Oakland, the prevalence ranged from .4 to two percent, but we found that facilitating testing with rapid tests in these care settings, again, found a higher proportion of individuals who were infected.
Virtually all of them received their HIV test results and, in most of our studies, 80 percent of the individuals found to be positive entered into care.
In non-clinical settings, we found that a number of demonstration projects -- this is the list of the community based organizations that are providing primarily outreach testing -- this is testing either in storefront sites or very often in mobile vans.
The overall prevalence is 1.4 percent in these kinds of settings, ranging from 0.7 to three percent, depending on the site.
Again, all the individuals received their preliminary positive test results. For many of these individuals, we do not have follow up information on how many received their confirmatory test results.
CDC also conducted a study of HIV screening specifically with the OraQuick test, and this was done prior to its approval, under a treatment IDE from the FDA, and the study was called MIRIAD, the mother infant rapid intervention at delivery.
It involved testing pregnant women in labor for whom no HIV test results were available. It was conducted in 12 hospitals in five cities, and a total of nearly 8,000 women were screened, of whom 54, or 0.7 percent, were found to be new HIV infections.
In this study, there were six false positive OraQuick tests and no false negatives. All individuals were compared to EIA and western blot.
There were 15 false positive EIAs in this population, seven with a P24 band only, and eight of whom were western blot negative.
So, overall, the specificity of the OraQuick in this population with 0.7 percent prevalence was 99.92 percent specificity of the EIA, and 99.8 percent. The predictive value, 90 percent for OraQuick, and 70 percent for the EIA.
We also conducted four studies comparing OraQuick with both whole blood and oral fluid to EIA and western blot.
One was conducted among known HIV positive persons recruited from a clinic in Los Angeles, and then prospective studies were done in an HIV clinic and STD clinics in Los Angeles, which were high prevalence, high risk individuals, another in Phoenix, which was low prevalence high risk individuals.
The pregnant women, there was a substudy done in the MIRIAD sites comparing blood and oral fluid, on an outreach setting study done in Minneapolis which, again, was high risk individuals in a low prevalence setting.
These are the data from the performance of the OraQuick in the known HIV positive persons. Overall, the sensitivity with whole blood was 99.5 percent, with oral fluid it was 99.4 percent.
However, stratifying these individuals by their therapy status, among the people on HAART, all the false negatives occurred on people who were on highly active antiretroviral therapy.
This finding has been previous reported in 2000 by the paper by O'Connell et al, in the Journal of Clinical Microbiology.
In that particular study, they had follow up. It was done in the air force hospital, and they had pre-treatment specimens from individuals who were found to be false negative on the OraQuick test.
These individuals' pretreatment specimens were positive. End point dilution titers on the individuals who were false negative showed substantial reductions in their GP 41 antibody, and decreases in intensity on their western blot bans. In the know positives not on HAART, 186 individuals, there were no false negatives.
In the prospective testing -- I am presenting data from the combined study population, of whom 327 were HIV positive by reference test and 12,000 were HIV negative by reference test, and I am sorry that have not stratified this by different study site, given the information that the committee is seeking from this study with respect to predictive value, because it would obviously depend on the prevalence at the different sites.
Overall, the sensitivity of OraQuick with hole blood in this group was 99.7 percent, and specificity 99.9 percent.
The sensitivity with oral fluid was 99.1 percent, and specificity 99.6 percent, and in this population the specificity of the serum EIA, compared to western blot, was 99.7 percent.
We have conducted post-marketing surveillance on the OraQuick test since it was introduced into HIV testing sites in 2003.
In this first start-up phase, there were 20,585 rapid whole blood rapid whole blood HIV tests evaluated. This was at the time when the test only had an indication for finger stick whole blood testing. Of those tests, 392, or 1.9 percent, were confirmed HIV positive.
We did follow up on all the individuals who had discordant confirmatory test results. So,in that group of -- it ended up being over 400 individuals who had reactive tests -- there were 21, or 5.4 percent, who had a reactive OraQuick test, and a negative or indeterminant confirmatory test result.
In subsequent follow up of these individuals, 10 resolved to be true positive on follow up, and four resolved to be false positive on follow up, and seven individuals, one of whom had a positive EIA, and two of whom had indeterminate western blots, were lost to follow up. So, we don't have specific information on what resolved.
We continued doing post-marketing surveillance on individuals with discordant teste results to see if there are any associations we can identify with false positive test results.
Obviously, because this is post-marketing surveillance, we aren't doing parallel testing. So, we don't' have information to validate any false negatives that may occur.
In the more recent post-marketing surveillance, when the test has been in more widespread use, with both finger stick hole blood and oral fluid, conducted in 2004 and 2005, we have data from 347 testing sites and 14 project areas.
This charge presents the median for confirmed HIV seropositivity, specificity and the positive predictive value.
So, for the rapid test, OraQuick on whole blood, 83,000 tests were done. The positivity median was 0.9 percent, ranging from 0.1 to 2.8 percent.
The estimated specificity is 99.9 and the predictive value in this group was 97.3, as a median overall, with a range of 76.5 in the area with 0.1 percent prevalence to 100 percent in higher prevalence settings.
With oral fluid, we have data so far on 17,000 tests, approximately, of which 1.2 percent is the median positivity rate.
The estimated specificity of the test in this group with oral fluid is 99.8, and the positive predictive value is 94.5.
In conventional testing done in these same sites, about 31,000 tests were conducted, and the HIV seropositivity was 1.5 percent.
An additional point of information was whether people received their confirmatory test results, after being tested with a rapid test.
These data compare rapid and conventional testing from post-marketing surveillance, again, from the 347 testing sites in 14 project areas.
With rapid testing, 99.7 percent of people received their negative test results, compared with 79.9 receiving their conventional results. . Remember, there is a substantial difference of approximately 90,000 people were tested with rapid tests and about 30,000 with conventional tests.
All individuals received their preliminary positive rapid test results. In 90.6 percent -- again, this is a median -- returned for their confirmed positive test results, compared to a median of 83.3 percent of individuals tested with conventional tests.
We have also done follow up quality assurance and outcomes on post-marketing surveillance done in this. >We have data only from 154 sites in seven project areas conducted from January to June of this year.
In those sites, there were 35,188 persons tested. There were reports of four invalid test results. I think that was a question that was raised earlier by the committee.
There was a total of 1,086 controls run for this testing, which represented a median of 2.7 percent with a range of 0.5 to nearly 10 percent of all tests conducted and, out of those 1,000, two controls were reported as invalid.
Two sites each reported testing clients on one day when temperatures were out of range, and one site reported one day when the test kits were stored outside the recommended temperature range.
This now is switching gears to the post-marketing surveillance on home sample collection for HIV testing, which was conducted between May 1996 and September 1997.
CDC has long been cooperating with FDA in evaluating these tests in post-marketing surveillance. There were two home sample collection kits that were approved.
One of the kits, which is no longer no the market, used a pamphlet in the kit package to provide pre-test counseling information.
The other required users to telephone in and register their test kit before the test would be performed. Those users were invited to provide certain demographic information and risk information by touch tone telephone when they called in to register their test kit.
So, out of 165,000 total users during this time period, we have data for just about 59 percent or 77,000 people.
This, we hope, will give us some information on who might be potential users of these kinds of tests. So, as you see, about 62 percent of the individuals were male, among whom prevalence we higher. Eighty-five percent of the users were white.
Although five percent of the users were African American and Hispanic, the prevalence among those groups was nearly three percent compared to a prevalence of about 0.7 percent among white individuals.
Seventy-seven percent of all the purchasers were heterosexual -- I should say of all those who responded to the survey -- were heterosexual, representing 23 percent of the HIV positive users, for an overall prevalence of 0.3 percent.
Eight percent of the purchasers reportedly were bisexual males, representing 27 percent of HIV positive users, with an HIV prevalence of about three percent among the bisexuals.
Ten percent of users were men who had sex with men, representing 38 percent of the positives, and 3.5 percent prevalence or only one percent of the users were injection drug users, representing six percent of the HIV positive users and an HIV prevalence of 4.1 percent.
One of the companies provided information in follow up that was collected in counseling and this is, again, in approximately 76,000 users.
Fifty-eight percent of all the purchases of the kit, and 49 percent of the users who tested HIV positive, had never been tested for HIV before.
HIV prevalence was similar, 0.8 percent among those who had no previous tests, and 0.7 percent among those who reported a previous negative test.
We also did an analysis of the call log that was kept by counselors, when people were provided with their telephone results.
The individuals had an opportunity to call in for their test results. If they were negative, they would receive a prerecorded message with an option to speak to a counselor. All HIV positive individuals were directed immediately toward a counselor.
Of these HIV positive users -- and I don't have a total number up here, there were about 1,600 -- 23 percent had a source of follow up care already at the time of their call. Sixty-five percent accepted referrals for care, and 12 percent were already receiving antiretroviral therapy, and had used the test kit to see whether it worked.
Among the individuals reporting psychological distress in the counselor's call log, seven percent of individuals who were positive expressed shock at the unexpected HIV positive test results. Five percent of the callers hung up immediately without counseling.
Among the HIV negative users, 82 percent opted to receive the recorded message only. Twenty-nine percent of people called more than once to receive the recorded message and post-test counseling, and only 12 percent of HIV negative people elected to speak with a counselor.
I wanted to just present some data overall on where testing is at, and testing practices in the United States.
These are two surveys, the National Health Interview Survey, and the behavioral risk factors surveillance system from 2002.
The National Health Interview Survey is a probability based household sample, and the BRASS is a state based telephone survey.
Overall, the indication from 2002 is that the estimate of either 38 or 43 percent of all individuals say they have ever been tested for HIV. Approximately 10 to 12 percent say they have been tested in the past year.
Testing among individuals who report one or more risk factors in the past year is about twice as high, 21 percent of the NHIS and 27 percent in the behavioral risk factor surveillance system, and pregnant women report testing approximately 50 percent of the time.
Both of these surveys include only non-institutionalized populations. So, they would not include individuals in corrections.
So, the estimate is that, for diagnostic testing, not including blood donation, is there are approximately 20 million HIV tests done annually in the country.
The tests in general, about 43 percent are reported to be done by private doctors in HMOs, about 23 percent in hospital emergency departments, 23 percent by public sources, and five percent at home.
Some of these report that they are being done by a health care provider in the individual's home. So, this does not reflect the use of home sample collection kits, and 4.8 percent of individuals report using tests at other locations.
CDC has had a number of changes or evolutions in their counseling and testing recommendations since 1996 when the home sample collection kit was approved.
Number one, we recommend routine HIV screening in health care settings in high prevalence communities or facilities. This we recommended since 1993.
We recommend opt out consent for pregnant women, where they would receive written or verbal notification that testing would be done, and consent is inferred, unless the woman exercises an option to decline testing.
We now recommend that people can be provided written or verbal information about HIV and, with increasing evidence, in particular, in health care settings, that prevention counseling discourages providers from offering the testing that we recommend, and from some surveys that individuals say that prevention counseling discourages individuals who have been previously tested but remain at high risk from seeking testing, the prevention counseling, in conjunction with HIV testing, is not required in health care settings.
We recommend re-testing at least annually for persons at high risk. Right now this recommendation is for men who have sex with men, and we are considering revisions of that. Obviously, we encourage insurance of linkage of care for persons who test HIV positive.
The rationale for these changes, and some changes that are being proposed, is that there are high levels of knowledge about HIV, about the availability of effective treatment, and experience with HIV testing.
The National Health Interview Survey stopped asking questions on HIV knowledge in 1998 because they were consistently higher than 90 percent.
Many HIV infected persons access the health care system, but are not tested for HIV until symptomatic. It was estimated in 2002 that 42 percent of the people diagnosed with AIDS reported that they had had their first HIV test within one year of their AIDS diagnosis, many at the time they were diagnosed with AIDS.
We have now determined there is inconclusive evidence about prevention benefits from typical counseling for persons who test negative. This is based on meta analysis of 17 counseling studies.
There are specific theory-based counseling interventions that require two 20-minute sessions, and there have been two reports of randomized controlled trials that do show benefit for persons who test negative. So, I stress the term typical in terms of these findings with respect to counseling for HIV negative persons.
On the other hand, we have noted substantial reductions in high risk sexual behavior among persons when they become aware of their HIV infection.
From another meta analysis, there was a 68 percent reduction in unprotected intercourse with partners not known to be HIV positive, among people who were aware of their HIV infection, compared to people who were unaware of their HIV infection.
We continue to encourage prevention counseling for high risk persons, but we believe it does not need to occur entirely in the context of HIV testing, that there are many other indications for prevention counseling, such as for STD clinic patients, who may not necessarily be tested for HIV.
In terms of the potential value that CDC perceives in over-the-counter versus CLIA waived tests -- and I stress potential because we don't have any experience with this yet -- but first of all, we have some qualitative information indicating, from these health care settings, when we have done focus groups among people who refuse testing, that they indicated they might be willing to use a home test. So, there is potential value for persons unwilling to be tested in other settings.
We believe there would be value for these tests among persons who retest frequently. In the National Health Interview Survey, among high risk individuals tested in the past year, the range of tests reported by individuals, the median was 1.2, and the range was from two to 50.
We think also there is a benefit of the knowledge of the partner's status as a prevention intervention. We have done an analysis of the relative reductions in risk comparing condom use, different sexual acts, and having sex with a partner who had recently tested negative for HIV, and having knowledge of a partner's status confers an obvious protective benefit for individuals when they are aware of both their own and their partner's status.
Finally, although Elliott Cowan described the CLIA requirements, our experience, since rapid test introduction has begun, is that there are numerous local requirements for laboratories beyond CLIA requirements that, in some circumstances, impede HIV testing.
For example, CDC has been unable to initiate rapid HIV testing in the state of Hawaii. In New Jersey, CLIA waiver is not recognized. They have enacted legislation in order to allow for testing, but we think that in some circumstances, an over-the-counter test might promote testing in areas where right now it is inhibited or restricted.
