 |
 |
 |
|
 |
|
FDA Home Page | CBER A-Z Index | CBER Search | Contact CBER | CBER Home Page  |
||||
Workshop on Development of Donor Screening Assays for
|
|
DEPARTMENT OF HEALTH AND HUMAN SERVICES
Bethesda Hyatt Regency Hotel Tuesday, November 5, 2002 8:00 a.m.
CONTENTS V. Pathogen Inactivation Targeted at WNV Chair -- Mahmood Farshid, FDA, Luiz Barbosa, NIH Viral Inactivation Methods in Blood Components Overview of Different Methodologies: Steve Wagner (ARC) Industry Representatives
Cerus/Baxter: Lily Lin
Viral Clearance Methods in Plasma and Plasma-Derived Products Role of Model Viruses in Current Inactivation Studies: Dominique Pifat, Bayer Corporation Robustness Data from Model Viruses: Albrecht Groner, Aventis Behring GmbH Equality of Model Viruses and Current Data for WNV: Thomas Kreil, Baxter Bioscience General Discussion VI. Proposed Studies on Prevalence in Donors (Chair--Mary Chamberland, CDC, and Liana Harvath, NHLBI, NIH) The NIH Collaborative Donor Prevalence Linked Study Michael Busch, Blood Centers of the Pacific Discussion (With Steve Kleinman, University of British Columbia and Susan Stramer, ACR) VII. Regulatory Issues (Chair--Paul Mied, FDA) Considerations in Developing Assays for Testing Donors For West Nile Virus: Robin Biswas, FDA Guidance for Industry: Recommendations for the Assessment of Donor Suitability and Blood and Blood Product Safety in Cases of Known or Suspected West Nile Virus Infection: Martin Ruta, FDA Discussion VII. Implementation Issues: Blood and Tissue Organizations (Chair--Alan Williams and Melissa Greenwald, FDA)
AABB: Steven Kleinman Discussion Panel Discussion (Panelists: Jesse Goodman, FDA; Hira Nakhasi, FDA; Liana Harvath, FDA; Mahmood Farshid, FDA; Darin Weber, OCTGT; Bill Hobson, HRSA; Glen Freiberg, AdvaMed; Robert S. Lanciotti, CDC; Lou Katz, Blood Banking Organizations; Mike Busch, Blood Centers of the Pacific; Steve Wagner, ARC; Thomas Kreil, Baxter Bioscience) Concluding Summary: Edward Tabor, FDA
PROCEEDINGS DR. NAKHASI: Let's get started. Happy Election Day to you all. We need to get out today quickly because we have Election Day today so we all need to vote, especially in Montgomery County. As you know, it is a very close race so we need to make sure every vote counts. Anyway, we will try to be on time as we did yesterday. I would like to welcome you all today again. We have a full agenda today which includes starting with the viral inactivation process, pathogen inactivation of target West Nile Virus, then proposed studies on prevalence in donors. We then will talk about the regulatory issues which a lot of you may have questions about and implementation issues on blood and tissue, and then, finally, a panel discussion. That will be a free-for-all because we need to make a definite outcome of this meeting. Before I pass on my podium to my esteemed colleague, Dr. Mahmood Farshid, there is a change in the setting here. Joe Wilczek tells me that, around 10:00, when the first break is there, we have to vacate this room because there is going to be a marriage ceremony going on here. We are all welcome to stay here if you want to, on a lighter side. But, anyway, seriously, we have to leave this room at 10:30 at the first break, go upstairs one flight. The meeting will be in Cabinet Judiciary Suites. If there are extra people there--I was told that it only holds 225, but I guess yesterday we had 300 people here--there is a room next to it where we can put the spillover and there is also a big screen like this so we can have that, also. See you then and I guess I will pass it on to Mahmood. Thank you.
V. Pathogen Inactivation Targeted at WNV
Chair: Mahmood Farshid, FDA
DR. FARSHID: Thank you, Hira. My name is Mahmood Farshid. I am with the Division of Hematology in the Office of Blood at CBER, FDA. The first session for this morning is pathogen inactivation methodologies as applied to West Nile Virus. I and Luiz Barbosa of NIH will be moderating this session. The session is divided into two parts. The first one applies to the methodologies which are applicable to the labile blood components, and the second part will be the established methodologies currently used in the fractionated plasma-derived product. We have a packed schedule and without further ado, it is my pleasure to introduce our first speaker, Dr. Steve Wagner, who is the Director of Cell Therapy and Blood Cell Therapy Development at American Red Cross, Holland Laboratory. Steve. Viral Inactivation Methods in Blood Components Overview of Different Methodologies DR. WAGNER: I would like to thank the organizers very much for inviting me to the meeting. I would also like to say that I heard that yesterday's meeting was very good. More than a half-dozen people mentioned that they didn't see me yesterday. I was hard at work at the office. I am going to talk today about different challenges and a broad overview for inactivation or pathogen reduction of West Nile Virus. [Slide.] There is a number of rationale for inactivation. The first rationale is to deal with residual infectivity that might be in the blood supply as a result of screening tests that are already in place. In addition, we all know that pooled products transmit a threat of infection if any member of the pool were to be contaminated with virus. Pathogen reduction also constitutes an additional layer of safety in addition to donor questioning and screening with respect to West Nile Virus. West Nile Virus is currently acknowledged, it has been around for a number of years, since the '50s, it has been described, but currently, there is no test that is available in a licensed form for blood centers to use, and I understand that you heard yesterday about some potential development of tests that are currently underway. In addition, pathogen reduction might be able to deal with variant agents, HCV, HIV, a number of the viruses that have the capacity to mutate to other viruses at a frequency that might be a bit higher than some of the DNA viruses, particularly appropriate when discussing variant agents, and then there might be new agents that come about. Of course, the public and the folks in Congress are very concerned about the safety of the blood supply, and any risk is of great concern. [Slide.] So, West Nile Virus, you have probably heard this, is a flavivirus. It is an enveloped, single stranded, positive stranded virus. Some very early studies in the 1950s in infected patients indicated that these patients, when samples of their blood were taken, could be diluted between 100-fold and 100,000-fold and inoculated in the susceptible animal in a bioassay, so titers are considerable although they are certainly not more than 106. The CDC has estimated that levels in asymptomatic donors may be between 103 and 104 genome equivalents. With that said, however, the relationship between the genome equivalents and the plaque-forming units hasn't been completely characterized yet and I imagine a lot of this work will go on in the next few months as we get more information about West Nile Virus. So, the actual log reduction right now necessary to prevent transmission from the blood of asymptomatic donors is really not known at this time, and this is some very important information that needs to be collected. [Slide.] With respect to taxonomy, West Nile Virus is related in many ways to other viruses that have been used for pathogen reduction experiments. We have used Sindbis in our laboratory with some experiments a number of years ago. Most of the companies and we now have looked at BVDV, however, perhaps hepatitis C is a more closely related virus than even BVDV. Of course, West Nile Virus itself is the best model, if you will, or the best virus to test for its inactivation, and this is somewhat hampered by the fact that it is classified as a BL3 virus at this point, and so that is going to limit the availability of different laboratories for working on the virus. [Slide.] The good news is that both flavivirus and togaviruses should be susceptible to pathogen inactivation agents. Just looking at some classic books by Block indicates that it is susceptible by UV light, gamma irradiation, a number of disinfectants including glutaraldehyde, hydrogen peroxide, chlorine-containing compounds, bleach, for example, and alcohol, as well as iodine, so it is a very susceptible virus. [Slide.] There is a number of approaches to inactivation. Almost all of these are led by company approaches. For platelets, there are psoralens S-59, also called now amotosalen. There is a group working with riboflavin and red cells. There are some alkylating agents. One is FRALE, which is called S-303, and then there is another agent called INACTINE. I am not going to be really going over plasma today, but plasma, as you know, dealt with solvent detergent, and there are other approaches, phenothiazine dyes, as well as a psoralen that is being investigated for plasma. [Slide.] S59 is a planar molecule, psoralen or a fucocumarin. Its status is that it has completed Phase III clinical trials in the U.S. and Europe. It is licensed in Europe with a CE mark. The buffy coat method has been licensed in Canada, and the plasma work is in Phase III trials in the United States. [Slide.] Psoralen is a photochemical. The first step is for the drug to intercalate between the bases of double stranded regions of DNA and RNA. Even RNA has double stranded regions. Upon the absorption of ultraviolet A light, psoralens make mono- and di-adducts with pyrimidine bases in nucleic acid. Diadducts and monoadducts can prevent the subsequent nucleic acid replication of the pathogen, and because pathogens contain nucleic acid, and platelets and red cells do not need nucleic acid for their storage and viability, this is an approach that all the companies take. [Slide.] There is a number of steps that are used for pathogen reduction for S-59. They use, at least the Phase III trials in the United States and Europe, have looked at apheresis platelets. S-59 is added to the platelets and then the mixture is transferred to a UV-permeable plastic container where it is illuminated with light, and then after illumination, the platelets containing S-59 are added to a bag that contains a resin that absorbs a lot of the free S-59. It stays in the resin for several hours before it is transferred to a container for storage. [Slide.] I am not going to go over any West Nile Virus data, I presume the companies will do that. I am just going to be talking about the published data on viruses that are close relatives to West Nile. There has been identification of inactivation of HCV in a chimpanzee model with a platelet suspension treated with S-59, as well as inactivation of bovine viral diarrhea virus in a platelet suspension. [Slide.] The other method for inactivating viruses and particular potentially West Nile Virus in platelets is riboflavin, some molecule riboflavin. It is a vitamin. Its status right now is in preclinical. [Slide.] It works also by a nucleic acid method. It binds to DNA by intercalation. Upon absorption of either visible or near UV light, the complex induces guanine oxidation, single strand breaks, and the formation of covalent bonds. [Slide.] There are some published data for pathogen reduction of viruses related to West Nile Virus in this respect, bovine viral diarrhea virus of more than 5 logs with riboflavin and light. Some of these data was presented just at the recent AABB conference in Orlando. For red cells, as I mentioned, there are some alkylating agents that are being used. One of the companies is working on a method using S-303. S-303 has a very similar structure to a compound called quinacrine mustard. These are acridine nitrogen mustard compounds, and one of the major differences between quinacrine mustard and S-303 is S-303 has this ester bond in the middle. Its status right now is in Phase III in the United States. The mechanism by which FRALES work is that the anchor, which is the acridine moiety of FRALES, intercalates between the bases of double-stranded regions of DNA and RNA, and the nitrogen mustard moiety or the effector of the FRALES makes adducts with nucleic acid bases. Diadducts, for example, form a cross-link between nucleic acid strands and again, like the psoralens, that prevents subsequent nucleic acid replication. The ester moiety in FRALES, which I pointed out before in