FOOD AND DRUG ADMINISTRATION
CENTER FOR DRUG EVALUATION AND RESEARCH
THE ANTIVIRAL DRUGS ADVISORY COMMITTEE
Wednesday, December 12, 2001
8120 Wisconsin Avenue
ROY M. GULICK, M.D., M.P.H., Chair
Associate Professor of Medicine
Weill Medical College of Cornell University
Director, Cornell Clinical Trials Unit
525 East 68th Street
New York, New York 10021
TARA P. TURNER, PHARM.D., Executive Secretary
Advisors and Consultants Staff
Food and Drug Administration
5600 Fishers Lane
JANET A. ENGLUND, M.D.
Director, Pediatric HIV Program
Department of Pediatrics, MC 6054
University of Chicago
5841 S. Maryland Avenue, Room C638D
Chicago, Illinois 60637
VICTOR G. DeGRUTTOLA, SC.D.
Professor of Biostatistics
Harvard School of Public Health
677 Huntington Avenue
Building 2, Room 439A
Boston, Massachusetts 02215-6096
PRINCY N. KUMAR, M.D.
Associate Dean of Students
Georgetown School of Medicine
Chief, Division of Infectious Diseases
Department of Medicine
Georgetown University Medical Center
Kober-Cogan Building, Suite 110
3800 Reservoir Road, N.W.
Washington, D.C. 20007
COMMITTEE MEMBERS: (Continued)
WM. CHRISTOPHER MATHEWS, M.D., M.S.P.H.
University of California at San Diego Medical Center
Professor of Clinical Medicine
Mail Code 8681
200 West Arbor Drive
San Diego, California 92103-8681
JONATHAN M. SCHAPIRO, M.D.
Clinical Assistant Professor of Medicine
Center for AIDS Research
S-156 Grant Building
Stanford University School of Medicine
Stanford, Connecticut 94305
SHARILYN K. STANLEY, M.D.
Disease Control and Prevention
Texas Department of Health
1100 West 49th Street
Austin, Texas 78756
BRIAN WONG, M.D.
Chief, Infectious Diseases Section
VA Connecticut Health Care System
950 Campbell Avenue (111-I)
West Haven, Connecticut 06516
LAUREN V. WOOD, M.D.
Senior Clinical Investigator
HIV and AIDS Malignancy Branch
National Cancer Institute
Assistant Professor of Pediatrics
Uniformed Services University of the Health Sciences
Building 10, Room 10S255
10 Center Drive
Bethesda, Maryland 20892-1868
KEITH A. RODVOLD, PHARM.D.
Professor of Pharmacy Practice
Associate Professor of Medicine in Pharmacy
University of Illinois at Chicago
833 South Wood Street, Room 164
Chicago, Illinois 60612-7230
LEONARD B. SEEFF, M.D.
Senior Scientist for Hepatitis C Research
National Institute of Diabetes and Kidney Diseases
National Institutes of Health
Building 31, Room 9A18
Bethesda, Maryland 20892
VOTING PATIENT REPRESENTATIVE:
MICHAEL H. MARCO
Treatment Action Group
105 West 73rd Street
New York, New York 10023
NON-VOTING INDUSTRY REPRESENTATIVE:
EUGENE SUN, M.D.
200 Abbott Park Road
Department 48U/Building AP30-3
Abbott Park, Illinois 60064-6146
JAY H. HOOFHAGLE, M.D.
Director, Division of Digestive Diseases and Nutrition
National Institute of Diabetes and Kidney Diseases
National Institutes of Health
Building 31, Room 9A23
Bethesda, Maryland 20892
FOOD AND DRUG ADMINISTRATION STAFF:
DAVE GREEN, PH.D.
LOUIS MARZELLA, M.D., PH.D.
WILLIAM SCHWIETERMAN, M.D.
JAY SIEGEL, M.D.
KAREN WEISS, M.D.
DR. JANICE K. ALBRECHT
DR. MARIELLE COHARD
DR. PENELOPE J. GILES
DR. KENNETH KOURY
DR. MARK LAUGHLIN
DR. JOHN McHUTCHINSON
MR. BRIAN KLEIN
MR. JULES LEVIN
KATHLEEN SCWHARZ, M.D.
C O N T E N T S
An update on the approval of
BLA 103949/5002, PEG-Intron
(PEG-interferon alfa-2b) powder for injection,
indicated for use alone or in combination with
Rebetol (ribavirin, USP) for the treatment of
chronic hepatitis C in patients with compensated
liver disease who have not been previously treated
with interferon alpha and are at least 18 years of age
- - -
AGENDA ITEM PAGE
CONFLICT OF INTEREST STATEMENT
by Dr. Tara Turner 9
by Dr. Penelope J. Giles 12
Advances in the Treatment of Chronic Hepatitis C
by Dr. Janice K. Albrecht 14
Risk/Benefit Decisions in Treating Patients
with Chronic Hepatitis C
by Dr. John McHutchinson 42
by Dr. Louis Marzella 45
QUESTIONS TO PRESENTERS 65
OPEN PUBLIC HEARING PRESENTATIONS
by Mr. Brian Klein 119
by Mr. Jules Levin 133
by Dr. Kathleen Scwharz 142
COMMITTEE DISCUSSION 154
P R O C E E D I N G S
DR. GULICK: Good morning. I was going to ask people to take their seats, but it looks like everyone already has, which is a good sign to say that I guess we can get started.
I'm Trip Gulick from Cornell, and I'd like to open this meeting of the Antiviral Drugs Advisory Committee.
For those of you who were at the last one, we've taken great pains to look at the plumbing system in this hotel.
DR. GULICK: And everything checks out really well, so I don't think we'll have any worries today.
I'd like to start by having the committee introduce themselves. Please state your name and where you're from, and we'll start with Dr. Sun over at this end.
DR. SUN: Eugene Sun, Abbott Laboratories.
DR. HOOFNAGLE: I'm Jay Hoofnagle from NIDDK at the NIH.
DR. SEEFF: Leonard Seeff from NIDDK and the VA.
DR. RODVOLD: Keith Rodvold, University of Illinois at Chicago.
MR. MARCO: Michael Marco, Treatment Action Group in New York.
DR. DeGRUTTOLA: Victor DeGruttola, Harvard School of Public Health.
DR. SCHAPIRO: Jonathan Schapiro, Stanford University.
DR. TURNER: Tara Turner, Executive Secretary for the committee.
DR. WOOD: Lauren Wood, National Cancer Institute, NIH.
DR. ENGLUND: Janet Englund, University of Chicago.
DR. WONG: Brian Wong, West Haven, Connecticut, VA and Yale University.
DR. KUMAR: Princy Kumar, Georgetown University.
DR. MATHEWS: Chris Mathews, University of California, San Diego.
DR. SCHWIETERMAN: Bill Schwieterman, Center for Biologics, FDA.
DR. GREEN: Dave Green, clinical trials, FDA.
DR. WEISS: Karen Weiss, Center for Biologics, Food and Drug Administration.
DR. SIEGEL: Jay Siegel, Center for Biologics.
DR. GULICK: Thank you, and by teleconference, we have Dr. Stanley.
DR. STANLEY: Hello. Dr. Sharilyn Stanley, Texas Department of Health.
DR. GULICK: Very good. We can hear you well. Can you hear us well? I guess you can.
DR. STANLEY: Pretty good.
DR. GULICK: Thanks.
Tara Turner will now read the conflict of interest statement.
DR. TURNER: The following announcement addresses conflict of interest with regard to this meeting and is made a part of the record to preclude even the appearance of such at this meeting.
Based on the submitted agenda for the meeting and all financial interests reported by the committee participants, it has been determined that all interests in firms regulated by the Center for Drug Evaluation and Research present no potential for an appearance of a conflict of interest at this meeting with the following exceptions.
In accordance with 18 U.S.C. 208(b)(3), full waivers have been granted to Dr. Keith Rodvold and Dr. Jonathan Schapiro which allow their participation concerning Biologics License Application 103949-5002, PEG-Intron, sponsored by Schering-Plough.
A copy of these waiver statements may be obtained by submitting a written request to the agency's Freedom of Information Office, room 12A-30, Parklawn Building.
In accordance with section 505(n)(4) of the FDA Modernization Act, Dr. Princy Kumar would like to disclose that she owns stock in a competing firm. The value of the stock is less than $5,000.
We would also like to disclose for the record that Dr. Jonathan Schapiro and Dr. Victor DeGruttola have interests which do not constitute financial interests within the meaning of 18 U.S.C. 208(a), but which could create the appearance of a conflict. The agency has determined, notwithstanding these interests, that the interests of the Government in their participation outweighs the concern that the integrity of the agency's programs and operations may be questioned.
In addition, Dr. Eugene Sun from Abbott Laboratories is participating in this meeting as an industry representative acting on behalf of regulated industry. As such, he has not been screened for any conflicts of interest.
Further, with respect to FDA's invited guests, Dr. Jay Hoofnagle, an employee of the National Institute of Diabetes and Digestive and Kidney Diseases, has received clearance from his ethics office to participate in today's meeting.
In the event that the discussions involve any other products or firms not already on the agenda for which an FDA participant has a financial interest, the participants are aware of the need to exclude themselves from such involvement, and their exclusion will be noted for the record.
With respect to all other participants, we ask in the interest of fairness that they address any current or previous financial involvement with any firm whose products they may wish to comment upon.
I have one announcement. I would like to note for the record that our consumer representative, Dr. Courtney Fletcher, had to cancel his participation in this meeting at the last minute and there was no time to replace him. We are, however, fortunate to have Michael Marco present as our patient representative to provide that special point of view. Thank you.
DR. GULICK: Thanks very much.
I'd like to call on Dr. Weiss for a few introductory comments.
DR. WEISS: Yes. I just want to welcome the committee members, our invited guests, and the public to this meeting and to let people know that the purpose of this meeting is to review and educate the committee and the public on the data supporting the safety and efficacy of the combination of pegylated interferon plus ribavirin for use in patients with chronic hepatitis C. That license application was approved by the Center for Biologics in August of this past year, and we believed it would be appropriate to update the committee on the database, as well as to inform the committee and the public about the issues that were left outstanding at the time of the approval that have resulted in a number of phase IV or postmarketing commitments, studies that are either planned or ongoing. So, we hope that with this presentation, we will just provide that information to the committee and to the public. I look forward to the presentations and the discussion.
DR. GULICK: Thank you.
So, we'd like to start with the sponsor presentation from Schering Corporation. Dr. Giles?
DR. GILES: Thank you, Dr. Gulick. Allow me to introduce myself. My name is Penny Giles from the Regulatory Affairs Department at Schering-Plough. We at Schering are pleased to be invited here to provide this committee an update on our progress since we were last here in 1998.
Prior to 1998, the only therapies available for the treatment of chronic hepatitis C were alpha interferons. Sustained virological response was low when Intron A was given for a 24-week period. We showed that by extending the duration of treatment to at least 48 weeks, that sustained response rate could be more than doubled.
When we were last here, we showed the data that when Rebetol was combined with Intron A, that same virological response rate of 41 percent could be obtained. This new combination therapy was a big step forward in the treatment of chronic hepatitis C.
Since that time, we have developed a pegylated interferon which has improved the same virologic response over that of the normal interferon, and today we will be talking about the addition of Rebetol to PEG-Intron which results in a sustained response rate of 52 percent.
The analysis that we will be presenting today shows that a 61 percent sustained virologic response rate is attainable if the ribavirin dose is weight-adjusted.
I'd like to introduce our primary speaker today, Dr. Janice Albrecht from Clinical Research, Dr. Garaud from Clinical Research, Dr. Koury from Biostatistics, and Dr. Laughlin from Clinical Pharmacology.
We also have with us today consultants: Dr. John McHutchinson from Scripps Clinic. Dr. McHutchinson will say a few words at the close of our presentation about some of the risk/benefit decisions confronting physicians in choosing to treat hepatitis C patients. Also with us is Dr. L.J. Wei from Harvard, Biostatistics, who has helped us with some of the analyses that we have done on the data that we will be presenting.
With that, I'd like to turn the floor over to Dr. Albrecht.
DR. ALBRECHT: Thank you, Dr. Giles, Mr. Chairman, members of the committee.
The therapeutic goal of the treatment of chronic hepatitis C is the eradication of the virus and a sustained virologic response that will result in the halt of the disease. Until recently Intron A plus Rebetol was the treatment standard for chronic hepatitis C.
Recently we have shown with PEG-Intron plus Rebetol that the efficacy can be increased, and as Dr. Weiss noted, this was licensed, approved in August of this year. The indication is treatment of chronic hepatitis C in treatment-naive adults. The approved dose is PEG-Intron 1.5 once weekly plus Rebetol 800 milligrams per day for 48 weeks.
The committee has been provided as background the key publications on the use of Intron A plus Rebetol, PEG-Intron monotherapy, and PEG-Intron plus Rebetol.
The PEG-Intron molecule is created from a combination of the parent, which is interferon alfa-2b. The molecule is achieved by attaching a 12,000 Dalton polyethylene glycol molecule which is straight chained to the alfa-2b molecule. The result of this attachment of the PEG molecule extends the half-life to approximately 40 hours compared to 3.6 hours with the parent molecule. What we achieve with this extension of the half-life is the ability to dose once weekly.
During the development program with PEG-Intron, we looked at our Intron database for the parent compound which is administered as a single dose 3 million units three times a week. We knew from the literature and from our own database that patient weight had an influence on sustained response when we use a flat dose of Intron A. What we found out when we did logistic regression on our database was that weight of the patient was a significant factor, and one of the questions we ask ourselves is should we weight-base our PEG-Intron.
When we look at our 800-patient database of Intron A monotherapy at 3 million units three times a week, all patients receiving the same dose, what we find are patients that weigh less, those small patients less than 55 kilos, those patients 75 to 55 have really quite a good response with monotherapy, 19 to 25 percent, as opposed to the rather low response rates that we see in those patients who weigh more, in particular more than 75 kilos.
The take-home message from this analysis for us was that one size doesn't fit all, and therefore we made a decision to weight-base our dosing of PEG-Intron. We dose PEG-Intron by microgram per kilogram dosing. Our phase I kinetic studies were done that way, and we then moved on to look at the safety and efficacy of PEG-Intron as monotherapy.
We conducted a large dose definition study, 1,219 patients. We compared three doses of PEG as weight-based dosing -- .5 micrograms per kilogram once weekly; PEG 1.0 micrograms per kilogram once weekly; and PEG 1.5 micrograms per kilogram once weekly -- to, at that time, the standard of care, which was Intron A, 3 million units three times a week. All patients were treated for 48 weeks, and they were followed for an additional 24 weeks when they completed their therapy.
The primary endpoint in the study was sustained loss of HCV RNA 24 weeks following the end of treatment.
I'd now like to show you the data that we saw during the course of the study. The y axis is the percent of patients that were HCV RNA negative using the National Genetics Institute assay, which has a lower limit of detection of 100 copies. For all of the subsequent data that I will show you, we have used the National Genetics Institute assay, and all studies were done by the same lab.
Across the bottom are the treatment weeks. As you can see, throughout the treatment period at the sampling times, 4, 12, 24, 36, and 48, the end of treatment, the three PEG doses, 1.5, 1.0, and .5, were superior to the Intron A 3 million units three times a week. At the end of the treatment period, the highest response rate was in the 1.5 dose.
However, with the alpha interferons as monotherapy, we're very much aware of the problem with relapse. In fact, relapse is about 50 percent, and it was when we added ribavirin to Intron A that we could decrease the relapse.
So, not surprisingly, what we saw was a very high relapse when we took these patients off drug. What was surprising is the 1.5 dose had essentially the same efficacy as the 1.0 dose. We tried to understand why this had happened, and what we really found and the best explanation we have is that in this increased response rate during therapy, there were more patients who were HCV-1 that actually responded. However, when we took these patients off of therapy, we had a higher relapse rate in the 1.5. 66 percent of the patients relapsed in the 1.5 dose as compared to 46 in the 1.0 dose.
The next thing we wanted to do is take a look at this study. As I told you earlier, we did a multivariate analysis on our Intron A monotherapy study, and what we found is that weight was a predictive factor. We did the same thing on this study trying to see if indeed we had a limited weight as the confounder when we use weight-based dosing with PEG.
What we found was that the significant factors associated with sustained virologic response were, not surprisingly, genotype and viral load. These are things that are well-known with the treatment of chronic hepatitis C; not surprising, the absence of cirrhosis or bridging fibrosis, and age has also been reported, younger ages with the patients responding better. What we did see is that body weight was no longer predictive. So, we appeared to have taken care of the confounding factor of weight when we dose on a microgram per kilogram basis.
The next slide shows you the doses that we selected for use in combination with Rebetol. I will comment prior to initiating the trial, I'm going to describe we did combination toxicology in monkeys, we did phase I studies to look at pharmacokinetics and assure ourselves that ribavirin in combination with PEG-Intron was appropriate to go forward. These were small studies and we won't take the time today to discuss them.
We selected our doses to be used with Rebetol at 24 weeks of treatment in the PEG-Intron study. We selected 1.5 micrograms per kilogram because it had the maximum antiviral activity at that time point and also because we saw this heightened activity in the HCV-1 patients. We selected a low dose. We selected the .5 microgram per kilogram dose because it looked very much like Intron A and it seemed to be better tolerated, and we thought perhaps that we would, indeed, have a regimen that was equivalent to Intron or maybe even slightly better.
I'd now like to describe the study on which the license is based. This is a study comparing the standard of care at that time, Intron A 3 million units three times a week with 1,000 to 1,200 milligrams of ribavirin administered daily. The basis for the ribavirin dose was patient weight. Those weighing less than or equal to 75 kilos received 1,000; those weighing more than 75 kilos received 1,200.
The first regimen that we looked at with PEG-Intron plus Rebetol was basically an induction regimen. In this regimen we used the high dose, 1.5 micrograms per kilogram once weekly for 4 weeks. We followed it by the low dose, .5 microgram per kilogram once weekly for 44 weeks. The ribavirin dose administered was the same as is with the Intron A regimen.
The basis for this regimen was the fact that in the literature at the time and in our own databases, we knew that if you used high-dose daily interferon, you would see a very, very rapid decrease in the HCV rate, and what you would see is a very early negative response with high-dose interferon. What we were trying to do is to get the patients to become HCV negative and then maintain them on a lower dose. As you'll see later, this induction strategy is really not effective.
The third arm in the study was PEG-Intron 1.5 micrograms per kilogram administered once weekly with Rebetol at a dose of 800 milligrams. All patients received the same dose of Rebetol.
As we look back, hindsight is 20/20. We selected the 800 milligram dose of Rebetol because there was some concern that there might be an additive toxicity with a higher dose of PEG with that very high dose of ribavirin. And in fact, as I will show you later, I think that concern was overrated at the time and I think we could use the higher dose of Rebetol with the PEG 1.5.
As in the previous study, the primary endpoint for the study, as typical for all chronic hepatitis C studies, is the loss of serum HCV RNA 24 weeks post treatment.
In this slide are the demographics for this study. These demographics are consistent for studies that have been conducted in Europe and the United States. In fact, this 1,530 patient study was conducted in 62 centers in the Americas and Europe. As you can see, they're balanced across the groups. The patients are predominantly male. They're predominantly middle-aged and caucasian.
I would call your attention to the weight range in this study. The mean was 82 kilos, and if you remember our studies from 5 years ago, the weight in the United States particularly has gone up. In fact, the mean back then was about 75 kilos. We have a wide range from around 40 kilos to almost 180 kilos. This becomes important later on.
The disease characteristics in these populations are also well balanced. They are also consistent with what we've seen in previous studies. The majority of the patients are genotype 1, high viral load. So, approximately 70 percent of our patients had HCV-1 with more than 2 million copies per milliliter.
About 30 percent of our patients had evidence of advanced fibrosis. We used the Knodell HAI and this is based on F3 and F4.
I'd now like to turn to the sustained virologic response in this study. That was the primary endpoint in the study. The primary comparison was to be the 1.5 group versus Intron plus Rebetol.
In this study we show the primary endpoint. As you can see, the basis for this was sustained virologic response 24 weeks following the end of treatment. This was calculated using the start date for treatment and going to 24 weeks post treatment. The assessment at 24 weeks post treatment had to be negative and it had to be within the 28-day window, as specified by the start date of treatment.
Using this analysis per protocol, we see that the response rate in the PEG 1.5 versus Rebetol compared to the Intron/Rebetol is 52 versus 46 percent with a p of .03.
In this population of patients, we had a few patients who did not meet the window of 28 days for the follow-up. They were, however, negative. We consider these patients responders. When you do that and look at the data, what you see is a 54 percent response rate in the PEG 1.5/Rebetol versus 47 in the other two treatment groups. For the subsequent analysis that I will show you, we will be using this patient population.
I'd now like to move to the additional analysis that we did. Since we had met the primary criteria in the protocol for efficacy, the 1.5 versus the Intron/Rebetol, we then proceeded to look at factors that might predict response. To do this, we used classic techniques. First, we used univariate analysis by logistic regression, and what we found is not surprising. It's what you see in all hepatitis studies: genotype non-1, lower baseline viral load, lighter weight, bridging fibrosis, cirrhosis, age, and gender to a lesser extent.
To determine whether these variables were independent, we then did multivariate analysis, and what we found is we retained all of the variables with the exception of gender. When you account for patient weight, gender drops out in a multivariate analysis.
However, we saw that we still had baseline weight in this analysis, and we knew from our previous analysis that we had probably removed as a confounding factor the PEG dose because we're basing on a weight-based basis.
So, we then proceeded to do some additional logistic regression, which I will show you, in an attempt to understand what was influencing our response rate with weight still there. We decided that we would look at ribavirin on a milligram per kilogram basis. So, what we did is we looked at the doses the patients received and then expressed them as milligram per kilogram.
This slide shows the regression analysis that we did in the PEG 1.5 group, as it is the superior group to the other three. On the left-hand side, is the percent of patients who had sustained response when we did this analysis. Across the x axis, you see the Rebetol expressed as milligrams per kilogram. The dotted line is the fitted regression line. The circles represent the patient data expressed as moving averages. The size of the circle represents the amount of data in that interval. As you can see, as the dose of ribavirin increases, so does the sustained response rate.
We then wanted to look and see what happened when you put our Intron/Rebetol data on the same kind of analysis, and what you see here is the Intron A 3 million units three times a week with 1,000 to 1,200 milligrams ribavirin. As you can see from the placement of the circles -- this is the patient data -- they are further along the axis on the milligram per kilogram. They received more ribavirin on a weight basis. As you can see here, with 800 milligrams of ribavirin and a heavier patient, they're further down on the axis.
We had anticipated that our patients would weigh about 75 kilos, so we wanted to look and see what an average 75 kilo person would have received as a dose. Basically you can see that most of our patients didn't receive that dose. They received less than 10.6 milligrams per kilogram. In fact, this is about 60 percent of our patients and this represents about 40 percent. The patients in the Intron/Rebetol group, in contrast, received about 13 milligrams per kilogram, as you can see looking up here.
It's difficult to compare data when patients have received different doses of ribavirin on a milligram per kilogram basis, so we wanted to look at the observed response rates, trying to look on a more equal basis the dose of ribavirin that was received by the patients. So, what we elected to do is use the break point of a 75 kilo man, because that's basically what we thought our patients would weigh, and look at these two groups compared for observed response rate and look at the response rate in this group. I will tell you there are very few patients in this group who received less than or equal to 10.6 milligrams per kilo in the Intron A/Rebetol group. In fact, there are 22 out of 511.