We recently have also been notified by some private providers that if they feel that their utilization of HIV tests in their practice was low, they would prefer to have an over-the-counter test in lieu of applying for a CLIA waiver and maintaining a status as a waive laboratory.
Our opinion on the potential for validation studies is that we would recommend observation of self testing at high risk venues.
We would select high risk venues in order that we would have the opportunity to identify reactions of people who test themselves and find out that they are positive.
So, the design we propose would be to have the counselor provide the client with an OTC testing device, observe the client, do the specimen collection and testing, and interpretation of the result, document the client's reaction to receiving the test result, and then, of course, verify the client's interpretation of the test result.
We would recommend that we select several settings serving clients with different characteristics, similar to the studies we did with rapid HIV testing, and test a minimum of 500 clients in settings with a prevalence of three percent to five percent, so that we would have enough opportunity to observe clients who tested themselves and obtained an HIV positive test result. Thank you very much.
DR. DI MICHELE: Thank you very much for that presentation. Questions for Dr. Branson?
DR. NELSON: One of your slides I may have misinterpreted, but where you talked about the testing of persons aged 18 to 64 in 2002, and you had the national health interview study, 48 percent. Is that 48 percent of pregnant women or, of the people who tested, were 48 percent pregnant?
DR. BRANSON: Those are, of the individuals who tested in the past year.
DR. NELSON: So, the denominator is not pregnant women.
DR. BRANSON: No, individuals who tested in the past year so that, of the women who were pregnant, 48 percent reported that they had tested in the past year. Does that make sense?
DR. BRANSON: Only 48 percent of pregnant women tested? That is very disturbing.
DR. BRANSON: We thought so, too, but this is survey based data, and we are obviously attempting to increase those figures. So, we don't have any way of verifying it.
DR. BRANSON: In Thailand it is over 90 percent. We are not doing very well.
DR. KLEIN: I have two questions. First, could you tell me how the phase IV studies were performed, and whether you think that there was some bias in the data that you collected?
DR. BRANSON: The phase IV studies were performed. I think there is definitely a high likelihood that there is some bias in the data that we collected on the post-marketing surveillance.
As I mentioned, there were two kits that were approved and only one collected information from clients, and not all clients answered all questions with the same frequency.
The most frequently answered questions, when they registered on the touch tone phone, was sex. The least frequently answered was zip code. So, people obviously had some concerns, even through the anonymity of that system.
DR. KLEIN: The second question was, do you know how many of the state public health agencies do contact follow up for HIV, and is that a problem, if more people are not presenting through the usual channels, so that contact follow up isn't done, or isn't contact follow up done for HIV?
DR. BRANSON: Many states do contact follow up. I think all states offer some kind of contact follow up service but, in most states, the contact follow up is only done for people who are tested in their publicly funded sites, like STD clinics, or HIV clinics.
Some states do offer those services to people who are tested by private providers, or in other circumstances, but it is by no means universal.
DR. KUEHNERT: I just had a quick question. On this slide on the performance of OraQuick in known HIV positive patients, there was a difference in sensitivity and specificity between those on HAART and those not on HAARt.
Since it is an antibody test, that didn't make sense to me. So, I just wondered what the explanation was and what you thought the reason for it was, or whether you thought it was not statistically significant.
DR. BRANSON: We documented this earlier and that was what I mentioned. We, in fact, did a case control study on individuals who were on HAART therapy who proved to be false negative on the test, and compared them to individuals who were positive.
There was clearly an association with reduction in antibody titer, in individuals who had started HAART early after their infection and maintained long term and effective viral suppression.
Some of these people, in fact, bordered on being indeterminant on the western blot because of the reductions in antibody levels.
DR. KUEHNERT: Would this also be a concern in, say, health care workers who are put on post-exposure prophylaxis and had a needle stick, and are trying to find out what their status is? Would this have an effect?
DR. BRANSON: Do you mean -- I am not sure what the question is. If you are testing the source patient?
DR. KUEHNERT: Yes, for health care workers using the OraQuick after a needle stick.
DR. BRANSON: I think that if you tested a source patient who was on HAART who might be false negative, you would know he is in HAART, most likely. There is the possibility, and that was raised in the paper and we discussed this, that you could potentially get a needle stick from an individual who had been on long-term effective HAART and failed to disclose that information.
On the other hand, if he is on long-term effective HAART with undetectable viral load, the chances of transmission are also lower, since the occupational exposure risk is related to the viral load in the contaminating substance.
For the health care worker themselves, I think we have shown that there are attenuated antibody responses, if you start the person on HAART. If you were doing follow up on a health care worker, I am not sure why you would do it with a rapid test.
DR. KUEHNERT: That was actually my point, was the second. They might not follow up. They might want to just do it themselves if they had an option of an over-the-counter option. I agree with you, that they should be following up, but they might not, if they had the option.
DR. SZYMANSKI: I was very interested in your statement that counseling might not be so necessary any more in hospitals.
Up until now, it is a very important requirement that the person who is tested for HIV will receive counseling.
Also, of course, you need to have permission of a person to test that person for HIV, and now if you have over-the-counter testing, you might have testing -- sort of coerced testing of HIV.
Let's say parents want to have their kids tested and they don't want to, or if you have your partner, and you will test the partner while he is asleep. You could use that without telling anybody or getting any kind of permission. These might possibly be risks.
DR. BRANSON: I am not sure exactly what part of that question to answer. I certainly suspect that somebody using an OraQuick test on my gums while I was sleeping, I would probably wake up.
I would like to address the first part of your question in terms of CDC's recommendation with respect to consent and counseling in health care settings.
Since April 2003, CDC has recommended streamlined procedures for both consent and counseling in health care settings, because many providers perceived that the old perceived requirement for counseling was a significant barrier.
They claim to not have time or at times they claimed to not have skill in order to perform that counseling.
As we continue to document that many HIV infected individuals who are undiagnosed pass through the health care setting without getting HIV testing, that has been a recommendation.
I don't wish to -- I don't remember exactly the way you phrased it, that CDC thinks counseling is not useful, but we are saying that it is not a requirement in health care settings.
We are doing that similarly, in particular, for pregnant women. We recommend the opt out approach to testing, which does not require signing a consent form. It requires -- and this is also recommended by the Institute of Medicine -- that the individual be notified that they are going to be HIV tested as part of a routine panel of tests, and it is clear that they have the option to define, but they do not have to affirmatively sign a consent.
DR. QUINN: Bernie, just to follow up on that, the proposed changes for testing and counseling, are those published now?
DR. BRANSON: No. We have a set of changes that are in process. We had a community forum. We recently had a panel of professional consultants review them, and they are in the process of modification.
The current set of proposed changes apply explicitly to health care settings, and then there is also a revision of the counseling and testing guidelines for non-clinical settings, but the revision of those is probably about 18 months away.
The reason I bring it up is, for the streamlined testing, it seems to be taking place in some arenas but not all, the emergency room at Hopkins, as you pointed out, is streamlined.
In the clinic, when patients are referred in, you need to get detailed informed consent, detailed counseling needs to be provided.
If you want to re-test them six months later, you need to go through the whole process all over again, and it is inhibitory to getting the testing done, both from the health care provider who doesn't want to spend that amount of time, and the individual or the patient doesn't.
It would be really interesting to collect data after CDC issues these guidelines, to see whether health care centers and other facilities are actually following through on streamlining the testing procedures.
DR. BRANSON: I would like to think that CDC's recommendations could solve your problems. On the other hand, at the current time, there are actually a substantial number of states that have statutory requirements for either counseling or consent, that would probably need to be addressed before the CDC guidelines could actually be implemented.
DR. QUINN: Labs are the same way. Labs are actually saying, we won't test that blood unless we get a copy of the informed consent and the counseling has been provided. So, laboratory settings as well.
DR. KULKARNI: I just have a question for you. I am wondering whether, in the different medical settings of prevalence of HIV, whether you look at HIV in sexually abused patients, whether they be adults or children, and would that population benefit from rapid HIV testing or, if a home testing is available, my question is whether they would ever report that abuse, even.
DR. BRANSON: We don't have any data on that, and the nature of our post-marketing surveillance is that we don't collect information that would potentially be that detailed.
DR. DI BISCEGLIE: You mentioned that individuals with a positive test sort of altered their behavior in an appropriate fashion.
Any evidence that somebody with a negative test might take that as a license to participate in high risk activity?
DR. BRANSON: There has been one study reported in the literature in -- don't quote me on this -- I think it was 1992, looking at a retrospective analysis of data from a Miami clinic, showing higher levels of risk behavior among people who tested HIV negative, after they received their test result, compared to the people who didn't receive their test result, compared to the people who didn't receive their test result.
An attempt was made to repeat that study in the same data set in the same clinic in a subsequent year, and the same finding did not persist.
So, this is raised very frequently. There was only one study that suggests the potential for the disinhibition effect that you are asking about.
MS. BAKER: At the risk of sounding like a broken record, are you aware of any other governments who are either in the process of approving over-the-counter HIV home testing, or who have approved it. If there are, what of their experience could inform?
DR. BRANSON: The only other place that I am aware of that has approved a home test for over-the-counter user, is Hong Kong, and I believe it was six months ago and we, so far, have no information on what has happened since it has been approved for over-the-counter use.
DR. DI MICHELE: Seeing no other questions, I will take the last question, if possible. When you spoke about the fact, initially, with conventional testing, that 31 percent did not return for positive results, for HIV positive results, how did that compare with individuals not returning for HIV negative results? I have a follow up question to that, but go ahead.
DR. BRANSON: Thirty-one percent of the individuals who don't return for their results usually averages between 44 and 48 percent. So, that 31 percent figure reflects the additional efforts that were taken to locate individuals with positive results, but those efforts were unsuccessful.
DR. DI MICHELE: So, there isn't, for instance, a prohibitive effect of someone who thinks that they may have engaged in high risk behavior and doesn't want to really know that they are positive. It sounds like there isn't a prohibitive effect in terms of coming back for confirmatory results.
DR. BRANSON: Since the people don't come back, we don't know why.
DR. DI MICHELE: The reason I am asking is, one of the things I wondered about it, in terms of the same thing with the home testing.
I mean, if people were really concerned or very nervous about a positive test, and knew that it wasn't confirmatory, the question is, psychologically, what would be the prohibitions of actually going and getting that confirmatory testing.
You have one follow up study that suggests that 90 percent did, if I was reading that right, which sounds like it is a whole lot better than for confirmatory testing. Is that correct? Am I reading those results right?
DR. BRANSON: Ninety percent of the people who received a rapid reactive test came back for a confirmatory test.
We have actually looked at that more closely in several studies, and we find that some individuals -- for example, when they are tested at a testing site that doesn't offer medical services, and they receive a positive rapid test -- elect to not get confirmatory testing at that site and choose to go to a source of medical care, so they can get their confirmation and have it in the record.
In several instances of people that showed up with a discordant test result, by the time we reached them for follow up, they had already had a viral load done.
I think there are many alternatives that people do, but one issue with respect to the return rates for after reactive test results in these settings is that, since many of these only offer testing services, we have some evidence that people, when they get a positive or a reactive rapid test, go to a care site for follow up testing, rather than return to the testing site and then go to the care site afterward, especially because, in these sites, the number and rate of false positives have been extremely low.
DR. DI MICHELE: Am I reading this right? Do your post-marketing surveillance studies suggest that there is better compliance with confirmatory follow up with the rapid test than with the standard confirmatory testing?
DR. BRANSON: Yes, and that has been a very consistent finding.
DR. DI MICHELE: Thank you. We are going to be under the constraints of lunch. The restaurant will only be open until 2:00, which means that we have to stop at 1:00, or shortly after 1:00.
What I think we will do is actually take one more presentation, and then we will break for lunch and come back for the remaining presentations.
So, I would like to ask Dr. Devery Howerton from the CDC to speak on the role of quality systems for diagnostic tests. By the way, thank you, Dr. Branson.
Agenda Item: Role of Quality Systems for Diagnostic Tests.
DR. HOWERTON: Good afternoon. I would like to focus your attention now on a little bit different topic. We are going to switch gears now.
I was asked to address the considerations of quality systems as they apply to, or potentially apply to, over-the-counter testing.
I know you are all anxious to get out of here and go to lunch, so I will try to speak quickly. The outline of my talk, just briefly, I will touch on what are termed quality system essentials, and then focus briefly on what we might think of as the basic components for testing.
Then, a brief discussion of CLIA waived testing in general, and then some issues to consider in test evaluation.
So, what do I mean by quality systems? This is a definition from a COSI document. That is the Clinical and Laboratory Standards Institute, that a quality system provides a basic framework for laboratories and other health care units to direct and control activities and functions along the path of work flow, with a focus on managing quality.
What is meant by the path of work flow here is, essentially, the sequence of steps or processes that are required to transform the decision to perform a test into the actual test result information.
So, if we can think of a path of work flow, typically we think of it, in a laboratory setting, as involving processes throughout the total testing process, that we describe as pre-analytic, analytic and post-analytic.
If you wanted to imagine this as a home testing process, we could also think of this path of work flow to involve the same processes or the same phases of testing.
In the pre-analytic phase, the individual would need to obtain the test, whether that be going out to their local pharmacy and purchasing it, for example, be able and willing to read the instructions, set up a testing area according to those instructions, collect a specimen and then, in the analytic phase, actually perform the test.
So, in the case of the test under consideration here, whether they would put the device in the developer and read it at appropriate periods of time, and be able to read the results, and then post-analytically interpret that result in light of their own personal health history information. Then, obtain the appropriate follow up testing or counseling as needed.
The quality system essentials, as described in a COSI document, cited at the bottom of this slide, describe 12 of what we call QSEs.