the alkyl region of the compound, is the frangible linker region, and it hydrolyzes forming negatively charged acridine compound that doesn't further interact with nucleic acid, and the rate of reaction of the ester linkage is slower than the nitrogen mustard, and that is how that compound works. The reactants produced by the FRALE that are free in solution, and not alkylated, can be potentially depleted by a removal device. [Slide.] S-303 pathogen reduction, they published on bovine viral diarrhea virus, and they see more than 5.6 logs inactivation. [Slide.] The other compound that is being studied in red cells, the company calls INACTINEs. This is a cartoon, a picture of INACTINEs. It has got a 3-cycled ring, which is a covalent modifying group. It has a cationic alkyl tail to it. The alkyl tail is positively charged, which confers DNA binding to nucleic acid by electrostatic interactions. It is said to stabilize molecule, and this molecule has a much smaller size than the others, so it can inactivate viruses whose capsid structure proteins are tightly interdigitated, that are somewhat resistant to inactivation by other agents. The INACTINEs have this azito [?] moiety at the end of the compound. This is an example. I believe the actual compound that is being studied is PEN110, so this is not it. This is ethylamine. So, it has a zerodino [?] moiety at one end, and then it has two or more nitrogens in the compound, separated by hydrocarbons. [Slide.] INACTINE reacts with the N7 bond of guanine, producing a monoadduct. This can serve as a stop signal to replication. Also, repair enzymes can recognize this and cause the loss of a base or a basic site, and once this occurs, there is a potential for strand breakage. [Slide.] Some work was done where people did typical sequencing of a template, of a normal template, as well as a template that has been treated with INACTINE, and as you can see here, at high salt concentrations, you can get some replication of the template, but in particular notice that in the C residues of the primer, that there is considerable stops, and this indicates that at the G residues of the template, replication has stopped probably because of adduct formation. [Slide.] Their process is adding the compound to red cells with an incubation period. I believe this is at room temperature. Then, the compound is removed by extensive automated washing. [Slide.] They have seen more than 6 logs of bovine viral diarrhea virus in red cell units. [Slide.] There are a number of challenges for pathogen reduction techniques. First of all, there is potentially a lot of transfers between bags. Every time you transfer a component from one bag to another, there is some loss, so there may be some unwanted reduction in cellular yields. Although some agents may be specific for nucleic acids, that is not universally true. There are going to be some side reactions that occur. The side reactions may be reactions to lipids with the compounds, reactions to proteins with the compound. In both of these circumstances, whether or not a particular method has a removal technology, they are not going to be able to remove the compound when it has reacted to lipids. If it has reacted with a cellular protein also, that is not going to be able to be removed, that is going to be transfused to the recipient. In addition, the photochemicals have the potential, even though they may be adduct-forming, to generate reactive oxygen species which can be harmful to cellular membranes, and that is of a concern, as well. These side reactions may be responsible for the loss of survival or function of blood components. Some of the loss of survival, for example, have been observed in clinical trials with some of the agents, so these things need special consideration and thought. [Slide.] There are some other challenges. The side reactions could be responsible for unwanted low-frequency adverse events. The search of things that might be important to look for, for these low-frequency adverse events might be immunological reactions, they might be allergies, but they could go to anaphylaxis. This could be 1 in 100,000 units, we don't know, but until a method is used extensively, that information won't be available. In addition, there may be increased sensitivity of blood cells to other pharmaceuticals. For example, if there is some singlet oxygen damage or oxidizing damage, the blood cells may be sensitive to oxidizing drugs. If a chemical interacts with glutathione, it also might be sensitive to oxidizing drugs. So, these drug-drug interactions may be of importance to patients when enough of them are investigated and more people are treated with these agents. In some agents, an unexpected, accidental exposure to people who are manufacturing the drugs or transporting the drugs or blood center staff could lead to increased genotoxic risk, and that obviously is of some concern. [Slide.] So, in evaluating pathogen reduction methods, it is important to pay attention to the potential for low frequency adverse events, so you really need to implement before you will able to see some of these effects if they are to occur. Without implementation and long-term study, it might be difficult to predict the risk to blood bank workers or recipients by accidental exposure or by residual drug. Without implementation and surveillance, it may be difficult to assess a risk of allergic or hypersensitivity or anaphylactic reactions in susceptible recipients caused by alkylations to proteins or by drug metabolites. Without implementation and long-term surveillance, it may be impossible to determine if the risk of a fatal outcome from an inactivation process is greater than the current risk of fatalities from infectious disease transmission. So, there has to be a good surveillance system put in place if these methods are introduced, so that these sorts of things can be measured. [Slide.] In conclusion, all methods that I have discussed today I believe target nucleic acid. These methods can reduce the infectious titer of extracellular and intracellular enveloped viruses. All available preliminary information suggests that West Nile Virus should be susceptible to all the mentioned pathogen reduction techniques. It is not known what level of West Nile Virus reduction will be necessary to prevent transmission from asymptomatic donors. Implementation and surveillance may be required to assess low frequency risks. In the low-frequency risk assessment, it is essential for establishing that fatalities from the pathogen reduction process are less than current fatalities from infectious disease transmission, and non-nucleic acid side reactions, as I call them, may be important to understanding some recipient reactions, as well as to explain any loss of cellular function, recovery, or survival. Thank you very much. [Applause.] DR. FARSHID: We will have a question and answer session at the conclusion of this session. Thank you very much, Steve, for an insightful overview. Next, we move to industry presentation and we hear from industry representatives, which currently they are working on this technology, and we are promised a adduct-driven presentation relevant to West Nile Virus. Our next speaker is Lily Lin from Cerus/Baxter.
Industry Representatives
Cerus/Baxter DR. LIN: I would like to thank the organizer for inviting Baxter and Cerus here today to present the data. [Slide.] In my talk, I would like to cover three areas, the first on the Helinx technology developed by a sponsorship between Baxter and Cerus, and secondly, I would like to discuss the inactivation of pathogens in general, and then move to the last part of inactivation of flaviviruses, and I would present data on inactivation of hepatitis C virus, the bovine viral diarrhea virus, BVDV, and inactivation of the West Nile Virus. [Slide.] The Helinx technology is based on the following. As Dr. Wagner already pointed out, that pathogens, leukocytes require nucleic acid for replication, and in contrast, blood components do not require nucleic acid for therapeutic function. So, Helinx technology relies on small chemical compounds that target and modify nucleic acids to prevent replication of viruses, bacteria, protozoa, and leukocytes. The compounds I am going to talk about today, the first one is amotosalen, also known as S-59, is developed to treat platelet product. The same compound is used to treat plasma product. A second compound was developed to treat red cell products called S-303. [Slide.] Both compounds operate under the same mechanism of action. These compounds are represented here. Each of them has two reactive sites, and because of the small size, they intercalate effectively and reversibly into the helico regions of both DNA and RNA. Only when activated, these compounds will react and form covalent bonds with bases of the nucleic acid. Amotosalen is activated by a long wavelength ultraviolet light UVA. S-303, on the other hand, is inactivated by a pH shift. After activation, because of the two reactive sites, it forms a permanent cross-link between the two nucleic acid strands, and these cross-linking products effectively prevent a replication of the nucleic acid. [Slide.] Now, the systems developed for treatment of platelet concentrates contains two components. The first one is a UVA illuminator that delivers the required dose of UVA, and the second component is an integrated container set that allows the addition of the amotosalen and treatment of the platelet concentrate in a closed system. [Slide.] The system developed to treat plasma also uses the same UVA illuminator, the same amotosalen compound, but the integral disposable set is slightly different from what is used with the platelet concentrate. [Slide.] The system under development for treatment of red cells will use a series of disposable containers, but all of these systems are designed to treat single unit products, a single unit platelet, plasma, or red cells. [Slide.] To validate pathogen inactivation using these systems, we have used full-sized therapeutic units of platelet concentrate, plasma, or red cells. Each unit was spiked with approximately 106 infectious unit of a pathogen per ml of product, all with the highest titer stock available. The contaminated platelet and plasma product were treated with 150 micromolar amatosalen and 3 joules of UVA light. Contaminated red cell units were treated with 200 micromolars of S-303. In many cases, inactivation kinetics were measured. The infectivity of each pathogen was measured using either culture methods or, in cases that culture methods were not available, we used animal models. [Slide.] The results demonstrate that Helinx technology inactivates high levels of a broad spectrum of viruses, bacteria, leukocytes, and protozoas. Here, I have only summarized a subset of the data. This table has a list of all the pathogens being tested today in blood centers including HIV-1/2, hepatitis B, hepatitis C viruses, HTLV-I/II, and a Treponema pallidum that causes the syphilis. The inactivation levels were expressed as log reduction, and the greater than sign demonstrate inactivation to below the level of detection. [Slide.] Flaviviruses, as you heard today and yesterday, they are enveloped, single-stranded RNA viruses. Examples include yellow fever, Japanese and St. Louis encephalitis viruses, border disease virus, hog cholera virus, the Dengue fever virus, and Usutu virus. Those three viruses highlighted here, the hepatitis C and BVDV virus, and West Nile Virus are flaviviruses, and I am here today to present you the inactivation data for those three viruses. [Slide.] The study design used for HCV inactivation in platelets is the following. We have spiked 4 1/2 logs of chimp infectious dose of the well characterized, the Hutchinson strain of HCV into three, full-sized units of apheresis platelet concentrate, and treated with 150 micromolar S-59 or amotosalen, and 3 joules of UVA. After treatment, the entire unit, approximately 300 ml was infused into a seronegative chimpanzee, which was followed for six months for development of hepatitis, as well as molecular and biological markers of HCV infection. As you know, infection of this Hutchinson's strain of HCV in chimpanzees has been shown to be uniform and consistent. [Slide.] Results of one of the chimps shown in this graph here, the arrow indicates