In the next few slides, I'd like to show you our categorical analysis adjusting for weight on a milligram per kilogram basis and using that break point of 10.6.
This slide is all genotypes. All of the next slides are set up pretty much the same way. On the y axis, percent sustained response; in the left-hand columns, the all-patient dose, Intron A 3 million units three times a week, 1,000 to 1,200 PEG 1.5, 800 milligrams once daily. This is the 47 and 54 percent you've seen previously.
Then what we did is, controlling for ribavirin use, we used the break point that I described, less than 10.6 milligrams per kilogram, greater than 10.6 milligrams per kilogram. And when you adjust and try to look on an approximately equal basis, what you see is the differential between the Intron A/Rebetol group and the PEG 1.5 group becomes wider, with a 61 percent sustained response rate in that group that received PEG 1.5.
The next slide shows genotype 1. For the all patients, what we see is the Intron A/ribavirin group, 33 percent versus 42 percent for the PEG 1.5/800 milligrams. This is statistically significant at the p .02 level. Again, when we control for the ribavirin dose and look at those patients who received at least 10.6 milligrams per kilogram, we see 34 versus 48 compared to the PEG 1.5 group. These sample sizes are small in these people that received less in the Intron A group, and I tend to think that we should not be looking at them as a comparison.
The next slide shows the response rate by genotype 2/3. We have not included 4, 5, and 6 in these slides because of the small number of patients in our study.
Again, for the all, 79 percent for the Intron A/ribavirin, 82 percent for the PEG 1.5/800. When we controlled for the ribavirin dose, you see a differential, 81 to 88 percent. With these drugs, we are seeing very, very high response rates in the 2/3 patients, and we are very close to properly reaching the maximum when we treat these patients. These are intent-to-treat analyses essentially, so we have not accounted for whether the patient received all their drug.
The next slide is fairly complicated, but it's a set of data that I think people would like to see. It's set up the same way except it's in table format because of the complexity of the data.
On the left-hand side, what we have done is we have controlled simultaneously for genotype and viral load. HCV-1 less than or equal to 2 million/greater than 2 million; HCV-2/3 less than or equal to 2 million, greater than 2 million. Intron A/Rebetol 1,000 to 1,200; PEG-Intron 1.5 microgram per kilogram/Rebetol 800; and then the PEG 1.5 group controlled for ribavirin less than 10.6 and greater than 10.6 milligrams per kilogram.
I'd like to work my way through this slide because I think there are some interesting things to be looked at.
First, in those patients that we considered to be low viral load HCV-1, there's a new finding that we haven't seen with Intron/Rebetol, and that is, when you add PEG 1.5, the response rate approaches that that we usually see with genotypes 2/3. Granted, this is a small subset of the population. It's about 20 percent of the HCV-1 patients. So, about 10 percent of the population overall. But essentially we now have a new group of patients, those patients who are infected with HCV-1, but in whom we have a relatively low viral load, we now have a fairly high response rate compared to the 45 percent in the Intron A/Rebetol group.
There doesn't seem to be much effect of the ribavirin dose here. I think that this probably needs further exploration. We had about 100 patients in each group.
I'd like to now turn to the greater than 2 million group, which are the HCV-1 patients that are most difficult to treat and also the most prevalent. When you look at the Intron A/Rebetol versus the PEG 1.5/800, the response rates are essentially the same. However, when you control for the ribavirin dose, what you do see is you see a differential that appears in the patients who got the higher dose of ribavirin.
Turning to the HCV-2 and 3 patients, those that have low viral load and are lucky enough to be both low viral load and 2/3, have a 91 percent response in the PEG 1.5 dose, as you can see, substantially higher than the Intron A/ribavirin dose. There's a small differential when you control for the ribavirin dose, but it's very small.
When we look at those patients with more than 2 million copies of virus, it's 77 percent versus 76, essentially the same, with an incremental benefit when you control for the ribavirin dose.
There's one more slide I'd like to show you. I keep forgetting that I have this slide. It's a very interesting slide and we put it in because we hadn't really look at this data in this particular way before.
One of the things that we saw when we looked at the overall database in the Intron/1,000 to 1,200 versus the PEG 1.5/800 is that the relapse rate appeared to go up. This was fairly disturbing because we were hoping that, indeed, if the relapse rate would stay the same with an incremental increase in the initial response, we certainly didn't want it to go up.
So, the first question we asked ourselves is how did dose of ribavirin affect this relapse rate? Because this is one of the primary characteristics of ribavirin, that it does affect relapse rate.
When we controlled for the ribavirin dose in the PEG 1.5 group, what we actually see is in those patients that got less than 10.6, we've got a fairly high relapse rate. In contrast, when we look at the 1.5 who got more than 10.6 milligrams per kilogram of ribavirin, we see very comparable relapse rates and even a bit lower than those in the genotype 1 patients than those we've seen with Intron A/ribavirin. This says to us that with the addition ribavirin to PEG, what we're actually seeing is an increase in the initial response rate with a similar relapse rate when we add ribavirin to the compound.
In summary, PEG-Intron 1.5 micrograms per kilogram of Rebetol is significantly more effective than Intron A/Rebetol and PEG 0.5 microgram per kilogram of Rebetol. The approved regimen is 48 weeks of treatment in treatment-naive patients. The regimen is PEG 1.5 micrograms per kilogram once weekly plus Rebetol 800 milligrams per day.
Further analysis of our database suggests that weight-based dosing of ribavirin, in combination with the weight-based dosing that we currently use with PEG, results in an improved sustained virologic response.
The other side of any therapy is safety, and when we looked at our safety database for these two compounds, what we saw was the types of side effects associated with Intron A/Rebetol and PEG-Intron/Rebetol were very similar. They're the same types. We see no new side effects. What we did see was an increased incidence with PEG-Intron/Rebetol. Therefore, we thought it was important to look at the differences and see where these increases were occurring, and that is basically the way we have setup our safety review for you today.
We've done two things. We've looked at the groups of PEG-Intron 1.5/800 versus Intron/Rebetol, and then we have also looked at the safety controlling for the ribavirin dose as we did in the efficacy.
The first slide addresses adverse events, and what we have elected to do is look at those adverse events in which there's a greater than 10 percent difference between the treatment groups. The slide is set up as for the efficacy slides. Across the top, the Intron/Rebetol 1,000 to 1,200 milligrams, the PEG 1.5/800 milligrams, and then PEG 1.5 adjusted for milligram per kilograms of ribavirin, less than or equal to 10.6, greater than 10.6 milligrams per kilogram.
When you categorize adverse events using the classic systems, they come out as body systems. What we have done here is identified the body system and the adverse events under the body system. Application site is essentially injection site reaction. And when we look at the Intron/Rebetol group versus the PEG 1.5/ribavirin group, what we see is an increase in the inflammation, and reaction is really nonspecific. That's the fact that the patient saw something there. What you will see is there's approximately a 1.5-fold increase between the two groups.
Now, interestingly in this study what we saw was a very high incidence of injection site reaction in the Intron A/Rebetol group. We think this is because we used a questionnaire to specifically ask the patient were they seeing anything at the injection site, and what we got was an incidence of about twice what we normally see. However, what we saw in the monotherapy study with PEG versus Intron was about the same increment of about a 50 percent increase between PEG and the Intron A. I will tell you that most of our injection site complaints were mild and there was very little pain associated with any of these, about 2 to 3 percent.
Not unexpectedly, what we call body as a whole, which is basically flu-like side effects, which includes fever, rigors, and weight decrease, we saw more than a 10 percent difference in these three side effects when we went to the higher doses of PEG. This is probably not surprising, given that 1.5 of PEG-interferon is a lot more interferon than you're going to get with 3 million units three times a week.
GI side effects. The same situation. There were more GI side effects with PEG 1.5.
And this was an interesting finding that we're not quite sure what to make of. Alopecia was more frequent in the high dose Rebetol group when we adjusted for weight if we looked at PEG 1.5, greater than 10.6 milligrams per kilogram. The only supposition that we have is this group may have had more women in it, and I can tell you women are much more sensitive to the alopecia than the men. So, that may be the reason for that higher incidence. We'll have to find out in future studies whether indeed this is true.
And on the bottom of this we have listed -- and you have it in your handout -- the incidence of any side effect that occurred in more than 10 percent of patients in any treatment group.
The outcome of side effects, as far as we're concerned, are primarily discontinuations and modifications. These are the important ones that we really want to look at, and I'd now like to focus on these for a few minutes.
The discontinuations across these treatment groups, Intron versus PEG 1.5/800 or PEG 1.5 controlled for the ribavirin dose, are very similar. I would comment, if you remember the Intron A/Rebetol studies, the dose discontinuation rates in those studies were about 20 percent. So, we've actually seen the dose discontinuation rate go down.
However, when you look at dose modifications, there are two things to be looked at. We have about a 34 percent dose modification rate in the Intron A/Rebetol group. If you look at the PEG 1.5/800, it's 42, and when we go over here controlling for ribavirin dose, we see it goes to 49 percent.
We think it's very important to take a look and understand what those modifications that are causing this increased incidence are, and in the next slide what we have done is we have looked at reasons for dose modification for adverse events that occurred with a greater than 2 percent difference between the groups. What we see on the top line I think is really the bottom line of why we're seeing dose modifications with PEG 1.5, particularly when we weight-adjust it for the ribavirin.
Neutropenia occurred in 8 percent of patients in the Intron A/Rebetol group, 18 percent overall in this group, and increased slightly when we adjusted for the ribavirin dose. The increase in neutropenia is probably not surprising. We actually expected it with the PEG 1.5 dose, and this is what we saw.
We also looked at anemia, 13 percent in the Intron A/Rebetol, slightly lower here, but that is really being driven by those patients that got less than 10.6 milligrams of ribavirin. When you look at those patients that got at least the lower limit of the ribavirin dose, in these patients it flattens out.
PEG-Intron monotherapy and alfa-2b interferon monotherapy are associated with drops in platelets. However, when you combine ribavirin with these alpha interferons, you get a reactive thrombocytosis that's actually due to the hemolysis of the ribavirin. We looked at simply the people who had dose modification for platelets, 1 percent here, 2 percent here. Interestingly more patients received it here probably due to the fact they were getting less hemolysis.
As I mentioned, body as a whole is basically flu-like symptoms, malaise, fatigue, and what we see is a slight increase over Intron/Rebetol in the PEG 1.5/800, whether it's weight-adjusted or not for ribavirin. This isn't surprising, again because of the higher dose of PEG.
GI side effects were slightly higher.
And I think importantly, a question that everybody always asks is what's happening with psychiatric events when you use another therapy. What we show here is basically in psychiatric events, which are depression, we see insomnia. Insomnia, in fact, is a big part of the psychiatric events with the alpha interferons. Basically the dose modification rates are flat.
I'd now like to just talk briefly about laboratory abnormalities, focusing on the hematologic adverse events. I'd like to start with neutrophils. I've shown you that we have neutropenia, and I'd like to talk a little bit more about what we actually saw.
This slide is set up as the previous slides with the treatment groups across the top, Intron A, PEG 1.5/800, PEG 1.5 controlled for the ribavirin dose.
In this protocol, patients were required to come into the study with a minimum of 1,500 neutrophils. Almost all patients dropped their neutrophil counts, not surprisingly. We know this happens with alpha interferon.
The protocol required that any patient that had a neutrophil count that dropped below 750 be dose-modified. The dose modification was 1.5 for the Intron A and .75 for the PEG 1.5. Patients who dropped below 500 neutrophils at any time were to be discontinued from the study. They were to have both drugs be discontinued and they were not to be restarted.
What we found was that looking at patients that had less than 750 neutrophils at any time, that there was definitely a difference between the Intron A/Rebetol and the PEG/ribavirin, 18 percent here and a slightly higher increase in the 1.5 where the ribavirin dose was greater than 10.6. As you'll notice from the dose modification slides, these numbers match almost exactly.
There were some patients that had a count of less than 500, 2 percent here, basically 4 percent here, and 7 percent over here. However, when you look at those patients who were discontinued for neutropenia, which they were to be discontinued at 500, these numbers don't match.
The reason for that being is we used a central lab for this particular study, and sometimes neutrophils don't travel well and you'll get back a result that says the white count is X and the neutrophil count is below 500. We allowed investigators in this study to call the patient back in, do a stat WBC, and then make a decision as to whether the patient had to be discontinued. And as you can see, we do have a discrepancy here.
Now, when you have neutrophils that drop below 500, you always have a concern about infection. What we did is we went back in our database and we assured ourselves that no patient whose neutrophil dropped below 500 had a serious or severe infection. We then looked at our serious and severe infections and then determined whether any of those patients had a neutrophil count that dropped below 750. None of the patients did. So, we cannot find a correlation between neutropenia that could be associated with infection and severe or serious infection. We had no patient die from infection during the study, and so we're pretty convinced that the neutropenia we observed here is not associated with severe infection.
I'd like to now take a look at hemoglobin and hemolysis. Patients were required to come into the study with a hemoglobin of a minimum of 13 grams per deciliter in males and 12 grams per deciliter in females. The dose reduction criteria for this study said that if your hemoglobin dropped below 10 grams, that you had to be dose-reduced. The reduction was to 600 milligrams per day of ribavirin. If your hemoglobin dropped below 8.5, you had to be discontinued from the study, and that included stopping both drugs. We didn't allow people to stay on their interferon if their ribavirin had to be discontinued.
And what we found was that 12 percent in the Intron A/Rebetol group, 9 percent in the PEG 1.5/800, but as I showed you previously in the dose modifications, these two become very similar when you adjust for the dose of ribavirin. Dose reduction was an adequate way of managing this toxicity. As you can see, we have very, very few patients who discontinued the study for anemia. In fact, these are the numbers of patients.
I would also comment to you that for both neutropenia and hemoglobin drops, they are very baseline-dependent, and those patients that come in with the minimum values often have to be watched. A patient that comes in with 12 grams of hemoglobin is more likely to drop to 10 than the male that comes in with 16.
In summary, the types of adverse events that we observed with Intron and PEG 1.5/Rebetol are similar, but there is a higher incidence of some side effects in the PEG 1.5/Rebetol group.
Neutropenia less than 750 is more frequent with PEG 1.5/Rebetol than with Intron/Rebetol for both the fixed dose and the ribavirin adjusted dose.
When we look at weight-based dosing with ribavirin greater than 10.6, we see an increased occurrence in anemia and neutropenia with those doses.
Discontinuations across the group were similar. They aren't any different. Dose modifications are certainly more frequent with the PEG 1.5 group and appear to be really related to the alpha interferon component of that combination.
The incidence of side effects that we see with the PEG 1.5 either as the 800 or the adjusted for weight-dosing were adequately controlled with dose modification.
Although I didn't mention it, people are always interested if there are deaths that occur in a clinical trial. We had two deaths in our trial. One was a motor cycle accident in the Intron/Rebetol group that we believe is unrelated to the study. The second one was in the PEG 0.5 group which was a suicide. In our protocol, we prohibited patients that had ever experienced suicidal ideation or suicide attempt from entering the protocol. This patient did not reveal to his physician that he had previously attempted suicide and committed suicide during the course of the study.
As part of the agreement for the approval of PEG-Intron plus Rebetol, we have agreed with the agency to conduct certain post-marketing studies. I'd like to now just briefly describe these.
Schering is supporting a study in approximately 4,000 patients. This study is a PEG 1.5 micrograms per kilogram study once weekly for all patients, combined with either Rebetol 800 milligrams per day as a flat dose versus the weight-based dosing of 800 to 1,400 milligrams per day. So, what we have essentially done is we have tried to achieve a dose of approximately 13 milligrams per kilogram plus or minus 2 milligrams per kilogram in all patients.
Within this study is a commitment to have at least 1,000 patients with favorable prognostic factors to evaluate the effect of duration, 6 versus 12 months.
I would also mention that there is a supporting study going on in Europe with approximately 500 patients. In this particular study, we're evaluating favorable prognostic factor patients, genotype 2/3 and HCV-1/low viral load with treatment for 6 months.
The second study that we've agreed to do is a study in approximately 1,500 patients. This will compare two doses of PEG-Intron, both compare PEG 1.5 versus PEG 1.0 once weekly administered for 48 weeks. The Rebetol dose regimen that we will use in this study will be determined from study number 1, whether it be 800 milligrams as a flat dose or a weight-based dosing regimen.
We have also agreed to evaluate within this study at least 100 African Americans for response. I would also mention to you that in study number 1, because of the large size -- it's probably the largest HCV study that's been done to date -- we expect to have about 400 African Americans.
Ribavirin has a food effect. Therefore, we have also agreed to further look at this food effect, fasted versus low fat versus high fat. Although I didn't mention it, because of our knowledge of the food effect, the clinical trial with PEG-Intron/Rebetol was done with all patients taking their doses with food.
I'd now like to take the opportunity to introduce Dr. John McHutchinson. Dr. McHutchinson is a well-known hepatologist and clinical trialist. He's led a number of large clinical studies. Dr. McHutchinson, in working with me, has been the principal investigator on the Intron/Rebetol study that was reported in the New England Journal in 1998 and most recently has been co-principal investigator on the PEG/Rebetol study that I just reported and that was published in the Lancet in September. Dr. McHutchinson will now speak briefly about the risk/benefit of treating chronic hepatitis C patients. Dr. McHutchinson.
DR. McHUTCHINSON: Thank you. I appreciate the opportunity to speak here, on the one hand, as a consultant but, on the other hand, as a hepatologist trying to care for many patients with hepatitis C.
The decision to treat patients with chronic hepatitis C involves many factors, and it's a complex decision making process. While the natural history of the disease is variable and somewhat controversial, some patients with chronic hepatitis C do develop progressive disease and can thus benefit from successful therapy as judged by viral eradication.
Host factors important in this decision making process include the severity of the disease as established by liver biopsy and comorbid conditions that, of course, might prevent a patient from being safely treated with their current therapies.
Likewise viral factors, particularly genotype, provide a guide to the likelihood of response, and they may influence the decision to treat. For example, since patients with genotype 2 or 3 are more likely to respond to therapy, they may be firstly more willing to undergo therapy, and their practitioners may be more favorably inclined to treat them because the likelihood of response is greater in that group of patients.
Finally, whilst our current therapies are imperfect, the efficacy of therapy, the side effects, and the costs also influence this decision making process.
So, as practitioners treating hepatitis C patients, we must weigh the risks of our current therapy, their likelihood of success in slightly more than half the patients, as you've just heard, the drawbacks and the investment of all concerned against the benefits of a sustained response. The latter include the normalization of ALT values, eradication of serum and liver HCV RNA, improvement in liver inflammation, and health related quality of life which have been shown to be durable.
Whilst we have no definitive evidence from prospective trials that therapy definitely prevents the development of liver cancer or the development of cirrhosis or decreases morbidity or mortality or delays the time to liver transplantation, we believe there is accumulating data to support these longer-term benefits, and we hope that in the future, as more data and more outcomes become available, we can more firmly establish these goals and benefits in due course.
Now, the decision to treat people with hepatitis C also includes two additional factors. First, the majority of patients who are acceptable for therapy have genotype 1 infection and some degree of fibrosis. Both are unfavorable factors in terms of their likelihood of response. Secondly, treatment involves a significant investment in terms of the duration of therapy. The more aggressive regimen of the drugs we're using now and the time and commitment required by the patient, the practitioner, and also the ancillary staff.
So, how can we achieve the greatest treatment benefit whilst diminishing the risks for these patients with hepatitis C who are considering therapy or for whom we are considering therapy?
Initially we should provide the best support and education available, both before and during a course a therapy. Secondly, we should prescribe the most effective and safe doses of PEG-Intron and ribavirin. Understanding that we will almost universally encounter side effects as you've heard, we should monitor these patients closely and dose-reduce when necessary rather than discontinuing therapy to provide the patient with a continued opportunity of responding. And finally, stopping rules allow us to discontinue therapy early in those unlikely to achieve a sustained response.
So, taking the data as presented by Dr. Albrecht and the issues I've touched on today as a practicing hepatologist into account, I believe the risk/benefit ratio for PEG-interferon and ribavirin is acceptable. While we all realize the need for more effective and safer therapies in the future, in the meantime the risks and benefits of our current treatment with PEG-interferon and ribavirin should be addressed as part of the individual doctor/patient relationship in an informed fashion, and the goal should be to provide the patient with hepatitis C the best chance of a response and its potential benefits the first time around.
DR. GULICK: Thanks, Drs. Giles, Albrecht, and McHutchinson.
I'd like to ask the committee if we could hold questions until after the agency presentation, which will be next. Dr. Louis Marzella from the agency will give the next presentation.
DR. MARZELLA: I think we're ready to begin. I apologize for that delay.
Mr. Chairman, distinguished members and guests of the advisory committee, ladies and gentlemen, good morning.
The objectives of the FDA presentations today are twofold. The first objective is to summarize the efficacy and safety data which led to the approval of PEG-interferon and ribavirin for the treatment of adults with chronic hepatitis C. The second objective of our presentation is to discuss the outstanding issues which remained at the time of the approval which led to the FDA request for additional postmarketing studies.
In our presentation, we will focus on what in our view are the main issues, namely the need for further dose optimization of PEG-interferon and ribavirin.
We will begin by reviewing the design of the phase III study and discuss the rationale for the selection of the PEG-interferon and ribavirin doses which were used in the phase III study. For the purpose of dose selection, in-treatment data from PEG-interferon monotherapy trials and small dose-ranging trials of interferon and ribavirin were used. As I will discuss, the dose in-treatment data turned out to be not very predictive.
We will then consider the summary of efficacy, and we will begin by considering the prespecified analysis focusing on the primary efficacy outcome, and we will then discuss the efficacy data in specific patient subsets focusing on weight-adjusted ribavirin dosage.
We will then briefly consider the safety profile of PEG-interferon and ribavirin and compare it to that of interferon/ribavirin. Again, we will review subgroup analysis to look at the effects of weight-adjusted ribavirin dosage on safety.
We will conclude then with a summary of the postmarketing commitments. As you've heard already, Schering committed to carrying out further studies to optimize the dosage of PEG-interferon and ribavirin, to define the optimal duration of treatment in patient subgroups, and of particular interest here are patients that have baseline characteristics which predict good response to treatment. Safety and efficacy characterization in African Americans is necessary because historically this patient population is known to not respond as well as other ethnic groups to treatment. And finally, Schering committed to further characterizing the effect of food on ribavirin absorption.
Let's begin by looking at the design of the phase III combination study. This was a 1,500 patient study. The design was multicenter, randomized, open-label. The active control was interferon and ribavirin in patients who are treated for 48 weeks and followed up for 24 weeks. The primary outcome measure, the loss of HCV RNA detection, was determined at 24 weeks of follow-up.
As you heard before, the three arms in the study were an arm in which a high dose of PEG-interferon was used. This was 1.5 micrograms per kilogram weekly. In this arm, patients received a flat dose of ribavirin, namely 800 milligrams per day.
In the low PEG-interferon arm, patients received 0.5 microgram per kilogram weekly, and ribavirin was given as either 1,000 or 1,200 milligrams per day. In other words, there was a crude dose adjustment for ribavirin, and note that the ribavirin dosage was higher in this arm. Of note is also the fact that in this arm patients received a 1-month induction treatment with high dose PEG-interferon.