What I have done here is gone through these 12 QSEs and highlighted what I think would actually apply to over-the-counter testing.
That would be personnel, obviously, the individual who is performing the test. The purchasing and inventory control would basically be performed by both the retail outlet, if you would, and the individual purchaser, to ensure that the product is used and stored appropriately, that it is not purchased or used beyond its expiration date, for example.
Then, process control refers to actually performing the procedure, following the steps as prescribed in the package insert, or documentation that comes with the product, whether they adhere to the timing requirements, et cetera.
Information management is something that we typically think of in the laboratory as being done through electronic data reporting systems, for example, where the patient information and test result information are managed.
In the case of over-the-counter testing, this would be actually done by the individual, obviously, and that would require the individual to be able to manage the information obtained from the test results, in light of their own risk behaviors and history.
Facilities and safety, I think has been touched on briefly. Facilities obviously are the environment where the test is stored.
That would be the retail outlet, for example, as well as where the test is performed. The requirements for safety with an oral fluid test would basically be precluded as typically, in a laboratory setting where you have to abide by OSHA standards, for example, for biosafety, which wouldn't really apply in this particular case.
Just distilling this down even further to what I would think of as the three basic components to performing a test, no matter where it is done, this would depend on the individual, the person doing the test, the test environment, and the test materials.
So, if we look at the characteristics of the tester, or the self tester, in this case, one would need to consider the ability of that individual, as well as the willingness of that individual to read the instructions, to follow those instructions, and then to evaluate results and take appropriate action.
Also, the need for that individual to be aware of the need to follow the instructions explicitly without deviation.
Looking at characteristics of the testing environment, I think these are fairly obvious. The controls for temperature, humidity, the need for adequate lighting, for example, in reading the lines on the device, and the level, stable work service without a cat around to knock things over.
The characteristics of test materials, the robustness of the test device at the temperature extremes. What would happen, for example, if it was frozen inadvertently, the shelf life, the impact of using tests after their expiration date.
The complexity of test instructions, I think that has already been touched on a fair amount, and how those instructions -- how readable they are, and how the user has the ability to follow them.
Then some human factor considerations that may come up in the next presentation as far as how the packaging and the device are configured, and how the specimen collection device is used.
Now, just briefly, touching on CLIA waive testing, I think this may be somewhat informative for the discussion today.
Elliott, in his presentation, did describe CLIA waiver. These are the criteria in the CLIA amendment for waiver, the considerations done by the FDA as to whether the test is FDA cleared for home use, employs methodologies that are so simple and accurate as to render the likelihood of erroneous result negligible, or pose no reasonable risk to harm to the patient if performed incorrectly.
Essentially, the limited requirement for a site performing only waive testing is that they obtain a certificate of waiver, follow manufacturers instructions, and permit inspections by HHS under certain conditions.
What I wanted to focus on here briefly is the requirement to follow manufacturer's instructions, which is the only quality system, if you will, requirement for waive testing.
The CLIA regulations, in general, for laboratories performing higher complexity testing, does follow a quality system approach.
These are some data from some surveys conducted by CMS during 2003, 2004. This represents, this is a little over 3,000 testing sites.
Currently, there are approximately 100,000 testing sites with a certificate of waiver. I am just showing this to give you an idea of what the most frequently performed waive tests are in these settings that were surveyed.
This represents a fairly cross sectional survey of the types of sites performing waive testing, and throughout the country.
As you can see, glucose is the test performed most often, about 45 percent of the sites all the way through prothrombin time, which is done in about five percent of the sites.
Interestingly, on this slide, of these top 10 tests, all of them are available over the counter except for the group A strep antigen, the only infectious disease type test here.
Prothrombin time test is available over the counter by prescription, I believe, although these tests performed in waive sites are not relegated only to over the counter testing. There are many, many other tests that are available for these particular analites that are not available over the counter.
Looking at the information that was collected during these surveys by CMS, just to focus in on whether or not the sites were following the manufacturers instructions.
During these surveys, this represents data from 2002 to 2004, where there were about 4,200 sites surveyed. During these surveys, these were on site surveys in which the surveyors had questionnaires that they filled out.
The purpose was primarily not just to collect information on the procedures and the processes being done in these sites, but as well to provide some educational benefit.
During the surveys, they found that 12 percent of these sites did not have current instructions. They weren't even available for all of the tests that they were performing.
Twenty-one percent did not routinely check for changes in the instructions. Just looking at whether they followed manufacturer instructions, about 21 percent did not perform the quality control as recommended in those instructions, and two percent respectively, did not adhere to expiration dates or use the appropriate waive test estimates.
Also, from CDC studies, through the laboratory medicine sentinel network, during a slightly earlier period of time in three different states -- Washington, Arkansas and New York -- just looking at whether these waive testing sites followed instructions for quality control, as you can see here, approximately 60 to 70 percent of the sites did follow instructions, which translates into the finding that 30 to 40 percent did not.
So, given what we know about waive testing, and these gaps that were discovered during these and some other surveys, a number of different approaches have been used to try to give some educational outreach to waive testing sites.
One, of course, is the first one that Elliott mentioned earlier, the FDA sales restriction for rapid HIV testing.
Let me back up just briefly. What I was presenting to you on waive testing really is not focused on rapid HIV. That is waive testing in general.
In fact, I think out of those 4,000 surveys, only about four of them, as I remember, were performing rapid HIV testing. So, I am just looking at this in a broad sense.
Although the rapid HIV waive tests are the only waive tests that do have sales restrictions, and those were basically by a CDC work group that met early, I think it was in 2002, to address quality assurance for rapid HIV, especially with the Oraquick rapid HIV test that had just been approved.
At that time, the CDC developed guidelines for quality assurance, to provide some educational information to sites who might want to perform the Oraquick test on basic quality assurance practices like training and documentation and record keeping, the essentials of quality control, and that kind of information.
These were published on the CDC web site in the summer of 2003. Currently, on the basis of the CMS data and other data collected by CMS, during those surveys, CLIAC, which is the Clinical Laboratory Improvement Advisory Committee, in reviewing those data, recommended that a document be prepared that would provide basic good laboratory practice information for waive testing sites.
This is actually in press currently. We anticipate it will be coming out some time later this month. It goes through the CMS survey data, as well as other data that we have on waive testing performance, and provides a fairly comprehensive set of guidelines for sites performing waive testing.
The intention here is to provide this as an educational outreach. In addition to the MMWR, we are looking at other kinds of products that can be provided as educational components or materials that some of these sites may be able to use.
As I mentioned, the CMS, during these surveys, also provided some educational documents. They have a one-page good laboratory practice document that is provided at these sites at the time of survey. Professional organizations, in addition, provide quality guidance for waive testing, such as COLA.
Now, switching gears briefly just to touch on some issues that might be considered in a test evaluation of a product like this, I have gone through a couple of CLSI guidance documents, again cited here at the bottom of the slide.
The first one is for users, basically for end users, on how to evaluate qualitative test performance. The other. on specifications for immunological testing for infectious diseases, is focused on the manufacturers, to the most extent.
This just lists some of the criteria that you can find in these documents that might be considered during the discussion today.
The obvious one has come up a number of times, the instructions, and the familiarization period with the device.
The evaluation materials, what are the appropriate materials, to evaluate a test in a home use environment, whether they would be actual test specimens or simulated, the process of specimen collection, reproducibility of results, and how the test performs at antibody concentrations near the cut off. That would be where the line would form in a positive test.
Comparison with existing methods or gold standards, and the clinical diagnosis, which would translate into the clinical sensitivity.
Continuing on, looking at test kit stability, as I mentioned earlier, the impact of the shelf life of the test devices, and the use of the test beyond the expiration date, and the variability and reagent lots and source materials, how that would impact the test.
The adequacy of specimen collection, obviously whether there is enough specimen on the device to adequately perform the test.
Test performance, as I mentioned earlier, the analytic sensitivity specificity which was touched on, I believe, in OraSure's presentation with interfering substances, or the limits of detection.
One consideration is the method for assuring quality in the absence of external controls. As these kits are packaged now, there is a stipulation that external liquid controls will be run periodically to evaluate the test performance and, in the over-the-counter packaging, these controls would not be required.
Then, continuing on to considerations by the tester, again, some of this has already been addressed, but whether the tester, when they obtain a reactive results, would seek confirmatory testing, or know how to go about doing that, obtain post-test counseling, and accessing care as needed.
For a non-reactive result, considerations that the individual would need to know as to whether they would need to be retested, and for how long a period of time they would need to wait to have a re-test.
The manufacturers oversight, adverse event reporting. I think I skipped over this in one of the earlier slides. I was looking at the QSEs that addresses the occurrence management.
When we talk about occurrence management, we are talking about the ability to document and report problems. So, one issue would be how would an individual tester or purchaser of this test, or even the pharmacies that might stock them report adverse events, or non-conformities or have questions or comments, how would those be addressed. Also, how the manufacturer would control for changes in production and lot to lot variability.
Just in summary, the basic quality system approach can be applied to home testing. Quality recommendations and guidance were developed, and are continuing to be developed to address gaps in CLIA waive testing, which may be something that might be considered for over the counter testing, some kind of educational outreach.
Lastly, recommendations for test evaluation, include an evaluation of the total testing process. Any questions?
DR. DI MICHELE: Thank you very much. Questions from the committee? Everybody wants to go to lunch. Actually, I have one quick question.
The performance statistics that you gave for CLIA waived facilities, were they meant to show us that, even in the hands of sort of professional users, so to speak, there isn't always good adherence to quality control and to using the test correctly.
Is the implication that in the home you would expect it to be even worse than this in terms of the compliance or better?
MS. HOWERTON: My purpose in showing that was, as you said, demonstrate that -- these are health care settings, primarily. It is a smattering of different types of settings.
About half of these sites are physician office laboratories. Many of them are clinic settings or nursing home settings, outpatient surgery centers, and those kinds of places.
They are typically staffed by health care professionals, although the individuals performing these tests are not laboratory trained.
So, the purpose was just to show you what we know about waive testing, and that in that testing environment the control requirements or just having the manufacturer's instructions available are followed.
How you can translate that into over-the-counter testing, I am not sure, but that is just sort of a measurement, I think, that you might want to have considered in the context of home testing.
DR. DI MICHELE: Okay, thank you very much. If there are no further questions, in closing this particular session, I would like to first of all thank all the speakers for adhering to time so extremely well.
It really allowed the committee to be able to ask a lot of questions and have a lot of their questions answered. So, I want to thank everybody.
We will break for lunch. I have been asked to make sure that everyone is still back here at 2:00, so that we can begin again at 2:00, and to notify the public that it is anticipated, given that we are running behind schedule, that the public hearing will begin closer to 3:00 p.m. Okay, thank you very much. Have a good lunch, everyone.
[Whereupon, at 1:15 p.m., the meeting was recessed, to reconvene at 2:00 p.m., that same day.]
A F T E R N O O N S E S S I O N (2:00 p.m.)
DR. DI MICHELE: Good afternoon, everyone. We are going to begin the afternoon session. I am pleased to invite Dr. Joseph Inungo from Central Michigan University to speak to us now about the psychological and social issues associated with HIV testing, and OTC home use HIV tests. Dr. Inungu, welcome.
Agenda Item: Psychological/Social Issues Associated with HIV Testing and OTC Home-Based HIV Tests.
DR. INUNGU: Thank you very much. The diagnosis of HIV is associated with ongoing deterioration of the quality of life, and a significant curtailment of life expectancy.
It is, therefore, not surprising for the general public to assume that an HIV test will lead to an increase in anxiety, depression and suicide. This concern has been supported by several studies that have been published in respectable journals.
After 25 years of HIV epidemic, there are 40,000 new cases reported every single year in the United States. The fact that the infection in these cases is being driven by people who are not aware of their HVI serostatus underscores the need for us to increase the number of people who know their HIV serostatus.
The question is, how can we meet this objective without causing undue pain. In the next few minutes, really, the objective of this presentation will be to answer the following questions.
The first one will be, what are the reasons people seek HIV testing. The second one, does notification of a positive HIV test result in adverse emotional consequences.
The last question is, does notification of a positive HIV test lead to a sudden and substantial rise in suicide deaths or adverse social consequences.
We will start with the first question, what are the reasons people seek HIV testing. Many studies have been conducted and addressed this specific question, but I will try to summarize the one that I published in the May 2005 issue of the Drug Benefit Trend and AIDS Reader.
We reviewed data from the 1998 and 2002 national health interview survey, to determine the percentage of adults in the United States who ever tested for HIV and the reason why they did so.
In 1998, about 31,000 adults tested were interviewed, and 31 percent of them reported to have been tested for HIV.
In 2002, about 31,000 adults were also interviewed, and 35 percent reported to have been tested for HIV.
Those who reported to have ever been tested for HIV were quite similar between 1998 and 2002, with regard to selected social demographic characteristics.
The main reasons recorded in 1998 were, in 1998, about 34 percent of adults reported to have been tested to find out whether they were infected or not, followed by 16 percent of adults reporting to have been tested because somebody suggested it.
The third most commonly reported reason in 1998 was because of pregnancy or delivery. In 11 percent of cases, it was because the test was part of a routine medical check up.
Then, in 10 percent of the cases it was because of health or life insurance, and then the remaining were less than five percent.
If we consider what happened in 2002, the most commonly reported reason for testing for HIV was because it was because it was part of a routine medical check up that was reported by 24 percent of the people, followed by pregnancy or delivery, and then they wanted to find out if they were infected or not.
So, if we were to compare the two years, what we can see here is people who tested because it was part of a routine check up, the number has gone up from 1998 to 2002.
We can see a similar trend for pregnancy or delivery, but people who seek HIV testing because they wanted to find out whether or not they were infected has decreased from 1998 to 2002. In this group, that we can consider as mandatory HIV testing, things remain about stable.