the time zero for transfusing or infusing the entire unit of treated and spiked platelet concentrate. This animal showed normal liver histology both before infusion and six months after infusion, and throughout the evaluation period, there was no antibody developed against HCV virus or by an RT-PCR methodology, there was no HCV viral RNA detected. Throughout the evaluation period, the liver enzyme ALT and AST were normal, and they were consistently at the background level. So, these results clearly demonstrate inactivation of 4 1/2 logs of the HCV virus as measured by an infectivity assay in chimps. [Slide.] Inactivation of BVDV uses the following design. Approximately 105 to 106 PFU per ml of BVDV was spiked into full size units of platelet plasma and red cells. The contaminated platelet plasma units were treated with 150 micromolar amotosalen and 3 joules of UVA light. Contaminated red cells were treated with 200 micromolars of S-303, and the viral titer in the sample was measured using a plaque assay bovine terminate cells. [Slide.] The results are summarized here, and as you can see, we have achieved consistently very high levels of inactivation. These are updated results, the number may look slightly different from what Dr. Wagner presented. For the platelet and plasma product, we have achieved a greater than 6 logs of inactivation, and in red cell product, we have achieved a greater than 7.3 logs of inactivation. [Slide.] Now, just to evaluate how sensitive BVDV is to Helinx treatment, we have done a kinetic analysis. As you heard, that would be 150 micromolar amotosalen combined with 3 joules of UVA is the process developed for treatment of platelets. Using that condition, we have achieved a more than 6 1/2 logs of inactivation. We have also taken out 30 ml aliquots after only half a joule of elimination, and demonstrated no recoverable viruses in the aliquot in four out of four experiments. These results demonstrate inactivation of more than 6.3 logs of BVDV. So, to borrow a phrase that is coined by Dr. Bernie Horowitz, these results demonstrate that the Helinx technology or the system has plenty of reserve capacity to inactivate BVDV. [Slide.] To look at the sensitivity of this virus to Helinx treatment, we have also looked at the dose-response curve. The results shown here demonstrate that we can lower the concentration of amotosalen from 150 micromolars to as low as 3 micromolars. With a combination of 1 joule over UVA light, we have inactivated more than 5 logs of the virus. [Slide.] So, finally, inactivation of the West Nile Virus. The study was conducted in collaboration with Dr. Kristin Bernard of the New York State Department of Health. The viral inoculum used in the study was prepared from the BHK cells infected with a full-length infectious clone of the West Nile Virus. The parental strain of West Nile Virus lineage 1 was isolated from the epicenter of New York City during the year 2000 outbreak. The infectivity and virulence of the cloned virus and the parental virus are similar. The plaque morphology of the cloned West Nile Virus is also indistinguishable from the parental virus, and we used the working stock has a titer of 108 PFU per ml. [Slide.] We have spiked approximately 106 PFU/ml of the cloned West Nile Virus into full size units of platelet concentrate or red cells. The spiked platelet units were treated with 150 micromolar amotosalen and 3 joules of UVA light, and the spiked result units were treated with 200 micromolars of S-303. The titer of the West Nile Virus in the sample was measured using a plaque assay on Vero cells. [Slide.] These are preliminary results and they are summarized in this table. For platelet units, the treatment volume was approximately 300 ml. An aliquot of the pretreatment sample confirmed the level of the inoculum at 5.4 times 105 PFU/ml. After treatment, we have seen no recoverable virus in 1 ml samples in two of two experiments, demonstrating inactivation of more than 5.7 logs of West Nile Virus in platelet concentrate. Similarly, for red cell units, the treatment volume was approximately 300 ml, and the pretreatment aliquot demonstrated the infectivity of the inoculum at 9.1 times 105 PFU/ml. After treatment, no recoverable virus was detected in 1 ml samples in two of two experiments. These results demonstrated inactivation of more than 6 logs of West Nile Virus in red cells. So, these preliminary results confirm our prior expectation that the Helinx technology inactivates West Nile Virus very effectively. [Slide.] In conclusion, Helinx technology inactivates a broad spectrum of viruses, bacteria, protozoa, and leukocytes in the three components of the blood, platelets, plasma, and red cells, and our preliminary results demonstrate inactivation of high levels of West Nile Virus in platelet concentrate and red cell components. Both the amotosalen and S-303 are effective against West Nile Virus. Thank you very much. [Applause.] DR. FARSHID: Thank you, Dr. Lin. Next, we hear about INACTINE technology by Vitex. Dr. Bernadette Alford will present the data. Vitex DR. ALFORD: Thank you very much. It is our pleasure to speak today about another pathogen reduction technology, and that is the INACTINE technology. I will focus most of my efforts specifically on West Nile Virus. [Slide.] What I would like to do first is just briefly introduce you to Vitex, if you are not aware of who Vitex is. Our goal at Vitex is to introduce a new safety barrier for red cell concentrates beyond donor selection serologic screening. What we would like to do is to build a safety into the manufacturing process rather to inspect safety into the product. Our focus is not to restrict the donor pool, but rather use the single step to address a wide range of viruses, as well as eukaryotic and prokaryotic pathogens. Our approach is chemical inactivation by a compound called INACTINE PEN110, which is combined with red cell purification, and the regulatory approach is through IND BLA, and we are currently in Phase III clinical trials. [Slide.] This is just quickly to depict our INACTINE-automated processes that we have just recently developed. We have INACTINE PEN110 delivery to the red blood cell, which is a completely automated system both for formulating our working solution from a concentrate involved with the delivery of PEN110. Then, there is an incubation step at room temperature and in washing, as was described by Steve Wagner, and after washing, we have a unit that is what we term "pathogen inactivated" and ready for transfusion. [Slide.] What this process does is really produce two particular steps. The process combines chemical inactivation of pathogens, and that is viral inactivation of both enveloped and non-enveloped viruses, as well as cell-associated and latent viruses. We also have shown studies to prevent bacterial outgrowth during storage, protozoan inactivation, as well as leukocyte inactivation. The contaminant removal by washing was developed to address PEN110 and potentially PEN110 adducts, but in addition, it is able to remove soluble prion proteins. We have demonstrated removal of immunoglobulins, cytokines, and other plasma proteins. [Slide.] This depicts a slide of some of the non-enveloped viruses that we are able to inactivate. It encompasses everything from human B-19 through various different size, genome sizes of non-enveloped viruses. In each case, if you notice on the far right column, there is a reduction in infectivity as a log of 2 CID50/ml, and this represents the highest possible spike that is able to able to be added to a full red cell unit, and this is to the limit of detection of that virus, so we can't add any more virus than what is shown here. At the end of our kinetic studies, we see no infectious particles. In addition, I should tell you that all of the viral studies are done without washing, so that we are really quenching right at the end, so what we are looking at is really the capacity and the capability of the INACTINE PEN110 to do inactivation. This is our focus of today, of course, is more on the enveloped viruses. This has listed some of the enveloped viruses that we are able to inactivate. Clearly, we are able to inactivate HCV, BVDV, as Steve showed us. We have also used Sindbis as a model, but the focus today is specifically on West Nile Virus. I am not going to go through the mechanisms of action. Steve Wagner now does a better job than I on describing the mechanisms, so I will leave it up to Steve to do that, but what I would like to do is focus more on the inactivation itself of West Nile. [Slide.] Before I do that, I think it is very important to call attention to the cell-associated viruses that we have studied. We have looked at both HIV and CMV in their latent and active forms, and this is a very important point to take as we move forward to look at what some of the preliminary results we have uncovered through studies on West Nile Virus. So, our research approach started with establishing collaborations and methodologies, and we established a collaboration with Dr. Fred Brown at the USDA Plum Island Animal Disease Center, with Tom Mather at University of Rhode Island, and Robert Tesh at the University of Texas. We also have an internal focused virology team looking at West Nile Virus, and we, of course, continue our close interaction with the FDA since we are in Phase III pivotal trials on our technology. [Slide.] This was the first inactivation study that we did earlier this summer and was presented at the workshop in August. This was inactivation of an isolate from a crow from New York. It was done in conjunction with Dr. Fred Brown at the Plum Island Institute, and we added about 7 logs of spiked to full units. We saw complete inactivation within about 15 minutes, and that was to the limit of the detection. We did this on two separate experiments. [Slide.] After we completed this experiment, we realized there is some fundamental questions that we have to really start to address to have a better understanding of both the virus and what the capabilities of our technology is. So, the questions included can the West Nile Virus survive in blood under blood bank storage conditions, can the virus exist in different blood compartments, is it cell associated, is it free, can the West Nile Virus then infect human leukocytes. Of course, we are very interested in the fact that can the technology effectively inactivate the West Nile Virus in blood, and then finally, a comparison of pathogen inactivation versus diagnostic testing. [Slide.] So, the first experiment we did following that initial study was to confirm the initial kinetics, and this was done with Tom Mather at the University of Rhode Island. In this case we used isolates from mosquitoes from New Jersey and a crow from Rhode Island. They were done in duplicate with two different isolates in both cases, and you will see the kinetics here. In all cases we had inactivation to limit of detection in samples collected 24 hours post-treatment, which is part of our treatment procedure. [Slide.] We then moved to study survival, and that was survival in the virus in human red blood cell concentrates because this is the focus of our technology, and this was using a high titer viral spike. "High titer," I mean a titer that is near a log of 7. [Slide.] That is what you should see here, and this is storage time in days up to 35 days. Of course, these are pilot studies, I want to remind you, so if you were to call on Thursday, we would have the 42-day data. This is as of last Thursday. You will notice in the supernatant, there really isn't any change at all in the survival of the West Nile and red blood cells. As you will note, it appears to be a decrease in the red blood cell concentrates, and this is suggested to be a consequence of cell association, but the experiment that was done at this point, which is one of our earlier experiments, wasn't done really to address cell-associated, but I will show you that data in just a moment. [Slide.] We similarly used a low titer viral spike, this is a spike of around 4 logs, and we got very similar results, and this experiment has gone up to now about three weeks or 21 days. [Slide.] We then wanted to study more in a natural setting, so we did move to an in-vivo model, and this in-vivo model was looking at the virus in golden hamsters. I am