Then finally, the standard treatment arm at that point for this particular trial was interferon 3 times 10 to the 6th million units three times weekly and ribavirin, again the "high dose" with crude dose adjustment based on body weight with patients receiving either 1,000 or 1,200 milligrams per day.
The most significant protocol amendment after the study began was a provision that patients take ribavirin with food, and the reason for this amendment was that data from a clinical study became available which indicated that food had a major effect on absorption of ribavirin, increasing the absorption as much as 70 percent in the presence of food compared to the fasting state.
Now, let's focus on the rationale for the selection of the PEG-interferon dosage. Let me clarify one point. The phase III combination study refers to the PEG-interferon/ribavirin study. This needs to be differentiated from the PEG-interferon monotherapy study which was performed earlier and led to the approval of PEG-interferon for monotherapy of chronic hepatitis C.
Now, this slide shows the treatment response at the end of 6 months of treatment, as well as at the end of the 6 months' follow-up at the end of the 1 year of treatment. As you can see, the in-treatment response tended to show a dose response in the range of between 0.5 and 1.5 micrograms per kilogram. This in-treatment data suggested that the high-dose PEG-interferon might be the most efficacious. And on the basis of this, therefore, the sponsor selected the 1.5 microgram per kilogram dose to study in the phase III study.
Now, unfortunately, at the completion of the PEG-interferon monotherapy study, data showed that whereas PEG 1.5 was superior to interferon, it was not actually superior to PEG 1 microgram per kilogram. And in addition, the high PEG-interferon dose showed an increased toxicity. For these reasons, the agency licensed then, because of the demonstrated efficacy, PEG-interferon monotherapy for the treatment of adults with chronic hepatitis C, and the agency recommended a dose of PEG-interferon of 1 microgram per kilogram.
Now, while these data were under review, the combination phase III study was already completed. For this reason then, the PEG 1 microgram per kilogram dose was never studied and no data about safety and efficacy of this dose was available for this review.
Now, let's consider the rationale for the selection of the ribavirin dosage for the phase III combination study. This selection was primarily based on a small dose-ranging study which included about 70 patients, and in this study the following range of PEG-interferon doses were used, between 0.35 and 1.4.
In addition, patients received a range of ribavirin dosages. I will not show the data because the numbers are so few. Suffice it to say that the results suggested that low doses of PEG-interferon tended to work only with higher ribavirin dosages; whereas in the arm where patients received 1.4 micrograms per kilogram, as low as a 600 milligram dose, flat dose, of ribavirin turned out to show evidence of virologic activity. For this reason then, the sponsor chose the 800 milligram flat dose hoping, in so doing, to minimize toxicity due to ribavirin without compromising efficacy. These data that led to the selection of the ribavirin dose were also based on in-treatment responses.
Now, let me focus then on the primary efficacy outcome of the trials, sustained virologic response. This was defined as loss of detection of HCV RNA 24 weeks after the end of treatment. The prespecified efficacy analysis stated that the high-dose PEG-interferon arm was to be compared to the interferon/ribavirin arm.
There were two stratification variables in this study. One was the presence of viral genotype 1 at baseline and patients were dichotomized into either genotype 1 versus non-1, and the other stratification factor was the presence of liver fibrosis. These stratification variables were then also used in a prespecified fashion in the efficacy analysis. The data was adjusted for these factors.
The initial study design foresaw an equivalency comparison and a non-inferiority margin was selected and prespecified, and it was designed to exclude an unacceptable loss of efficacy of the new treatment compared to the old treatment.
As you can see here, the proportion of responders in the PEG 1.5/ribavirin arm was 52 percent compared to 46 percent in the interferon/ribavirin arm. So, the treatment difference was 6 percent. It was modest. The PEG 0.5/ribavirin arm was not superior to the interferon/ribavirin arm.
Now, let me then move on and consider treatment outcomes based on patient subsets. Let me clarify again that the prespecified subgroup analyses were, as I indicated, presence of viral genotype 1 and presence of cirrhosis. Post hoc analyses were based on viral titers, age, gender, and ethnicity, geographic location, and body weight.
Now, let me, in passing, cite the effects of age, gender, and ethnicity on the treatment outcome because I will not dwell on those. There was a correlation between a younger age and a higher treatment response. There was also an effect of gender on treatment response with women having apparently a higher treatment response. Treatment responses also tended to be lower, as it's known, in African Americans.
Now, let's consider the effect of baseline viral genotype on treatment response. In this particular slide, what we have done is looked at patients with genotype 1, subdivided in patients that had high viral titers at baseline or low viral titers. High viral titers is defined as greater than 2 million particles of HCV RNA per ml of serum.
As one can see here, the responses tended to be greater in patients with low viral titers. Looking at a comparison between interferon/ribavirin and PEG 1.5, there appears to be a higher response to treatment in patients who have genotype 1 and low viral titers.
Now, it's somewhat not clear whether or not this indicates that patients with specific prognostic factors are more likely or less likely to have superior responses to PEG-interferon compared to interferon. As you can see, the patients with genotype 1 and high viral titers, who were the patients who have the worst prognostic outlook, really essentially have similar response rates to PEG-interferon/ribavirin and interferon/ribavirin.
Now, let's also look at treatment response in patients with other genotypes, genotypes 2 to 6. Again, this slide subsets these patients based on viral titers at baseline. As one can see here, there's essentially no suggestion of difference between interferon/ribavirin and PEG 1.5/ribavirin based on these prognostic indicators. So, if one looks at the patients with the better prognostic factors, there's no clear indication that PEG-interferon is likely to result in higher response rates. If one looks, as we did in the previous slide, at patients with the worst possible prognostic outcome, again there's no difference. So, it's not clear with what assurance to look at the data that shows that the patients with genotype 1 and low viral titers have apparently superior response rates.
Now, let's consider treatment response by geographic location. For the purpose of this analysis, the patients are divided in patients seen in U.S. centers and non-U.S. center. Non-U.S. centers were primarily centers in Europe, as well as a few centers in Canada and Argentina. The general considerations are that patients in U.S. centers tended to have lower response rates than patients in non-U.S. centers. The reasons for this might be related to the fact that prognostic factors were less favorable in patients in the U.S., things such as incidence of viral genotype 1 tended to be higher, incidence of high viral titers tended to be high, and interestingly also, body weight, which was a factor in the treatment response, was also considerably higher in patients in the U.S.
Another comparison of interest is that in comparing across treatment arms in the U.S., the overall difference in treatment response indicating a superiority for PEG 1.5/ribavirin seems to be supported.
Now, let me then turn to another major issue which is the issue of performing efficacy and safety data based on weight-adjusted ribavirin dosage. The sponsor in the presentation I think has done a very good and balanced job of presenting the data based on regression analysis, as well as analysis based on categorical cuts of the data. However, we feel that caution is called for in interpreting the results of these analyses.
The first point is that these analyses are post hoc and were not prespecified.
The next point to consider is that the true variable being considered in these analyses is really body weight, and that the hypothesis being considered is that body weight is in fact a surrogate for ribavirin dosage.
Now, there were substantial differences in dosages across study arms, and we think that this is a fundamental problem which makes it very difficult, if not impossible, to compare treatment response across arms. This slide essentially compares the PEG or interferon dosage across treatment arms, and it shows that the PEG-interferon dosage was weight-adjusted in the PEG-interferon arms but was not weight-adjusted in the interferon arm. Ribavirin dosage was lower in the PEG 1.5 arm, and there was no weight adjustment for the dosage. The ribavirin dosage, however, was higher in the PEG 0.5 and interferon arms, and a crude adjustment based on weight was performed.
So, for this reason then, these analyses are essentially based on nonrandomized subgroups that differ, as I will show you in a minute, very substantially in terms of numbers body weight, and may well differ in other unknown factors. However, as I will show later, within-arm comparison is suggestive. It indicates that weight is certainly a factor predictive of response, but the data are too few and inconsistent. Again, the basic point is that across-arm comparisons are not appropriate in our view.
These are the data showing treatment response by ribavirin dosage. In this particular slide, the ribavirin dosage is divided in dose quartiles. The numerator shows the numbers of patients responding in that particular subset, and the denominator is the overall number of patients.
I think that the point that I want to emphasize is that there is a rather large difference in the number of patients within each subset. There tended to be very many more patients in these two arms in this subset, in this higher exposure subset. Essentially a lot of the data for the 1.5 came really from patients who received relatively lower doses of ribavirin.
So, the point to emphasize is that there's extremely limited data upon which to really base an analysis of safety and efficacy of higher doses of ribavirin in the licensed PEG 1.5 interferon group. As you can see then, the median dosage was quite different between the groups, as well as there was a large range of differences within arms.
So, let's then look at, in fact, adjusting for weight for ribavirin, what the treatment outcome was. We did a number of analyses, and the analyses of subgroups were not really consistent, particularly looking at patients in the U.S. The reasons can be multiple. There are differences in prognostic factors, differences in body weight. There is apparently no evidence of interactions, for instance, in the low PEG 0.5 arm. So, there are trends basically but the trends are very difficult to interpret.
If one looks, however, at dichotomized groups based on body weight, we feel that the important point to emphasize is that whether you look at the low body weight patients or higher body weight patients, that the relative difference between arms is essentially the same.
Now, I also need to briefly mention at least the population PK and PD study which was done as part of the pivotal trial. Serum samples were obtained and analyzed for ribavirin dosage, and modeling was done to look at clearance of the drug, as well as analyzed virologic response and safety. For the purposes of safety, anemia was the only parameter looked at.
There were some shortcomings to these analyses. There are some remaining issues which are still being discussed between the agency and the sponsor. The main issues are that basically the simulation of safety and efficacy did not follow the proposed dosing that the sponsor proposed, ribavirin dosing. There was little or no data for patients at the higher exposures, and then there were issues related to whether ideal body weight should be used in the analysis, as well as pooling the data and not performing the analysis based on separating out subgroups by gender.
Let me move next to a consideration of the safety data, and I will be very brief here and focus primarily on the comparison between the high dose PEG-interferon and ribavirin arm and interferon and ribavirin group.
I think the basic message here is that PEG/ribavirin compared to interferon/ribavirin is associated with a higher incidence of toxicities. I think it's particularly noteworthy to look at the number of dose modifications. In the clinical trial, there were very strict entry rules which excluded patients who would have a high likelihood of having adverse reactions. There were also very specific dose-modification rules which governed dose reductions, as well dose discontinuations for patients who experienced toxicities. And of course, monitoring was very intensive as appropriate in an efficacy trial.
Particularly as related to the issue of the unproven hypothesis that higher ribavirin doses might increase response rates, concerns we would have would be that these increased dose modifications might be associated in actual medical usage with increased toxicity because it's common experience that follow-up and dose modification might not be as tight outside of clinical trials.
There were a number of serious and severe adverse events which were greater in the PEG/ribavirin arm compared to the interferon/ribavirin. A particular concern here is the suggestion of a synergistic effect between PEG-interferon and ribavirin on bone marrow toxicity. So, of particular interest are the issues related to neutropenias and of infections.
Then looking at overall adverse events of all severities, as the sponsor has previously suggested, there was a tendency for specific adverse events to have a higher incidence of occurrence.
Now, let me then focus on the subset analysis based on ribavirin weight-adjusted dose and focus again on these issues, dose reduction and adverse events. I would like to throw out for your consideration the possibility that there might be actually a suggestion of a threshold where the increase in ribavirin toxicity may be steeper, but these are just suggestions.
Focusing then on serious infectious adverse events, these are listed by the classic clinical trial mode of -- I'm blocking now the classification. I think the point that is important to make is that there's a suggestion here of a dose response in terms of the incidence of serious infectious adverse events.
Now, as the sponsor discussed in their presentation, there's no evidence in the trial of an association between serious infectious adverse events and actual decrease in neutrophils, but obviously this is not cause for reassurance. In looking at the overall safety database, we have seen this as a concern, this incidence of serious infections, including lethality, in interferon products.
Now, let's look at the issue of dose modification in patient subsets defined by body weight, and let's focus on the PEG 1.5/ribavirin arm. These are the body weight categories. Again, this is the variable that we're looking at. This variable translates in these ranges of ribavirin dosages. As you can see here, there's a tendency for dose reductions to increase.
I should also caution you that numbers are progressively fewer as we go towards lower body weight. So, one has to take it with a grain of salt the actual incidences in these groups. But again, there's a tendency for not only classic ribavirin toxicity such as anemia to increase in patients with lower body weight compared to patients with higher body weights, but also things like neutropenia show a tendency to increase.
I'm sorry. I misspoke. This actually doesn't look at incidence of adverse events. This looks at modification of dose.
Now, this slide actually then compares the actual incidence of anemia and neutropenia. Again, the same trends can be seen here. Patients with lower body weight appear to have a higher incidence of anemia compared to patients with higher body weight. In the severe to life-threatening category, there also seems to be a general trend. So, this reflects then potentially increases in toxicity due to at least anemia and neutropenia with potentially higher exposure to ribavirin.
To conclude then, the review of the data showed that PEG-interferon and ribavirin, the 1.5 microgram per kilogram plus 800 milligram ribavirin, dose is more effective than interferon plus ribavirin for inducing sustained HCV response.
I didn't discuss the data but most responders, about 95 percent, to PEG-interferon/ribavirin do so by week 12.
Sustained response rates are higher in genotypes 2 and 3 and lower with genotype 1. Patients with genotype 1 and high viral loads have the poorest response of all.
As I indicated earlier, in our view there's no clear indication that particular subsets of patients based on prognostic factors are more or less likely to have higher responses with PEG-interferon/ribavirin compared to interferon/ribavirin.
PEG-interferon plus ribavirin is associated with a higher number of adverse events, for example, infections, neutropenia, and injection site reactions, compared to interferon/ribavirin.
Few safety or efficacy data exist for African Americans, a group with a high incidence of HCV hepatitis and a group known to have a poor response to interferon.
Compared to patients with higher body weights, patients with lower body weights tended to have higher response rates and higher rates of toxicity. However, as I indicated, a number of factors could account for this apparent effect of body weight on treatment response and toxicity and due to the study design, analysis of these subsets are particularly troublesome. So, for this reason definite conclusions about the safety and efficacy of PEG-interferon plus ribavirin as a weight-based regimen cannot be drawn based on these data.
This then leads me to conclude by then describing the postmarketing studies that are designed to assess the safety and efficacy of PEG-interferon plus ribavirin as a weight-based regimen and the safety and efficacy of shorter durations of PEG-interferon plus ribavirin in patients with high likelihood of response.
Now, for the members of the advisory committee, in your briefing package -- as well as for the public, posted on the web -- are the studies that the agency and the sponsor agreed to, but the actual design of the studies is in evolution for reasons of increasing the efficiency of the studies. For instance, these two aims were combined in a large trial, which is a multicenter, randomized, open-label trial in approximately 4,000 treatment-naive patients with chronic hepatitis C.
In this particular study, there are two main arms. One arm will receive a fixed dose of ribavirin. The other arm will receive weight-adjusted ribavirin. Within each study arm, then the other variable to be looked at is the duration of treatment. So, in arm A, patients will receive 1.5 micrograms per kilogram of PEG-interferon plus ribavirin 800 milligrams as a flat dose for either 24 weeks or 48 weeks. In arm B, the patients will receive 1.5 micrograms per kilogram of PEG-interferon and ribavirin roughly dose-adjusted to provide around 13 milligrams per kilogram daily, again for either 24 or 48 weeks.
And the two lines below show essentially the groups. Patients with less than 65 kilograms would receive 800 milligrams. Patients in this weight range would receive 1,000; this dose range, 1,200; and then patients with body weight greater than 105 would receive 1,400 milligrams of ribavirin per day.
Then in the next study, two objectives were pooled together, and this particular study then will look further at the issue of dose optimization of PEG-interferon by comparing the 1.5 micrograms per kilogram dose to the 1 microgram per kilogram dose. In this study, approximately 1,000 patients with chronic hepatitis C of genotype 1 will be studied.
As the sponsor indicated, the dose regimen of ribavirin will be determined from in-treatment data in the ongoing ribavirin dose optimization study.
Then the final objective of the study is also to assess safety and efficacy of the treatment in African Americans.
Now, this slide compares the response to treatment at the end of 6 months of treatment, as well as at the end of 6 months of follow-up in the three treatment arms. This is data from the phase III study. Essentially what this data indicates is that the in-treatment responses at week 24 tend to be predictive of the sustained viral responses essentially in all the dose groups, indicating that, for instance, the PEG 1.5 arm is the highest apparent response rate. Again, this leads us to use the in-treatment data from the ongoing study to decide which ribavirin dose regimen to use in the PEG-interferon optimization study.
This is my final slide. Finally, the final commitment is to look at the relative bioavailability of ribavirin compared to the fasted state after a high fat meal and non-fat meal.
DR. GULICK: Thanks, Dr. Marzella.
We now have an opportunity for the committee members to pose questions either to the sponsor or to the agency. Dr. Mathews will start us off.
DR. MATHEWS: I have a couple of questions for the sponsor. The first one relates to any analyses that you've done that showed -- I suppose more of an on- treatment analysis -- whether there was a decrement in response rates in patients who had to be dose-reduced for ribavirin and/or interferon during treatment.
And the second question relates to constitutional symptomatic toxicity. I believe what you showed us was the proportion who had flu-like illness, myalgias, and so on, but did you do any analyses on number of treatment days that were symptomatic or severity of symptoms when you compare the dosing with the longer-acting preparation to the three times a week?
DR. ALBRECHT: Let me address the first question that you had with regard to the effect of dose reduction. One of the analyses that we've done, and in fact has been submitted for publication, is an analysis that looks at the ability of the patient to take his medication. We call this the 80-80-80 analysis. Essentially what it says is the patient was able to receive at least 80 percent of his prescribed drug without dose modification for 80 percent of the duration. What we see is that in those patients, the response rates are higher.
It's very difficult to look back and look at the effect of the two different drugs individually and what dose reduction to the patient does.
I think it's important to mention that with the 1.5 dose, we're reducing those patients to .75, and if you looked at the response rate with the PEG .5, it's very similar to the Intron A/Rebetol. So, you're basically sitting with a patient on a PEG dose that is probably still effective. What we do see is when we reduce the ribavirin dose to 600, we do see a decrement in response rate.
You asked if we looked at number of days of symptomatology with regard to side effects with the longer-acting preparation versus the shorter-acting preparation. No, actually we didn't look at the data that way.
DR. MATHEWS: Maybe Dr. McHutchinson could comment based on your experience. If you treat them once a week is the duration of the symptoms throughout the dosing interval or just in the beginning?
DR. McHUTCHINSON: My observations might not be representative of the whole study, of course, but with that caveat, I think the most information that's come in respect to whether the long-acting, once-a-week interferon is more difficult for the patients in terms of symptoms is to look at people who've been involved in this trial and particularly in other previous treatment trials where they've already been exposed to three-times-a-week interferon and now they've been subsequently treated with once-a-week interferon.
To summarize, I think there are two groups of patients. There's a group of patients who prefer the once-a-week interferon. It sort of smooths out the side effect profile over the week. They're not getting that hectic fever after the night following the three-times-a-week injection. And there's a group of patients that seem to feel worse on the PEG-interferon.
So, I mean, I cannot say it's 50/50. I specifically ask them about this because it's of interest to us. Many of the patients feel it's smoother with the PEG-interferon, and some of them don't. They don't fare as well. That would be my answer.
DR. GULICK: Dr. Schapiro?
DR. SCHAPIRO: Along those lines --
DR. McHUTCHINSON: I'm sorry. And whether it's dose-related in terms of low dose/high dose PEG, I think it is obviously dose-related. The side effects are less in the lower doses of PEG-interferon than they are with the high doses. That's my clinical observation as you asked.
DR. SCHAPIRO: Along those lines, were formal quality of life assessments done? Is there data to actually look at the quality of life of the different regimens?
DR. ALBRECHT: As published, for the other studies, we used the same quality of life instrument, the SF-36, in these patients. Surprisingly, when these patients are on therapy, their quality of life gets worse, and it happens whether you use Intron plus ribavirin, whether you use Intron alone, whether you use PEG. So, there is a decrement in their quality of life.
The only thing that we can show is in a subset analysis which people basically do not agree with, and that is, if you look at the baseline quality of life in the patients who become sustained responders and then look at their quality of life after they've been off the drug for 6 months, these patients in general are doing better. But that's probably related to their sustained response. So, I think quality of life with these kinds of drugs are very difficult because, obviously, there's a very big decrement in quality of life during treatment.
DR. SCHAPIRO: So, you didn't detect a difference between the arms.
DR. ALBRECHT: We did not detect a difference between the arms in this study.
DR. SCHAPIRO: The other question was on autoimmune side effects. I don't think we saw the data. Most of these side effects seemed to be reversible when drug was stopped. Sometimes with autoimmune disorders, that can be an issue. I didn't see any of the data on the autoimmune disorders, and to what degree they were in the different arms.
DR. ALBRECHT: As you see with alpha interferons, we did have patients who developed thyroid dysfunction during the course of the study. Some of these patients were treated successfully and stayed on their drug, having their thyroid dysfunction treated. In general, when you look at the database, when these patients come off drug, they return to baseline.
Other autoimmune disorders we didn't see -- Dr. Cohard? No. We didn't see. We looked in the database very clearly. We don't see any difference in other autoimmune disorders.
I will say, however, the protocol clearly excludes those patients that have autoimmune disorders from coming into the trial. In fact, it's excluded in our label for both of the alpha interferons.
DR. SCHAPIRO: So, you didn't see irreversible autoimmune disorders.
DR. ALBRECHT: That's correct. Dr. Cohard, yes? That's correct.
DR. MARZELLA: To be more precise, even after the treatment is discontinued, some of these events do continue, for instance, the thyroiditis. Others, for instance, colitis, resolved promptly upon discontinuation of treatment. So, it's a mixed picture. But discontinuation of treatment does not necessarily lead to resolution of the autoimmune phenomenon at least during the observation period which is 6 months following the end of treatment.
DR. GULICK: Dr. Kumar.
DR. KUMAR: I would like to specifically ask regarding new psychiatric issues that occurred during the follow-up phase of the study. Page 21 of the briefing material that was given to us said that some patients experienced ongoing or new serious adverse events during the 6-month follow-up period. 13 patients experienced life-threatening psychiatric events, including suicidal ideation or attempt.
My questions to you are you specifically excluded patients that physicians knew had underlying psychiatric illness before entry. So, were there any other risk factors that predisposed these patients? And why do you think that even after the treatment was completed during the 6 months of follow-up, that they were at increased risk?
DR. GULICK: Just to clarify, that's page 21 in the FDA briefing document.
DR. MARZELLA: Just to comment, I think that there was a progressive increase in incidence of psychiatric adverse events. It did decrease following the discontinuation of treatment, but it was still higher compared to baseline.
DR. KUMAR: No. My questions are again for clarification in my own mind. After you stopped the treatment, did you still see psychiatric events, and if so, how do we explain that? And were there risk factors that predicted who would have new suicidal ideations once you stopped treatment?
DR. SIEGEL: I would note, over the years for a variety of interferon products, we have consistently observed reports of new suicides or suicide attempts or ideation occurring in the several months following treatment. It's not easy to know the association or the reason, but I think we would expect that it's a real phenomenon, that it is treatment related based on the numbers we've seen and the screening going into the trial.