Although we know that adolescents or young adults represent about 50 percent of the newly diagnosed cases in the United States, they happen to be reluctant compared to adults to seek HIV testing.
When they do test, the return rate is quite low. So, we wanted also to understand the reasons why youth seek HIV testing.
We rely on a study by Murphy that was conducted in 2000, where they studied 246 HIV infected adolescents compared to 141 high risk, unaffected, in 15 cities in the United States.
In 73 percent of the cases, adolescents sought HIV testing out of fear, following high risk behavior. The second most common reason reported among adolescents was because it was recommended by somebody else. Then there was a small percentage that seek HIV testing because they were feeling sick.
These findings seem to support others that were reported by Bogart in 2001, were a recent high risk exposure was mainly the reason that brought youth to seek HIV testing.
When asked why they did not seek HIV testing, again, the most commonly reported reasons had to do with trying to avoid having to deal with distress following a positive HIV test.
The second question that we are going to examine is, does notification of a positive HIV test result in adverse emotional consequences.
Here, the majority of the studies that we reviewed agreed on one thing, that people do experience a high level of distress following at the time of HIV testing.
Once the test result is made available, those who tested negative experienced a significant relief of distress.
When they got a positive test, we found some discrepancies. There was a group of studies that reported an increase in the level of distress following a positive HIV stress, and there was a second group of studies that found a more stable or non-significant change in the level of distress.
Interestingly enough, there was even a third group that reported a decrease in the level of distress following a positive HIV test.
Among people who experienced a significant level of distress, they were further analyzed, and it came out that people who experienced averse emotional reactions were likely to have advanced HIV disease or AIDS.
They were also more likely to have a previous history of psychiatric disorders. They were more likely to be female, low income, to be an injection drug user or a heterosexual, compared to men who have sex with men.
Significantly, they were also more likely to have lack of social support and family. They were likely to be African American or people who were of young age.
Now, we wanted to see what is happening among young adults. Unfortunately, here, I would like to acknowledge that there are limited longitudinal studies on changes of psychological symptoms among adolescents.
All we know so far is really based on cross sectional studies, studies that showed a high prevalence of psychiatric disorders among adolescents.
These will range from anxiety, depression, substance abuse, homelessness, runaway, and all sorts of abuses.
We also know, from cross sectional studies, that adolescents do not handle stressful life events very well. In fact, the younger they are, the more poorly they react to stressful life events.
Further studies were done to review all the studies that reported an increase in the level of emotional reactions following a positive HIV test.
Chesney and associates reviewed all the studies, and what they noticed is that the majority of studies that reported an increase in the level of emotional reactions were conducted in the 1980s, early in the infection, and the numbers of that type of study decreased over time.
That change in the perception really has been attributed to good counseling but also to all the improvement in the HIV management that has taken place in the last 10 years.
I would like to skip the next six slides for the sake of time. These were just a review of specific studies, and they will not affect anything in this presentation.
We will now move to the third question which is, does notification of a positive HIV test lead to a sudden and substantial rise in suicide death.
Here, we need to recognize that there are about 30,000 cases of suicide reported in this country every year, and that number has not changed a lot in the last 10 years.
We also recognize that people who suffer from cancer are maybe four times more likely to commit suicide compared to the general population, whereas people who suffer from AIDS are 66 times more likely than the general community or the general population to commit suicide.
What is interesting to remember is, even if they commit suicide, 90 percent of the people who do commit suicide do have a history of psychiatric disorders.
The question now is, does notification of a positive HIV test lead to increased suicide death. To answer that question, we will review a study by Perry and associates conducted in 1990, where they compared 244 men to 56 females.
They compared the level of suicide ideation two weeks before notification of a positive HIV test, one week after, and then two months later.
What they found is, two weeks before notification, there were 28 percent and 27 percent among HIV positive and among HIV negative who had some suicide ideas. Here, it is again two weeks before notification.
Following notification, the percentage of people who had suicide ideas went down, to the extent that, by the time we reached two months, both of them went down to about 15 percent.
If I were to go back, four percent of people had what we called a suicide which, which is a strong feeling for committing suicide. Interestingly, among this four percent, all of them had clinical symptoms of depression as well.
These findings were also reported by Grassi and associate in 2001, where they compared suicide ideas among HIV positive, HIV negative, and people who had hepatitis C also.
Again, in this group of injection drug users, they did not find a significant difference in the proportion of suicide ideas between HIV positive and HIV negative before and after notification of an HIV test.
The third study that we will review is an interesting one, and here I would like to mention that this study was done in the Netherlands.
So, taking into consideration the difference of the country and culture, I would like you to factor that into the interpretation of what we are going to find.
This study was conducted on injection drug users who were followed for four years. After four years, there were 17 cases of death. Seven were suicide, 10 were due to overdose.
Comparing the two groups, what they found was there was a higher risk of committing suicide among HIV positive than among HIV negative. However, the difference was not statistically significant, as you can see by the 95 confidence interval shown here.
The other finding is, they did not report any single case in this study. They did not find a single case of suicide immediately after notification.
As a matter of fact, the first case that they reported here occurred after six months. So, that in itself is probably suggesting that notification is not directly associated with a behavior that was explained here.
So, based on studies that we reviewed, notification of an HIV positive serostatus does not appear to lead to a sudden and substantial rise in suicide death.
However, the development of HIV symptoms, and the presence of depression are probably the two major factors that lead to depression in all the cases that we review from the literature.
This will lead us to our last question. Does notification of a positive HIV test lead to adverse social reactions.
We also here need to recognize that increased violence after HIV infection is a reality. Considering a study conducted by Zieler on about 3,000 HIV positive people, 21 percent are women, 12 percent were men who have sex with men, and eight percent of heterosexual men did report some type of physical harm.
In half of all these cases, notification of HIV positive results was considered as the major cause of abuse that they sustained.
In the United States, violence is quite common among women. About one third to one fifth of women in the United States will experience some sort of violence during their life time.
It is now established that gender oriented violence is, in many cases, the cause of HIV infection, that sometimes, as in these cases, it can be the consequence of HIV infection.
Another study conducted by Sewell in 2002, based on 275 HIV positive women, showed that 33 percent of all these women reported physical abuse following an HIV positive result.
There was a study by Koenig that reviewed all the studies that reported violence among women, but this time they took into consideration the demographics and the behavior.
When they stratified by these groups, they realized that really, in the final analysis, there was no significant difference between women who were HIV positive versus those who were HIV negative.
Here, the author did recognize that violence in women does exist, and this occurs mainly following disclosure, or around condom negotiation. Those would be the two incidents where violence occurred in HIV positive women.
I wanted also to mention another study, and this one by Klimax that was conducted in 1998, that showed, in that group, that four percent of people did lose their job following HIV testing.
One percent were asked to move out of the house by the landlord, and then one percent did report a case of physical abuse.
So, now we come to our summary. People seek HIV testing for various reasons. What caught my attention here is what is happening among adolescents who happen to be seeking HIV tests out of fear after high risk exposure.
You saw the study by Pollack that showed a recent high risk exposure. That, in itself, is a matter of concern that needs to be addressed because, as previous speakers mentioned, the importance of the window period that will affect the results, and may be misleading for some of the teenagers.
Another concern here is, I would like to point out here that the majority of the studies on psychosocial effect were conducted in the 1980s and the 1990s.
There are very limited studies conducted during the highly active retroviral era. So, there is a need for us to conduct the studies during this time to see how people really react.
The majority of studies that we reviewed are unanimous about the fact that, as soon as somebody gets a negative HIV test, that will lead to a significant decrease in the level of anxiety.
As we mentioned, the result is not that clear for a positive HIV test, and there will be many reasons that will explain the discrepancies that we found here.
The first one would just be based on the way the studies are designed, or the type of instrument that was used to measure the anxiety, depression and other disorders.
Also, one may think that notification of an HIV test may probably not, in itself, be a strong predictor of suicide in this specific case, that other factors, like psychiatric disorders, lack of social support, the fact that somebody has symptoms may play a more major role than notification itself. Again, we need studies, more studies, to confirm what we are just mentioning now.
Although death from suicide is common among people with notification of a positive, it does not appear to lead to a sudden substantial rise in suicide death.
To conclude, social adverse reactions do occur following HIV diagnosis, and it has been shown that these reactions most of the time are associated with lack of knowledge and fear. This is the end, and I will be pleased to take a few questions.
DR. DI MICHELE: Thank you, Dr. Inungu. That was excellent. Committee questions?
DR. KLEIN: I may have missed this, but I am presuming that most of these, if not all of the people in the studies, were actually notified by a medical person and had relatively immediate access to some kind of explanation or counseling, rather than self testing, where there might be no knowledgeable person available, or were there some studies where this is not true.
DR. INUNGU: That is correct.
DR. SCHREIBER: I was wondering if you could comment on the length of follow up for suicide. It seems to me, from my reading, that suicide is contemplated, and is usually a planned event, and perhaps the follow up period is not long enough to see the true effect of being notified that you have HIV.
I don't think I would expect to see it within the first couple of months, but I might expect to see some impact further down along the line.
DR. INUNGU: What you said is correct. In that case, it becomes even more difficult for you to link notification with the occurrence of suicide, and that is a difficulty that we are now facing.
DR. SCHREIBER: I think the other comment that I would make is that a lot of these studies seem to be very small and, while there appear to be perhaps some trends, they are probably underpowered to be able to detect any differences between the two test groups, when you have 100 or so, and you have a rate of occurrence of less than five percent.
DR. INUNGU: That is what we know. For every single study, the sample size is really key to determining the power of the test.
At the same time, things have changed with the management of HIV. So, there is really a need for us to see what is happening now that we have effective antiretroviral therapy in place.
DR. NELSON: You mentioned one, I think,key issue, and that is that people who were symptomatic or had more advanced HIV were more likely to commit suicide than those who were asymptomatic and were early in the infection.
In fact, since we know that nearly everyone will eventually become symptomatic, I think the real issue is perhaps less the testing and, in fact, early testing and perhaps management might actually prevent some suicides that might occur if people were untested until they became symptomatic.
That is a complex situation to evaluate, in a research protocol, but I think it makes some sense that that might be the case. Could you comment on it?
DR. INUNGU: In fact, the literature supports exactly your statement. Many researchers even pointed out that physicians or health care providers will start playing a major role now to prevent suicide from occurring.
It is not just something that happens. It happens in a specific group of people, those who have symptoms, and people who have a history of depression.
With these characteristics,if somebody were to come and see you, you can at least probe whether or not there are suicide ideas, and take action to prevent suicide from occurring.
DR. SZYMANSKI: I think that suicide is one of the extreme emotions that could happen after being informed about HIV status.
There are other very adverse feelings and anxieties that would be created when you learn whether you are HIV positive or negative. Those things, I think, are important as well, in addition to suicide.
DR. INUNGU: Yes, while preparing this, I was specifically asked or tasked to look for anxiety, depression, suicide, and those were the three that I looked at.
I definitely could have expanded that and come up with different conclusions, but we wanted just to start, for this one, and show what is happening really, because suicide and depression were the most -- those are what people refer to when they fear adverse emotional reactions in HIV positive people.
DR. DI MICHELE: I have one question. I don't think you have the data to answer this, but I am going to ask it anyway.
Is there anything, from what you have gleaned in the literature, that would allow you to have some conjecture as to how the availability of over-the-counter home testing might impact on the emotional climate that you just conveyed to us today, and the potential emotional ramifications of HIV testing, particularly in the high risk groups such as the adolescents and young adults.
DR. INUNGU: I would probably say that my answer probably will be just pure speculation, and I would rather keep quiet than go on the record with something that has not been proven.
DR. BRANSON: In response to your most recent question, I think the only thing we are familiar with in the literature that has bearing is based on now three published studies looking at increasing numbers of places that perform HIV testing but provide results by telephone, because that has been shown to increase people's receipt of results.
Many people have expressed that they prefer to receive positive test result in their home situation where they are comfortable, or where they may be someone around, rather than in a clinic situation with a stranger.
So, I think that potentially we should look at the telephone notification of results while people are at home to give us some advice or parallels to what might happen with over-the-counter tests.
DR. SIEGAL: Just a comment, that it is all well and good to show that this very crude measure of distress, which is suicide, doesn't change much, but we shouldn't lose sight of the fact that notification of HIV seropositivity is a life changing event.
If you actually talk to human beings who have HIV infection, you certainly know that. It is hard for me to believe that people's anxiety levels, over all, go down. That is just a personal guess.
DR. DUFFELL: To ut that in context, I think you would have to agree that there are a lot of things in medicine that patients are informed of, find out about, that create stress as well.
DR. SIEGAL: I agree completely, but you have to keep it in perspective.
DR. DI MICHELE: Any other comments for Dr. Inungu? Hearing none, I want to thank you very much. We will go on to our last formal presentation by Arleen Pinkos, on human factors in OTC testing.
Agenda Item: Human Factors in OTC Testing.
MS. PINKOS: Thank you, Madame Chair, and good afternoon to everybody. I want to start off by just giving you a frame of reference.
In vitro diagnostic devices are reviewed within the office of in vitro diagnostic devices, and that is in CDRH, and I am a member of that group.
Today, I will start you off with a little background information, and then cover the three major elements of our review practice, and end up with a few conclusions.
I think we have talked about this already today, but there are about 817 over-the-counter devices that we have cleared, and they are presented before you. Most are on this list. Of note is the absence of any infectious disease products.
Although IVD has reviewed three types of infectious disease products -- self diagnosis for strep A, influenza A and B, and the performance of these devices in professional hands were sensitivity around 55 to 80 percent, and specificity averaging 80 to 90 percent, this type of performance.
In combination with the high risks associated with false positives and negatives, it led the center to determine they were not suitable for home use.
Before CDRH can start their review, they really have to know the who, what, where, when and why about a test, because that is so important.