just depicting for you here the day after infection, and you could see what the highest level of viremia, which occurs about day two-three in this particular model. [Slide.] The West Nile levels and blood fractions then from naturally infected hamsters, what I showed you on the previous slide, was carried out, and this was using a West Nile isolate from Snowy Owls that were isolated from the Bronx Zoo in 1999, which was the onset of this virus in the United States specifically in the New York area. [Slide.] If you notice here that we looked at each of the blood fractions, both whole blood, plasma, PBMCs, and red cells, and it was day three post-infection based on the model that I just showed you with either one-day storage or four days storage, and the hampster data indicate that there is a cell-associated and a virus-free form, but the duration of the viremia in the cell-associated form is currently unknown. [Slide.] We went further to look at the kinetics of inactivation of whole blood to confirm the initial cell-associated virus that I was speaking to, so we used a couple--this is a very similar model to what I described before--it is two groups of hamsters that were infected with 104 TCID50. One group, the blood was collected three days post-infection CPD and stored for four days at 4 degrees, The other, similarly collected, but after storage for just one day. Our standard inactivation technology involves 0.1 percent volume to volume with PEN110. We incubate up to 24 hours and then we did infectivity of TCID50 in Vero cells. [Slide.] This is the results of the initial kinetics. I do apologize, there is only a couple points here. I do want to remind you that it is early pilot studies, and we will be confirming the full kinetics of these studies, but you could see that within 24 hours, both groups had complete inactivation--and this is of whole blood--to the level of sensitivity of the assay, and the assay now is becoming much more sensitive. [Slide.] The next question, one of the last questions we want to answer was can West Nile Virus infect human leukocytes, and we did a study obviously with media control, non-simulated PBMC, an IL-2 simulated, but the most important fraction that we studied here was a monocytic cell called THP-1. We collected samples once a week and fed fresh media new cells if it was necessary to retain the cell level consistent throughout the samples, and then we tested for infectivity of the cultured cells after trypsinization and washing, and that is a very important point. The trypsinization is very necessary here to remove any free viruses, so specifically we are looking at what is infected in the leukocyte. [Slide.] This is the infection of West Nile Virus in human leukocytes, and the important point that I really want to draw you to is obviously in the non-stimulated cells. This is just PBMCs in the media. We see a reduction in titer, just what you would expect, but this is that monocytic cell line that I was referring to, and there is no reduction, so clearly, there is a potential through this model that one can see that West Nile infection does occur in human leukocytes. [Slide.] So, what about this transmission by the mosquito versus blood? Well, the mosquitoes, as we know, can transmit West Nile to humans, and, in fact, the mosquito contains about 103 to 104 PFUs, and this was by communication with Dr. Tom Mather at Rhode Island, who we are collaborating with. Therefore, one could envision that about 104 PFUs would be enough to infect an individual. So, a blood unit that is contaminated with 10 PFUs/ml, and it is about a 350 ml unit, would likely transmit the virus. [Slide.] So, the conclusions from these studies, and these are the questions that I started out with, is does the West Nile Virus survive in blood under blood bank storage conditions, and the answer is yes, the West Nile survives in storage conditions at least for 35 days, and the study, as I said, is ongoing. Where in the blood does the West Nile Virus exist? Can the virus infect human leukocytes, and not only is the virus harbored in red cells, plasma, and platelets, but from our studies, it also appears that it is in leukocytes. Since the West Nile is present in plasma, the leukocyte filtration will not address the health risk of West Nile Virus. Further INACTINE PEN110 inactivation of West Nile is not sensitive to the presence of these leukocytes. [Slide.] Can our technology effectively inactivate the West Nile Virus in blood? The answer is definitively yes, INACTINE PEN110 can effectively inactivate even 1,000 to 100,000-fold of the amount of West Nile Virus that has been reported during human infections in the literature. Insofar as I have demonstrated to you, we have tested four different isolates, and as Steve Wagner told us earlier, we don't know yet what the log reduction is necessary. That is why it is very important for us to have a tremendous fold inactivation above what we think could be occurring. [Slide.] Obviously, there is additional studies that are ongoing, but for a moment I would like to compare for you pathogen inactivation versus diagnostic testing. This is a paradigm scenario that one would consider. Obviously, there is a donor, and the concern is donor to a recipient. Right now the donor questionnaire, as we heard yesterday, there is about 80 percent that are asymptomatic or there is no specific risk factor, so it is very hard to identify who that particular donor would be and how to exclude that donor from the pool. We talked yesterday about West Nile donor screening and there is going to be a lot more discussion. It was a very good presentation yesterday and a lot of wonderful interaction, and we were pleased to be party to that, but some of the questions that we have seen come up and are addressed as we move forward in the future, is there is very low viremia. There were questions raised yesterday about that, so there is a potential for false negative. There is a high infectivity, which means there is a very low infectious dose that is potentially required. It appears from our data that the virus is cell associated, can NAT address a cell-associated virus? What about the persistence of cell-associated virus? It is very different. Then, finally, of course, the IgM antibody, how do you address a window period with a test of that nature? One would offer an alternative. I believe that alternative appropriately could be pathogen reduction, and not only does it inactivate West Nile Virus, but in addition, it offers a broad spectrum inactivation of other viruses, parasites, bacteria, and even leukocytes. Our concern, of course, together is that recipient, that recipient who potentially is a very high risk. We understand about 20 percent of the blood transfusions occur in immunocompromised patients and about 70 percent are in patients that are 65 years or older, so it is very important to address this specific population. I offer that just as a point of interest and discussion, and it is based on a lot of the interaction in the discussions we heard yesterday. I thank you very much for your time. [Applause.] DR. FARSHID: Thank you, Dr. Alford. The last speaker for this session is from Gambro. Dr. Ray Goodrich was supposed to give the presentation, but he called in sick, and Robert Antwiler will substitute for him, and they talk about their technology using riboflavin. Gambro DR. ANTWILER: Thank you. I am going to be speaking today on the reduction of the West Nile Virus in packed red cells, single donor platelets, and plasma using the riboflavin and light technology. [Slide.] As Steve Wagner showed earlier, this is a picture of the riboflavin molecule. Riboflavin is commonly known as vitamin B2. It is an essential nutrient and it is a dry powder in the solid state. We use it packed as a liquid in a saline-riboflavin solution. [Slide.] Riboflavin has the following absorption curve. It absorbs in both the UV and in the visible region. In both regions, it activates in the same mechanism. We use the UV region for treating platelet and plasma products, and we use the visible region for treating the red cell products. [Slide.] I have included a picture of our illuminator. The illuminator is this white box that has a drawer that you simply pull out. You place your units to be inactivated on the shelf. The shelf oscillates back and forth to provide mixing of the product. In use, of course, you close the door, you turn the switch on. It starts the process, monitors the amount of light delivered, and then stops automatically when the proper light delivery has been achieved. We do have a system for recording the data and logging all of the parameters during the inactivation process. [Slide.] Today, I am going to focus on the West Nile Virus study that we just recently did. This study used the TCID50 method using Vero cells. There is some additional analysis in process using the TaqMan PCR testing. That is underway and I do not currently have the data to share with you on that. All of the procedures and the assays were performed at the CDC laboratories in Fort Collins, Colorado. [Slide.] On the West Nile Virus plasma study protocol, we used the UV process. We did a kinetic study. The energy points that were chosen for this study were based upon our previous experience with BVDV. I am going to show you one slide just to illustrate where and how we pick those energy points. We did an N of 3 for plasma. The riboflavin concentration was our standard 50 micromolar, and the product volume was 250 ml of plasma. [Slide.] This is the curve I said I would show you. Plotted here is data from previous viral inactivation experiments that we have done, plotted as the log virus, and this is log/ml reduction versus the energy delivered. The purpose of this graph is simply to show you the BVDV, which is where we ended up picking our energy points from, fairly linear kinetic down to the limit of detection and then it becomes flat. [Slide.] This is the data using the same energy points that we had chosen from the previous BVDV data. This is the N of 3 for the log reduction of West Nile Virus in the plasma product. Again, this is plotted logs reduction/ml. The open symbols represent data at the limit of detection. You can see that we do reach the limit of detection, around 6 joules/cm2. [Slide.] We followed that with a study looking at the kill of West Nile Virus in platelet products. Again, the platelet product uses the UV process. We did a kinetic study. Again, the energy points were based upon our previous experience with BVDV. Similar conditions as with the plasma, N of 3, 50 micromolar, riboflavin concentration and a product volume of 250 ml of platelets. [Slide.] The kinetics for the platelets is very similar to that with the plasma, reaching the limit of detection around 8 joules per cm2. Again, log reduction/ml is the expressed reduction factor. [Slide.] We then did a study of the West Nile Virus reduction in our RBC products. That uses the visible light process, similar experimental design, kinetic study, energy points chosen based upon the BVDV experience, N of 3. The riboflavin concentration with our RBC product is 500 micromolar, and the product volume was 266 ml at a hematocrit of 30 crit. [Slide.] Shown here is the data for those three experiments. You can see that we achieve log reductions in the 4 1/2 to approaching 7 again depending upon the limit of detection here. [Slide.] In conclusion, we have shown that the West Nile Virus is reduced by the Gambro inactivation system, greater than 5.1 log/ml inactivation in plasma products, greater than 4.8 log/ml inactivation in a single donor platelet product, and greater than 4.0 log/ml inactivation in packed RBC products. We would like to acknowledge the cooperation of the people at the CDC and Fort Collins, and thank you very much. [Applause.] DR. BARBOSA: The morning session on pathogen inactivation targeted at West Nile Virus was predominantly focusing on cellular components of blood. Now, we are going to switch to plasma, viral clearance methods in plasma and plasma derivative products. Before we start I wanted to make a request of the speakers to limit their time to 10 minutes, no more than 10 minutes. We are going to start with Dr. Dominique Pifat from Bayer Corporation. She will be talking about the role of model viruses in current inactivation studies.