Going back to my medical school days -- and this is highly speculative, but it's been suggested that there's often a higher risk of suicidal attempts as people are coming out of depression as they have more wherewithal to consider actually doing things than in depression. Again, that would be speculative. And I think any other response as to why we see that in the months following treatment would also be equally speculative.
We are inclined to believe it's real. It's not specific for this product. It's not particularly higher post treatment than during treatment, but it appears to be higher than one would expect to be occurring had there not been treatment.
DR. KUMAR: May I ask a follow-up question?
DR. GULICK: Sure.
DR. KUMAR: Was this higher in patients that received the higher doses of interferon? Was there a dose relation?
DR. MARZELLA: No.
DR. KUMAR: Thank you.
DR. GULICK: Did the sponsor wish to comment further?
DR. ALBRECHT: No.
DR. GULICK: Dr. Wood.
DR. WOOD: My first question is for the study sponsor regarding dose modification. There is, from the FDA data, a lower incidence of neutropenia, but we didn't get the breakout in terms of whether or not there was less dose modification for the dose intensification arm where they got 1.5 for 4 weeks and then 0.5 thereafter. So, that's the first question, whether or not there was less dose modification for any reason, either neutropenia or anemia.
The second --
DR. GULICK: Do you want to tackle one at a time?
DR. WOOD: Okay. We'll tackle one at a time.
DR. ALBRECHT: Dr. McHutchinson can also address this in that we worked on the data together quite extensively, if you'd like to.
In the group to which you refer, there was an induction arm, the 1.5 for 4 weeks. Dose modification in that arm was very, very high. Patients didn't like that high dose, it seemed, in that arm. So, you have to separate out the dose modification, the first 4 weeks versus the next 44 weeks. If you look at the first 4 weeks, it's equivalent to the 1.5 arm that received 48 weeks of the 1.5. So, the dose modifications look similar. If you look at the next 44 weeks, the incidence of dose modification looks similar to that seen with Intron/Rebetol. In fact, .5 and Intron/Rebetol looked very, very much alike. So, when you think about that particular arm, you have to think about the first 4 weeks and what happened with dose modification versus the next 44 weeks, and the next 44 weeks looks similar to the Intron/Rebetol.
We haven't recommended using that dose, and we didn't bring the exact numbers with us.
VOICE: We have it by drug.
DR. ALBRECHT: We have it by drug? Okay. Can you show that?
No. That's not the one I want. That doesn't help us, no. Thanks. You can take that off.
DR. WOOD: The next question I have is for Dr. Marzella of the FDA. During your presentation, you made two comments. There was one slide that said that at 12 weeks it was predictive of whether or not individuals were going to have a sustained virologic response, and then you also alluded to 24 weeks. So, my next question is given the data that we've heard, clearly the quality of life for patients is compromised during therapy. Should there be a recommendation that if patients have not responded by either 12 or 24 weeks, that therapy should be discontinued since they are unlikely to derive any further benefit from it?
DR. MARZELLA: Yes, I think so. I would agree with that. I think that the label does make that recommendation. I think in actual practice there might be even more leeway and perhaps even shortening that. But I think in the label we have a cutoff of about 6 months.
DR. SCHWIETERMAN: It might be worth mentioning that with this particular product, the number of people who responded but had not reduced HCV viral load by 12 weeks was higher. In other words, if you were going to respond, you didn't always do it by 12 weeks with this product versus the other products, interferon, for example, with ribavirin or interferon monotherapy. You had close to 95 percent of the patients by 12 weeks responding. Here it was closer. I think the number was 91 or something like that. So, 10 percent of the patients didn't do so until the latter half of the first 6 months.
DR. GULICK: Thank you.
DR. WEISS: Our label does indicate that patients be discontinued from therapy if the viral loads remain high. The label does indicate or suggest that if patients still have a high viral load after 24 weeks of therapy, that consideration should be given to discontinuation.
DR. GULICK: Thanks for that clarification.
Mr. Marco. I'm sorry. The sponsor.
DR. McHUTCHINSON: It seems an appropriate time to show this data in relationship to this comment and which I mentioned also in terms of early stopping rules to discontinue therapy in those unlikely to respond.
This data shows the three treatment groups in this large 1,500 study in the ribavirin weight-based dosing group. It's a complex slide, but from the database, looking at week 12, or 3 months of treatment, HCV RNA, the prediction or the ability of week 12 HCV RNA to predict sustained response. You can see that patients who have not had a 2-log decrease, irrespective of the treatment arm, have very little chance of achieving a sustained response; whereas, patients who have alternative -- looking at it the flip side, the other way, patients who have a large more than 2-log reduction in therapy within the first 3 months, but they remain PCR positive because they've started off with very high viral load, have some chance of a sustained response, about 20 percent overall. And those who are PCR negative after 3 months of therapy have a much better chance of achieving a sustained response overall. So, I think this is important data in addition to the week 24 data.
DR. GULICK: A follow-up question?
DR. KUMAR: Yes, a follow-up question. Do people with genotype 1 take longer than 12 weeks to respond?
DR. McHUTCHINSON: Yes. There are a small group of individuals -- I don't recall the percentage off the top of my head -- in this trial who are what we call late responders to therapy. They lose HCV RNA between week 12 or week 24, even some, a very few, after week 24 of therapy. They are usually the genotype 1 infected patients.
DR. GULICK: Mr. Marco and then Dr. Hoofnagle.
MR. MARCO: I guess my first questions are really for the agency, and everybody from Dr. Siegel, Dr. Schwieterman, or Dr. Weiss could answer this. But I sort of would like to almost know the ground rules and what questions are we allowed to ask.
For example, are we here just to truly discuss the weight-base dosing of ribavirin? Because as we see, the genie is out of the bottle and the combination has been approved. So, are we allowed to discuss questions about sort of the dose of the PEG either 1.0 or 1.5 and what really should be used?
DR. SIEGEL: Absolutely. Let me just make clear that we propose questions to an advisory committee. It is at the discretion of the chair who gets recognized and what can and should be discussed. We pose questions in those areas where we're seeking input, and the area you mentioned is one of them. We've talked about postmarketing commitments both on interferon and on PEG-interferon dose.
We certainly appreciate input that you may feel is valuable to us in other areas that we might not have recognized the need for input or there are other areas outside the regulatory realm which can be discussed about these products that are less useful to us but many areas that you feel that we can contribute to, as far as we're concerned, we welcome.
MR. MARCO: No. I just asked that because we're even seeing here during this presentation, which was excellent, by Dr. Marzella, that there is some question about whether the PEG combo is actually more effective than standard interferon/ribavirin combination in certain patients.
But I guess my really only question is for Dr. Marzella. Even though you say in your conclusion number 5 that there's not enough safety or efficacy data for PEG/ribavirin for weight-based dosing -- it just can't be drawn yet -- do you think that the postmarketing study that was designed with the agency, if it's done correctly, there's proper follow-up, could answer that question?
DR. MARZELLA: I think so. I think that the sponsor has really taken the commitment to heart to try to optimize dose. I think that while the study is run by an investigator, that the sponsor has committed enough resources to ensure that not only the efficacy data, but also the quality of the safety data will be such that we will be able to make a risk/benefit assessment. So, there is adequate power in the study and it's large enough to also assess the safety concerns.
DR. GULICK: Let me just remind the committee at this point this is a good opportunity to continual informational questions, but let's try to avoid the tendency to get into the discussion part, which we're going to do after the open public hearing.
Dr. Hoofnagle and Dr. Wong.
DR. HOOFNAGLE: Well, I think you've nicely documented the epidemic of obesity that's struck the United States in the last 10 years.
DR. HOOFNAGLE: And yet, the very strange thing about these data is how much improved therapy is. Even standard interferon/ribavirin therapy has improved considerably from your previous studies reported by McHutchinson and Poynard. It is really quite striking, and I wondered if you had an explanation. That was one question.
But it goes to the central issue. The one group that seems to have increased benefit from PEG/ribavirin over standard ribavirin are patients with genotype 1 and low viral load, which is a very big surprise. But it's in that group where there is a very major increase in response rate. For instance, in the Poynard study, that group had a response rate of 36 percent. In the current study, with standard interferon, that was 45 percent. I would say that that's significant improvement just with time. And with the PEG-interferon, that's 73 percent. Whereas, in all the other subgroups, and I know it's post hoc analyses, but these are analyses that have been somewhat routine since the original publications by Poynard and McHutchinson.
How do you explain that and how does this fit into your concept that the ribavirin dose was inadequate in this group? Does it seem to be adequate in the group of patients with low viral load? Is that the issue? Do you have an explanation for this major change?
DR. ALBRECHT: I agree with all of your observations, Dr. Hoofnagle. I had looked at the same thing and was very surprised. I have to tell you when we first analyzed the data and saw this 47 percent response rate in the Intron/Rebetol group, I was surprised because I was planning on about 42.
I think there's a factor that's occurring. We had a period of about -- what -- three years in between the Intron/Rebetol studies and this study, and I think what we have is a greater comfort level with this particular drug. As I mentioned to you, the dose discontinuation rates in those first studies were 20 percent. If you look at this study, we're looking at someplace between 13 and 14 percent. I think that in the very first study, we were extremely worried about what we would see with the combination of Intron/Rebetol and we were quick to discontinue if we thought there was any problem.
So, I think what we have is a more experienced set of investigators doing these trials. They were routinely using Intron and Rebetol for the treatment of their patients not in studies, and I think they were more aggressive about dose reducing than stopping the drug. So, I think that's why we're seeing these increased efficacy rates in some of these populations.
I think the HCV-1/low viral load is very interesting, and I agree with you. I cannot comment as to why we see that. Although if you do look at the analysis controlling for the weight, there doesn't seem to be a big impact of the dose of ribavirin, and I can't explain that. I think the best thing we're going to see is in a 4,000 patient study that we will see whether that indeed is confirmed.
Dr. Koury, did you want to comment from a statistical point of view?
DR. KOURY: Yes. I don't have a backup slide for this, but as you might have imagined, when we investigated the attempt to control for ribavirin dose using weight, we did lots of analyses to show that that trend was similar across many different patient subgroups and subtypes. One of the ways we helped to assess that is in terms of calculating some odds ratios which estimate the effect of a ribavirin dose when it's expressed this way. The odds ratios are a little difficult to interpret, so I have to give you a couple of them to try to put it in perspective. But I think what was shown in the categorical cuts was a little bit of an artifact of that particular cut as opposed to using all the data and running the regression analyses to estimate these effects.
For example, when we use all the patients, our odds ratio for the ribavirin effect is 1.09, and now that is expressed relative to a 1 unit increase on a milligram per kilogram basis. But, for example, when we specifically look at the genotype 1/low viral load, that odds ratio actually increases to 1.2. So, we don't really have any evidence that the effect of ribavirin is less in the genotype 1/low viral loads. It may have been a bit of an artifact of that particular cut of the data.
In fact, when we look beyond the 1.5 group but look in similar analyses which try to estimate the effect of this ribavirin dose even in the other treatment groups, which of course was partially controlled by weight, as Dr. Marzella indicated, the odds ratio is exactly the same. So, we have no evidence of really a different trend with the ribavirin dose in the other interferon groups.
So, we think that there is still evidence of a ribavirin effect no matter how we look at the data, and that's one of our bottom line conclusions. There is a tendency for association with ribavirin dose when expressed as milligram per kilogram, but we have to be cautious about looking at some of these subgroups, and we think that's going to be the benefit of the postmarketing study which will help us get enough patients in the various subgroups to get better estimates of what the actual effect is when you look at it by important prognostic factors.
DR. HOOFNAGLE: You know the dosing group of .5 microgram of PEG actually did weight-base both drugs, maybe not as much as you wanted, but it did weight-base the drugs. And was there not an effect of obesity or weight on response rate in that group?
What I'm trying to say is that weight is an independent predictor of a response to antiviral therapy in this disease. It's true of many diseases, including hepatitis B, and we don't know the reason for it. So, that would, it would seem to me, be the group that would best show this effect.
DR. KOURY: Right. And in fact, this is actually the figure that's shown in the Lancet paper because it is the comparison of the regressions of the 1.5 to the .5 dose. So, the effect of the Intron weight adjustment is accounted for. And you can see when the ribavirin dose is expressed as milligrams per kilogram, which is reciprocal of weight for the 1.5 dose but is a little bit less clear for the other group, you can see the same general trends. You see a clear dose response with both components of the combination. The 1.5 is clearly better because the line is above the .5. So, there's the clear dose response for the PEG formulation, and we're also seeing a ribavirin dose response that's suggested by this analysis, again with the caveat that it's not based on a randomized treatment assignment.
DR. HOOFNAGLE: And also with the caveat that this Rebetol milligrams per kilogram is just a surrogate for weight. That line can be drawn just with weight. You don't need --
DR. KOURY: Well, it won't fit as well. It's not quite a surrogate. Even the 1.5 group is reciprocal of weight, and statistically it fits a little better than simple weight.
DR. HOOFNAGLE: But the ribavirin dose was fixed, so it's not an independent variable.
DR. KOURY: Right. So, it's a reciprocal of weight. Reciprocal of weight and ribavirin dose expressed this way are statistically identical. But that's not true for the .5 group.
DR. SIEGEL: I'd like to address this concern because it's been one that we paid a lot of attention to, this concern raised by Dr. Hoofnagle.
It's true in the PEG/ribavirin arm that weight is essentially a perfect surrogate for ribavirin dose, and the 10.6 cut that you saw the data from is essentially a -- 75 kilograms divided by 800 is 10.6. And that's why we present the data 75 kilograms versus less.
There are major artifacts that occur when you compare, as you've seen in several of the Schering slides, the patients who are above or below 10.6 in different arms. Those arise from the fact that in the other two arms, there is both higher ribavirin, 1,000 or 1,200, and some level of weight adjustment. So, if I could call your attention to the top four slides on page 4 of the Schering handout, you can see in the first slide, which has a vertical line and it's at the 10.6 level. So, if you're looking at the heaviest patients, which are the ones to the left of that line, because again it's an inverse level -- they're getting the lowest of Rebetol per body weight. You're talking about two-thirds of the patients on the PEG/ribavirin arm, or 320 out of the 500, and a very, very, very few percentage of patients from the other arm. You see just those few dots. It represents about 25 patients total. When you break them down by genotype, you get to their fourth slide, you see a 50 percent response rate in those patients less than 10.6. That represents 3 of 6 patients.
The weight surrogacy is different, therefore, in each of these arms. What you're looking at in the Rebetol arm, when you look at less than 10.6, is people who weighed more than 75 kilograms, or 165 pounds about. In the other arm, you're looking at people who weighed more than 113 kilograms, or about 250 pounds. So, in the second and third graphs, if you look at the middle bar pairs, if you're comparing a 50 percent rate to a 27 percent rate, you're comparing people who weighed more than 75 kilograms on one arm to people who weighed more than 113 kilograms on another arm.
Aside from the artifacts that arise from confounding by weight, in this case you're also confounding by interferon dose because in that arm, interferon was not weight-adjusted either in the Intron arm. So, lower effects in patients who got less Rebetol could be because they weigh more, as you pointed out. It could be because they got less interferon per body weight, or it could be because they got less ribavirin per body weight.
But also, aside from those multiple contributions, you're talking about nonrandomized comparisons. You're talking about one subset of the patients in one arm to the another subset of patients on the other arm. That does some oddly artifactual things, and the top right slide on page 4 can point out those artifacts. The net treatment effect between the two arms on this slide is 54 versus 47. It's a 7 percent effect. That's because of the modified response rate which includes people who are outside the predefined window. It would be 6 percent, as you know, if you look at the defined.
In the overall population, though, you're looking at a 7 percent treatment effect. This is the next slide in your presentation, if you want to be able to project it.
But then when you subset that into these two groups, which are not comparable subsets, you have a 7 percent effect here, and then you divide each of these groups into two groups, but you divide them differently because of the dosing regimen. You see in one group it's a 23 percent difference, and in the other group it's a 14 percent difference. Well, those 23 and 14 percent differences are apples and oranges. They're comparing different groups of patients. You can't have any subset of a total of 7 percent that gives you two subsets, one at 23 and the other at 14, and when you put them together, you get a 7 except when you're throwing in different cut points.
So, those are some of our concerns with this analysis, the potential for weight interaction, the potential for interferon interaction. I think our overall gestalt, as explained by Dr. Marzella, is that there is some significantly suggestive evidence that a dose adjustment of ribavirin may get better response rates, and there is some suggestive evidence that it may get toxicity. But there's neither conclusive evidence on either of those points nor is there evidence as to whether the extra response rates, if they do occur, outweigh the extra toxicity. And that's where we see it.
I want to just come back and underline the point you said. Among the potential explanations, in addition to interferon, weight, and artifactual problems, it's this issue just of weight. You didn't see the data on all arms done by weight, but in all arms, heavy people do not as well as light people on the approved Intron regimen, on all arms in this trial and other trials as well.
DR. GULICK: This is a follow-up comment, I take it?
DR. KOURY: Yes, I think so.
DR. GULICK: Okay.
DR. KOURY: And I think it's fair to say that we agree that there has to be a lot of care in interpreting these kinds of analyses because of the potential for residual confounding variables to influence these results.
So, what we did, in an attempt to try to understand whether other variables could explain these apparent treatment effects, we did this in a series of analyses in which case we compared PEG 1.5 to Intron while controlling for the ribavirin dose on a milligram per kilogram and then examining, one at a time, the effects of additional covariates such as genotype, baseline viral load, gender, and age, and the residual effect of weight.
What we can see in these cases are that the important prognostic factors, genotype and baseline viral load, remain significant. However, the treatment effects remain significant once those variables are accounted for in these regressions. Similarly, things like gender is not significant anymore and age is, although if you control for it, you still get significant treatment effects.
And then we just systematically stepped through this one at a time and did it with the PEG 1.5 versus .5, getting similar results, and stepping through now combining the two control groups saying there's a little more data if we combine the .5 and the Intron -- they are very similar in their responses. We get very similar results.
If we step through to the next slide, what we did was say if we now simultaneously control for the key prognostic factors such as genotype and baseline viral load -- in the first case, genotype is in the model, and we looked at the others one at a time. In the second case, genotype and viral load are in the base model, and we looked at the residual effects then of gender and weight and so forth. And we're getting consistent conclusions, that these other known variables don't really explain away the treatment differences, and yet they do seem to account for whatever the apparent gender and weight effects were in the univariate data.
So, again, you got to be careful, but we think we've looked at some of the usual suspects here and they don't really explain what we see in the data.
DR. GULICK: Let me stop for a second. Drs. Wong, DeGruttola, Seeff, and Mr. Marco have all raised their hands. Do any of you want to directly address this point before we move on to other points? Dr. DeGruttola first.
DR. DeGRUTTOLA: I just had a question here. I think that Dr. Siegel did raise the important question of residual confounders. You basically have noncomparable subgroups, and you've shown that you've tried to look for some of these residual confounders. I'm wondering if you just did any subgroup analyses, similar to what the FDA did, just based on weight rather than the dose issue so that you do have comparable subgroups in those analyses and don't have to worry about all the residual confounders that you may or may not be able to identify.
DR. KOURY: Well, there would still be some potential confounding by weight because we didn't randomize that way either, and our impression of the data was that the ribavirin, when it's expressed as milligram per kilogram, explained the apparent effect of weight and gender. But we didn't actually do the same series of analyses using weight as the covariate.
DR. DeGRUTTOLA: I'd have to say that I think that if you just use weight as the covariate, then you have comparable subgroups. So, in the FDA's analyses, they're easier to interpret than when you have noncomparable subgroups and a lot of factors being entered in, the main one I think Dr. Siegel mentioning that you have very heavy people in one group compared to only moderately heavy people like me in another group, and they may not be comparable, and understanding how to do appropriate adjustments for that situation I think can be complex. So, again, the question was just about if there are analyses just done by weight, which I think would be of interest to see.
DR. SIEGEL: Just as a quick comment, since we did a lot of those analyses. Certainly for the overall treatment effect and in many of the critical effects, when you subset by weight, the treatment differences between the interferon/ribavirin and between PEG-Intron/ribavirin tends to remain in the 5 to 7 percent rate in both heavy patients and light patients. You see both rates higher in the lighter patients and lower in the heavy patients, but the rates tend to run in that range.
DR. KOURY: But that doesn't totally control the ribavirin dose because in the Intron group, the heavier weight patients got an even higher dose of ribavirin.
DR. SIEGEL: You're absolutely right.
DR. KOURY: So, the trouble is you can't do it both, and that's the fundamental dilemma. So, we agree that things have to be looked at cautiously and we think this is pointing us into a direction. We agree that the postmarketing studies will provide substantial additional information to help further characterize what's going to happen in the important subgroups.
DR. GULICK: Mr. Marco, did you have a follow-up?
MR. MARCO: No. It was basically I wanted a number because it looks like approximately 70 percent of the patients in the study had genotype 1 and approximately 30 percent had less than 2 million copies of virus. So, how many are we talking about that were genotype 1/low viral load out of the approximately 1,500?
DR. ALBRECHT: I think I mentioned this when I talked about it. I said when you're talking about genotype/1 low viral load, you're talking about probably about 10 percent of the total population. It's about 18 percent of the HCV-1 patients. I'm referring to the 1.5 in the Intron A groups. There's about 90 patients in each group.
DR. GULICK: I have Dr. Wong, Dr. Seeff. Then we're going to need to move on.
DR. WONG: I'm going to return to a request for information.
DR. GULICK: Thank you.
DR. WONG: I don't really have a good handle on the safety profile of ribavirin with respect to dose. The data you showed for a general summary of adverse events was just one table that summarized the events in which there was a greater than 10 percent difference between groups. Could you just give us a little more information about just what is the safety profile for ribavirin at various dose levels, not just in this study but in the previous studies you've done also, so we have a better idea, as we go up in dose, what sorts of things would be expected to be seen? And then where does the combined exposure to interferon plus ribavirin fit into this?
DR. ALBRECHT: Can we have master backup number 18, which are the most frequent adverse events?
I think there are a number of questions I think here that I'll try to answer one at a time, and if I'm not answering them, please help me with what you asked.
I think one of the questions that we need to answer is that the defining toxicity for ribavirin is hemolysis. And there are really two components here. As the dose of ribavirin increases, the amount of hemolysis increases. We've seen that in our studies where we looked at lower doses of ribavirin. So, you do see a proportional increase in the hemolysis.
The other part of the hemolysis question is that it's very much a concept of baseline hemoglobin. If you start with a low baseline hemoglobin, you're more likely to go down below 10 grams per deciliter or to the 8.5. What we see is there's a very high incidence for need for dose reduction in those patients with low baseline. Women at 12 grams are particularly susceptible. You get a man in at 16 or 18 grams, they usually go through the study with no dose modification. So, hemolysis is the defining toxicity for ribavirin.
As I mentioned earlier, neutropenia from a laboratory side effect for the alpha interferons is certainly defining. Almost everybody has a drop in their white count. Almost everybody has neutropenia, and again it is baseline dependent. If you come in with a lower baseline, you are more likely to reach the levels I described for dose reduction.
Now, if you look at the slide that I showed you. I had showed you a slide on the dose reduction and the dose discontinuation. There does not seem to be an interaction between the dose of PEG and ribavirin for hemolysis. There doesn't seem to be any increase. There is a slight increase in the need for dose modification when you look at what we used as greater than 10.6 milligrams per kilogram of ribavirin with the PEG 1.5. There seems to be a small increase in the need for dose modification for neutropenia. So, we're going to find out in our 4,000 patient study whether an interaction is really there, and we'll be looking at that very carefully.