That not only drives the protocol that needs to be done, but the questions that are going to come up and the data that is likely to be needed.
There are three primary review elements, and I am going to cover each of these in detail. Can the device be accurate and reliable in the hands of lay users? Is it adequately labeled, and do the risks outweigh -- hopefully not the risks outweighing the benefit.
So, the first element is whether the device is accurate and reliable in the hands of the user. As we talk about this element, we are going to talk about five different topics, whether the test can provide the right answer, whether the lay user can get the right answer, a little bit about human factors and risk mitigation, and then end up with some information about stress studies.
Okay, can the test provide the right answer. This question is answered by evaluating the analytical and clinical performance parameters of that test.
Almost all of the over-the-counter devices that we receive have previous clearance for prescription use. So, that information typically comes from that earlier submission.
So, I am not going to spend a lot of time on that, but you can see that generally the analytical and clinical sensitivity and specificity of the device, or predictive values, cross reactivity and interferences, both of which can cause false positives or false negatives, the precision.
We also look at some environmental factors, both of the storage requirements and the requirements for when the test is actually done, and stability of both the reagents and the sample.
Also, as these unitized disposable devices become more and more popular, the actual manufacturing or lot to lot variability in those devices becomes much more critical.
Now we know how it works under optimal conditions in the hands of professionals. The next step is to perform consumer studies.
The goal of a consumer study is to demonstrate that the device has comparable performance when it is used as intended.
So, here we are looking at not only is the stud enrolling representative lay users, but also is it in a representative home environment.
A home is not always a home. As we said earlier, it could be in a mall, in the back room at a doctor's office or a clinic.
One very, very important thing is that you want to capture the entire testing process, from collecting it to running it, to interpreting it, and maybe sending a sample back to the lab for confirmation testing.
It is important that that testing take place unassisted, because that is how it is going to happen in real life.
The studies should be using representative labeling, and that is usually just a single package insert, but the most important thing is that the study is developed and packed just like it is going to happen in real world.
So, if there are educational materials or training, like a video, or an 800 hot line available or something like that, that should also be part of the study, but nothing more.
There are a few things that we need to consider about the sample in the consumer study. It might be a clinical sample or it might be something that is prepared but, if it is prepared, it obviously has to have a matrix, and whatever you are adding to it, such that it is going to act like a clinical sample.
The concentrations of the analite in that whatever it is has to be challenging to the medical decision points of the test.
If you are only testing negative or sky high, everybody is going to look good, and you are not going to know who is good and who is bad.
Sometimes getting clinical samples challenging to the cut off is difficult. What we always have to remember, whether we are using banked samples or selected samples or prepared samples, is that when we review those results, we are not looking at clinical performance. We are really looking at analytical performance. We also want to see that the sample concentrations span the range of expected concentrations.
Okay, prospective sample collection does not always have to occur, but there are certain conditions that would dictate that, like if you needed to re-establish sensitivity and specificity in a different target population, or to determine whether they can actually collect the right sample, or if the sample is not stable.
We also want to avoid exposing participants to biohazardous materials whenever we can, but that is not always possible.
Also, the sample has to be well characterized, whether it is clinically characterized by a diagnostic criteria like WHO, or if it is compared to another credible test.
When we look at the protocol, the protocol should be comprehensive. It should certainly enroll enough participants or samples.
They should be using the correct statistical tools, like when they estimate the sample size, or how they are analyzing the data.
They should be employing masking and randomization techniques to avoid bias. Last but not least -- which you would be surprised about -- but that the protocol was actually written before the study was done, because often that happens, and that certainly biases the test results.
When we consider all devices, but particularly those that are going to be used over the counter, it is important to think about human factors.
Human factors is a discipline that is involved in how people interact or interface with the device. It is why we do the little things we do, and that we don't do.
They are so very important, because we know that users have varying ability, both mentally and physically, and they have a variety of behaviors.
It might be an individual trait, like being color blind, or it might be a group tendency, like not wanting to read the directions.
There are also -- these factors are all very important. They impact the safety and effectiveness of the device.
Human factor-related errors occur when users fail to perform a necessary step or they add something extra, they choose something wrong, make the wrong decision, or if they perform an action poorly, like if it requires a certain technique.
Who is responsible for these user errors, if you want to call them user errors. This illustration, the instructions are telling people to read the directions or read the result from right to left, and the sponsor is wondering, well, how was I to know they were going to do that.
That might be an exaggerated example, but it really stresses that manufacturers need to anticipate all the potential errors. I always like to think of that as applying both common sense and Murphy's law in your thinking.
All user errors cannot be prevented. We found that out when diabetics were cutting their glucose strips in half. I don't think anybody could have estimated or anticipated that. Certainly some, like this example, can be avoided with thinking ahead.
Okay, what do we expect from sponsors regarding errors? We are not talking about just human factors errors, but a mechanical failure, an electrical failure.
We are really hoping, and we try to stress to manufacturers, to have an effective and systematic approach to identifying and dealing with those errors.
The first step is usually identifying potential errors, and then identifying the risk, and then eliminating or reducing those risks to what is seen by most people as an acceptable level. Then finally, to verify the effectiveness of those mitigations.
Okay, here are some examples of the common types of safeguards that manufacturers might built into their device.
They might have a built in mechanism or control mechanism to look for short samples, whether it is the wrong sample, reagent integrity, or maybe they have a temperature strip that is sensitive in the box, to know whether the product ever got outside the required temperature range.
Internal QC is often limited in scope. The best example of this is the little immunochromatographic test strip for pregnancy and things like that.
Those built in controls are really not doing anything more than monitoring the volume of the sample, and nothing else.
People often think that they are monitoring the integrity of the reagent, and they are not doing that at all.
A device might have a shut off mechanism. Manufacturers love using labeling to mitigate errors. There are some labeling techniques, like bolding or putting a box around something.
That can be very effective, but sometimes it is sort of iffy whether that is an adequate mitigation, and we often find out about that after the product gets on the market.
The goal in all of those mitigations is to head it off at the pass before it starts identifying the wrong results.
A sample detection system that allows a result to drop to half the real value before it kicks in, is not really very good, but there are some like that on the market, unfortunately.
Although external QC is very beneficial in professional settings, it is not always a good fit for over the counter products.
It is ont always available or practical. IT is not always adequate. It doesn't necessarily always monitor the entire analytical process, and unfortunately there is no external control that I know of that monitors the sample collection procedures.
Is external QC actually needed? There are a lot of people that, because of the developing technologies and the advent of these little idiot proof devices, so to speak, that a lot of people argue that there is not a lot of value in it.
Lastly, will consumers even do it. Even if you built the best mousetrap in town, there is no guarantee anybody is going to do it. We see that with glucose meters a lot.
After manufacturers develop safeguards, we need to see if they are effective, if they are working. This is often achieved by running stress studies.
What are stress studies? They are basically challenges to the system in order to characterize performance under various conditions of use.
As an example, if the instructions say, read it in 10 minutes, well, you might be anxious and read it in eight, or the phone might ring and you don't read it for 15 minutes.
How are stress studies performed? You artificially simulate a potential error or deviation in some condition, and you are concentrating on the areas that are most prone to errors or failures.
For instance, if there is a certain technique needed, they might shake it instead of inverting it. If the directions are complicated or too long, they may not be able to follow it.
We are always interested in looking at deviations in the temperature, because it is not uncommon for somebody to stop at the mall on the way home from the drug store and leave the product in the hot car for three hours.
Then, after we do the studies, we look at the effect of those deviations on the test result. Okay, here are some stress study examples.
If the procedure were to say to store the kit at 60 to 80 degrees, you might store it that way, but also store it with intermittent freezing, or a little lower or a little higher.
If the procedure says to add three drops of sample, you might add two, three, four, five and see what the effect is.
The key to interpreting studies is, if a deviation that you can reasonably expect to occur is having a result that is clinically significant, then it might not be suitable for use. I saw you crinkle your head. Did that not make sense?
Okay, now let's move on to the second element of review, is the device adequately labeled. Are the steps and instructions simple, concise.
If they are too long, too complicated, you are going to lose your readers. .There have also been a lot of studies that indicate that the level of the writing should be at a seventh grade reading level to be most effective for over-the-counter use.
Do the consumers know who and how they should use the device and when, and are there clear instructions for interpretation.
Sometimes reading result is counter-intuitive. I don't know -- in many pregnancy tests, for instance, the absence of a line is actually a positive result, and the presence of a line is a negative result. That can be confusing.
Do consumers know the limitations of a test? Do they know that no device is ever manufactured perfectly each time or that there are biological variations that could affect the results.
Do they know when to call the doctor? Do they know what to do if the device didn't work. Do they know when and if they should repeat the test.
How do you know when it is good enough? This generally verified in the consumer study. If you get good results, you can assume everything went well.
Consumer studies are limited in their value. There are not usually a large number of individuals, and we observe a trial effect.
For instance, if somebody knows that they are being watched, or the results they get are going to be scrutinized, they are not necessarily acting like themselves. Again, will they read it.
Another tool that we use is a questionnaire. What that is, after the consumer does the testing, you give them a series of questions to ask.
Those questions should be very challenging. They shouldn't be open ended questions like, do you like my device or something like that. It should really hit them and really give you an idea of whether they got it.
The third and final element is where we balance risk versus benefit. As we do this, it is important to remember that lay users are very unique.
They are operating in an unregulated environment, and they may lack medical and technical training. They do not have the access to other medical information like other lab tests or physical examination.
They may not be able to collect sample or follow the directions. They may or may not follow up. They may not do what they are supposed to do.
One thing for sure, there is a high amount of variability in lay users, both in terms of their skills, abilities and knowledge.
We have internet junkies out there, but then we also have people who are very limited in their motor skills. Obviously, a good over-the-counter device is one that is suitable for use by everybody.
When we think about risk and benefit, we need to consider medical benefits and risks to both the user and the public.
Here are two questions that might best capture how we think about risk. What is the impact of a false positive or a false negative, such as in no following up or adverse medical conditions or unnecessary treatment.
What are the medical risks if there is a delay in obtaining professional examination or treatment, or if the device gives an equivocal result.
Here are also two questions that capture the benefit. What is the clinical benefit in terms of screening or diagnosing or doing whatever it is doing. What is the benefit, and what is the medical benefit to having the test available for use at home as opposed to having to go in to a professional's office.
In the end, we end up with the same three questions. Is it accurate and reliable in the hands of the lay user.
Is the labeling good enough, and do the benefits outweigh the risks. If the answers to these three questions are yes, we are generally going to find the device suitable for home use, and I thank you for your attention.
DR. DI MICHELE: Thank you, Ms. Pinkos. Questions for the speaker from the committee?
MS. PINKOS: Oh, I did want to answer about that pregnancy test question. Pregnancy tests were actually pre-amendment.
They snuck in right before 1976. So, they were grandfathered in and we didn't have to worry about it. So, we lucked out on that one.
DR. KUEHNERT: I just had a quick question about you had group A strep and flu in there. Were those discussed in any public meetings, or were those discussed internally only?
MR. GUTMAN: This is Steve Gutman from Arleen's work group, and the strep went to a public panel twice, the influenza never made it to a public panel.
DR. KUEHNERT: It might be useful, if there is a public record for that, to have that, if this comes before the committee again, or when it does, to have those available.
DR. MICHELE: Good comment.
MS. BAKER: You mentioned that over 800 OTC products have been cleared for home use. Do you have any sense of how many have not been cleared for home use and what the general trends were as to why they were not.
MS. PINKOS: Gosh, I don't know if I could even venture a guess. Steve, do you have any good idea?
DR. GUTMAN: I don't actually think we have that tracking information.
MS. PINKOS: Obviously, the newer types of devices are the ones which may not necessarily get a stamp. If it has been around for a long time, it is going to have an easier path.
DR. QUINN: I Just wanted to again get a sense for how this would proceed. The test seems to be fairly accurate. How the benefits might outweigh the risks is the big issue, and how well the individual will actually do the test.
Will a company, such as OraSure, work closely with the FDA? In other words, there was a discussion of some studies that might need to be performed. Is there an interaction with the FDA saying, here are the studies we are going to perform. If these come in and they look good, positive --
MS. PINKOS: Oh, yes, absolutely.
DR. QUINN: So, it comes to your work group?
MS. PINKOS: I am not sure exactly where it is going to go to be reviewed, but it is going to be right here.
DR. COWAN: To clarify that, there was an inter-center agreement which was written, which said that most in vitro diagnostics would be reviewed and handled by CDRH.
However, all blood donor screening tests and any HIV diagnostics would be handled by CBER. So, CBER would be the one to actually do the review.
Of course, we would be consulting with our colleagues in CDRH to get input based on their experience, and that is a critical part of this.
At the same time, the question you ask is a very good one, because the questions you are asking are the questions we are asking of you. We want you to tell us. I didn't want you to get away without -- so, we will be coming back to that in a little while, but I just did want to remind you that, yes, we are going to be working very closely with a company to determine suitability but, at the same time, we need to know what sort of information we would need, or a company would need, to validate for over-the-counter use.
DR. KULKARNI: I was just wondering if there were any requirements for disposal of OTC products. For instance, this OraSure, as somebody said, you can throw it in the trash. I wonder if a roommate looks at the trash and finds out this test, what is the implication of that.
MS. PINKOS: I think OraSure would probably have to answer that, because I am not positive. I don't know what chemicals or reagents are included in their test.
DR. KULKARNI: Or would it be like Mission Impossible and it just disintegrates.
MS. PINKOS: Do you all want to take that question?
MR. CARDOS: From everything we have looked at, a device like this would just fall under standard municipal waste and be disposed of that way.
MS. PINKOS: So, I guess the user just has to use his or her discretion.
DR. SCHREIBER: Does the FDA have performance guidelines for devices? In other words, do you have set, or in the back of your mind, sensitivity, specificity and other measures that you use to evaluate against so that something with an 80 percent is not accepted, but something with a 99 is?