Viral Clearance Methods in Plasma and Plasma-Derived Products
DR. PIFAT: Thank you. Good morning. I am going to be speaking to you this morning on behalf of the Plasma Protein Therapeutic Association, and I would like to thank the FDA for giving us an opportunity to speak this morning. [Slide.] I am going to be talking about manufacturing processes for plasma-derived products. We have the opportunity and the obligation to demonstrate that these manufacturing processes can provide effective inactivation or removal of both enveloped and non-enveloped viruses. [Slide.] Now, when we evaluate our manufacturing processes, and I am going to concentrate this morning on enveloped viruses because obviously, West Nile Virus is an enveloped virus, and when we develop manufacturing processes and we evaluate them for their ability to inactivate or remove viruses, we choose a panel of viruses that are very varied in terms of their genome, RNA, DNA viruses, their size, their shape, and this is deliberately done to address not only the removal of inactivation of known pathogens, but also to potentially address unknown or emerging viruses. So, the enveloped viruses, of course, that we choose in our studies are different, but they have, of course, their envelope in common. [Slide.] One of the reasons that the envelope is an interesting target, of course, is the lipids from the envelope are not derived or not coated for by the viruses themselves, so they are less susceptible to antigenic variations, so the envelope is usually the target of inactivation. We have shown that actually, inactivating the envelope is very robust and effective for a whole variety of viruses. Again, the inactivation processes are not influenced by subtle antigenic variations. [Slide.] When we look at inactivation of envelope viruses we have a number of tools at our disposal. The most common tools that are used throughout the industry are solvent detergent, heat pasteurization, caprylate, low pH. Now, most of the manufacturing processes include at least two significant inactivation steps for envelope viruses, and by "significant," we mean that they provide at least 4 logs reduction and overall our manufacturing process provide at least 10 logs reduction for envelope viruses. [Slide.] I am going to address now the use of model viruses to evaluate the safety of the products. This is a compilation of published data with a whole variety of envelope viruses, clearly of different families. Some of them are Flaviviridae, but there are other families represented here clearly. The three processes that are evaluated in these data are pasteurization, solvent detergent, and caprylate. You can see that these three different methodologies provide reduction in all these different viruses from all these different families to the limit of detection in all of these experiments, so a priori there would be no reason to believe that West Nile Virus would behave significantly differently from all of these other envelope viruses. In another talk, my colleague from Baxter will actually confirm that that is the case, that West Nile Virus behaves very similarly to all the viruses described here. [Slide.] We have seen this slide more than once already during these two days. Of course, Flaviviridae are broken down into three different genera. One of the viruses of interest in the industry is clearly hepatitis C virus, and what is commonly used in industry is bovine viral diarrhea virus as a model to look at the inactivation or removal of hepatitis C virus because hepatitis C is not easily grown in tissue culture. So, by this reasoning, if you use a pestivirus as a model for hepatitis C virus, there is some logic to using bovine viral diarrhea virus as a model for West Nile Virus, which, of course, belongs to another genus. [Slide.] These are some of the data that are compiled just for comparison purposes for the inactivation or removal of HCV, so since we can't cultivate HCV, various viruses have been used as models. BVDV, of course, I already mentioned, yellow fever virus, Sindbis virus, tick-borne encephalitis virus. If you look at this table and look at all these Flaviviridae, you can see that again pasteurization, solvent detergent, and caprylate provide inactivation to the limit of detection for all these different Flaviviridae. [Slide.] Justification for using BVDV as a model for West Nile Virus, clearly, BVDV closely resembles flaviviruses including West Nile Virus. BVDV has been successfully used as a model for HCV inactivation. There is a very large body of inactivation data that exists for BVDV, and these data should provide assurance that West Nile Virus can be inactivated during plasma-derived manufacturing processes. Again, we will show that there is some evidence that that is actually correct. The second speaker in this series is also going to present a compilation of the data with various model viruses that have been obtained throughout the industry, so all of the members of the Plasma Protein Therapeutic Association have provided data to show the effectiveness of our processes. [Slide.] There is always potential limitations to using model viruses, but actually, the safety of our products is really the evidence that the use of BVDV as a model for HCV, for instance, is a valid thing to do because of no seroconversions to HCV. Again, it is the safety record of our products that validates the clearance studies using model viruses. [Slide.] The safety of our biological products today is assured because we can demonstrate clearance with model viruses when it is not possible to use some of the viruses of interest. [Slide.] I have told you that we have done preliminary, we are going to show preliminary data on West Nile Virus itself, but the PPTA members will continue to conduct studies on the specific viral inactivation and removal steps in the manufacturing processes for plasma derivatives with West Nile Virus itself to confirm the reliability of the predictions that are based on model virus studies. Thank you. [Applause.] DR. BARBOSA: Additional data on the model viruses will be presented now by Dr. Albrecht Groner from Aventis. Dr. Groner. Robustness Data from Model Viruses DR. GRONER: Thank you very much for the opportunity to discuss with you the effort of the PPTA member companies to demonstrate the effective virus inactivation and the removal capacity by selected steps. [Slide.] The virus validation studies are essential to document the effective virus inactivation capacity and therefore we have to evaluate a wide range of viruses with known physical/chemical properties. There will be, of course, a quantitative estimate of the overall virus reduction capacity of the process, and that will also have an indirect evidence that the process will inactivate or remove also novel or emerging viruses. [Slide.] In these viral validation studies we are performing, we are using, of course, known viruses, and these known viruses should resemble the viruses of interest. They should represent the widest range of physical/chemical properties of viruses, and they should include, of course, laboratory strains which can be easily assessed and used. We have to consider that, of course, these laboratory strains have or mainly have different properties from the natural occurring viruses and therefore we conclude that all viruses used in various validation studies are in principle model viruses. [Slide.] As I already said, we have to use in our validation studies, viruses which can be reliably assayed in an infectivity assay. Therefore, we have to have the appropriate system in place. This system is, of course, reliable, sensitive infectivity assay, and the virus we are using should go to high titers to document high virus inactivation capacity of the manufacturing process. If there are two similar viruses available which could be used because they represent, say, the target virus, then, the more robust virus should be used, the more resistant one. There is also no question about we are looking for the lowest human hesset [ph] virus to avoid some negative impact on our stuff. [Slide.] The viruses which are used within the different companies are, of course, viruses HIV-1 and HIV, and the specific model viruses - BVDV and sometimes Sindbis virus as a model for HCV, as well as the parvoviruses from porcine and canine as a model for parvovirus B-19 and as a nonspecific model virus, often the herpesviruses are used in here especially the rabies [?] virus. [Slide.] Now, I would like to discuss with you the compiled data from PPTA member companies on the different products and to document the robust virus inactivation capacity after flaviviruses. The albumin, as you can see here, is used in the pasteurization at 60 degrees centigrade, and there were different concentrations from 3.5 to 25 percent protein, and under these production conditions, BVDV, as well as tick-borne encephalitis virus, as well as Sindbis virus are removed below the detection limit. The range you can see here is the range which are supplied by the different bumper companies, and that is just due to the fact that the amount of viruses added in the process is different when you have a higher titer to spike this definitely the virus reduction capacity will be higher in that respect. We are looking now for out of specification temperature and an intrastabilizer concentration, nevertheless, BVDV will be inactivated below detection limit. [Slide.] Similar data are true also for the SD treatment of the factor VIII product. It is also inactivated below detection limit, the BVDV, as well as the Sindbis virus using robust conditions, with only 50 percent of the SD concentration. We also have definitely a very effective BVDV inactivation. [Slide.] That is just to demonstrate the BVDV as well as the Sindbis virus inactivation capacity within a very short period of time we are reaching the limit of detection of the assay. [Slide.] Now, when we are looking for very robust concentration, I just showed you that 50 percent of the SD concentration will have no impact on the virus inactivation capacity. Here, in this slide, you see that even one-third has no impact on the excellent virus inactivation capacity. When we have 1 to 9 and higher dilution of the SD concentration, then, we see an effect at that concentration which is more than out of specification will be not inactivating the BVDV as expected. [Slide.] When we are now going for factor VIII and pasteurization, we have again BVDV, yellow fever virus, Sindbis virus, as well as Semliki Forest virus, inactivated below detection limit under all conditions, and that is also true for robustness when we are using temperature below the specification and stabilizer concentration above specification, as well as different protein concentrations. [Slide.] That is demonstrated here. In a graph, you see within fairly short period of time, the virus is inactivated below detection limit, and you can see from the graph that there is room enough to further inactivate an even larger amount of virus. [Slide.] When we are now going for factor VIII and by heat treatment, you see also under all conditions excellent virus inactivation capacity. [Slide.] That is also true for factor IX solvent detergent treatment, BVDV, as well as Sindbis virus. [Slide.] And in pasteurization of factor IX also excellent virus inactivation capacity for BVDV, as well as tick-borne encephalitis virus. [Slide.] In immunoglobulin preparations, we see again the same data. The solvent detergent treatment is inactivating the flavivirus model and the togavirus, which is used, under all production conditions, which differ, but they are always inactivating the virus below detection limit. [Slide.] We are now going for immunoglobulin and pasteurization, again, we see an excellent virus inactivation capacity at what action conditions, as well as be on the specification of the production. Here is just a graph showing the different parameters which were tested. That means standard stabilizer concentration, increased stabilizer concentration, decreased temperature, different pH values, as well as different concentrations. In all conditions, these virus inactivation capacity is very effective for BVDV. Now, we have, despite virus inactivation, we have also virus removal, dedicated virus removal steps in our production process, and there is, for instance, the nanofiltration. [Slide.] You see here different products from different companies. You will see different filter devices, and you will see here again an excellent virus removal capacity for BVDV, always below the limit of detection of the assay. [Slide.] When we are now looking for further manufacturing steps, which are used to purify and concentrate the protein, as caprylate and octatonic acid treatment as acetone suspension and chromatography, whether it is affinity chromatography or hydrophobic interaction chromatography, as well as the cold ethanol precipitation steps, we have always very good virus removal capacity of these selected steps. They may not be effective as the dedicated virus inactivation and removal steps, but they certainly contribute to the virus safety. [Slide.] I now would like to conclude that BVDV, tick-borne encephalitis virus, as well as yellow fever virus and Sindbis belong to the Togaviridae, are model viruses closely related to the West Nile Virus, and these are inactivated as we documented in our virus validation studies, very effectively, and we could demonstrate in the robust inactivation and removal of these enveloped viruses by the plasma derivative manufacturing processes. Thank you. [Applause.] DR. BARBOSA: Thank you, Dr. Groner. The last presentation on model viruses for validation and evaluation of inactivation procedures will be given by Dr. Thomas Kreil from Baxter Bioscience. I invite all the speakers to come to the podium immediately after this last presentation for the general discussion. Thank you. Equality of Model Viruses and Current Data for WNV DR. KREIL: Good morning, ladies and gentlemen. [Slide.] I would like to use the next 10 minutes or so to share with you the results of an investigation that we have performed at Baxter Bioscience in verifying that West Nile Virus indeed is just one of the flaviviruses and particularly so with respect to its being inactivated through the major inactivation processes which have been implemented by the plasma products industry in their respective