There's an abstract been submitted to DDW that at the moment has got several thousand patients in it, and we don't seem to see an increase in neutropenia or hemolysis with the high dose of PEG.
Now, subjective side effects. With alpha interferons, one of the things that we worry the most about, as I mentioned, are psychiatric side effects. In the group of psychiatric side effects, or the way they're classified, there are a number of things. Depression. It occurs in about 30 percent of patients. We did look at our database based on ribavirin dose, and we can't see that there's an exacerbation of depression with the addition of ribavirin.
We do see insomnia. Going back to the studies that we did in 1998 when we received the license for Intron A/ribavirin, I think Dr. Schwieterman will remember better than anyone else, we see an increase in insomnia when we add ribavirin. This is consistent with it being a nucleoside analogue. So, in that psychiatric category, we see an increase in insomnia when we add ribavirin. It's no different when we use it with PEG. We see the same thing.
I think this doesn't exactly address your question, but it does help you see again where the differences between the drugs are. What we did here is we looked at most frequent adverse events. That's any adverse event that occurred in more than 10 percent of patients.
In these large trials, the way you collect adverse events is the following. You ask the patient at every visit, have they had any problems since the last visit and try to get them to tell you how they feel and what they've experienced. That is then assessed by the physician and graded as to severity. The other thing that we do at each visit, which probably prompts side effects, is we say to them, well, last time you had X. How are you doing? Is it any worse? Is it gone? Whatever, and we actually record this data. So, what we see are treatment emergent side effects that we report. If you had an appendectomy while you're in the study, that's a treatment emergent side effect. So, we have lots of side effects reported.
What this shows are side effects that occurred in more than 10 percent of patients and for which we could see a difference based on the ribavirin dose, when we adjusted for dose over here, and between the two major treatment groups, as studied. As I indicated before, flu-like symptoms or body as a whole, as we call them, asthenia, fever, rigors, arthralgia, myalgia. I will mention that asthenia is a uniquely European term basically for the flu-like symptoms and those kinds of things. You can see the increase in the PEG doses.
I would mention to you here there seems to be a higher incidence of asthenia with the increased ribavirin dose. We'll have to find out if that's true in the new study.
GI side effects are clearly related to the PEG dose. We don't seem to see any effect really when we use a higher ribavirin dose.
DR. WONG: Could I just interrupt for a second? I guess what you're showing us again is the subgroups based on dosage versus body weight in this study, but you must have done some preliminary dose-finding studies just with ribavirin before you ever got to this study. I guess that's really what I was hoping to see. What's the safety profile? Can you just give a brief summary of the safety profile of ribavirin as a drug per se without interferon to begin with? Then put that into the context of the studies with interferon.
DR. ALBRECHT: I guess I could ask maybe Dr. Hoofnagle to give you a profile of ribavirin as monotherapy. As you probably know, ribavirin as monotherapy does not have activity in the treatment of chronic hepatitis C.
DR. WONG: No, I understand, but it has a safety profile.
DR. ALBRECHT: Right. But there were some studies done early, one of them that Dr. Hoofnagle did, where they looked at ribavirin monotherapy. So, Jay, would you mind kind of summarizing that?
DR. HOOFNAGLE: The studies that we did were quite small really. You do see hemolysis even at 600 milligrams. It's usually quite mild. We were never brave enough to go above 1,200 milligrams because the hemolysis becomes considerable.
But in the early studies, everybody received 1,200 milligrams of ribavirin with interferon, and it was Schering that introduced this idea of the two levels of 1,000 versus 1,200, which is reasonable because if you use the higher dose, you'll have to start reducing the dose in these small weight patients.
But I wondered whether Schering had data on just pharmacokinetics. Instead of doing a big study, 4,000 patients with all these dosings, can't you do a small study with pharmacokinetics showing what would be the appropriate dose in different sized people? Wouldn't that be so much easier?
DR. LAUGHLIN: We actually can address that two different ways. If I could have slide number 20 first. Perhaps to get directly to your question, this may be the best way to get to it.
DR. WONG: I like that question. I've been on this committee for quite a few years now and I can say that this is the first time I've ever received a briefing book that didn't give a general introduction on the drug, its pharmacology, its experience in preclinical and early clinical trials, et cetera before getting into a big 1,000 subject comparative treatment trial in which my ability to try to dissect out what's the safety and efficacy of the individual components of a combination therapy -- I couldn't do it.
DR. LAUGHLIN: Let me try and address it at least with the specific toxicity of anemia. As we said, because there's not an indication, at least in hepatitis or in this population, to look at ribavirin monotherapy, it has to be in the context of co-administered interferon.
If you look in this study, this is a dose-finding study which was conducted with about 40 subjects per group either with placebo, 400, 600, 800, or 1,000 to 1,200 milligrams of ribavirin. Again, if you just focus on antiviral effect, the antiviral effect measured at 12 weeks, in terms of what was the magnitude of viral reduction, there's an increase in that efficacy. Not surprisingly, there's also a price to pay in terms of reduction in hemoglobin. This is probably the best data that will tell you what is the isolated ribavirin toxicity, and here it's measured primarily --
DR. WONG: And this is really the parameter that we should focus on. Anemia is the dose-limiting effect for this drug.
DR. LAUGHLIN: Right.
DR. GULICK: Dr. Rodvold, a follow-up comment and then Dr. Seeff.
DR. RODVOLD: I've got a couple questions on the pharmacology as long as you're here. So, it's a good time.
DR. LAUGHLIN: Sure.
DR. RODVOLD: Is this linked to area under the curve or Cmax or anything? Can this be linked to concentration dependency? Does that lead you to concentration controlled trials?
DR. LAUGHLIN: Yes. The anemia is in fact the toxicity, and we've sort of modeled it in various different dose regimens. But the toxicity -- the two factors that control for the level of anemia are baseline hemoglobin and concentration of ribavirin.
Now, can you develop a concentration-based trial? I think not because the variability in that -- there is clearly an association on a population basis. On an individual patient basis, that correlation is weak, and I'm not sure, because the toxicity is recognizable and manageable and reversible, that you would necessarily want to trade potential efficacy just to keep a hemoglobin where you want it.
DR. RODVOLD: But I'm wondering more in the sense of a trial fashion to find that dose versus putting people into all these different dosage regimens you're proposing that you try to get everyone to a set dose hunt, so to speak. So, the trial design would be to find the magical dose of the ribavirin, not necessarily to propose that to physicians in clinical practice.
DR. LAUGHLIN: Well, we have attempted to define that relationship both in terms of efficacy and safety. Again, I think the problem is it's a shallow relationship. It is clearly there for both efficacy and toxicity, perhaps more for toxicity. It's clearly there. But on an individual patient basis, I think the data just isn't strong enough to direct you toward a concentration based clinical trial.
DR. RODVOLD: My final question is, is there any data that you generate from a pharmacology point of view, particularly PK, maybe PD as well, where they absolutely studied people that were, say, less than 100 percent ideal body weight, 100 to 150 percent body weight, and then a group that definitely was big, like 200 percent, an obesity group?
The reason I asked that was that in the clinical trials, you didn't have a lot of people in the extremes, particularly the upper extreme. So, I would guess in this trial you'll run into the same problems. Would a trial that would sort some of that out to really look at weight as the key factor, control all the other variables out, and maybe throw gender in there as well because that keeps tripping occasionally as well. Have you ever done that?
DR. LAUGHLIN: We have not prospectively done that to look at light, middle, and heavy folks. What we have done -- and I'll need the backup slides. If I could have slide number 18.
We did go back and look at sort of the other side of this. We've been talking mostly this morning about weight-based dosing and efficacy. We looked at weight-based dosing on a milligram per kilogram basis and exposure as measured by AUC. Each one represents an individual from one of the earlier clin pharm studies. Again, if you sort of break it where we've been talking about all morning, there is a relationship that exists that appears to be linear, at least within fairly wide ranges.
DR. RODVOLD: My question, though, is actually to ask you to look at the data like as a percentage of ideal body weight on the bottom x. Then go get your PK parameters. We've done that with some of the chemotherapy agents where again you have the problem of intracellular concentrations and all these issues. What you find is that actually there are relationships there that trip out clearance, for example, or half-life but not volume. And that starts to make you rethink about who should be getting the right doses because there is a PK parameter.
I understand this part, and I saw that in your insert, but I'm wondering, does one of the parameters come out -- is clearance different in people that are fatter or smaller? And is that what's driving this dose? So you can search for that. Maybe you've done that.
DR. LAUGHLIN: We can go back, if we have it. That's certainly something we could look at in the database.
DR. GULICK: I'm seeing lots of hands, although we really need to end. A follow-up question from Dr. Englund.
DR. ENGLUND: I have a follow-up pharmacokinetic question, and that is certainly in earlier studies in pediatrics we've studied ribavirin and have noted the presence of ribavirin triphosphate in the red cell fraction, which really is where we believe -- some people believe -- the toxicity derives from with very high intracellular RTP concentrations.
My concern for getting the pharmacokinetics is that we have a large population of, for example, HIV-positive patients that I want to be able to use a drug like yours on, and if we could get levels and get some pharmacokinetics, we may be better able to translate it.
So, I have a question. Do you have plans to study not just ribavirin levels, which are not that difficult to do but perhaps not going to relate to your toxicity as much as your RTP or perhaps other levels, which I'm not so aware of?
DR. LAUGHLIN: If we could have slide number 8 please.
I think you raise very good points. However, I think that we probably need to approach the different toxicities differently.
I think in terms of red cell toxicity, we don't know precisely what the mechanism is for that toxicity, but certainly one of the defining characteristics of that is for nucleated cells, this equilibrative transport system will generate mono-, di-, and triphosphate that is reversible. And it uses the same enzyme systems as adenosine to generate ATP. In the red cell, by contrast there's a massive accumulation of RTP, about a 60 to 1 ratio between intracellular red cell and plasma. And we have looked at that.
If I could have the next slide. Perhaps one of the speculations for the ribavirin-induced anemia is in fact this competition for ATP generation where you basically generate a nonfunctional energy store within red cells that can't be used, and that energy store is critical for the oxidation reduction of proteins, which when they get overexpressed on the red cell membrane, that's the signal to come out, for extrahepatic hemolysis.
DR. RODVOLD: But I think you've got a great opportunity here.
DR. ENGLUND: I'm just saying I don't see any data.
DR. RODVOLD: It would be a shame if you passed this up on 4,000 patients that you're giving multiple different levels of this drug. This is the opportunity of a lifetime.
DR. LAUGHLIN: To measure what?
DR. RODVOLD: Well, to take a measure of not only whole but also one or all these functions. I mean, you'll never get a database this big again.
DR. GULICK: Dr. Siegel, a follow-up?
DR. SIEGEL: Yes. Well, just an addition to the answer to Dr. Hoofnagle's original question, which opened this field, which is why do a large clinical trial. Can't you get the weight adjustments you need from PK?
I would say PK studies can be very important, can answer a lot of questions, but there are some important parts of the question they're not going to answer. So, you could, from a PK study, learn how to adjust the dosage so that large people have the same level as small people as a population, recognizing individual variations. Once you know that, you could decide, as Schering has proposed, to give small people 800 and move up to 1,400 in large people, or you could give large people 800 and adjust down for smaller people, or you could try to get the average, which this study did, by giving 800 to the average sized person and adjust up and down from there. So, you can adjust to achieve approximately similar exposures, but what the optimal exposure should be can only be done I think through a clinical trial.
So, what the nature of the trial that's underway will tell us is, at least in larger people, will a more intensive regimen actually provide, as hypothesized, better responses in a tolerable range. And you're not going to learn that from PK adjustments.
DR. HOOFNAGLE: But you could from PK adjustments learn whether this is the right scale. For instance, they're proposing to use 85 kilograms, whereas everyone up to here has used 75 kilograms. And actually if you use 75 kilograms and then your endpoints -- you're dealing with very few people who are more than 250 pounds and very few people anymore who are less than 120 or 110. And no one would argue that with the small people that you really should go down. The large people you're a little bit concerned with, and at what cut point should you increase the dose of ribavirin.
DR. SIEGEL: Right.
DR. HOOFNAGLE: 75 or 85 kilograms?
DR. SIEGEL: You're right, and this scale was chosen -- and I believe that's data driven, although I don't have the PK data before me -- to weight-adjust on milligrams per kilogram, so to keep in a range of approximately 13 milligrams per kilogram, regardless of your weight, with some variability because of the discontinuity of the cut points.
Some drugs, if you want to get similar exposure are better adjusted by milligrams per meter squared and some by powers in between a power of 2 as in surface area or a power of 3 as in weight. Those decisions can and often are made on the basis of PK data, and I believe in this case were. I don't know if we have the people who reviewed that here at this meeting.
DR. HOOFNAGLE: Well, ribavirin is distributed with water and comes out in the urine. So, it's not fat mass. It's lean body mass that should be the determinant.
DR. RODVOLD: I think that's an important part. It would help you understand the mechanisms. That's what I'm after so that we can make sense of this. And the question the FDA presented about lean body weight, there's an adjusted lean body weight, there's a total body weight -- you can do some of that but you can't -- at the same time as I was mentioning and followed Dr. Englund's question, I think that in this big controlled trial you have the chance to do PK/PD. I think the sponsor is right. What do you actually measure is the question. But that will also be important there because now you can go linking, hopefully, those two in addition to this raw milligram per kilogram basis.
I get a little worried we're only looking at one finite thing here, and when you're going to put that many people into that much pharmacology potentially, to pull out, man, I think it's a once-in-a-lifetime chance. It will help in peds potentially too.
DR. GULICK: There's clearly enthusiasm around the table to further discuss the design of the postmarketing studies. Let me hold us back and try to finish off the questions here because we are running a bit behind schedule.
Dr. Seeff has been waiting very patiently.
DR. SEEFF: Well, I'd like to make comments as a practicing hepatologist. Perhaps what I'm going to say is not for now and it may be a discussion for later and you'll stop me because this may be really the issue that Mike Marco brought up a little earlier.
In looking at the data that we see in the Michael Manns paper, there's clearly evidence that the drugs that we used were significantly better than the combination of conventional interferon plus ribavirin, and the difference is 6 percent or 7 percent significant. But, of course, that's an overall number.
And the question is how do we apply that to the patients that we actually take care of. When I address the patient and I see the patient, I have to say, you have an X percent likelihood of responding.
If you look at the data, it's quite clear that there are only subgroups in which there is a significant response rate and other subgroups in which there is no difference between the two drugs. For example, two-thirds of the patients are genotype 1 and have high viral load and the response rate is the same between the pegylated interferon and the conventional interferon together with ribavirin. It's only in the low titer group where you actually see a significant difference in which nobody could argue that you should be using the pegylated interferon.
Now, there may be a variety of reasons to choose one drug over the other, but I'm not sure that there's enough information here to say that one is significantly better than another in the high titer group.
I'm intrigued with the difference between the U.S. and non-U.S. people, and I'm more interested in the fact not so much that there's a difference in the U.S. group -- and there is and clearly that's important and probably will work out to be perhaps weight-based -- but that there was no difference in the non-U.S. group, absolutely no difference. 57 versus 59 percent.
And then from my point of view, in my own personal practice, I'm at the VA hospital in Washington, D.C. 85 percent of my patients are African Americans. 98 percent of them are genotype 1, and when we measure their titers, about 75 percent, maybe 80 percent have got high titers above 2 million copies. I have to tell you that we have found that using the conventional interferon and ribavirin, that our response was less than 10 percent.
So, now I am faced with the fact that we have data that's been reported that there's a much higher response rate overall. And obviously, this study can't answer it. I'm delighted to hear that Schering is going to do a study looking at African Americans, and of course, we are doing a study at NIDDK to look at the issue of lack of response.
But my point is I guess to cite a figure as an overall figure that can be applied to patients that we're taking care of is probably not accurate and not correct and is not applicable in our discussion with patients. I cannot go to my African American patients and say, when we start you on this combination, this is what we can anticipate. We have to find that out.
So, I think that we can say that there are subgroups in which there is clear-cut significance with respect to a treatment response and others in which we either don't know and we can't make a recommendation one way or another, or that we have to find out more information in the studies that are being done. So, that's just a comment and that may be one that should have come later.
What I'm also intrigued about is that the criteria used here to exclude patients was very carefully selected at 1,500 I guess for neutrophils and 1,000 for the platelets. Perhaps I shouldn't be saying this in front of the FDA, but I'm hearing that there's a tendency for people to be bringing patients into treatment with lower levels. Now, while I recognize that the more important concern is the anemia, I guess that we don't really know what would happen if we treat patients who have 1,000 instead of 1,500 or 750 instead of 1,000 using this particular combination.
So, I think that we have to be very cautious until the studies have been done to identify that, and I think we need to make sure that whatever recommendations are put in, that it makes clear that we have to stick with the numbers that were selected because we are concerned right now about the potential side effects, and this may become more evident if we start off with lower values.
So, that's just a comment rather than a series of questions.
DR. GULICK: Again, we really need to finish up, and if there are some burning questions -- Dr. Wood has a burning question.
DR. WOOD: And it's kind of a segue and a follow-up to the issue that Dr. Seeff raised regarding neutropenia. This is for the FDA.
While there clearly wasn't a direct correlation in terms of incidence of neutropenia at the time of the serious infectious adverse events, I'm curious as to whether or not the individuals in the PEG 1.5 arm had been dose-reduced for neutropenia. So, even though they did not have neutropenia at the time of their event, they had a history of requiring a dose reduction for neutropenia. Is there any information about that?
DR. MARZELLA: I don't have that information, but I'll have to look at the data. Does Schering have that information?
DR. WOOD: That would be helpful just because there were 17 serious infectious events in the PEG group. We clearly know that the PEG 1.5 group had a higher sustained dose reduction for neutropenia. So, that's the reason for wanting that information.
DR. ALBRECHT: I was just speaking with Dr. Cohard, who is the director of my group, and we looked at this data. We can't exactly answer your question, but what we did is, as I said. None of the patients were below 750 neutrophils during the course of the study. She remembers that 1 or 2 patients may have earlier had a dose modification for neutropenia, but there doesn't seem to a clear association with either dose modification at the time or around the time associated with those infections.
DR. GULICK: Dr. Marzella.
DR. MARZELLA: I just wanted to return to a comment that Dr. Hoofnagle asked about a relative comparison of treatment response of the interferon/ribavirin arm in later studies and also to some comments that the sponsor made about higher adverse event rates in the interferon/ribavirin arm in this study and whether or not those are due to different ascertainment methods for adverse events.
I think it's important to emphasize in this particular study, in contrast to all others, that the ribavirin was given with food, whereas in previous studies it was given without regard to food and that there's data that the sponsor has which indicates that the absorption of ribavirin increases dramatically, as much as 70 percent. So, it's another factor to be considered in that issue.
DR. GULICK: I'm going to take the chair's prerogative and ask the last question. That's strongly worded there.
DR. GULICK: Could you please compare the demographics in your pivotal phase III study with the demographics of hepatitis C infection today in the U.S. and address any differences?
DR. ALBRECHT: These are the demographics that were in the study. Now, as I commented earlier, these demographics are really pretty consistent with what we've seen in the previous study in 1998. The weights are not. The Americans have gotten heavier, and if you'd like, I'll be happy to show you the distribution on the weight in the Americans. We had a large number of patients over 100 kilos.
Maybe, Dr. Hoofnagle, would you be willing to comment -- and we have this demographic slide, as well as disease characteristics -- about what your opinion is and whether this is reflective of really what we see in the United States population?
DR. HOOFNAGLE: Well, the mean age is probably about correct. You did exclude children and the elderly, though. So, it's not completely it.
The gender is the approximate breakdown in gender distribution of hepatitis C in this country.
The race is not. Our estimates are that African Americans represent 22 percent of chronic hepatitis C in the United States. Hispanics you don't break out, but they're also over-represented as having this disease.
As far as weight, I have no information.
The other issues, though, are socioeconomic class. This is a disease that's much more common in the poor. It is, of course, a major risk factor today, as injection drug use, and I'm sure in this study, you probably excluded patients who were using drugs or had used drugs in the previous year. So, that is a major difference, and they represent, of course, the large reservoir and the difficult reservoir.
You also excluded HIV positive patients. This is an important disease in those patients.
So, no, it doesn't represent the demographics of this disease, but it certainly represents the demographics of people enrolling in large clinical trials no matter what the disease is.
What we need to do now is go back and pick up these special groups and do special studies in them. They have special concerns. Like in HIV-positive patients, you have to be very careful with the ribavirin dose I think. With injection drug users, you have to use all kinds of other special monitors of compliance. Actually PEG-interferon will be a major advance for that group I think because of the ability to administer the drug actually. So, this isn't representative, but it's hard to say that they could have done differently.
But I think it's interesting. The patients are getting heavier and yet your response rates are getting better.
DR. HOOFNAGLE: This is the conundrum that occurs in all obesity studies. Americans are becoming more and more obese. Deaths from heart disease and cancer are decreasing. How do you explain this dichotomy? We haven't got an answer.
DR. ALBRECHT: And I know the chairman wants to stop. I have two brief comments. Dr. Hoofnagle touched on something very important.
In trying to have a more homogeneous group in these trials, we do exclude the HIV-positive patients and current injection drug users. Schering is supporting trials in these patient populations. In the HIV group in particular, we're doing a study with the ANRS in France, looking at PEG/ribavirin versus actually Intron/ribavirin. We're also supporting studies in patients in methadone clinics. Indeed, the once-a-week dosing allows the patient to get their drug at the clinic in certain cases as opposed to self-injecting, which is a problem with these patients. So, we are supporting studies in these small populations.
Not everything is done by the Research Institute, but we do have another group within our company where we support these studies, many of which are under the INDs or under the purview of the individual investigator.
DR. GULICK: Thank you.
We're going to take a 10-minute break, 11:25. I recommend if you do eat something, that it be low fat.
DR. GULICK: Could people please take their seats? We're ready to go.
Welcome back. I'd like to open the open public hearing portion of the meeting. There are three people that have signed up and one group that asked that a letter be read. First is Brian Klein from the Hepatitis C Advocacy and Action Coalition. You can either use this mike or you can come to the podium, whichever you prefer.
Dr. Stanley, are you with us?
DR. STANLEY: I am here.
MR. KLEIN: Good morning. My name is Brian Klein and I'm a community patient advocate and a founding member of the Hepatitis C Action and Advocacy Coalition, or HAAC. HAAC is a grassroots, all-volunteer group of individuals committed to non-violent direct action for access to treatment, funding and resources for care, prevention efforts, and sound public policy in addressing the hepatitis C crisis. We accept no money from the pharmaceutical industry and that includes no industry funding for my appearance here today.
I am also a person living with co-infection of HIV and hepatitis C.
I thank the chairperson and the advisory committee for this opportunity to provide testimony to you today. There are written copies of my complete testimony, but I will try to abbreviate it a bit and try to keep it down to about 10 minutes. I know the panel wants to probably move on this morning.
I want to address three concerns related to Schering-Plough's PEG-Intron and PEG-Intron/Rebetol post-approval. These issues are concerns on the weight-based dosing studies of Rebetol in use with PEG-Intron, current off-label promotion of weight-based dosing of Rebetol, and the Access Assurance PEG-Intron Registration Program.