MS. PINKOS: I wish we did.
DR. GUTMAN: Obviously, Elliott pointed out in biologics there are. In the center for devices, there is a handful.
There are well established standards, for example, for cholesterol testing as an international standard for glucose testing.
If any of you have spent much time pursuing the general laboratory literature, you will be astounded by the paucity of information there is on performance of devices, and how good is good enough.
Our work group takes a somewhat contingent approach, as Arleen pointed out. Where it is going to be used and why it is going to be used might determine what kind of performance we would expect, but there isn't a rule of thumb, and there isn't anything cast on gold tablets.
DR. COWAN; If I could just add on to that, question one, I am going to be asking of the committee, the BPAC did concur with the standards for performance that we had recommended back in 2000, and that is 98 percent as the lower bound of the 95 percent confidence interval for sensitivity and specificity.
That was based on state of the art for tests such as that up to that point. The first question that we are asking, of course, is, is the standard which we are currently applying to rapid HIV tests appropriate for use with an over-the-counter test as well.
The two centers do differ, depending on what the analite is, depending on the public health significance, for what the minimal acceptable level of performance actually is, that we chose to have as a criterion.
DR. DI MICHELE: Okay, thank you very much. The bad news is we are behind time. The good news is that, on the revised schedule, we are on time. I will hand this over to you, Don, to initiate the public hearing.
Agenda Item: Open Public Hearing.
MR. JEHN: As part of the FDA advisory committee meeting procedure, we hear open public hearings to give members of the public an opportunity to make statements concerning matters pending before the committee.
Madame Chairperson, at this time, we have 12 written submissions. These statements, when they are cleared, they will be posted on our web site.
We also have received 19 requests in advance for oral presentations, and we are asking that they be held to five minutes each, and do you want to go ahead and read your statement?
DR. DI MICHELE: I need to read, verbatim, a statement prior to opening the public hearing session, and it is the open public hearing announcement for a particular matters meeting. It goes as follows:
Both the Food and Drug Administration -- the FDA -- and the public believe in a transparent process for information gathering and decision making.
To ensure such transparency at the open public hearing session of the advisory committee meeting, FDA believes it important to understand the context of an individual's presentation.
For this reason, FDA encourages you, the open public hearing speaker, at the beginning of your written or oral statements, to advise the committee of any financial relationship that you may have with the sponsor, its product and, if known, its direct competitors.
For example, this financial information may include the sponsor's payment of your travel, lodging or other expenses in connection with your attendance at this meeting.
Likewise, FDA encourages you, at the beginning or your statement, to advise the committee if you do not have any such financial relationships.
If you choose not to address this issue of financial relationships at the beginning of your statement, it will not preclude you from speaking. With that, we will open the public hearing.
MR. JEHN: There are three microphones behind the committee, and we ask -- I am going to call off several names to begin with, so we can sort of queue so we can sort of speed things up. Again, everybody is entitled to their time. Mr. Elliott Millenson, Dr. Wendy Strong, and Mr. Wesley Rodriguez.
Agenda Item: Remarks of Elliott Millenson.
MR. MILLENSON: Hi, I am Elliott Millenson. I was founder and chief executive officer of Direct Access Diagnostics, which was a Johnson and Johnson company which developed Confide, the first home AIDS test approved by the FDA.
I no longer have any financial interest in AIDS testing, but I do have an interest in public health. I am here to offer some history on home AIDS testing, a perspective on how FDA's actions have been perceived by industry, the folks who must commit the resources to bring a home test to market, and some thoughts, as an armchair quarterback, on where we go from here.
I am a businessperson, but I worked with the U.S. Public Health Service in the 1970s, helping educate Americans of the health consequences of smoking.
I have seen first hand what those in public health face. Strong science sometimes needs stronger politics. When I had the idea for a home AIDS test, I thought of developing a rapid test, since the technology was available.
It was clear FDA would not approve such a test. So, we developed a blood collection kit instead, and sought FDA approval.
We expected, after approval, we would quickly transition to a rapid home AIDS test, which we always knew had much broader appeal.
Even getting Confide approved turned into a nine year battle. It started with our 1987 application, to which FDA responded by placing a ban on all home AIDS tests.
FDA announced, in March 1988, by letter to our company, which they copied other companies on, as I gather from their history, but we were the only company which had actually submitted data, based on clinical trials at Johns Hopkins.
FDA responded that they would only consider AIDS test applications -- quote -- for professional use only in a health care environment.
Industry believed FDA's ban was politically motivated. FDA was pressured by AIDS activists, who feared the social consequences of making AIDS tests too easily acceptable, and clinics who feared the financial consequences, reduced government funding, if home AIDS testing caught on.
Opponents have lobbied aggressively against home AIDS testing at all levels of government, with particular focus on FDA.
Opponents conjured up the scare tactic that face-to-face counseling was essential to prevent suicides, a claim that, as you heard earlier, had absolutely no scientific basis.
In fact, for well over a decade, studies have revealed the overwhelming majority of people getting testing for AIDS haven't received counseling anyway.
For perspective, in 1992, CDC conducted a survey and, as part of that survey, at my company's request, they asked about counseling.
That survey showed, in public clinics, 79 percent of people received no counseling, no pre-test counseling or no post-test counseling, 79 percent.
If counseling was so important to avoid this grave social consequence, to avoid suicides, why didn't CDC threaten to cut off funding to those public clinics who didn't provide counseling?
That same survey showed 89 percent of people getting tested in private settings -- at their doctor -- 89 percent received no pre-test or post-test counseling.
The same question for you to ponder. If counseling was so important, why didn't FDA put a boxed warning label on HIV test kits, warning doctors of the grave consequences if counseling weren't provided. Why haven't either of those agencies done anything to this day?
It took two lawsuits from my company just to get FDA to consider our application. Finally, in 1994, FDA brought Confide, our company's product, before this advisory committee, the predecessor committee, which expressed support.
In a subsequent private meeting, documented in a smoking gun memo aired by CBS news, CDC opposed Confide which, I should mention, they had publicly supported.
In private, they opposed it, arguing that approval could lead to -- quote -- a sudden increase in referrals to already over-burdened health clinics.
Two more years passed. About 80,000 more Americans became infected before FDA finally approved Confide in 1996, nine years after FDA refused even to consider our applications.
That is all history now, but past actions affect future actions. AIDS is an incurable sexually transmitted disease that has already killed half a million Americans.
Every day, over 100 more men and women in America become infected with HIV, most from sex with an infected person.
That is a wake up call. Some people don't abstain, aren't faithful, and won't use condoms. Actual use data represent a 10 percent failure rate for condoms, and that is a conservative estimate.
You have been talking about labeling today. If you look at the labeling for condoms, you won't see that mentioned, but it is a reality. Any number of studies have shown it.
That is particularly worrisome, since an alarming number of people who know they are infected with HIV don't tell their partners.
Encouraging people to know not just their HIV status, but that of their partner as well could save many lives and reduce much suffering.
Let me give you some perspective, because we were talking earlier, the question about people testing their partner while they are sleeping in bed.
I am worried about what people do in bed when they are wake. Forty-two percent of gay men self report that they haven't informed their partners of their HIV status, 21 percent of heterosexual men and 19 percent of women.
The only true safe sex is sex between two uninfected partners. We teach our children to put condoms on cucumbers. We need to teach our children the importance of AIDS testing.
In our sexually active society, testing must be accessible in many places, especially the home. The majority of people don't want to go to a doctor's office of clinic for an AIDS test.
Let me go back to some CDC data. CDC -- and you haven't hard this here this morning -- they have known it for well over a decade.
They know a rapid test could quadruple the number of Americans getting tested for AIDS. It is a good sign that a company presented here today. I can prove I have no financial relationship. I haven't mentioned their name.
FDA needs to encourage many companies to develop and market home AIDS tests. Competition means less expensive, faster, more reliable, and more user friendly home AIDS tests will be developed quickly.
Companies skilled in reaching consumers will educate Americans about the benefits of AIDS testing, and responsibly market their tests through a broad range of distribution channels including, no doubt, public clinics, who can give the test away to those with limited resources.
Senior executives at major health care companies believe that FDA has had a strong bias against home AIDS testing. How could they not? For two decades.
America's most innovative companies will be slow to commit their research and development dollars to home testing, unless it is clear that FDA's long-standing opposition has truly softened.
For perspective, I don't know if those of you on the committee know this, but 15 years ago, when the FDA held its advisor's meeting, it was held as part of a settlement agreement.
I sued them. They proposed settlement and said, Elliott, we will hold an advisor's meeting to consider your product.
You know what they said before that meeting? They said, in The New York Times, we are holding the meeting because new treatments mean people are less likely to have adverse reactions to an AIDS test result.
I think the politics have really changed this time. It is too late for a lot of people, but I urge you, the advisors in this room, recommend FDA send a clear, credible, decisive message to industry that FDA is not just accepting applications. It is strongly encouraging them. Thank you very much for your time.
DR. DI MICHELE: Thank you, Mr. Millenson. I appreciate your comments. By the way, I just wanted to indicate to the committee, because there are so many statements being read at the public hearing, that we won't specifically ask questions of any of the speakers, if that is okay with everyone, but if there are any questions, we can ask them at the time of our committee discussions. Is that all right with everyone? Okay, thank you again. We will move on.
Agenda Item: Remarks by Wesley Rodriguez.
MR. RODRIGUEZ: My name is Wesley Rodriguez, and I am from the Latino Commission on AIDS in New York. Thank you for this opportunity to testify today.
As the Blood Products Advisory Committee hears a proposal from OraSure for its oral fluid OraQuick advanced rapid HIV antibody test, the Latino Commission on AIDS is a national and regional organization dedicated to addressing HIV and AIDS in the many Latino communities throughout the country.
We have never testified before this or any other panel convened before the FDA. We are here because almost one half of Latinos who are positive do not know their status.
An over-the-counter home test that provides rapid results, that is available in any drug store or on line, would have direct and indirect benefits that would result in more Latinos knowing their status and getting connected to care.
In addition, the mere fact that the test would be available just as simple as glucose tolerance and pregnancy tests would help to de-stigmatize this HIV test.
The stigma associated now with testing is that only bad people get HIV and AIDS, men and women who use drugs, homosexuals, and/or seen as promiscuous, are viewed by many Latinos as persons who become HIV infected.
So, just taking the test is seen, in many parts of the Latino community, as admission of engagement in this bad behavior.
Making a test that produces rapid results over the counter would help to routinize HIV testing, but we all share the responsibility for fostering and even encouraging the stigma associated with HIV testing in the past.
First, many state laboratories make it very difficult to obtain the limited waiver necessary for rapid testing currently.
Some jurisdictions require that a nurse or a medical technician administer the oral rapid test. Others charge exorbitant fees for these community based organizations that want to offer the rapid test.
Still others make the application process needlessly complex. This contributes to the inaccessibility of the rapid test for those organizations that truly know the communities they serve best.
This patchwork of regulatory requirements only serves to collect revenue for state governments and protect jobs.
The requirements have little to do with public health, especially with the enormous investment that they have made in training community organizations to offer the testing.
Second, many HIV and AIDS organizations and clinics have made a sizeable real estate and personnel investment in the testing process.
The possibility of a home, over-the-counter quick test threatens their revenues and grants of these organizations.
Many of them will try to block the approval of this over-the-counter rapid test for reasons that may be public health oriented, such as proper counseling or coercive testing, but in reality, they turn on financial concerns for their organization.
Third, the FDA and CMS have contributed to this mythology that only certain people can perform the rapid test, with the licensing requirement and the requirement of control tests.
Hundreds of thousands of dollars have been spent on control kits that many in government and industry will tell you are completely unnecessary.
All of these very powerful forces have combined to make testing as something other than routine. They have helped to perpetuate this hysteria that surrounds HIV to this day, surrounds the HIV positive test result, and helps to reinforce fear that many Latinos feel when they get a positive result.
Still, there are legitimate issues that must and can be addressed in any over-the-counter rapid test. First, the package inserts need to be written in a very simple language that explains the necessity of a confirmatory test, also follow up medical care, the time lag between infection and the production of an HIV antibody.
While our concern is that the inserts be easily available in Spanish, other languages are also as important for the African and Asian communities.
Second, the telephone service that comes with the home test needs to be comprehensive and in several languages.
We conducted an informal survey at the commission of the home access telephone service, and found it to be excellent in providing needed referrals, access to medical and mental health care, the importance of partner notification, and the need for a confirmatory test. So, we know it can work.
We would only recommend that some way be found for the company offering the service to obtain a HIPAA waiver from the purchaser, that would enable them to contact a person testing positive in some manner that respects privacy, but ensures that there is follow up. We need to think outside the box to make sure that persons testing positive are connected to care.
Third, state and local health departments need to collect information on persons testing positive in their jurisdictions.
This can be done through emphasizing that the over-the-counter test is only a screening device that requires a confirmatory test.
Through the confirmatory testing, or subsequent medical visits, the government can collect the necessary information it needs.
The confirmatory test and follow up medical visit is a challenge to the current system and will be probably a continuing challenge for the rapid over-the-counter test.
The advancing HIV prevention initiative of the CDC is an important step in reducing the number of HIV infections in the Latino community.
Testing, condom use, monogamy and abstinence are all critical to lowering the number of new infections, but any testing, whether over the counter or in person, needs to be culturally and linguistically responsive.
It also needs to be responsive to the rigid gender roles that impact on men and gay women, and contribute to accessing care and testing.
It also needs to be responsive to the Latino family realities that often fail, the man or woman testing positive for HIV, because of the stigma surrounding an HIV test that is positive.
It also needs to be responsive to the immigration realities confronting many Latinos, making their accessing medical service, housing, employment, and stabilizing their immigrant status more problematic.