manufacturing procedures. [Slide.] As you have heard from the two previous speakers, this industry really knows a lot about flaviviruses in general, obviously driven by hepatitis C virus here, a virus of potential concern for transfusions, and a virus which is unfortunately not available to us experimentally. That is why this industry has resorted to the use of model viruses, most notably the bovine viral diarrhea virus here, but then also some similar viruses, such as tick-borne encephalitis virus here, which really is almost a twin brother of West Nile Virus. [Slide.] This is the result that we have generated at Baxter for these viruses. I think at first glance it is fair to say that all of these viruses are very susceptible to the virus inactivation processes that this industry typically uses, that being, for example, pasteurization here for human serum albumin, then solvent detergent for factor VIII, also solvent detergent for intravenous immunoglobulins, and then vapor heating for here, for example, are factor VIII inhibitor bypassing activity product. It is also important to note that when you compare the inactivation of the different viruses used, that being flaviviruses--here are some related togavirus--they are all very similar one to the other. [Slide.] Now, we tried to verify that West Nile Virus indeed would not be behave any differently to what we knew about the other flaviviruses and therefore we have obtained a West Nile isolate from the 1999 New York outbreak. The virus was isolated from the liver of Snowy Owl and provided to us by Dr. Robert Shope, and I want to thank him for that. The virus was then characterized by sequencing the genome, and the spiked virus that we have used for subsequent studies was prepared as a supernatant of Vero cells, serum-free Vero cells in this instance. The assay we have used is a Vero cell assay, and the cytopathic effect can already be read after three days. [Slide.] This is some results from the assay setup experiments. You can see that we have had eight operators titrate the virus on seven different days, and as you will be able to appreciate, there is no variation almost between days and also between operators. The titer of the stock virus available to us at 8.9 log 10 or, in other words, a billion infectious units per ml is very significant and lends itself ideally to do some studies with that virus. [Slide.] This is the same table that you have seen before now including the data that we have obtained for West Nile Viruses in these processes here. You can see that West Nile Virus was inactivated to below the limit of detection in all instances, just as we would have predicted by the data known to us from other flaviviruses. Well, obviously, the reduction factor is not everything you can investigate to make a side-by-side comparison of the different flaviviruses, and so we went into some further detail. [Slide.] You can see here we used a downscaled version of our large-scale manufacturing process for albumin. This is only the critical process parameters for pasteurization, that being the process temperature here, the treatment time, and then the protein concentration, the lowest and the highest end. In our downscale, we have used a temperature just below the process temperature and the large scale, the shortest possible time, and then we have bracketed the protein concentrations. [Slide.] This is what the results look like for a number of different flaviviruses. While this is a pretty busy slide, I think the message is quite clear in that all the viruses that we have tested are very, very quickly inactivated to below the limit of detection, that being probably the most widely used model virus BVDV, then followed up by tick-borne encephalitis virus, a twin brother, as I have mentioned before, to West Nile Virus, now complemented by West Nile Virus data itself. Here we have a togavirus in there, a Sindbis virus, but again no difference to be seen between all of these viruses. [Slide.] This now is another albumin preparation that we commercialize, a slightly different composition of intermediate, and therefore, we tried to also generate some data on this preparation again using the same downscale setup. [Slide.] As you can see here, again indistinguishable kinetics of inactivation between bovine viral diarrhea virus and West Nile Virus itself, also not dependent on protein concentration in this specific process. [Slide.] This is another product of ours, the anti-inhibitor product FEIBA, which is subjected to vapor heat treatment process, and I need to walk you through the manufacturing process just briefly here. The product is heated at 60 degrees Celsius for 515 minutes. Then, it is brought up to 80 degrees Celsius for another 65 minutes, and all that with the lyophilized product at between 7 and 8 percent residual moisture content. The downscale has used temperatures just below that specified for manufacturing, the shortest possible incubation times, and then at a residual moisture content of the lyophilized product similar to the manufacturing process. [Slide.] This is what the data looked like. This is again the process. You take the product up to 60 degrees Celsius, treat it in this instance for 510 minutes. Then, we bring it up to 80 degrees Celsius for another hour, and then we cool it down. The first thing I want to mention is that all of the viruses tested in this downscaled model were inactivated to below the limit of detection already within the 60 degrees Celsius treatment phase. In other words, the additional 80 degrees Celsius phase here only provides additional safety margins to this product. Also, you should be able to see that all the viruses were inactivated with very parallel slopes of inactivation, which indicates that they have a very similar sensitivity to this particular heat treatment. The reason why they are only parallel is that for tick-borne encephalitis virus here, in orange, and then West Nile Virus here, in red, we have higher spiking titers here in the range of 8 logs for the spiked product, which is some 2 logs higher than what we were able to use for bovine viral diarrhea virus earlier. [Slide.] Solvent detergent, yet another inactivation procedure widely used throughout the industry, the key process parameters being the temperature at which the treatment is performed, the time obviously, and then the concentration of SD chemicals. This is really an important point to notice because if we use the nominal concentration, such as those given for manufacturing, then, what we get is instantaneous virus inactivation, so you are not able to demonstrate the kinetics of inactivation. That is why in the downscale, we resort to using a drastically reduced SD concentration, here, only one-tenth of the nominal concentrations of the components for this SD treatment. [Slide.] What you can see is that West Nile Virus still, despite the fact that we are using only one-tenth of SD chemicals in this instance, is completely inactivated instantaneously. The "B" here indicates that we have used bog titrations, so that meaning 10-fold bigger sample sizes to determine whether there was any residual infectivity, and there was not. In this instance, really, BVDV is somewhat more resistant to that treatment although I should reemphasize that at the nominal concentrations, BVDV follows a course of kinetics just like this, so this is 10-fold reduced SD chemicals. [Slide.] Here, we have another SD treatment as applied to our intravenous immunoglobulin product Gammagard. Again, you have given here temperature and time of treatment, which are similar between downscale and manufacturing procedure. Here are the SD chemicals and again we have to use a drastically reduced concentration of SD chemicals to be able at all to demonstrate kinetics. In this instance, we have reduced it by 20-fold from the nominal concentration, and the reason why we went further down even here is that this is a tricomponent SD treatment using two detergents and one solvent here, which is even more effective than the SD treatment that I have shown to you before. [Slide.] This is what the result of this experiment is. You get instantaneous inactivation almost for both West Nile Virus and BVDV even at a 5 percent only SD chemicals, so I think very nicely supporting what it is thought to be the code case anyway, that this is one of the most effective inactivation procedures really. [Slide.] This is the results now from a study which has been performed by Alpha Therapeutic Corporation on their alpha-1 proteinase inhibitor product. The step investigated here is a 15 Asahi nanofiltration. You can see a comparison here between manufacturing parameters and then here the equivalent downscale parameters, and again the two mimic one and the other very nicely. This is the results obtained in a run using, as an assay system, the NGI SuperQuant PCR system that has been introduced to you yesterday, and as you can see, from the spiked greater than 9 log genome copies per ml, and that is log tens, I need to emphasize, the filtered alpha-1 product is below zero log, 10 copy numbers, giving you a very high reduction factor of greater than 9. [Slide.] As of the end of last month, the FDA has provided guidance to the industry in relation to the blood product safety, and I would like to quote from that. "The FDA has reviewed the viral reduction processes in place for all plasma derivatives. The methods in place have been validated to inactivate flaviviruses related to West Nile Virus." [Slide.] I guess it is fair to say that the West Nile data presented here support the FDA's conclusion and that it verifies that West Nile Virus does not behave differently than other flaviviruses. Also, the concept of using a range of physicochemically diverse model viruses for the validation of virus reduction steps has been verified in that the behavior of a virus of interest, which here obviously is West Nile Virus, has been adequately predicted. Thank you very much. [Applause.] DR. FARSHID: Thank you. Now, we open the floor for questions and I invite all the speakers to please come up. General Discussion DR. FARSHID: While people are making their way to the microphone, I would like to start with a question that is in reference to use of the model virus for performing inactivation studies. The principle that we apply in FDA in evaluating viral validation studies is that if the relevant virus is available and is feasible to use, i.e., if there is a high-titer stock and there is a quantitative infectivity assay present, the relevant pathogen must be included in validation studies. Use of specific model viruses should be justified, for example, in case of HCV or HBV, because the culture is not available, therefore, we basically have no choice but to use the specific model viruses which resemble HCV or HBV. Therefore, we encourage the manufacturer to validate their process and show its capacity to inactivate, and one other reason for that is because these viruses, they behave differently, and this has been shown even if you use one virus, for example, in the case of hepatitis A, it has been shown one virus with different strain may behave differently. Therefore, to come out with high degree of assurance that the manufacturing process or the inactivation process can inactivate the virus, that specific virus needs to be included in validation studies. Dr. Tabor. DR. TABOR: Ed Tabor from FDA. To underline what Dr. Farshid just said, I would just like to comment on a statement made by Dr. Pifat. Dr. Pifat said that the safety of her company's product proves that BVDV is a good model for showing the inactivation of hepatitis C virus. Now, whereas, BVDV has been used as a model for the inactivation of hepatitis C virus because hepatitis C virus cannot be grown in cell culture, the fact that her company's product is safe does not prove that BVDV is a good model. It only proves that her company's inactivation procedures are effective for eliminating the hepatitis C virus. DR. NAKHASI: Hira Nakhasi from FDA. I think I would like to congratulate all the speakers this morning. We heard very excellent presentations from how the industry has really gone and done excellent work on inactivation process, and I think I would like to thank you, congratulate you all. The question I have is well and done, you know, the activation processes work very well. How does it affect the product itself, because many of these are, as some of you alluded to the fact that yes, it intercalates the DNA, you know, many of them are very dangerous stuff, and I guess Dr. Wagner alluded to the fact that there may be some adverse effects. So, the question to the panel is have you looked at any adverse effects or anything like that of the products after treating with this stuff. DR. ALFORD: What we have done are several things. First, we have conducted two, a Phase I and a Phase II clinical trial, and we are in pivotal Phase III clinical trials to address any related safety issues as it applies to man. We have not seen any safety issues at all or adverse effects in the studies we have done to date. Then, in addition to address what you are talking about in the quality of the red cells itself, we have done survival studies and additionally to that, we have done a series of studies on the red cells from osmotic fragility to ectocytometric assessments, even we have subjected it to sheer force under simulated extracorporeal circuitry, so we have looked at it as much as we can right now prior to completing our Phase III trials to see what the quality of that red cell is, as well as any adverse effects. Of course, related to that is an extensive toxicology program which we have published on in part and presented, and which we are completing, so those are I think the three major focuses that we are relying on to answer the question that you are speaking to. DR. ANTWILER: I can speak to the case of riboflavin. It, of course, is a natural vitamin that we all have. The photo byproducts of riboflavin are in each and every one of our bodies right now and it is produced every time we walk out into the sunlight, lumichrome being the photo byproduct, so we have currently natural mechanisms of dealing with its photo byproducts. We have done extensive toxicology, genotox mutagenicity testing in addition, and all of that has shown negative. So, I hope that is at least a partial answer to your question. DR. LIN: I would like