Most of my comments on the concerns for the weight-based dosing studies will concur with much of what the FDA has said this morning. I was very relieved to hear that presentation and most of the comments this morning.
We support the current FDA labeling of PEG-Intron with a fixed 800 milligram per day dosing of Rebetol. That is what the true intent-to-treat analysis of the registration trials demonstrated as effective. The study showed equivalent, not superior virologic or histological efficacy to standard interferon/ribavirin treatment for the vast majority of patients, including genotype 2/3 patients and genotype 1/high viral load patients. These patients do no better on PEG-Intron/Rebetol therapy.
Anecdotally, we have been getting more reports than we received with standard interferon/ribavirin of patients who have to dose-reduce the PEG and/or the ribavirin due to respective adverse events. Also, in regards to the convenience issue, we've been getting reports of dosing errors due to the confusion over the reconstituting of the product since it comes in a powdered form. This is difficult for many patients as well.
Be that as it may, I do question whether this drug was even ready for approval when we are not sure of the dosing and it offers such disappointing results. But as was stated earlier, the genie is already out of the bottle. Therefore, weight-based dosing of Rebetol should definitely not be approved unless or until the prospective study demonstrating statistical significance of the efficacy and safety of such dosing is completed and clearly demonstrated.
I am very concerned that this haste by Schering to have weight-based dosing of Rebetol is an attempt to market higher doses and sales of ribavirin and bring back a bundled product. There are several pertinent points here.
One, we still don't clearly know the mechanism of action for ribavirin. No one is really sure that high intracellular levels are really needed for improved efficacy. It is unclear whether there is a strong dose-dependent efficacy effect with ribavirin, but we certainly know there is a dose-dependent toxicity effect.
I believe the situation is similar to that which we experienced with AZT, a nucleoside analogue, as is ribavirin. Only later it was seen that half the original dose of AZT would be as effective and far safer. As with other nucleoside analogues, we are seeing reports of mitochondrial toxicities with increasing ribavirin dose, particularly in highly sensitive HIV/HCV co-infected patients. There are also reports of decreased bone mineral density after therapy containing ribavirin.
Next, it is well established that increased body weight, as has been talked about a number of times this morning, is an independent negative prognostic indicator for response to any HCV treatment to date. This effect is all that the retrospective analysis in the Manns data demonstrated.
As was also stated earlier, ribavirin is known to be as much as 70 percent more bioavailable when taken with a high fat meal. Mo mention or attempt was made to control that within this particular analysis.
The weight-based dosing conclusions hinge on really the results of 58 genotype 1 patients at that 10.6 milligram per kilogram dose out of over 1,500 patients in the study. I have no idea what statistics would show that to be significant.
When looking at the lower dose of PEG in the study, the ribavirin milligram per kilogram dosing scheme seems to be a moving target, up to 13 to 15 milligrams per kilogram. What I really believe we're seeing here is an effect of the PEG-Intron dosing, not ribavirin. If one were to believe these efficacy numbers that are in table 4 of the Manns study to be valid, it would mean concluding that ribavirin had a larger effect on sustained response rates than the PEG-Intron dose itself.
Whenever patients have had to be dose-reduced on ribavirin for safety reasons, data from Schering and from Roche have demonstrated no significant loss of efficacy at the lower doses. One question that was asked earlier and was answered by Dr. Albrecht regarding the 80-80-80 studies that were done, at AASLD last month, Dr. Forensi did show that dose-reducing of ribavirin did not affect the efficacy. So, I think these kinds of things need to be looked at.
There are studies that compare various doses of ribavirin. Bonkovsky in the October 2001 Digestive Diseases and Sciences concluded that a lower 600 milligram dose was better tolerated and was equally effective in bringing about a sustained virologic response. Schering has conducted at least two major studies that have compared various doses of ribavirin with their interferon products. I have learned from investigators that one multicenter study used Intron A with doses of ribavirin varying from 600 to 1,200 milligrams per day. The results of that study have never been made public.
Another has been published by Paul Glue in Hepatology 2000 and was a phase II study for PEG-Intron/Rebetol combination approval. In that study, doses from 600 to 1,200 milligrams per day were used. While safety data is presented in this study, efficacy of the various doses is omitted. When Dr. Glue presented this data at AASLD in 1999, he said they collapsed the data into one set because there was no statistical difference between the doses of ribavirin. In fact, it was this study that was initially used by Schering to design the phase III study using a fixed 800 milligram ribavirin dose. Now the company doesn't like the results of the trial, it starts to manipulate the data, making retrospective analyses.
I ask this committee to recommend that Schering be required to supply it and the agency with its entire ribavirin dose-ranging efficacy data with Intron A and PEG-Intron to consider while reviewing prospective weight-based dosing data.
It is my belief that while the company is conducting frivolous patent infringement lawsuits to delay a generic version of ribavirin from coming to market, this weight-based dosing marketing attempt is not only an effort to sell more Rebetol to patients, but also may be an excuse to bundle Rebetol with PEG-Intron in a new co-packaging scheme. Once again, this type of marketing strategy will restrict choices by medical providers and patients by tying the sale of one product to the sale of another. Three and a half years after the introduction of Rebetron, the FDA still does not have a comprehensive policy on bundling. And without one, we are going to see this anticompetitive, anti-patient market manipulation perpetrated on the public again in the name of safety for the patient, while safe and effective individualization of treatment is lost. Rebetol changes from being safe or unsafe for separate sale depending on market advantage to the company not on scientific grounds.
I ask the committee to recommend that the FDA develop a policy preventing the marketing of products in a bundled only form on safety and efficacy grounds as it purposely hinders and prevents individualization of treatment for the patient. Dosing and product choice should be up to the physician working with the patient, not the whim and revenue projections of the drug manufacturer.
My next issue is concerning the current off-label promotions of weight-based dosing of Rebetol. Sadly too few health care providers are reading the intent-to-treat analysis in the labeling of this treatment. Schering marketing efforts are currently doing everything to avoid showing that label and are instead working with national pharmacies and physicians to promote off-label weight-based dosing of Rebetol. Some physicians are paid by the company to promote the weight-based dosing as the new gold standard. There are already slide presentation programs from the company promoting this weight-based dosing under the guise of certified medical education, CME. One such presentation is available for free to anyone with Internet access at projectsinknowledge.com.
Schering is working with national pharmacies to promote off-label dosing of Rebetol. These pharmacies include at least Caremark, CuraScript, CVS Procare, Priority, and Walgreen's. These pharmacies carry dosing tables and/or prescription forms with the weight-based formulas already written on them and marked as optimal or other such dosing labels. While the approved labeling for PEG-Intron/Rebetol combination shows data and guidelines for patients only up to 95 kilograms in weight, these dosing tables manage to go up to over 105 kilograms of weight and up to 1.4 grams per day of Rebetol. Only in the tiniest print is the approved dosing marked. These materials are purposely written to mislead the reader into believing weight-based dosing of Rebetol is preferred or optimal.
These pharmacies did not independently cook up this promotion. It is all coming from Schering for mutual benefit. Schering sales reps can go handing out the pharmacy marketing pieces pretending they are from an independent source having nothing to do with Schering.
The well-intended FDA disclaimer that accompanies the Manns article in promotions is not enough.
At the AASLD meeting in November at the Schering exhibit booth, the only copies of labels on display were PEG-Intron monotherapy labeling and separate Rebetol labeling. This was purposeful evasion of having the full actual combination use label and data readily available.
While HAAC supports doctors using their expertise to prescribe off label in order to individualize patient treatment needs, we also support regulations that intend for pharmacies and drug manufacturers to not promote such off-label uses. And we believe the FDA needs to rigorously enforce these policies with Schering and all manufacturers and pharmacies. With all such product promotion activities, the entire pertinent labeling must accompany all marketing materials used by a company regardless of their source. We ask this committee to recommend that the FDA compel Schering to live up to the letter and intent of such marketing guidelines and stop this off-label promotion by them and their pharmacy surrogates.
The last issue I want to address many on the panel may not be familiar with. I hope some of you are. It's the Access Assurance Registration Program from Schering. This is one of the largest concerns of patients because where do patients go to have their privacy rights protected here.
This program, instituted in late October, requires patients to register with the manufacturer. Patients will not receive medication if they do not register.
Patients register by divulging the following confidential information: name, address, phone number, best time to be contacted, gender, weight, date of birth, their doctor's name, referred to on forms as the investigator as if the patient were signing up for a clinical trial, health plan, and anticipated start and completion dates of therapy. Company representatives claim this has been tinkered with in recent weeks and pared down to main identifying information of name or name, address, phone number, and permission to be contacted or not. But whenever I call the number, as recently as yesterday, to check, the same full set of information continues to be required.
Once the company has all the information, only then is the patient given a unique identifier number to give to the pharmacist to use to get the prescription filled, which seems to take anywhere from two to five days. There is no guarantee from the company as to what will be done with this information.
While Schering publicly states that a third party administrator McKesson is handling the information and this constitutes a firewall protection for the information, when patients call the access number, all they are told is that they are speaking to a Schering representative or a PEG-Intron representative, which makes it purposely sound that they are giving the information directly to the company or a representative working for the company. There is no mention or guarantee of any firewall to the patient calling.
The FDA did not request, require, or approve this program for safety or for any other reasons, but when a patient calls the access line, they are told the FDA approved the program. Schering unilaterally initiated it.
Now, Schering built great anticipation for the approval of this combination therapy. This resulted in patients and physicians deferring treatment and creating a backlog of patients until approval. There has been significant misinterpretation and mistrust of the program and significant anxiety that patients won't be able to get drug. The company initiated this plan without appropriate and timely education of any kind. In fact, Schering started it with no warning or consultation with the patient community, pharmacists, payers, health care providers, or the FDA.
If Schering-Plough really could not meet production, launch should have been delayed until such issues were resolved. Standard interferon/ribavirin therapies work as well, so there was no patient crisis. I suspect if there were current competition in the marketplace, this program would instantly vanish. Yet, this program was only instituted after Rebetol was approved and made available to use with PEG-Intron. PEG-Intron itself has been on the market, sold with no such program since February 2001, plenty of time for the company to anticipate any potential increase in demand so many months later.
Schering has never shown any public proof of the need for a registration program. It is outrageous that the burden of proof has to be on everyone else to prove that they do not. Even with a legitimate need, there are easily ways to have set up this program to maintain confidentiality. One simple example would be the use of bar coded or preprinted unique identifier number stickers for the physician to put on a prescription. That only takes one or two seconds. Then all confidential information stays with the doctor and pharmacist, as the intent of federal and state patient health information laws intend. This program was purposefully designed to skirt these laws by making the patient contact the company and its representatives directly.
Schering has managed to compromise with the Veterans Administration and, I understand, at least one prison system, that of the State of Pennsylvania that all identifying information stays with the VA or respectively that prison system. Suddenly when the threat of losing this revenue stream became clear, Schering managed to find a way. If Schering can do it for the VA and this prison system, then the company is perfectly capable of doing it for the rest of us.
A Schering representative has been quoted as saying, "Well, you cannot expect us to make up drug that is just going to sit on the shelf." So, I do not believe that this registration program is to assure patients access to drug in an environment of potential scarcity of the product. Rather, I believe it was designed to assure the company quick and free access to patient confidential information in order to ensure itself that it does not manufacture too much product. The safety concern of drug access is of artificial creation by the company. It is time for patient rights to come before marketing needs.
I ask the committee to recommend the following. That the FDA put in place a more comprehensive oversight on this Schering program and any other such program from other drug manufacturers. FDA approval should be required before initiating any such program. These kinds of programs should only be allowed when proof of clear safety need is demonstrated to the agency by the company and then done in such a way as to preserve that all patient identifying information remain with the primary health care provider as the law intends. Schering and any other drug manufacturer should and can easily comply with such a policy.
In closing, I want to say that the FDA is one of the only lines of protection patients have from these kinds of tactics from drug manufacturers. Schering-Plough has continued to show its disregard for patients and health care providers by engaging in these tactics and refusing to engage in any meaningful dialogue on them, using misleading information wherever it can. Community meetings and communications with this company continue to be the worst in the industry. Schering's community meetings are simply a window dressing exercise conducted in the hopes that patient advocates will simply rubber stamp what the company has already decided upon. Or as in the case of the Access Assurance Program, the company doesn't even bother to consult the community whose privacy rights it intends to violate. Most companies with HIV and hep C products that I work with make some level of attempt to balance patient needs with those of investors and profits and to seek substantive input from the community they wish to use their drugs. We don't always agree but there is some level of meaningful dialogue and real work to compromise. This company continues to not even care to try. In the absence of competition in the marketplace, regulation by this agency is the only way to protect patient rights and needs in drug treatment. I hope this committee will consider the requests I have made.
I thank you for your time and attention today.
DR. GULICK: Thank you.
Our next person to speak at the public hearing is Jules Levin from the National AIDS Treatment and Advocacy Project, NATAP.
MR. LEVIN: Hi, everybody. I'm pleased to be here today, although it was tough for me to get here. Some of you know me and some of you don't. I'm pleased to see some of the press here today even though it's somewhat of a relatively small audience.
I think that it should be clear to everyone in the room today why this hearing took place today, why this public hearing took place today. The PEG-Intron and ribavirin or PEG-Intron alone was approved in August. The community, of which I'm one -- I'm a person that's had HIV and hepatitis C for 18 years, and actually I've been on pegylated interferon and ribavirin for 7 months now, which is why I have this big tummy here. It's not from lipodystrophy. I'm the only person who gains weight on interferon. Everyone else loses weight.
MR. LEVIN: Although I do have lipodystrophy. And these are issues that need to be addressed.
The drug was approved in August and the community, of which I'm a member, of infected people put together a consensus statement to ask the FDA to meet with us to talk about issues and concerns we have regarding treatment and drugs and various issues related to hepatitis C and co-infection. And the FDA responded, and we had a meeting with them during the summer. One of the requests we had was to have a public hearing like this regarding PEG-Intron and ribavirin.
I think that it's clear to me and I hope to everyone in the room that the FDA and one of our supporters -- I'm going to mention his name -- Richard Klein, who works in the FDA, deserves a lot of support, a lot of thanks and credit. The FDA gets credit for having this meeting here today. And I've been a critic of the FDA in the past, and there are times when I'm sure in the future I'll be criticizing them. But I think that the FDA gets credit for this.
And I think that one of the important things that we had here today -- because in my opinion this -- and I'm going to talk policy here for a few minutes. This is a very key meeting here today, to bring public scrutiny to hepatitis C and to co-infection and to new drugs coming to the market and to use for patients who have HCV and, perhaps more importantly, co-infection. Without this hearing, a lot of what you heard here today would not have been brought to public scrutiny. Key HIV treaters and researchers who are on this panel here today would not have had the opportunity perhaps to ever hear some of the issues.
Co-infection and HIV treaters becomes a crucial part of where we go from here, due to the fact that overall 30 percent of everybody with HIV has HCV. Perhaps more importantly is 80 or 90 percent -- or it's estimated 60 to 90 percent of everybody who got HIV through IV drug use also has HCV. That's the population that is mostly what we're talking about here. We're talking about former and current IV drug users. We're talking about African Americans. We're talking about people who are poorly educated and, as one of the persons on the panel mentioned, of low income. That is most of the group we're talking about here, and I suppose we're also talking about, as was brought out here, a high viral load and genotype 1 patients.
And that would not have been emphasized as much publicly in front of all of us if we didn't have this hearing. So, I think that it's really important that we have this hearing here today.
One of the reasons that we requested this hearing is because we're entering into a new era now. We're entering into an era where there's a lot of HCV research going on. A lot of the drug companies are going to be doing a lot of research in hepatitis C and hopefully a lot of new drugs will be coming to market. The major companies are all showing attention. All the major HIV drug companies are showing major attention to hepatitis C research.
And this meeting here today sets a precedent. We need hearings. We need discussion. We need to scrutinize publicly the data and where we need to go from here.
Thanks to a lot of HIV advocates and activists over the years, going back many years, 10 years, 15 years, the FDA has been sensitized to the consumers and the community in having public hearings and to addressing our needs and issues related to HIV. We need to do this now in hepatitis C. HIV advocates and activists are bringing this now to the co-infection era and to HCV.
And we need to continue this so when Pegasis hopefully gets approved, the Roche product, next year we have the same hearing, and when other HCV drugs come before the FDA, we have hearings because look at, in my opinion, the important tone and questions that came out of this and the important information that was discussed here today.
I want to talk just a little bit about co-infection. I'm co-infected. There's a key group of people who are co-infected. There are about 900,000 people in the United States with HIV. A third approximately have HCV. The key figure is perhaps, like I said, 60 to 90 percent of people who got HIV through IV drug use has HCV as well. And that's the population that we're mostly dealing with here. This is going to be a population that may be much more difficult to treat and certainly to educate.
That's what my organization does is treatment education for HIV and for co-infection in affected communities and people who work as professionals in the field. And I can tell you that they are not well informed about the issues about treatment. It is difficult to educate these individuals. There are a lot of issues surrounding this. And co-infected people -- we don't know how they're going to respond to the treatment. There are a lot of variables here.
And this needs to be addressed this afternoon by you guys when you have your private meeting with Schering and by the FDA, and in the future hopefully public hearings that we'll have not just with the FDA but the NIH, to talk about these issues and get them out on the table. How can we treat and handle co-infected patients better who are suffering with HCV than we are up to now? This is going to take a lot of public policy, a lot of public discussion.
Before I forget, let me mention that I'm certainly glad that Dr. Sun is on the panel and someone from industry is on the panel because in my opinion every HIV drug company has a stake in this and needs to be more interested in this discussion because it's their clients, it's their patients that are suffering with HCV and have co-infection. They have a stake in this. The HIV drug companies need to take more interest in this and to do more. And I'm not talking about one company. I'm talking about all of them. In my opinion, the HIV drug companies have neglected this subject and not because I haven't tried and others haven't tried to bring it to their attention.
So, what is needed here? Well, we need more and better studies in co-infected patients. I know that Schering has a study. I know that Roche has a study. In my opinion, one study doesn't answer the questions. As you can see here that was discussed today, there are a lot of questions that are not addressed by the big studies, the one big study that the companies do. We need a lot of studies to look at all the variables because I'm telling you now that the HIV activist community is going to bring this to public attention. And that's what this hearing hopefully starts us doing, to bring this to more public scrutiny and attention. In my opinion, in the HIV community the studies are better, they're more scrutinized, and they certainly deal with all the minor issues a little better.
So, there are some things that we need to look at I think that are very important for us who are co-infected. One was mentioned, the food effect with ribavirin. I think that the product label mentions that a high fat meal increases ribavirin blood levels or intracellular levels significantly. And that's key when we talk about the food effect and timing of dosing for patients who are co-infected who are on all sorts of other drugs. Eating food, can't eat food. I mean, it's a problem for people who are infected with HIV, and now we're adding something else onto the regimen. That needs to be talked about. And this was mentioned before by some other people, I think by Brian or some other people.
The issues of diabetes, the issues of lipodystrophy, the issues of bone mineral density, these are all issues that HIV patients suffer with. This is going to be somewhat increased, the problem, with interferon and ribavirin. We need to sort this out a little bit more in studies and look at this.
There haven't been enough studies. One study of retrovirus showed that patients who had co-infection had more of an incidence of lipodystrophy, body changes. Certainly bone mineral density has become a major issue. It's been an issue ongoing in interferon and ribavirin for years and in hepatitis C, but it needs much more study and to be looked at much more carefully to understand it better. It's never really been done that much before, and hopefully now we can start to look at this stuff.
And what about diabetes? Diabetes is associated with HCV. It appears to be associated with HCV treatment. It certainly is associated with HIV treatment. This needs to be better understood. Are co-infected patients who go on interferon or ribavirin going to have more problems with lipodystrophy and diabetes and bone mineral density? And how can we understand this better and deal with this all?
I think lastly what I want to mention is -- and this is one of the issues that we brought up in our consensus statement in the meeting during the summer with the FDA, which the FDA nicely agreed to have with us. I think it was a key moment and this is a key moment today. The issue of fast track approval for hepatitis C drugs. You know, we've had this in HIV for years, fast track consideration, fast track approval, close attention paid to study design and outcomes and so forth. We need to have this in HCV now. We need to discuss more openly, the FDA needs to consider more carefully where we're going with policy with drug development, how the studies should be designed, what kind of outcomes should we be looking for in these studies, and what about fast track approval for hepatitis C drugs?
I'm certainly not going to recommend policy on all this stuff today, but these are issues that need to be considered very carefully because there are certainly some patients who are going to need fast track access to certain hepatitis C drugs if we make progress in research. And I think we will. Those things really need to be carefully considered by this panel, by all of us in the community, by the FDA, and all interested parties, including the drug companies.
So, just last, I hope that this meeting today will be the beginning of better cooperation between the industry in hepatitis C, the industry in HIV, the public interested parties, the FDA, the HIV researchers. I'm very pleased that we have the ACTG represented here today. The ACTG has a liver disease focus group where they're doing a bunch of hepatitis C studies. It's a good place to do a hepatitis C study. So, I hope that this all will foster better cooperation and collaboration in trying to understand and to do better studies, do better research, respect community issues and concerns, and certainly to pay the most attention we can to co-infection issues.
DR. GULICK: Thank you.
Our next speaker is Dr. Kathleen Schwarz, who is the Chief of Pediatric GI and Nutrition at Johns Hopkins.
DR. SCHWARZ: Mr. Chairman and other members of the committee, I really appreciate the opportunity to provide some pediatric perspectives this morning. My name is Kathy Schwarz. I'm Chief of the Division of Pediatric Gastroenterology and Nutrition at Johns Hopkins.
I'm also a member of an informal international group of pediatric hepatologists who care for large numbers of children with hepatitis C. Both we hepatologists and also the very concerned families who have children with hepatitis C have been anxiously awaiting a statement from the FDA following the pediatric presentation on April 23rd requesting a pediatric mandate for interferon, pegylated interferon, and ribavirin trials.
What I would like to do today is to go over the rationale why we as a group believe that it is inappropriate to extrapolate from adult efficacy and safety data and that we need properly done, randomized, controlled pediatric pegylated interferon and ribavirin trials and we need the mandate from the FDA to do these trials.
My clinical research has been supported by both Schering-Plough and by Roche Pharmaceuticals, both of which manufacture pegylated interferon products.
So, what I'd like to do very briefly is to share with you the magnitude of the problem of hepatitis C infection in children and then to go over these five points, natural history, safety, pharmacokinetics, efficacy, and some public health considerations as to why children with hepatitis C should be regarded differently than adults with hepatitis C.
Hepatitis C is a big problem in pediatrics. Dr. Miriam Alter did a very nice epidemiology study of unselected 6- to 14-year-olds showing that the prevalence rate was .2 to .4 percent. However, more data is needed. There is only one study of prevalence in a high risk group. This was 5 percent of adolescents in an Oregon shelter.
We have a study, which has been funded by NIDA, in which we have the privilege of doing epidemiology of hepatitis C in homeless children in Baltimore. Although our study has been underway only a couple of months, we are very alarmed by our findings. These are very preliminary. About 35 percent of the mothers in the homeless shelters and 7 percent of the children 2 to 18 years of age that we have surveyed are hepatitis C positive. There are about a million homeless children in the United States ages 2 to 18. If our figures are anywhere close to the mark, this may mean that there are 50,000 to 70,000 high risk children with hepatitis C.