It also needs to be responsive to the religious context of many Latinos that fosters the stigmatizing of so many behaviors associated with HIV infection.
Finally, it needs to be responsive to the sexual silent imposed on so many Latinos, which make it so difficult for Latino parents to discuss HIV with their children, or openly confront homophobia and sexism.
All these challenges can and must be met by any counseling offered by the company offering OTC rapid HIV testing, in partnership with medical and community based social service providers.
Ultimately, only a relatively small number of Latinos may take advantage of this over-the-counter rapid test but, for those persons, we need to respect their choices for home testing, and the fact that HIV infection is no longer a death sentence.
By providing a rapid test over the counter, we are one step closer to making it clear that HIV and AIDS is just one more chronic disease. It is manageable with education and care, and preventible with education. Thank you.
DR. DI MICHELE: Thank you, Mr. Rodriguez.
Agenda Item: Remarks by Patrick Keenan.
DR. KEENAN: Good afternoon. I am Dr. Patrick Keenan with the University of Minnesota, Department of Family Practice. Thank you for allowing me to make a statement at this meeting. It is exciting to be here discussing even the possibility of an over-the-counter rapid HIV test.
Financial disclosure, OraSure Technologies paid for my flight and hotel so that I could attend this meeting. Otherwise, I have no disclosures.
My background, in Minneapolis, we have done operational research on the use of OraQuick in the outreach setting, using both whole blood and oral fluid.
We did a CDC funded project from July 2002 through August 2004. When we started the project, OraQuick was still an investigational test.
We have had ongoing rapid HIV testing projects with the Minnesota Department of Health. In the past three years, our outreach workers have done over 5,000 OraQuick tests in diverse settings, such as homeless shelters, chemical dependency programs, patient home visits, et cetera. So, I feel we know this test well, and I would like to tell you about some of our in-the-field experience with OraQuick.
I would like to comment on the technical simplicity of the test. We found that, with a minimum of training, basically the time it takes to demonstrate one run through of the test, about 30 to 60 minutes, people from diverse backgrounds, such as being a printer, a bus driver, chef or bartender, were able to do the test well. We trained 14 workers with the test.
Of course, to train a good outreach worker, we had to cover a lot of other topics, such as universal precautions, HIPAA, pre and post test counseling.
In fact, we felt that the teaching of the OraQuick testing procedure itself was the easiest part of the outreach worker training.
Certainly even this brief amount of training is a lot different from the anticipated over-the-counter use we are talking about today.
How people would do with no previous training certainly needs to be studied extensively, but it is a fact that you don't have to be a laboratorian to perform this test.
Second, I would like to say that OraQuick is a robust test. Our team demonstrated, sometimes inadvertently, that this test held up to temperature extremes in Minnesota weather without loss of accuracy. I don't want to go into that further.
We had very few problems with invalid tests or controls. We did external controls with each worker every day for two years, and we never had an invalid control.
In the 5,000 tests we have done in the outreach setting, we have had only seven invalid tests. Five were blank with no lines at control or test line areas. The other two had been dropped on the floor. I haven't heard whether there was a cat involved or not.
Sometimes, for training purposes, we made use of test devices that had expired dates. We found that the control and test lines became gradually fainter after several months. So, use of expired test devices is a real concern with OTC use of this test, because it could lead to false negative tests.
Post-test counseling, the information that accompanies the OTC test must emphasize the possibility of a false positive test.
In Minnesota, we tested the highest risk persons we could find. We had about a one in 200 to 300 rate of true positives.
Our rate of false positives was just about the same, about one in 200 or 300. So, our positive predictive value was roughly 50 percent.
I think that, in a lot of populations with OTC testing, this is the kind of positive predictive value that would be expected.
I would also like to talk about what I call a knowledge gap. >From personal experience, I think there is a knowledge gap between the public health community that knows a lot about rapid HIV testing, and the rest of the medical community which does not.
I speak of this coming from a primary care department in an academic center. My concern is that a person with a reactive rapid HIV test might see their doctor and receive inappropriate confirmatory testing or counseling.
The post-test element of the packaging should specify that a western blot is needed for confirmation, and the person with a reactive test should be encouraged to impart this information to his or her physician.
If over-the-counter testing becomes a reality, I would recommend an effort to educate primary care physicians about rapid HIV testing, for example, talks at state and regional meetings, and perhaps a mailing from their professional organizations.
Conclusion, the United States would benefit, I feel, from having an over-the-counter, technically simple, rapid HIV test.
There remain many questions to be answered and much work to be done before implementation of such a test, but I think the time is right to start addressing those issues. Thank you.
DR. DI MICHELE: Thank you, Dr. Keenan.
Agenda Item: Remarks by Freya Spielberg.
I am pleased to be here today, to be a part of this discussion. I would first like to say that I don't believe that I have a conflict of interest with OraSure technologies. I am not a consultant and, sadly, I do not have any stock in the company.
I did, however, accept travel expenses, so that I could be part of this today, and I appreciate this.
I am an assistant professor of family medicine at the University of Washington, and a researcher in the Center for AIDS Prevention Research there, and have spent 15 years, actually, studying rapid testing and alternative HIV counseling and testing strategies, both in developing the test and in looking at use in the field, and acceptability of different strategies.
I have done a small study recently that looked at acceptability and feasibility of self testing for HIV that I want to share a little bit of data with you from today.
I don't need to go over the need, because we have heard it over na dover again today, but I think that certainly there is clear reason to believe that the epidemic is being fueled by these 250,000 people who are unaware of their status, and that an over-the-counter test could actually have an impact in changing that steady rate of 40,000 new infections each year.
When people find out they are HIV positive, it actually is the most powerful prevention intervention we have to use, even in all the behavioral interventions that are being promoted, many of which have been proven effective in good studies. Still, finding out that you are positive is more positive than any of those.
So, why an over-the-counter rapid test? Studies have shown that self testing is actually preferable to people at risk, who have never tested.
I have done survey studies among over 400 high risk people. Also, a study in UCSF found the same thing. People who have never tested are more likely to prefer an over-the-counter test.
Also, I think it is important to realize that self testing or an over-the-counter test is more preferable than a home specimen collection test.
I did qualitative studies to find out why people preferred different strategies, and learned that people with the home specimen collection test didn't like a few things about it.
They had to collect it themselves. So, they could make mistakes, but also, of course, they had to wait for results, and they had to put it in the mail, and no one really trusts the mail. So, you could wait for a really long time, and that caused a lot of anxiety.
Then, getting the results over the telephone, even though it is anonymous, people can figure out where you are calling from. So, it is not actually so anonymous.
People were actually 25 times more likely to prefer a self test than a home specimen collection kit, and I think we can expect the market for this test will be much greater.
I think it is also important to think about over-the-counter testing in the context of stigma, and that in areas and communities where there is high stigma around HIV testing, that may be keeping people out of the clinics. This is an option.
Also, the safety. I think you have already heard the data, so I won't go into that, but there are a lot of fears that haven't really been confirmed by evidence.
We need to do the right studies to make sure that getting the test results doesn't cause increased risk, but we don't have any reason to believe that it would.
So, the questions that have to be answered are: Can people perform the test accurately? Will they understand the meaning of the test results? Will they follow up for counseling?
I am really interested in finding out what type of education and counseling will be necessary to ensure accuracy, safety and follow up.
One of the other research projects that I am involved in is, we have created an interactive computer counseling tool that we are using in an urgent care, where the staff couldn't do HIV testing, because they didn't have counseling.
We have found that, with that counseling with the computer, people have good levels of understanding about the test and we are hopeful that it will also result in risk reduction.
So, something like that could also be used in the context of an over-the-counter test, either through the internet or in a kiosk in a pharmacy. Clearly, for post-test counseling, 24 hour telephone counseling will need to be available.
I think it is also important to think about how much people will be willing to pay. In the home specimen collection kit that was priced $35 to $50, that is clearly high above what the people who really need the test are willing to pay.
So, we are not going to have much public health benefit if the test is priced so high that no one is willing to use it.
I conducted a preliminary study of self testing to see just what people thought of it, whether they would be able to take a kit without any instructions except for written materials, and perform it, what the difficulties were in performing the test and interpreting the test.
Initially, what we did was, we performed this in the Madison Clinic, which is an HIV clinic in Seattle. It is an urban, kind of underserved setting. So, very diverse, low socioeconomic status, low education level.
I wanted to find out, among people who already were HIV positive -- so they knew what it was like to get a positive test result -- what they thought about an over-the-counter test.
We did seven waves of 20 people. We did observations to see what the difficulties were. We adjusted the instructions to try to address some of the difficulties, and then we held the instructions fixed and performed another 100 tests. We asked about preferences, and about willingness to pay.
What is important to think about with this data is that this is probably the most difficult population, and with the most difficult climate for the test, I suppose, in that 70 percent of these people were on antiretrovirals.
You have already heard data today about how that decreased the signal level. So, I think that this is a very low estimate of the sensitivity.
So, we saw, for finger stick and oral fluid, there were five to six false negatives out of 100. All of those people were on antiretrovirals.
We also saw almost 10 percent invalids. What the invalids were due to -- and this is something that will need to be addressed by the instructions.
We did make some dent, but there were people that would actually take the device, and they would put it in the fluid, then put it in their mouth, then lie it down, or they would put blood right on the device and then lay it down, and they wouldn't interact with the fluid in the right way. So, that led to invalid results.
The invalids, however, I think we can tolerate a certain level of invalids, because those people realized, oh, the test didn't work. I need to do it again, or I need to go somewhere else for testing.
Preferences, this was really interesting, I thought. As far as location, where would you prefer to test.
If you didn't know your HIV status, where would you prefer to get your test results, and find out you are positive.
Sixty-one percent said they would prefer to get it at home. These are people who knew what it was like. Privacy was considered much better with the home kit, convenience obviously much better. They were more comfortable doing the home kit, it felt safer to them.
The only thing that wasn't as good with the home kit was accuracy, but clearly, a lot of the people who felt that it wasn't as accurate still preferred to do it, because their other priorities were more important.
Willingness to pay. There was a clear cut off in what people are willing to pay. If the test cost about $15 around 70 percent of the people would be willing to buy it and use it.
When you go above that, you see a marked drop off. So, I think that is something that will have to be considered, and I think we also need to be creative about how this test is gotten out there, and that there may be a place for public health to subsidize use of these kits, if it makes it possible for people to afford them and to use them.
Clinical trial considerations, as others have said, I think it is really important to do the clinical trials among the populations where the marketing is planned over the internet and the drug stores, et cetera, and in high risk outreach venues, so we get some data on what the effectiveness would be if we did use it in public health settings.
This question of how sensitive does an over-the-counter test have to be is, I think, a really important one. If the main goal is to increase the number of people in this country who are aware that they have HIV, then we can actually tolerate a lower sensitivity in this test than in a test that we might use in a clinical setting.
I say that because there are a substantial number of people out there who won't access testing in other places.
So, those are people who are only going to be identified if you have a home test. The risk, of course, is if there is a lower sensitivity, then there is going to be a certain population who would have tested in the clinic and gotten accurate results, but switch over, use the over-the-counter test, and may be one of the people who are false negatives.
I think that is the balance that really has to be weighed out. Looking at just sort of modeling some of the considerations -- this is just a rough estimate -- but say that the sensitivity is greater than 80 percent even.
The benefits of a test would outweigh the risks if, for every one first time tester who tested positive using an OTC, there were less than five people that switched over, that would have gone to a clinic but, instead, used an over-the-counter test.
If you assume that over 25 percent of the testers will be first time testers -- which is conservative, because with home specimen collection, it was 49 percent that were first time testers -- then there may be a ratio of one new tester to less than four to five that switch over.
I think, in the least favorable circumstances, from the study that we did, the sensitivity was round 93, 94 percent.
I think that -- well, I think two things. I think that it is very likely that they will be able to design a test that has much higher sensitivity than what I saw here, given the limitations that we discussed.
I also think that, even at lower sensitivities, it is pretty clear that the benefits will likely outweigh the risks.
Post-marketing studies we have already talked about. We need to see what happens to these people. Do they follow up for care, is there increased risk of adverse events.
There will be good mechanisms, I think, to do some of this follow up by integrating into some of the existing behavioral surveillance that is going on, but also I think a longitudinal follow up should be requested among the subset of the people that test using the product.
So, in summary, I think the main concerns, especially of suicide, the risk of self testing remain unproven.
I think actually the concerns about increased anxiety with positive test results, I think that is rather beside the point.
I mean, we wouldn't say don't do a home self breast exam, because if you find something it might make you anxious, and we don't want that to happen.
The ultimate goal is to get people so that they are aware of their status. .There may be some anxiety associated with that, but if they find out they are positive, they can access life saving treatments, and stop infecting others.
So, I think that the FDA should move forward quickly to encourage clinical trials for an over-the-counter rapid test.
I think that it really is one of the few tools that we have to change the course of the epidemic right now, anyway. I am hoping that it is fast tracked. Thank you.
DR. DI MICHELE: Thank you, Dr. Spielberg. Dr. Waheed Khan?
Agenda Item: Remarks by Waheed Khan.
DR. KHAN: Good afternoon, and the honorable committee members. My name is Waheed Khan. I am a retired physician and scientist from Children's National Medical Center here in Washington, D.C., and now presently I am president of ASCO, Inc., who is involved in research and education for HIV/AIDS.
Before I start saying anything, let me mention that I have no interest in any of the companies dealing with this product.
So, my comments will be not as highly educational as the colleagues and previous speakers have presented here, but I am also a world traveler, and I have traveled practically all over the world, and I have seen the devastation of AIDS, especially in Africa and in India.
I know all the extent of tests which are available to people over there and here. Last year, I was in South Africa, and I saw every test available in this country and also abroad. They were available in the pharmacy for people to use it, buy it.