to make a comment on the Baxter-Ceres system. We have completed the Phase III clinical trials for both the platelet and the plasma product in the Phase III clinical trials for the red cell system is ongoing, and much of the data for platelet and plasma have been published. As far as the efficacy of treated platelets and plasma, we have demonstrated almost equivalence in terms of stop bleeding for the platelet product although a secondary endpoint platelet count increment demonstrates slightly lower for the treated products. My second comment on the toxicology study is that we have conducted a very comprehensive list of studies for both the amotosalen and S-303 that includes the acute toxicity, genotoxicity, total toxicity for amotosalen and carcinogenicity, and also all the absorption distribution, metabolism, and excretions in both animal studies and in clinical trials, and we have not seen relevant results in the treated. DR. GOLDING: I just would like to continue on the same vein as Dr. Hira Nakhasi. Clearly, for these products to be approved, one would have to do some kind of risk-benefit analysis. On the one side, we are talking about removing West Nile Virus or other viruses, and on the other side, we are talking about what happens to the products in terms of safety and efficacy. The third part of the equation that we haven't discussed yet, it is on the agenda, is what is the prevalence of West Nile Virus in these products in the first place. So, I think somewhere along the line, maybe not today, we are going to have to take all this information and look at it in a comprehensive way and make a risk-benefit analysis. The other point I would like to make, one of the speakers on pathogen inactivation referred to the fact that while if you can remove the pathogen from the blood, maybe you do not have to worry about other things. I am not sure if that was really the implication, but that is how it came across to me. I think that the FDA's approach to it and I think part of the industry's approach to it, as well, has been that we have to look at multiple levels, we cannot just choose one approach when it comes to viruses especially life-threatening viruses, that we have to look at the donor screening and maybe pathogen inactivation and viral inactivation removal steps, testing of minipools and plasma pools and maybe final product, and not look at just one part of the process and say that is sufficient to provide safety in this arena. DR. DHAWAN: I would like to make two comments actually that relate to the questions by Dr. Nakhasi and Dr. Golding, again, the safety issues. I can see these cross linkers, 303, INACTINE and PEN110, and while they can be used to inactivate West Nile Virus and other pathogens, in the systems like plasma and serum, I don't see how it can be used to inactivate whole blood. In red blood cells, okay, but if they intercalate the nucleic acids, when you feed the whole blood with these inhibitors or intercalator agent, I don't see how you can use the whole blood for transfusion. That is one. Number two, have any of you studied the effect of all these agents on differentiation of bone marrow, progenitor cells or other cell types? DR. ALFORD: Maybe I could start. I presented some preclinical data in the hamster on whole blood. I am not suggesting at all that INACTINE PEN110 is the mechanism to inactivate pathogens in whole blood. That product was specifically designed to address pathogens in red cells. That was just a preclinical experimental design, so if I was unclear, I apologize. The technology is only for red cells as an example. The second part of your question, we have not yet addressed a bone marrow in the manner you are speaking, but it is a very good question. DR. DHAWAN: One of you, I don't remember who mentioned about the infection of leukocytes. DR. ALFORD: That was me. DR. DHAWAN: And you said, well, the filtration of blood or centrifugation and filtration will not be an effective way of eliminating viral components or virus, whole viruses, then, how would you eliminate cell-associated virus with any of these agents? DR. ALFORD: I don't think I was addressing filtration or centrifugation. What I was suggesting is that if the virus, and there is further confirmation that would be necessary that it is cell-associated, my questioning is if we do a diagnostic screening, will that be able to, as we move forward, address the cell-associated virus. It may be that there is more than one barrier that is necessary, and maybe screening is one barrier, and a pathogen reduction may be an additional barrier, to have both address the safety issue. That is what my suggestion is. DR. LIN: I did not have time to present all our data. Maybe results have been published or presented elsewhere. The Helinx technology has been shown to inactivate not only cell-free viruses and cell-associated viruses, as well as a provirus. My second comment, the question of whether any of these technologies works for whole blood, initially, we are developing a separate technology for platelet products and red cells. Our company is working on a methodology to treat whole blood, and as soon as the results are available, you will hear about it. DR. PETERSEN: Lyle Petersen. I have two questions. One question is, is what is the anticipated cost of implementing these technologies, and the second question is, do these technologies produce a toxic waste problem? DR. ANTWILER: I can speak for the riboflavin technology. I cannot tell you what the cost is on that, but it does not produce any toxic waste products. DR. ALFORD: Speaking on INACTINE, we are developing right now a pharmacoeconomic analysis to determine what the costs are as we are finishing our development of our process. As soon as we have that ready, we will be glad to share it with you. On a waste disposal, if you will, we do have disposables. I believe all the technologies would have a disposable involved at some time, and I guess I would consider that a waste, because it is a disposable that has seen blood. We do washing and, as you know, we have some liters of wash solution, and what we are addressing is quenching that solution, so we could go directly to drain, but otherwise, it is really a disposable that you would still be looking at for a potential waste product. DR. LIN: The comment I would like to make is that we have MSDS information for amotosalen, and since it is treated in a closed system, the way to treat the waste is no different from your biological waste. DR. FARSHID: Dr. Wagner, would you like to comment on that question? DR. WAGNER: I think this is an issue that needs to be looked at pretty carefully by parts of the government that are involved in environmental safety, both with respect to workers, as well as what goes down the drain. I think these are issues that are probably being grappled at to some extent now, but a lot more thought and work needs to be placed on these issues. DR. WILKERMAN: I would like to come back with a short remark and a question to the comparison of the properties of the behavior of some viruses and the question of model viruses. I think that Thomas Kreil presented very nice data I think demonstrating that West Nile Fever Virus is as sensitive as we thought or as we predicted from other studies and that we can assume that the manufacturing processes which are in place for inactivation are really factors. I think this comparison of the kinetic even at conditions which are a little bit strange to these procedures give excellent information about it. I have two questions. First, can we expect that such data are soon published, and the second question is what is with the other methods which are working according to another mechanism? I mean INACTINE and psoralen and riboflavin. If you look directly on the kinetic, do you see then differences between West Nile Fever Virus and BVDV or the other viruses which have been used, and I mean mostly BVDV has been used, do you see differences if you look really in detail on the kinetic? DR. ALFORD: I will answer your second question first. We see a little bit of difference between the kinetics of BVDV, Sindbis, as examples, HCV if you are going to look at it, but specifically BVDV and the different isolates, in fact, on the West Nile Virus that we are using. We have used four different isolates to date. As I hope I portrayed, our information is pilot, we have just started. We hope to get much more information in the next few months. On a publication perspective, one of our goals is to publish it as much as possible. In fact, in the most recent issue of Transfusion, we have three publications in that. One of our three collaborators is struggling very quickly because he wants to be the primary author, and we are going to get that paper out as soon as possible, so that we are sharing not just the results, but the methodology that is used. So, if there is comments on the methodology or critique on our methodology, we would be glad to discuss that further. Thank you. DR. WILKERMAN: Thank you. And PPTA, will PPTA publish it, the data of PPTA? DR. KREIL: I can obviously not comment for PPTA, but the data that we have seen from Baxter Bioscience, they will be published very soon. DR. FARSHID: Dr. Lynch. DR. LYNCH: Tom Lynch, Clearant. I had an editorial comment actually on the use of model versus relevant viruses. Your statement that relevant viruses must be used when they exist and when they are available as laboratory strains is I think an accurate reflection for the most part, but not entirely true, a little bit oversimplistic. There are some viruses that are simply considered non-issues for plasma derivatives even though they exist in the source material or blood itself. Some are simply not considered important enough to assess and are not looked at. There are many, many viruses that one could isolate from a human blood or plasma donation if one wished to, and the idea that each and every one of those, where they are available for cultivation in the laboratory, must be evaluated, is I think too broad a statement, but the principle derives from a time when the tools that were available to assess these methods were far less sophisticated than they are now. The focus 20 years ago on these very specific risks has shifted today to a much broader assessment of the capabilities of these technologies. This idea of robustness that came out of Europe to assess the breadth of effectiveness, not just the effectiveness to one, or two, or three or four specific viruses has proven to be very valuable, and the data that I have seen today on West Nile confirms the value of that approach. The fact that these techniques have been evaluated for their robustness in this sense of being broadly effective, and then the confirmatory tests that says this virus does, in fact, behave as we expect it to, would suggest to me the need to assess whether or not all the resources to revalidate all these products and all these processes is putting the resources where they matter the most. There are other problems, there are other viruses that may be more important to focus on if West Nile does, in fact, prove to be representative similar to other flaviviruses. DR. FARSHID: If I may say that I did not imply that every relevant virus and every virus was confined in plasma should be included in validation study, but if we determine a virus is relevant and pathogenic, then, and is available to do the experimentation, I don't see any reason not to use it. So, if we are faced with an application and the manufacturer comes and say we want to do the validation studies, why not ask them to include West Nile Virus basically, as I indicated, to increase the degree of assurance in regard to capacity of the end process to inactivate the virus. It is pathogenic and a virus of concern, and should be included. If it is not feasible to do that technically, that is a different question. DR. WILKERMAN: I completely agree with your position in general. I think in the case of West Nile Fever Virus, it is I think relatively difficult to give the general or to require in general that all methods are revalidated toward this virus. I think it lasts relatively long. I mean if it is possible to compare those viruses and to demonstrate by, of course, not only with one method, to demonstrate on a broader basis that the behavior of this virus is really sufficiently reflected from the model viruses which we have used already, then, I think this is also an acceptable approach. West Nile Fever Virus, we could ask the question is it possible to replace maybe BVDV by West Nile Fever Virus because, of course, if West Nile Fever Virus would have a similar behavior, maybe that is similar to HCV, we cannot check it because we have HCV not in our hand. But I think we should also consider that this virus is not so easy to handle. It is a virus which requires laboratory Level 3 for the general use, and so I think we have to be a little bit cautious with such recommendation. On the other side, we are in a status or we are in the situation that we have already I think a lot of data which demonstrates the robustness of manufacturing processes which are in place, of course. I think if it can be demonstrated that for these processes, which we use already a longer time, and for which we have some really good databases, if it is possible to demonstrate that West Nile Fever Virus fits very well into this databases and confirmed what we know already, then, from my perspective, this would be sufficient, or sufficient maybe is not a good term, maybe this would convince me at least. So, it is a little bit different if you are looking on other methods, so it should be demonstrated again, I mean if we look on other mechanisms of action, then, this should be investigated I think. But on the other side, it is impossible. I think we have to trust a little bit on some robustness studies which we have seen already if it is equivalent with the viruses, to the other virus, and can be demonstrated by basic experiments. That is my present opinion. DR. FARSHID: Dr. Busch. DR. BUSCH: Focusing on the cellular products, I think if we were three or four years down the road where we had inactivation of both the platelet and the red cell products, as well as the FFP, which I think is where we have to be to be relaxed now and not consider testing, it is interesting to think about whether we would be here today and whether West Nile