Furthermore, we have extrapolated from the prevalence data that Dr. Alter supplied and what we have, the 1990 Census data of children in various age groups, and have estimated that in addition to these high risk children, there are probably about 150,000 children in the U.S. now with hepatitis C.
And then Dr. Eve Roberts from Toronto Sick Children just published an article in Hepatology last month making some projections about the new patients with hepatitis C who will be born each year using world prevalence rates, 35 percent of women in the child-bearing age, annual fertility rate of 2 percent, maternal/fetal transmission rate of about 4 percent. So, we anticipate that somewhere in the neighborhood of 10,000 to 60,000 newborns will be born every year in the world with hepatitis C.
We do not have elegant natural history data in unselected populations of children with hepatitis C comparable to the wonderful studies that Dr. Seeff has done in adults. However, there are a few clues. There is the New England Journal paper by Vogt, et al. in a 20-year follow-up of newborns operated on for cyanotic congenital heart disease and transfused with contaminated blood. There was a 55 percent persistence of hepatitis C RNA.
On the other hand, there is another study by Toyo, et al. in infants who had maternal/fetal acquisition of hepatitis C showing a 90 percent chronicity rate.
In most published series -- and admittedly these are university selected patients, so they would tend to be more severely ill -- those children who have undergone liver biopsy have mild to moderate hepatic fibrosis, and then a few do have cirrhosis, up to 8 percent in a paper from Harvard from Dr. Maureen Jonas.
Dr. Guido, et al. showed that as hepatitis C progresses in a child, the fibrosis scores increase. We are very concerned about this data because we believe that there is little rationale for just waiting until a hepatitis C positive child becomes an adult before treatment is considered.
Then finally, although it is rare to have a child with such aggressive hepatitis C disease that he or she requires transplantation, it does occur. Every university center, including ours, sees children who have aggressive end-stage liver disease from hepatitis C.
In the 1998 SPLIT database, which is a pediatric liver transplant database -- this was 1998 data, and at that time SPLIT only represented a fraction of the transplant centers in the United States. There were 10 children who had either undergone liver transplantation or who were awaiting a transplant for hepatitis C cirrhosis. However, given the long life of a child, the cost burden of just one child to have a liver transplant for a preventable disease or a treatable disease is unacceptably high.
And then finally, there's very interesting data to suggest that in general children have lower viral loads than adults. Now, this is important, as you know, because lower viral loads are one of those factors that predict a good response to therapy.
This is our own data which we have published showing hepatitis C viral load in different pediatric cohorts, and those who acquired hepatitis C by a blood transfusion, those who acquired hepatitis C by a maternal/fetal transfusion had viral loads of below a million per ml. This is good news. The hemophiliacs had very high viral loads, but very little liver injury, suggesting that they are a population that needs to be treated differently.
Then there are safety issues, different in adults and children we believe.
Dr. Deidre Kelly of Birmingham presented data from an uncontrolled study of interferon and ribavirin of 61 children treated with this regime and varying doses of ribavirin. She presented this data in the liver meetings this year in Dallas and did show that during the therapy there was weight loss and there was a reduction in linear growth velocity. Now, after cessation of therapy, these problems resolved, but it does call to mind that these are important pediatric issues. So, we need to know how to give these drugs to children. What about the effects of the treatment regimens on reproductive capacity and on puberty?
One of the important issues that shows how children differ from adults in this arena is that they do appear to respond better just to interferon monotherapy, and if they respond better to interferon monotherapy, they may very well respond better to pegylated interferon therapy. The reason that this is so important is if it's not necessary to give a teratogenic drug to children with chronic hepatitis C, then we shouldn't be doing it, and until we have the support to do the appropriate prospective randomized, controlled trials, we're not going to be able to answer this very important question.
Now, it's also interesting to note the toxicity. Hemolysis, of course, is a known toxicity of ribavirin. In the data supplied by Dr. Kelly in the liver meetings this fall on the interferon/ribavirin therapy, there was actually less hemolysis from ribavirin in children compared to published data in adults. So, this is good news.
Our group has reviewed all of the reports in the English literature of interferon monotherapy of children with hepatitis C that had what we thought were valid virologic endpoints. This summary will be published in the Journal of Pediatric Gastroenterology and Nutrition in the spring and addresses results in about 300 children. It's very interesting in that both the end-of-treatment results and the sustained response results in the group of children were two and a half to threefold better than the comparable published results for adults treated with interferon monotherapy.
Finally, what about the public health considerations? Well, the American Academy of Pediatrics in 1998 recommended the screening of high risk infants, infants born to high risk mothers, for hepatitis C. So, by the mandate of the American Academy of Pediatrics we are going to be identifying a large number of children and we need to know how to treat them.
So, there's a large number of children available for study and treatment. Our preliminary uncontrolled data suggest that the children may actually respond better to treatment than adults, and this is not surprising because they have lower viral loads, they have less fibrosis, they've had the disease less, they weigh less. So, if we're able to eradicate this large reservoir of hepatitis C, then the hepatitis C burden is going to decrease.
Finally, if we have properly done randomized controlled trials, supported by the FDA, then this is a way to address the increase in proportion of new hepatitis C patients who are going to be continuing to occur because of maternal/fetal transmission because right now we don't have any way that we know to prevent maternal/fetal transmission.
And finally, because we would be treating children, because we hopefully would be eradicating hepatitis C in a proportion of children, there is a better cost/benefit ratio per year of life saved.
For a parent to have a child with hepatitis C is emotionally devastating both for the family and for the child. So, this is an emotionally charged issue both for the physicians who care for these children and for the parents of the children. As of last night, there were over 110,000 Internet sites where hepatitis C in children was mentioned or addressed.
And these parents are pressuring us, a group of pediatric hepatologists and pediatricians, to come up with appropriate treatment. There is a very active support group, Parents of Kids with Infectious Diseases, which publishes support materials for families of children with hepatitis C. And scheduled for release in January 2002, supported by the Centers for Disease Control, is a pediatric handbook which addresses all of the viral hepatitis problems in children. It contains the statement, currently there is no FDA approved drug to treat children with hepatitis C. Most physicians use experimental drugs that have been used in adults to treat liver damage in children. That's the best we can come up with.
So, as I said, we as a group are very concerned that we need a mandate from the FDA to inspire the pharmaceutical companies to do the appropriate prospective randomized, controlled trials in children with hepatitis C.
I have given you the letter. I just excerpt a couple of quotes from it. We're concerned that since the approval and release of one form of pegylated interferon and ribavirin, that there will be indiscriminate and unregulated use of these agents in children without adequate scrutiny by the FDA, and given the differences in published trials between adults and children, we do not believe that simple extrapolation of adult efficacy data to children is appropriate.
Finally, we believe that properly done, prospective, randomized, controlled trials would be most consistent with the intent of the pediatric rule. Such trials would optimize drug development and safe and appropriate use in children.
We would very much appreciate your directive to the pharmaceutical companies to comply with the pediatric program requirements. Thank you very much.
DR. GULICK: Thank you.
Lastly we have a letter from Glenn Eichhorn who's a Pharm.D. and Marshall Flam who's an M.D. from the Hematology-Oncology Medical Group of Fresno, Incorporated. It's a short letter, but they've asked us to read it at the open public hearing.
DR. TURNER: Thank you. All of the committee members should have a copy of this letter in your folders.
It says, Antiviral Drug Advisory Committee: This concerns a newly established drug distribution policy by Schering-Plough for PEG-Intron.
On Monday, November 5, 2001, we attempted to place our weekly order of PEG-Intron as we had for the last year. At this point was referred to the "PEG-Intron Access Assurance Program" (1-888-437-2608). The staff at this program required we have each of our patients call and register with the program to be assigned a patient ID, and this program claimed the Access staff asked the following: patient name, patient address and home phone number, physician name, stated the Access staff would only contact the patient to inquire as to the reason for stopping therapy with PEG-Intron.
Upon securing a patient ID number, we were informed that the patient needed to obtain an authorization number. This number could only be released to a retail pharmacy with a NCPDP number and that our clinic could no longer purchase PEG-Intron. The issues that concern us surrounding this policy are:
Number 1, lack of adequate notification of medical providers, threatening to interrupt patient therapy.
Number 2, violation of patient/physician confidentiality.
Number 3, intrusion by Schering-Plough into patient disease management by controlling access to drug therapy.
Number 4, redirecting therapy provided from major medical to retail pharmacy. This will subject patients to higher insurance deductibles, transporting this refrigerated medication for physician office injection or potentially self-administering an incorrect dose without home health care support.
Number 5, no clear rationale for instituting this Access Assurance Program.
We are concerned that Schering-Plough, by claiming a product shortage, may gain a monopoly of the interferon alfa market for the treatment of hepatitis C patients. The FDA can assist in resolving this potential shortage of PEG-interferon by giving fast track approval to Roche's PEG-interferon, Pegasis. This move will relieve any potential shortage and may provide a cost-competitive market for the hepatitis patients.
Additionally, we support the request to unbundle ribavirin with Intron (Rebetron). We have been disappointed with the bundled product Rebetron because it does not allow for dose-adjustment for either the interferon or ribavirin. Bundled products only benefit the manufacturer, not the patient or clinician. We would appreciate the FDA not allowing the bundling of products in the future.
We thank you for your time and efforts concerning this matter.
Sincerely, Glenn Eichhorn, Pharm.D., and Marshall Flam, M.D.
DR. GULICK: Thank you.
If there are no further comments for the open public part of the meeting, then we'll go ahead and close that and turn to the discussion by the committee.
I think it's important just to remind us why we're here and what the purpose of the meeting is. There will not be a formal vote taken today, and we've been asked by the agency really to focus on two very specific questions. During the question period, we've actually begun to consider both of these questions, and although our time is relatively short, I think we could turn to both the questions, get input from the committee, and try to come up with some consensus.
Dr. Weiss or Dr. Siegel, anything to add to that?
DR. WEISS: No.
DR. GULICK: Okay.
I thought it would be helpful actually to address the first question. If we could actually display the designs of the postmarketing studies proposed by Schering, and then I'll go ahead and read the first question. And I should say they're summarized in Dr. Marzella's presentation on page 11, and we'll try to get the slides up there too for the committee's benefit.
So, the purpose of the meeting today was to update us on the approval of PEG-Intron in combination with ribavirin. Question number 1 to consider as a committee is, please comment on the nature and design of the postmarketing studies outlined in the August 7, 2001 approval letter. Let's go ahead and look at the design of those studies again to refresh people's memories.
So, Dr. Marzella actually summarizes it nicely on page 11 of his handout, slide number 2, that the postmarketing studies were to look at the safety and efficacy of PEG-interferon and ribavirin as a weight-based regimen and the safety and efficacy of shorter durations of PEG-interferon and ribavirin in patients with a high likelihood of response, specifically genotypes 2 and 3 or genotype 1 with a low viral load.
Why don't we read them together as we get this up there? So, this is the third slide on page 11. This is the initial large postmarketing trial proposed. The optimization of the ribavirin dose and treatment duration. So, both of those combined essentially into one study. Multicenter, randomized, open-label trial in 4,000 treatment-naive patients with chronic hepatitis C. Arm A is fixed dose ribavirin, PEG at 1.5 mgs per kg, plus ribavirin at the standard dose of 800 milligrams for either 24 or 48 weeks of duration. Arm B is the weight-adjusted ribavirin arm, again using PEG 1.5 in combination with ribavirin at 13 mgs per kg plus or minus 2 for 24 or 48 weeks, and then the doses as outlined on the slide.
So, let's open for comments about that study. Dr. Wong.
DR. WONG: I guess I'd just ask. It's hard to comment because, I mean, what's the hypothesis here? Is this a superiority or a noninferiority trial? Why 4,000 subjects? I mean, I think we have to be told a little more before we can comment.
DR. GULICK: Could the sponsor address those particular issues? So, I guess what's the primary objective of the study and then what's the power calculation in terms of supporting that objective. And how does that relate to the sample size of 4,000?
DR. KOURY: Well, it's certainly a superiority trial because we've controlled the dose of PEG to be 1.5 in both groups and then we're testing two different dosing strategies with ribavirin.
With that sample size, it's powered to detect a 5 percent difference in response rates, and perhaps just as important is to try to get additional information in important patient subgroups. A subpart of that study will be to test the duration effect in the genotype 2's and 3's with the better prognostic factors, and we're also hoping to get a substantial number of African American patients in order to be able to assess the response rates there. So, it's a large study powered to detect differences in the ribavirin dosing strategy and to obtain important information in patient subgroups.
DR. DeGRUTTOLA: What's the endpoint?
DR. KOURY: The endpoint is the standard endpoint of sustained response 24 weeks following the end of treatment.
DR. DeGRUTTOLA: So, the endpoint is measured at different times depending on whether patients are randomized to 24 or 48 weeks of therapy?
DR. KOURY: Right.
DR. HOOFNAGLE: I'm sorry. I don't know what you mean. You mean 72 weeks for the 1 year and 48 weeks for the 24-week patients?
DR. KOURY: It's 24 weeks following the end of treatment. So, that's correct.
DR. HOOFNAGLE: And you're not going to ask for liver biopsies at follow-up?
DR. KOURY: No.
DR. HOOFNAGLE: So this is factorial design I take it.
DR. KOURY: It's partially factorial. In the genotype 1's there's only treatment for 48 weeks, but within the 2's and 3's there's the cross with the durations.
DR. HOOFNAGLE: So, you're not going to do a 24 weeks in the low level genotype 1 patients. Is that correct?
DR. KOURY: That's correct.
DR. HOOFNAGLE: Just in 2's and 3's.
DR. KOURY: Yes, and part of that is because I think this is an investigator-sponsored study and there had to be agreement with the investigators. And there was not an agreement to look at the genotype 1/low viral load in that way for duration. But it will be assessed in another study that's being carried out in Europe.
DR. WONG: Can I just ask? Can you explain in a little bit more detail what the design of this study is? I'm afraid that this one slide, just showing what the two arms are, doesn't do it for me. I mean, what are the target groups? What are the inclusion and exclusion criteria? What are the hypotheses? What is the procedure going to be? It's very hard to talk about it without seeing a little bit more. In fact, quite a bit more, not a little bit more.
DR. ALBRECHT: We need slide 52.
This is a Schering-sponsored study under the direction of Dr. Ira Jacobson at Cornell University. The enrollment for the study was opened in January of '01 and as of December '01, we have 4,000 patients screened into the study. There are 26 regional PIs who oversee a region of sites that are participating, and there are a total of 225 sites.
As we mentioned, the study design, when it was initiated, was 1.5 micrograms per kilogram PEG-Intron once weekly plus either of the fixed dose of 800 milligrams per kilogram or an arm looking at 13 plus or minus 2 milligrams per kilogram of body weight administered once weekly. It was set up for 12 months of therapy.
The primary endpoint, as indicated, was sustained virologic response 6 months following the end of treatment, which is standard for all chronic hepatitis C studies.
As you have seen here, this is the dosing regimens we're using. It's not showing the 800 milligram regimen that's in there, but this is the distribution of the patients.
Now, after we had our discussions with the agency, there was a desire to include into this study -- may I have slide 56 please -- genotype 2 and 3 patients with a shorter duration of therapy looking at this. We amended the protocol, or Dr. Jacobson amended the protocol later this year, and all subsequent patients with genotype 2/3 are going to be randomized to either 6 or 12 months. We have assured that we will enroll at least 1,000 patients into this cohort to look at the randomization. So, in a sense, this is a study that has been amended to look at the additional 6 versus 12 months' duration.
DR. WONG: How about African American subjects? Are HIV co-infected people eligible?
DR. ALBRECHT: The African American subjects, based on what we know about the database, right now for demographics will run about 10 percent of the trial. Most African Americans are HCV-1, so we can expect about 300 patients in this study. There will be very few type 2 and 3.
HIV co-infected patients are not eligible for the trial. These are simply hepatitis C patients that have as their primary disease HCV with compensated liver disease. We have no decompensated patients in this study.
The same criteria that I described for our trials apply. Females, 12 grams of hemoglobin; males, 13 grams. WBC, 3,000; neutrophils, 1,500; contraception, all the standard criteria that you'll see both in the Intron/Rebetol trials and in this recent PEG-Intron trial.
DR. GULICK: Dr. Englund.
DR. ENGLUND: Are there any ribavirin drug levels going to be performed on this?
DR. ALBRECHT: No. This is a clinical trial to determine, as the primary endpoint, sustained virologic response rate 24 weeks following the end of treatment.
DR. GULICK: Dr. Hoofnagle.
DR. HOOFNAGLE: So, you're treating patients with genotypes 2 and 3 with 1.5 micrograms per kilogram per week of PEG-interferon.
DR. ALBRECHT: That's correct.
DR. HOOFNAGLE: And you've already shown .5 milligram per kilogram per week is equivalent. So, it seems to me the patients with genotypes 2 and 3 have gotten that bad vote twice, first on the original trial and this one as far as being given a lot of interferon, more than they need. In the previous trial, you were treating people for a year with your regular product you call Rebetron I guess, when you've already shown that 6 months was as good as a year.
DR. ALBRECHT: I don't know if the committee has time to see this, but we have done regression analysis looking at genotype 2/3 for both the .5 PEG and the 1.5 PEG. And all of the curves are shifted up simply because of the fact they respond better. If you look, there is still the differential between the .5 and the 1.5. I think this is probably a different question. I really think based on those regressions that we don't know whether .5 would be the appropriate dose for the genotype 2/3's.
I do agree with you. I think the 6 months in the 2's and 3's is probably going to be equally effective. However, it's just like the relapsed patients. We didn't do a study in relapsed patients. We're not licensed to treat relapsed patients because we haven't prospectively assessed it. So, now we are prospectively assessing the 2/3's and I think it's probably a good guess that those 2/3's will do well with 6 months. But I don't think we know yet and we will know when this trial is completed.
DR. GULICK: Dr. Schapiro.
DR. SCHAPIRO: I'm also actually very concerned about the PEG doses. This is a very large, long study, and it's assuming again the 1.5 dose. I think that dose has been shown to be the most toxic. It's been shown not to be superior to the 1.0 and to now proceed with this huge trial, assuming that dose, and then in a second phase to look, I think that's probably not a good approach. Based on this strategy, it will be years until we actually work out the PEG dose.
DR. SIEGEL: Let me comment on where we are historically.
It's, of course, as your comment reflects, a complex system where you're trying to optimize dose of two different agents, and there's a real potential and an expectation of synergistic toxicity, as well as the hope for synergistic effects, so that more of one may allow toleration of less than another.
So, when the question was asked earlier, a few hours ago, are we comfortable that at the end of this we'll know the optimum regimen, I had a little more doubts in my mind than the answer you got because there are a lot of different regimens and dosing levels you could be testing and you could test different ones in different populations and they interact with duration and with drugs and whatever. And optimally a very large trial in which both are married together in a factorial type design would probably be more informative.
There's an historical perspective here which is that at the time the application came in and the results for this trial became known and then PEG-interferon got on the market, there was a great deal of interest from a number of investigators in studying or in using -- in some cases studying, but in many cases using -- this combination therapy. And we received many applications for its use. Now I'm talking about a year ago now. Every week we'd get in recommendations for hundreds or sometimes thousands of patients with very little in the way of hypothesis testing.
So, we talked to Schering about the fact if you're going to support these studies, then there are so many unanswered questions, you should start using them to address many of the questions that are outstanding. And they were quite agreeable to that.
But the result is that there are timing issues. This study, for example, was done at a time before we had even substantially progressed in the review in looking at what the dosing issues were and what the subgroup issues were and whatever. And in interest in not letting take forever, it's hoped that the ribavirin trial will be able to utilize some of the data from this trial, at least the 24-week virologic data from this trial in optimizing.
That's how we got where we are. I'm not sure necessarily that we shouldn't be doing something else, but just as an explanation.
DR. SCHAPIRO: Just from what we've seen this morning where it actually did not pan out that the 1.5 was superior to the 1.0, we did not see any data to say that 1.5, by the endpoints that were defined, was better. So, this a dose which is not more efficacious.
We have seen data that there's more toxicity. This is not trivial toxicity. This is very significant toxicity. So, if we have a dose which is not more effective and has serious toxicity, to now start a trial knowing that with this many patients, I mean I'm compelled to think with this patient number, maybe a factorial design looking at both of the parameters would be more appropriate, especially with the time line. I think some of the input -- we had all these patients who were waiting for treatment. They're all going to get this dose now. That's the question.
We focused on ribavirin this morning appropriately, but we didn't really spend enough time looking at the PEG, and that had probably a greater deal of toxicity. We're used to seeing the numbers, but this is very toxic.
DR. SIEGEL: The plan, of course, is to study the PEG 1.5 versus 1. This trial got started because it's what this group of investigators was interested in. But it compares the regimen that was applied for by Schering to the regimen that they had actually studied, which was a logical comparison to do. It's enrolled. It might be possible to add on PK. I'm not sure if that is or isn't. There's advice we could take.
But there are other questions. The data from this will be used to determine whether the PEG-interferon dose should be studied in a dose response -- or at least the preliminary data should be studied with the higher ribavirin or the unadjusted ribavirin, although arguably you could say, because of interactions, the higher ribavirin may not be tolerated well in the study, but with the lower interferon, it might be tolerated and it might not get studied.
DR. SCHAPIRO: Right. It almost looks like we're zipping over some of the key issues because we already got this far. We didn't even see the non-pegylated, the Intron A, used at different exposures and seen how that would compare. It seems that we're at a certain point, so we're moving forward, but this is still very toxic and in many patients not efficacious, and we're sort of jumping forward without knowing maybe you can use less of the interferon. It's a little concerning since I think we're getting to that answer very late in the game.
DR. GULICK: Mr. Marco.
MR. MARCO: I think I share Dr. Schapiro's frustration.
I find it sort of interesting that this is the Antiviral Drug Advisory Committee under CDER but that we're sort of a little angry at CBER who, it appears, rushed to approve this combination therapy at a dose that we're really not sure should be the dose. And it looks like the sponsor has come in, at least on both of their pivotal studies, giving us dose-ranging phase III studies and not telling us exactly what dose should be used. So, it looks like many things here were rushed.
So, I really think that we need to find out which dose of interferon should be used before we really jump the gun in looking at the doses of ribavirin, and if it can be done in a factorial design, I think that would great.
I also think that in the year 2001, excluding co-infected patients, is just wrong. We've seen data from so many studies that have looked at regular interferon/ribavirin and even pegylated interferon/ribavirin in co-infected patients, and there truly is no major difference in response rates and little difference in safety.
DR. GULICK: Dr. Wood.
DR. WOOD: I was curious if Schering representatives might be able to add to the logistic regression analysis that we have presented in our handouts. It initially had just the 1.5 and then the Intron A, and then in the Lancet article, there's the higher dose of 1.5 and then the lower dose. What I'd be really interested in seeing is to see if you could add to that analysis the 1.5 who got dose modification and to see how that fell out and then the 0.5 who got a dose modification to see what the difference would be between those lines. That might help give us information to weigh the risk/benefits regarding toxicity, as well as antiviral response, which is one of the primary measurements that we'd like to see. I don't know how hard that would be to do.
DR. GULICK: Other comments from people? Dr. Hoofnagle.
DR. HOOFNAGLE: The stop rules that you showed us were very interesting. Are patients going to be stopped if they're PCR positive at 24 weeks in this trial?