I went to the pharmacy and they offered me -- I inquired, and they wanted to sell me. So, what I am saying is that all the technologies which are developed here, they are not available to the benefit of the people, average people here, whereas the rest of the world is taking advantage for themselves, and for their children and also for their spouses, making good use of whatever is developed here.
So, if I can put in a similar example, that all the medications which are available for AIDS -- ARVs, which are available in other countries at a very small fraction of the cost that it costs here, we have a situation where many of the states here are trying to get all medications, including the ARVs, from Canada with a 30 or 40 percent discount, and there are people in the government who are not in favor of that.
Anyway, I am very much in favor of these home tests. Last April, I submitted a paper in a conference which was going to happen, arranged by CDC and, unfortunately, it was not accepted.
By the way, I have circulated an abstract of that thing here, but what I would like to emphasize is, to give you an example, I am a diabetic for the past 20 years, and I have used the home glucose test.
I quoted my numbers as 16 million people in the United States doing this home use glucose test, and I am not aware of any single -- I am an infectious disease person. So, I am not aware of any single case where, by finger pricking everybody, that anybody has gotten an infection or gotten into trouble.
Similarly, now that this new company, OraSure, have come up with an oral test, which is far simpler than doing even glucose tests, so again I would say, it is highly recommended that this should be followed and, if possible, I would recommend to FDA to speed up this process.
I did learn that, even for a simple pregnancy test, FDA took 12 years to approve it. Now, today, I learned that it was a grandfather clause that brought it in rather than -- so, as a practical advantage, if I may make one suggestion.
Recently, I read in the papers that circumcision gives 60 to 70 percent protection against HIV. I would recommend to my non-Muslim or non-Jewish people, to get ready for that protection.
I must also tell you today, today is our end of Ramadan, which makes it even fitter. This is a very big family affair, and the equivalent of Christmas, that kind of celebration.
I knew this committee was holding this session, and I had presented my little thoughts on the subject. So, I forego the family get together, but I would recommend very highly that, for the human family, let's get together on this thing, and approve all of this home testing as soon as possible.
DR. DI MICHELE: Thank you for being here, and thank you for comments, Dr. Khan.
Agenda Item: Remarks by Patricia Charache.
DR. CHARACHE: I am Patricia Charache, and I represent the American Society for Microbiology. My formal title at Johns Hopkins is program director, quality assurance and outcomes research.
The position of the American Society for Microbiology is that the home use of HIV antibody testing removes all responsibility for result interpretation and counseling from trained professionals, and that erroneous test performance represents a significant risk, not only to subjects, but to society at large.
Such consequences, we felt, could not be justified in the name of increased susceptibility at this time.
The requirements for the waive test is where we have started, and that is that there are essentially two, to employ methodologies that are so simple and accurate that erroneous test results are negligible, an they pose no reasonable risk of harm to the patient if the test if performed inaccurately, and I am going to just address each of those very briefly.
I am going to go quickly, because I will try not to repeat those points that have been made by others this afternoon and this morning.
The risk of error varies with the nature of the population being tested. I think important there is the fact that, with the home test kits, the purchasing population cannot be accurately controlled.
Now, all submissions -- this I haven't heard before here -- of waive tests perform studies. They perform studies to show that the untrained can give accurate results.
I think we should emphasize that the results from such studies need not reflect the accuracy when the test is put out on the market.
The experience is that frequently it does not. In part, that may be because, in the test setting, you have people who know they are supposed to read the instruction and follow them closely, and in the market that doesn't happen.
For the ASM, I think the best example of this is the other infectious disease test that was licensed for influenza.
In trials, it had a very high sensitivity, which is important for that particular assay. In the market place, its sensitivity fell to about 50 percent, meaning that patients would be returned to a nursing home who, in fact, had the disease and perhaps should have even had an antiviral agent.
So, the tests that show the comparisons of the ability of an untrained person to do the test may not parallel out on the market.
Now, in this case, with the OraQuick, we saw that only 12 percent of the patients tested in the untrained studies had 11th or 12th grade educations. Four percent had a GED. Everybody else had advanced learning beyond high school.
That is not likely to be the only population that purchases the test. So, even in the study groups, there may be disparity.
Now, I am not going to go over many of these because Dr. Howerton reported on many of these, but it certainly is the ability of the purchaser to read the directions, ability to follow the directions they have read, ability to monitor time and other test requirements, and we are particularly worried about the time test, because 20 minutes is a long time, and they will get false negatives if they don't follow it. We are worried about whether or not they have cats.
Now here, also, I would emphasize that compliance with confirmatory testing cannot predict compliance with confirmatory tests that are done by looking at the results of the current OraQuick waive test, and trying to predict what we are going to see with an over-the-counter test.
I think we have to remember that, with the waive test, the patients are not simply deciding whether it is a yes or no. Somebody is telling them the results, advising them on how to interpret the test.
They may not see a particular counselor, but somebody is answering their questions and usually making sure that they know where to go to get a confirmatory test, and in some settings, particularly the hospital settings, they may get appointments to clinics or, in other places like our own institution, they will get the blood drawn for the confirmatory test.
So, you can't predict from current confirmatory studies what would happen when there is no consultative support from anyone.
This is just taken from the study that was already quoted here by CMS, showing the compliance and the ability of waive test laboratories to do a waive test correctly.
This happens to be a very minor modification of Judy Yost's slide, that summarized these studies for CLIAC. What they did was look at about five percent of all the waived laboratories in each of eight states, and they found the important factor here would be that 48 percent of the waived laboratories had quality testing problems. We can anticipate that an over-the-counter -- a home sale product would not be any better than that.
Now, we have talked about some of the very simple things with methodology that are challenges, if it is an over-the-counter test, but I would like to comment now on just a few of the second aspect, the no unreasonable harm.
I think what we have to emphasize here is that the risk of harm will vary with the likelihood of a purchase from a high risk population versus a member of the worried well, and that will have to do with the predictive value of the test, as has been discussed.
I can certainly speak from personal experience with working with patients who had a false diagnosis of HIV when they didn't have the disease.
I will just give you one example, a woman and her spouse both sold their private businesses, bought a Winnebago, and decided to drive around the United States until they found she was too sick to travel. It was a false test, and it actually destroyed their lives and their marriage.
The demographics of the purchaser, we said, cannot be predicted, but a subject from a low risk population has a high likelihood of a false positive, such as the worried well, and it can lead to substantial harm, just as a false negative would lead to delay, or less likely to get the test done elsewhere, and delay in appropriate management strategies.
Of highest concern to the ASM is the fact that HIV is a contagious disease, and that the error can harm not only the subject but also partners and others who have someone who is unknowledgeable about his status, but thinks that he knows it. It isn't that he doesn't know if he knows or not. He thinks that he is safe.
Now, we do believe that the way to control such an epidemic is access to quality testing, but we think access to testing which has major questions of accuracy is a greater risk than developing strategies for making controlled tests available, and tests in which there can be some type of a professional, whether it is someone in a group that is interested in HIV, or whether it is a physician or nurse, who can help guide the patient. Thank you very much.
DR. DI MICHELE: Thank you, Dr. Charache.
Agenda Item: Remarks by Tracy Powell.
MR. POWELL: Good afternoon. My name is Tracy Powell. I am the chairman of Home Access Health Corporation. We are the company that is being referred to as collection technology in the market place.
Let me first have a disclosure, that I am not financially involved with OraSure, although if they are looking forward to reimbursing my expenses, it would be appreciated.
I have the challenge to convey everything I have learned in the last 10 years in the next five minutes. Because I am not a scientist, regulatory person, what have you, my brain is a lot smaller than all of you sitting around the table, so I think I can handle it.
We are honored to have received FDA approval from both CBER for HIV at home testing, as well as CDRH for hepatitis C at home testing, and we are privileged to service hundreds and hundreds of thousands of Americans, who have placed their trust in us for education, testing, counseling, referrals, confirmation testing, et cetera, under a world class medically directed system.
It is important for me to point out that what we do is not a kit, a test. It is a subset. Testing is a subset of what we do.
Our system works in a way that there is a collection device that allows an individual to take a micro sample of blood very simply, and I say simply because 96 percent of people who take our test do, in fact, get their results.
A small sample is submitted into a return mailer, to a laboratory, and there is a code number that is attached to that testing system.
That code number gives access to a medically directed system, not a test, but a system. When we tell someone that they are positive for HIV, they are positive for HIV. I think that is extremely important and germane to this discussion.
Now, with respect to my opinion, or Home Access' opinion of rapid testing, at home testing, I am all for it. I think it is a great, great thing, but done so in a systematic way.
There is no reason why standards should be lowered. The system that is in place today is bullet proof. It allows people to get tested in a very highly, highly, standard environment.
I would like to clear up what I think are a couple of misconceptions about rapid testing, over the counter. First of all, it is going to open the flood gates, and we are going to find 300,000 people, like that, that are infected with HIV.
I don't buy it. Two reasons. This is an OTC approval. It is a low risk population. Secondly, when we are out in high risk settings, and we work with public health departments and community based organizations, and when we identify many, many high risk individuals, and we ask them to go ahead and take this test free, sometimes they will literally back up, like we are projecting, magnetizing them.
That is not about time for results. That is about stigma. If there were a cure, I would be altogether different about my opinion about rapid testing and lowering standards that exist.
There is another misconception. A rapid test for HIV is just like rapid testing for pregnancy. Well, I think a pregnancy test, a positive result, is a happy thing and something that is life beginning.
A positive test for HIV is earth shattering. It is life threatening. In the words that our counselors tell me, from many, many individuals, it represents social death.
In terms of efficacy, my view, strongly, is that the general population that we encounter, there is no difference today than there was 10 years ago that we see, with respect to the confusion about HIV/AIDS. In particular, what do you mean, window period. If I am negative, do you mean I am not really negative?
By the way, learning that one is positive or, even worse, indeterminant for HIV is the same catastrophe today as it was 10 years ago.
So, in summary, I truly believe that rapid testing is a good, good thing, but I believe that it is a good thing if it is done in the context of a system.
I agree, it should be approved if, systematically, all of the elements that are such gold standard, high standard, that exist today, will be demonstrated tomorrow by a rapid test that could theoretically be a subset of an existing system.
There are a couple of things that concern me, about lowering standards. When you deal with negative test results, I can't even imagine -- in our case, under FDA guidance, we made sure that every single negative test encounter goes along with an educational interaction that says, you may not be negative, even though your test says negative.
Why? Because there is a window period. This is critical, to provide people, as opposed to having them feel like, time to party.
Next, when someone gets a negative reaction, I can tell you that our counselors say that the most traumatic thing that they deal with is an indeterminant result, because people don't know. They are in never never land. They are actually better to get finality around a positive result.
An indeterminant result, it is not good. We went to the IFA method of testing as opposed to western blot for one reason. We wanted to avoid as many indeterminant results as we could. I, as I did 10 years ago, believe that confirmation testing is a critical and needed test.
Lastly, people who are told that they are positive, or may be positive, for HIV, they need immediate, one on one, professional assistance right then and there.
I am not suggesting that they are going to commit suicide if they don't get it, but there is a system that is well proven, that it delivers information very professionally, and it allows people to get guided into care.
The system that is in place today absolutely works and, again, in my opinion, the rapid test as a subset to a system like that, by all means, should be expedited for approval. Thank you, on behalf of Home Access Health, for your attention.
DR. DI MICHELE: Thank you, Mr. Powell, and for sticking to the time limits.
Agenda Item: Remarks by Ernest Hopkins.
MR. HOPKINS: My name is Ernest Hopkins, and I am the director of federal affairs out at the San Francisco AIDS Foundation, and I would like to thank the Blood Products Safety Advisory Committee for the opportunity to provide some remarks today.
First, let me state that the San Francisco AIDS foundation generally supports the availability of at home HIV rapid test kits.
We believe the availability of such a product could play a key role in increasing access to HIV testing, and in helping many Americans more rapidly determine their HIV status.
With the CDC estimating that fully 25 to 30 percent of the estimated 1.1 million Americans living with HIV are still not aware that they are infected, clearly, we must continue to develop innovative and creative approaches that will significantly reduce the number of people in our country who are unaware that they are HIV positive.
At home rapid HIV test kits should be a part of a larger effort, however. In particular, home HIV tests could be especially helpful and useful for individuals who are uncomfortable seeking an HIV test at a physician's office of HIV testing clinic, but who desire speedy test results, and prefer the privacy and anonymity of testing at home.
They could also be used by those who feel they do not have the time to spend waiting at a physician's office, or HIV testing clinic, and would far more likely get tested if they could get speedy HIV test results at home.
We recognize the potential misuses of this product, as described by members of the panel and some of the presentations that have been provide earlier today, such as a person somehow coercing or forcing a spouse or sexual partner to take such a test, or an individual attempting to gather a sample from someone else to be tested without his or her consent, and we understand that those concerns pose significant issues that must be addressed in the approval process.
We believe, however, that the benefits of such a product outweigh these risks. Now, having said that, we also believe that the FDA needs to apply rigorous criteria in reviewing such products, to ensure that they are as effective as possible, and do not produce negative health outcomes for those who purchase and use the products.
So, specifically, we recommend that the committee support the following criteria:
Number one, the kits must ensure that those who purchase and use the products are able to easily and correctly interpret the results.
It is critical that the person taking the test is able to easily understand the results, and has ample and easily accessible information that helps him or her understand what the results mean.
The fact that a negative test result could still mean that the person is actually infected because of the window period, for example, must be clearly explained in language that is easily understood.
Similarly, the remote chance that the test results could be inaccurate needs to be explained in clear terms that will be understood by all who use the product.
Two, the companies who sell these products must ensure that the individuals who test positive have immediate access to emotional support and counseling.
For many individuals, testing positive for HIV can create significant emotional distress, and the individuals who find out that they are HIV infected should have 24 hour phone access to trained emotional support counselors, who can assist them with dealing with the array of reactions that they may have.