DR. ALBRECHT: No, not according to the protocol.
MR. MARCO: I just have a technical question. I'm used to when the Antiviral Drug Division often gives HIV drugs accelerated approval, they ask for postmarketing studies, and that's understandable. Why are we doing postmarketing studies when this has been granted full approval? What if the sponsor decides not to do it?
DR. SIEGEL: First, I do want to comment quickly while I have the microphone on your earlier comment and just make it clear to the members of this committee that you are an advisory committee to the FDA. You are managed by the Center for Drugs, but we look to you for advice as well and very much appreciate your advice. So, don't feel odd about giving advice regarding biologics. That is well within your purview and role.
Of course, here we're talking about a combination therapy that involves a drug and a biologic, and as I'm sure you all recognize, there are many members of the division at the Center for Drugs that has responsibility for ribavirin that are here. And we have worked in close coordination on these therapies. So, let me make that clear.
We not uncommonly ask for postmarketing commitments for outstanding questions even when there's not accelerated approval. I have to say that we have somewhat less leverage in those cases. We can't make them happen. We can't threaten to withdraw their drug, but we work with companies and most companies, including this one, are willing to work with us and commit to do studies that they recognize as important.
Similarly, here we're not only at the time of approval but we're several months past approval, but our anticipation is that if there is -- and indeed there is -- important advice from this committee as to what other questions need to be studied, we will be discussing those with Schering. We don't have the leverage to say you have to do this or we'll withdraw you from the market as a condition of approval. We couldn't have said that even if we hadn't approved it yet. But we do work in a cooperative fashion to try to implement the advice and get the appropriate information.
DR. GULICK: Thanks.
Could I ask the sponsor to present the design in the detail that you just did for the second large study that you propose postmarketing? And then maybe we can talk as a group about that one next.
DR. ALBRECHT: The second large study, we will select the dose as to whether it's to be weight-based 800 to 1,400 based on the interim 24-week data as currently planned from the first study. We'll select a ribavirin dosing regimen. It is planned to use 1.5 with whichever regimen of ribavirin we select versus 1.0 of PEG with the selected ribavirin.
It will be randomized 1 to 1 in HCV-1 patients. The treatment duration will be 48 weeks. We will use the standard endpoint for determining response. It will be 6 months post treatment.
There are currently scheduled to be approximately 1,500 patients. The sample size is 1,500, 750 per group.
The inclusion criteria will be those studies that were used actually in the publication that you saw on PEG/ribavirin. We feel that HIV patients, for example, and some of the other subgroups need to be studied separately and those are separate studies. These will be treatment-naive patients with chronic hepatitis C with compensated liver disease, meeting the criteria that I previously described for our studies.
So, that study will be implemented once we pick out the regimen of ribavirin to be administered.
DR. SCHAPIRO: That's only genotype 1?
DR. ALBRECHT: Only genotype 1.
DR. WONG: Why does the other study require 4,000 but this only 1,500? It seems to me that in basic design they're equivalent?
DR. KOURY: Yes. That's a logical question.
In this particular study, the way we set it up was not totally a conventional comparison of two treatment groups. In this case what we do is we set up a decision rule saying that in order to claim that the 1.5 had an advantage over the 1.0, we had to have an observed difference of at least 4.25 percent. So, that puts the burden back on the 1.5 in order to kind of stay in contention instead of the 1.0. So, it's a little unconventional, but we mapped out the statistical properties of a decision rule like that and showed that we thought the risks for the sponsor were worth taking and it protected very well against falsely claiming that the 1.5 was better than the 1.0.
So, it's a little unconventional. It's made the keep the sample size within a reasonable framework, and effectively what it does, it says that if 1.5 is not performing reasonably better than the 1.0, we will agree that the 1.0 is the dose to go with. So, it's not that a simple lack of statistical difference will result in saying that 1.5 is okay. We sort of characterized it in a slightly different way saying that we have to have enough evidence from the study to really continue the recommendation of the 1.5 dose.
DR. WONG: So, what you're saying is that an absolute difference of less than 4.5 percent is not interesting to know, would not be a sufficient demonstration to --
DR. KOURY: We're agreeing that that's a reasonable cutoff.
DR. WONG: I don't know. I mean, an absolute difference of 4 percent. I guess I'd say that that's not a trivial difference.
DR. KOURY: That's a nontrivial difference.
Well, we agree that that's a possible interpretation, but the only way to get the study more sensitive is to now start substantially increasing the sample size and this is what at the time we agreed was a reasonable way to go.
DR. GULICK: Dr. DeGruttola?
DR. DeGRUTTOLA: Can I ask you a question? I still am a little confused about the 4.5 percent. Are you saying that the study is designed to exclude a 4.5 percent difference, in other words, to demonstrate that the 1.5 is not only better than the other arm, but is better by 4.5 percent, so that the lower bound of the confidence interval will exclude 4.5 percent? Or are you saying that you have power to detect --
DR. KOURY: No, no. It's not power. We actually have to observe a 4.25 percent difference which turns out corresponds to a one-sided .05 percent confidence interval or a hypothesis test.
DR. DeGRUTTOLA: So, what you're saying is you have to see a point estimate that's at least --
DR. KOURY: 4.25.
DR. DeGRUTTOLA: -- 4.5 percent.
DR. KOURY: Yes.
DR. DeGRUTTOLA: So, with a requirement that the point estimate be 4.5 percent, what effect are you actually powered to detect?
DR. KOURY: Probably about 8 percent.
DR. DeGRUTTOLA: Okay. So, that explains the difference. It's powered to detect a smaller effect.
DR. KOURY: Yes.
DR. DeGRUTTOLA: I have one further question on the other study. Sorry to go back to that. But in your factorial design, when you do your analysis of the main effects, will you combine over the levels of the other factors?
DR. KOURY: That's what we intend to do. So, we intend to keep the precision as high as possible in answering the main point of the study which was the ribavirin dose. And I guess we'd have to look to see if there was substantial interaction. Then maybe we'd have to reconsider that. But in the absence of that, we would intend to combine over the durations to test the main effect of the ribavirin dose and to keep that as precise as possible.
DR. DeGRUTTOLA: And it sounds like you would have good power for that question.
DR. KOURY: That's right. So, there you do have the traditional power at a more modest difference of 5 percent.
DR. GULICK: Dr. Kumar and then Dr. Schapiro.
DR. KUMAR: In the design of the study, my question is why are only genotype 1 patients being included and not genotype 2 and 3? Because here, at last, you have a dose comparing the 1.5 versus 1, and we know, at least from everything that we have heard, that genotypes 2 and 3 in people with lower viral load probably will respond to the 1 microgram per kilogram dose and have much less toxicity. So, since we did not have the benefit in the earlier study and it's up and running to change the design, can we add the genotypes 2 and 3 to this thing, at least then to say, at least for those genotypes, a lower dose would be adequate?
DR. HOOFNAGLE: They've already shown the lower dose is the same as the higher dose in genotypes 2 and 3.
DR. KUMAR: Right, but I think, if I remember, you had just said in some of the models that you're not sure that the lower dose is adequate for genotypes 2 and 3. Or maybe I misunderstood what you said.
DR. ALBRECHT: The reason we're using HCV-1 in the 1 versus 1.5 is that's going to give us the most opportunity to see the point difference we actually described because that's where we think we will see a difference.
I think that there are two interpretations of that PEG 1.0 versus 1.5 data. If you look at the PEG 1.5 throughout therapy, we have higher initial response. If you go and look at the relapse rates between Intron A and Rebetol and the 1.5 PEG, when we adjust for the dose of ribavirin, what we see are very similar relapse rates. It is my interpretation of the data, which you may not agree with, that when we use 1.5 in the HCV-1's compared to 1.0, we will see a differential. And I think if we want to see a differential in those two doses, this is where we will see it. That's why the agreement was made that we will look at HCV-1. We're not going to find out until we run the study, but that's the reason for looking at the HCV-1's.
Whether the HCV-2/3's respond to a lower dose, as I mentioned, the regression suggests that there is a difference between the .5 and 1.5. Again, that is, if you will, secondary analysis, so we can't prove it. I think probably more importantly with the 2/3's will be the duration of therapy.
DR. GULICK: Dr. Schapiro.
DR. SCHAPIRO: I think the process of generating a hypothesis and then a protocol and then proving it -- I think we should stay along those lines. The study did not show that 1.5 was superior to 1.0, and we've seen data here. I think it's been mentioned that actually the lower dose PEG showed good results with 2 and 3, and that's the one hypothesis which is not being studied in these studies. So, basically the approved dose now for 2 and 3 is the same dose, 1.5, and none of these postmarketing studies are addressing the possibility that a less toxic dose would be as effective and less toxic.
DR. SIEGEL: I'm not sure I understand that, or if I understand it, that I agree with the premise. The 1 will be compared to the 1.5 in all genotypes.
DR. SCHAPIRO: No.
DR. KUMAR: No.
DR. SCHAPIRO: It will not. That's my point.
DR. HOOFNAGLE: This study that you propose will be actually larger numbers than you have in the current study with genotype 1 treated with these two regimens. It will be almost twice as large.
DR. KOURY: [Off microphone.]
DR. HOOFNAGLE: Well, what you're saying is 8 percent difference is the difference between Rebetron, is it called, and the PEG-interferon in the current study. It was 41 percent versus 33 percent in genotype 1. So, that's what you'll be studying. Again, you have twice the number of patients.
DR. GULICK: Dr. Seeff.
DR. SEEFF: If I read this correctly, are you intending to study 100 African Americans? Is that right? Would that be sufficient to show any difference that you're looking for?
DR. GULICK: Dr. Seeff, we didn't catch the whole question.
DR. SEEFF: I'm again going back to the issue of the African American. The total number that they look to be studying is 100, and I wondered whether this was enough to be able to show a difference or no difference.
DR. ALBRECHT: There aren't going to be any statistically significant differences in that 100 patients. What we were asked to do and what we agreed to do was study 100 patients to characterize these patients. And I think it's important to note that in the large study, we think there will be about 300 African Americans. Now, we have about a 10 percent incidence in the big study which we currently know about because we're screening. And so I think 300 is a fair estimate. We add that to the additional 100 that we're going to do. We're right at 400 African American patients.
But to do a study to really compare differences in the African Americans would be a huge number. So, we will look at African Americans in the large study as a subgroup and try to understand more about these patients. But again, I don't think that we're going to ever be able to do a study of the size we need in African Americans.
DR. GULICK: Dr. Mathews.
DR. MATHEWS: Was the issue of dosing of the ribavirin on an ideal body weight versus given body weight resolved in this first study? Is the dosing going to be on total body weight?
DR. ALBRECHT: We're dosing on total body weight.
DR. MATHEWS: I think this may be appropriate to go back to the discussion that Dr. Rodvold had raised because I thought those were very important points. If the pharmacokinetics would suggest it should be based on ideal body weight, why proceed dosing it this way?
And secondly, also based on that previous discussion, it's not clear to me that this dosing algorithm based on total body weight, whether it's known that that is actually going to provide comparable drug exposure across the weight categories.
DR. ALBRECHT: Do you want to comment, Dr. Laughlin?
DR. LAUGHLIN: I think Dr. McHutchinson has made the suggestion that we will, in fact, go back and look at some of the earlier data on an ideal body weight basis as a first step in that, and I think that can be done relatively quickly. In terms of modifying this present study, I would guess that would be a difficult thing to do at this stage.
DR. McHUTCHINSON: Anecdotal, but we have looked at about 200 patients from our own center, not necessarily in these trials, looking at this issue, ideal body weight, lean body mass index, body mass index, et cetera. We haven't found anything. We may not have looked at enough patients. But so far we haven't been able to find any other than body weight. So, anything better than body weight. That what we've done so far. That was published in abstract form. It hasn't been submitted for publication. That's the only data I have.
DR. GULICK: Dr. Englund?
DR. ENGLUND: Well, my concern is the problem with much of the dosing with ribavirin is the diet dependence on the levels. It really would be helpful to have a known meal and a known blood value. And one could do population pharmacokinetics with as little as two blood draws on one patient at a supervised setting, and it could be a subset of the patients. And I would just say that I would feel that would be potentially valuable information.
DR. LAUGHLIN: I guess the one caveat to all of that is it is very difficult to -- I don't think any of us in this room maintain the same diet day after day for an entire year, and this is a year of treatment and then 6 months of follow-up. It's very difficult to quantitate and correlate the Christmas season with other times of the year, and those are very difficult things to control.
DR. ENGLUND: But, for example, RTP which accumulates over a month really could be perhaps a mean index as a better marker for overall body mass.
DR. LAUGHLIN: You mean specifically within red blood cells.
DR. ENGLUND: Yes, specifically looking.
DR. LAUGHLIN: Because it only accumulates to that extent in red blood cells. In nucleated cells --
DR. ENGLUND: That's right, and it's tricky and I can talk with you about it later. It's tricky to do but it can be done. I just offer that as a subset of patients that would potentially give you some data on which to proceed in future studies.
DR. RODVOLD: Yes. I would encourage that. Again, I come back to it because you're going to have a dose range in patients. This is a nice setting to start getting pharmacology, and you're drawing plenty of blood in this trial. Whether or not you can knock off a tube or two along the way and freeze down and do a POP analysis of this, I'd really look at that or look at the tail at the end of therapy when you're stopping up.
And you've already done some elegant PK work that you've published that you've already got a start, plus you have your own models already in place so that you don't have to do an elaborate study here other than a population type of analysis and link it to PK, both the toxicity and efficacy issues, because I think the dose ranging is where everyone is frustrated and so are you and so are we. And I think you just have a chance here to capture some things. Please, please consider doing that.
DR. GULICK: Dr. Weiss, these were really the two studies you wanted us to consider as a committee. Is that right?
DR. WEISS: That is correct. Those were the two main studies.
DR. SIEGEL: We've heard a number of ideas about other issues that might require other studies.
DR. GULICK: And that's where I want to turn next, but I just kind of want to sum up what we said as a group.
I think as a committee we thought it was valuable -- and it was echoed in the community comments -- to really take a close look at the data with pegylated interferon and ribavirin. To have access to all the studies to be able to discuss this as a group was something that I think was valuable for the committee to do.
I think we appreciate in the design of postmarketing studies -- and John summarized it best -- that there were hypotheses that came out of the previous study, and what you've done is to take these and propose prospective studies to carefully test those hypotheses. And as an approach, again I think that the committee is enthusiastic about that approach.
One difficulty I think we've had today is to be asked to comment on a study that's already designed and enrolling patients. Our input into such a study is relatively limited. Nevertheless, we --
DR. SIEGEL: I would simply note, though, that I think your comments in that regard are very useful because they help define not only what those studies will and won't tell us, but also your thoughts as to how important the questions are that won't get answered. I suspect that Schering, as well as the FDA, will find that very useful, not perhaps necessarily in changing that study, but in figuring out, not just for this drug perhaps but other drugs as well, what are the critical questions.
DR. GULICK: And that's exactly what I was going to say next. So, thank you.
DR. SIEGEL: I'm sorry. I'll just express our appreciation.
DR. GULICK: It's a love fest here.
DR. GULICK: I'll be brief, but clearly the committee thought that the important questions were to nail down this ribavirin dose strategy. Should it be one-dose-fits-all or weight-based?
At the same time, we agreed that the interferon dose clearly needs to be detected, and then one of the major concerns that we had as a committee was can you really separate out those two questions. They might have different answers and different interactions. That's something to think about for future studies.
I think we were also pleased to see that the duration question was being addressed in the genotype 2/3, and that adequate consideration was being given to subgroups of patients, particularly African Americans and contrasting genotypes 1 and 2/3 in their responses.
A number of concerns were voiced around the table. Again, I think the most important is, can you really dost-adjust ribavirin first and then go on to look at the interferon dose question, or do those need to be carefully asked at the same time, perhaps in a factorial analysis?
A number of comments about subgroups came up, whether we already know some of the answers for the genotype 2/3 group in terms of duration and dose of interferon. Do we already know enough from the published studies to date?
Some concerns about special groups, like the HIV population, intravenous drug users. Not mentioned was the Latino population and might there be different responses in those groups.
And then a sincere plea to consider this as an opportunity to assess population pharmacokinetics and the importance of that and, along with that, the idea of body weight dosing.
So, that's a brief summary. I think we have limited time, but we want to turn now to the second question that's been posed to the committee. Please comment on other issues regarding PEG-interferon and ribavirin that could be evaluated in further studies. Some have already been mentioned.
DR. HOOFNAGLE: Well, I'd like to ask the sponsor what they're doing in pediatric studies. This is very critical actually, and they're really good patients to treat too. So, the results are usually much clearer.
DR. ALBRECHT: I think Dr. Schwarz mentioned the program that's ongoing. We have a program in Intron/Rebetol that has been ongoing. In that program, we conducted pharmacokinetics with Intron A and ribavirin. We selected a dose of 15 milligrams per kilogram in that study to be used in combination with Intron A 3 million units per meter squared. We have enrolled in an open-label efficacy trial about 120 pediatric patients. Those patients have completed study. They are completing follow-up as we speak, and we will be providing to the FDA in April the final dossier on that product.
I would mention that the first study we did we did with ribavirin capsules. We had 50 milligram capsules that we made as an interim thing to use with the pediatric patients. The FDA asked us to try very hard to develop a formulation, and this is actually the slide that shows this.
In part 1 of the PK study, we used children 5 to 16 years of age. We looked at 8, 12, and 15 milligrams per kilogram of ribavirin. We selected the dose based on the hemoglobin and the antiviral effect in treatment at 12 weeks.
In part 2 of the study, we treated 35 more children.
In the next study, which is an open-label -- I think it should be on the next slide probably -- we treated younger children. As I said, the FDA asked us to try to develop a pediatric formulation, a liquid formulation that the smaller children could take and that we could more closely regulate the amount of drug they were getting. We did this. We have a liquid formulation. And it was actually used in this study, and we treated 70 children with the formulation. Now, we did treat the older children with 200 milligram capsules, but a good proportion of these kids were treated with a liquid formulation.
At the moment, we have actually all of the follow-up data, and as I said, FDA will receive the entire dossier on these two studies in April. That's where we stand with the data right now, and it is just now being finalized. I can tell you I think Dr. Schwarz spoke a little bit about what was presented at AASLD. Children have the same side effects as adults. You have to be careful with adolescents in psychiatric side effects because they are more prone. They actually tolerate it from a hemoglobin point a little better. They don't have as much hemolysis, and they actually have a little less neutropenia. So, all in all, from a response perspective, they look quite similar to the adult patients with Intron/Rebetol.
I will tell you these patients really had no fibrosis. So, these were patients that basically had inflammation. We did require that the patients have an elevated ALT in the first study, and in the second study, because many of these children have normal ALTs, we did allow those kids in as long as they have a liver biopsy that showed there was inflammation. We did not biopsy the children post treatment. We did biopsy them pre treatment to make sure they actually had inflammatory activity in the liver. So, that's being submitted in April.
DR. GULICK: We're actually going to need to wrap up the discussion. Let me just mention that we've talked about different groups over the course of the day that the committee was interested in seeing, pediatrics clearly just addressed, the HIV/hep C co-infected person, the methadone or intravenous drug using population, Latinos and African Americans. Hemophiliacs came up on one slide as having perhaps a different response rate, and then we've talked a lot about genotype 1 versus 2/3.
DR. MATHEWS: We didn't talk about patients with end-stage renal disease, which is an important group because of the prevalence of infection and dialysis issues.
Then the other issue I wanted to bring up was an issue of toxicity management because in the trials that we reviewed, patients were discontinued for hematologic toxicity, and I know in clinical practice many people are using cytokine support, and we didn't see any data on the use of erythropoietin and white cell support factors.
DR. GULICK: Were there other specific groups that people wanted to mention? Dr. Englund?
DR. ENGLUND: I don't want to mention, but I just want to more strongly recommend. I didn't hear anything about PEG-Intron for the pediatric patients. We've got to have it. We have to have that available to our patients. And for my HIV patients at home, I have to say that our patients want it. They really want it. So, we have to really strongly as a committee say that we need it.
DR. SIEGEL: May I ask a question about that in terms of as we discuss this with the company? We're often, as we've seen in the dose situation, faced with the issues of starting trials before we know the results of other trials to get the answers sooner or waiting for results so we can design the trials better or address the right issues.
Given the program in place with Intron A and the expectation that over the next several months substantially more will be known about risks and benefits with an Intron A/ribavirin approach in children, would you suggest that the company consider and we speak with them about starting a pediatric trial now with the pegylated or wait till there's some of those data to look at to determine what are critical questions, whether they're genotype related or age related or certain toxicity concerns or whatever?
DR. ENGLUND: I think no matter what, you're going to not necessarily want to use ribavirin in teenage girls. No matter what. I mean, I bet. So, you should be trying pegylated --
DR. SIEGEL: Oh, you're talking about PEG-interferon monotherapy.
DR. ENGLUND: I want PEG-interferon monotherapy. Maybe there's data but I haven't heard it. So, I think there's room to go with PEG monotherapy first and then, sure, let the adults get the ribavirin doses down, and then we can translate that. But we don't even have PEG data, unless I'm mistaken, in kids.
DR. GULICK: So, besides the subgroups and the pharmacokinetic analyses that have been suggested, as well as the other questions posed by the first two studies, are there other areas? Let me stick with committee members right now, Jules, and if there's time, we'll come back in just a minute. But are there other areas that people would like to suggest?
DR. STANLEY: Trip, this is Sharilyn.
DR. GULICK: Oh, great.
DR. STANLEY: I've been listening and I just didn't make comments because most of my concerns were articulated by somebody.
But I would like to raise the issue that one of the public raised which is the whole issue of the marketing strategies and campaign. Again, I know that we as a committee don't have a lot to say on that, but I can tell you that in my view Schering has pushed the envelope of ethical marketing in a lot of areas. I've seen it here in Texas. So, I just would urge the FDA, whatever pull they have over that, to look at those issues.
DR. GULICK: Thanks.
Any last comments from committee members? Dr. Hoofnagle?
DR. HOOFNAGLE: Well, I'm not sure what you meant about monotherapy with PEG-interferon, but one of the frustrations was that once we had interferon/ribavirin combinations which were so much better -- that really was the breakthrough adding ribavirin -- then to have go back to monotherapy studies was really painful. And I would think in the children studies that you don't have to go back to PEG monotherapy studies unless there are specific contraindications. Certainly that may be the case in renal failure patients, in some HIV-positive patients. But the next study in children really should be PEG-interferon plus ribavirin and not PEG-interferon monotherapy. Please.
DR. GULICK: Any last comments? Jules, did you have a suggestion about an issue that hasn't been raised?
MR. LEVIN: Yes. We still don't really know the effect of HART on liver disease progression in co-infected people. We know hepatotoxicity can occur with elevated ALTs. We don't know the clinical significance of that. That's a question that remains unanswered.
Lastly, I think it would be very helpful for the FDA to consider approving Pegasis or reviewing it and, if appropriate, approving it as quickly as possible. I think it would be good to have a competitive marketplace.
DR. GULICK: Thank you.
Would anyone like to have the last word? Besides me?
DR. GULICK: Okay. Then we'll close this section. Thanks to the sponsor, to the agency, to the audience, and to the committee members.
(Whereupon, at 1:19 p.m., the committee was recessed, to reconvene in closed session at 1:45 p.m., this same day.)