would cause still a concern. I think it would. I think we have all discussed and I think at the FDA Workshop on Pathogen Reduction, there was pretty much of a consensus that once these methods are introduced, we will still need to serologically screen, we will still need to NAT screen, because your methods, although very impressive, they still may have problems with very high titer viremia, so it is really an additive safety issue. We are hearing today on the derivative side, enormous kill, and yet there is still concern, there is still debate. We are still hearing the consideration of testing of plasma for fractionation. We saw NGI's presentation yesterday, they seem to be moving toward introducing West Nile Virus NAT even though their focus is strictly derivative manufacturing. So, in just trying to think forward, I think we would still be here if all of your methods were in place and we still might need to test or at least do a lot of studies just as is being discussed for the derivative side. DR. BULT: Jan Bult, PPTA. I would like to add to the comments made by Dr. Lynch and Dr. Wilkerman. There is no way that I could have said it much better than they did, but I would like to remind FDA on the presentations that were made on behalf of the industry, where you can see that first efforts have been made to test for the actual virus, and as you have seen at the presentation of Dr. Pifat, there is a commitment of the PPTA member companies to perform additional studies, and if we add that together with comments made by Dr. Lynch and Dr. Wilkerman, then, I would encourage FDA to look at this data before final decisions are being made about full validation studies. DR. FARSHID: Actually, I need to indicate that the view I presented here is not FDA's views, they are mine, and we definitely would look at those data, and there is more deliberation needed to be done before the final conclusion can be reached whether the products currently on the market need to be validated and their validation study needed to be supplemented using West Nile Virus. In my opinion, with this data, that may not be necessary. As I indicate again, this is my view, it is not FDA's position. We have not reached a decision at the agency level how to approach that. However, the comment that I made is in regard to a new product. If a new submission is being sent to the FDA and we are asked to evaluate the model viruses which are used for doing the inactivation study, I think it is reasonable to request that West Nile Virus, which is pathogen, is of concern, and is available and can be tested, should be tested to assure that the method that is being used has the capacity to inactivate the virus. This is simply for the new application and if someone wants to do the viral inactivation. I think it makes more sense to use the real virus rather than extrapolate. We have a large amount of data which shows that the viruses, they may behave differently, and we are dealing with a complex biological product, and the manufacturing processes are different, and these manufacturing processes have their own effect on how the inactivation methodology will be effective, and this has been shown. Therefore, just to be assured that basically, that the relevant pathogen need to be included if possible. DR. NAKHASI: I think I just want to say again I would like to compliment to the industry that they have done the studies. I think nobody denies the fact, and the processes are ongoing. I think the very fact that this virus is dangerous, and somebody made a comment it is very difficult to get the BSL-3 facilities, and the very fact that this virus is dangerous and it requires BSL-3, begs the fact that it should be shown that it is inactivated in the process, and I think ensure the safety of the process even though, you know, the data showed this morning that the inactivation process is the same as the other viruses, but since again I would like to reiterate the point. As long as there is a virus available and can be grown, I think it begs to the point that it should be shown that it is inactivated. Thanks. DR. FARSHID: One more question, then we break. DR. FITZPATRICK: I was just curious about the slide on washing and the statement that further reduction of viruses or prions was provided by the washing steps when we know from the past that we thought washing removed viruses, and Dr. Alter showed that we really didn't do that, and I was just curious as to how you were substantiating that washing was further reducing the load. DR. ALFORD: I am sorry if you gather that from my presentation. I didn't speak to that washing had any effect on the viral load whatsoever. In fact, in the experiments that we do on viral inactivation, we stop the experiments by quenching, so it purely is the inactivation step. We have not studied the viral inactivation due to washing. I am purely representing it as chemical inactivation. When I talked about prions, we have done some spiking experiments with some soluble prion proteins both platelet derived, as well as the alpha and the beta recombinant forms, and with those spiking experiments, we have just demonstrated a log removal of the soluble proteins, and they seem to track with other plasma proteins like albumin and serum. So, what that demonstrates is that there are proteins that are removed, specifically, some serum proteins. We study IgG and albumin obviously also during that washing process. DR. FITZPATRICK: Thank you. DR. FARSHID: Thank you. Now we take a 15-minute break. Please don't forget to go upstairs for the next session.
VI. Proposed Studies on Prevalence in Donors
Chair: Mary Chamberland, CDC
DR. CHAMBERLAND: Thank you for your patience. In addition to a little rearrangement in the room accommodations, we have also rearranged this session a little bit at the request of the participants. This is Session VI, Proposed Studies on Prevalence of West Nile Virus, and Liana Harvath from NHLBI will be co-moderating the session. The participants have requested that instead of three individual speakers, Mike Busch is going to speak about the project that is being proposed, and he will then be joined in a panel discussion by Steve Kleinman and Sue Stramer from the Red Cross. Mike will be our first speaker, Mike Busch from the Blood Centers of the Pacific. The NIH Collaborative Donor Prevalence Linked Study DR. BUSCH: Thanks, Mary. I am happy to represent the study group, which actually formed essentially the day that the first potential transplant cases were reported by CDC, we began to call around, convene conference calls and begin the process that is still evolving as you will see to develop the best study we think we can to try to understand, not only the prevalence of West Nile viremia in the donor pool, but also scientific questions, transmission rate, cofactors that determine transmission, and disease outcome in donors and recipients. As I think you will see, it is a struggle. I remember Indira yesterday talking about the virtual repository and in a sense this study is like a virtual study because we are really trying to chase an epidemic that is very regional and very temporal, and as I think we saw yesterday, by the time the first cases were reported, the epidemic had essentially already peaked, so the effort to try to capture specimens that are optimally relevant to determine prevalence has been a challenge. I just want to first again acknowledge the work of a number of people. This effort is kind of co-sponsored by NHLBI and CDC, George Nemo and Liana Harvath from NIH and Mary and Lyle, who you know, but also Ken Clark, who is really the person at CDC who has been focused on developing it originally, as we will talk, probably later, there was a concept of a Phase I link study that Ken was driving, and then this larger link study, and now essentially these two have merged. Within the REDS radar group, I particularly want to acknowledge Simone Glynn, who is really a wonderful physician and statistician, who has driven the Westat team to put together the protocol, and then, as you will see, the study that we will be pursuing involves a combination of some existing ongoing donor/recipient cohort studies called the RADAR and TRIP study, and within that radar program, Steve Kleinman is the lead on that project for REDS and Dale Truzey [ph] who is not a primary REDS investigator, is leading the site in Pittsburgh that CDC is supporting, the Institute of Transfusion Medicine. As you will see again, the TRIPS study is kind of a clustered study within this larger donor/recipient follow-up study to Harvey Alter is leading. It is essentially a continuation of Harvey's historical donor/recipient follow-up studies, but involves much more frequent samples and characterization of genetic markers of all the viruses in the serial samples. But as we will get to, in addition to these historical repositories or these repositories that are being built prospectively, we realized quickly that the samples in those repositories would not be adequate, so we engaged Sue Stramer and Chyang Fang and Roger Dodd, and as you will see, Red Cross is contributing a large number of additional specimens for use in this study. In addition, we have additional donations that we have identified from some of the ABC Centers, Sally Caglioti from UVS, Darrell, actually representing the Life Source Chicago collections, and Mike Strong who is a representative from the Roche Group and, as we will talk later, there is now an effort to capture specimens at Roche collection Sites This really is a study in progress, and we are engaged very carefully with the NAT testing manufacturers to try to identify and employ the best test we can and the blood screening assays as they are being developed in the study. The objectives of the study, first, we have an understanding that we really have the challenge of assessing the performance of the assays, particularly the RNA, but also the IgM assays, so really the first objective is to establish an analytic sensitivity panel that we can use to validate the RNA tests that will be the primary screening assays for the study, and our target is to have assays that achieve or exceed a less than 50 genome per ml detection limit. There is now work in collaboration with the CDC Fort Collins Group to really get these panels put together and distributed, so we can really understand the relative and absolute sensitivity of these assays initially focusing on the assays, the candidate assays to support the link study, but then subsequently or in parallel, build larger panels, perhaps in collaboration with some of the commercial companies like BBI that are building these panels already, established panels for both RNA and IgM assay assessment that we can use as the blood industry to compare these assays and understand the relative performance and role of these assays, both the screening tests, as well as confirmatory tests, if donor screening moves forward. The second set of objectives relate to defining the prevalence of viremia and the disease outcomes and the donors, so here, we will be testing archived donor samples for RNA using the most sensitive RNA tests we can identify and as single donation testing, so that we can identify as many viremic samples as may exist to the limit of detection of the assays. There again, this study will exploit the specimens that are available regionally, as well as temporally, to maximize representation of hot zone specimens, but also with the inclusion of samples from other regions of the country, so we can generalize a national prevalence estimate, and that will be sort of linked with Lyle Petersen's estimates of the relative prevalence around the country of West Nile disease. Once we find these viremic donation samples, we will characterize the viral load, the IgM status, and they will be subjected to culture to try to understand the determinates of transmission and also to guide the decision about what will be the advantage of trying to move to an individual versus bringing up West Nile on minipool NAT and is there ability of IgM to detect a subset, a significant subset of these viremic donations. The donations that are found viremic, this is a linked study and the protocol and consents and follow-up materials are in place, so that we will recall the donors, verify the infection status through both IgM and RNA testing, and also to recall the donors administered questionnaire as to whether they recall having developed symptoms shortly after the donation. We will also look at the blood centers routine callback records to see if any of these donors in fact did call back with the enhanced FDA recommended callback system to find out, you know, again, was there disease in any significant fraction of these donors. The other thing we realized we needed to understand was the background prevalence of infection in the donor pool, as well as in the recipients. Most of these samples we will be testing actually will not have pre-transfusion samples from the recipients, so we will have recipients we are recalling three, six months after they received a blood transfusion, and we will be assessing the transmission rate from viremic donors by determining the IgM rate in these recipients. That needs to be viewed in the context of a background rate within a recipient population to determine the transmission rate. So, we have designed two sort of control populations. The first will be to look at we are projecting about 10,000 representative allogeneic donors from different West Nile activity regions, do IgM, and with that rate of IgM positivity, we can I think work with Lyle Petersen to feed his model to estimate the incidence rate based on the crude idea of reactivity, and knowing the incidence rate, one can estimate the transfusion risk given the understanding of the duration of the viremic window period. So, that is one piece, the allogeneic donor piece. But, secondly, rather than trying to enroll a large population of control recipients, recipients who did not get viremic donations, we are actually going to use autologous donors as a surrogate for recipients. We have done work recently, Steve Kleinman presented at AABB an analysis. Autologous donors are essentially people who are coming in to get a transfusion in the next week or two, and they are just giving to themselves. What we know is that the rates of all the markers are very similar in autologous donors in pre-transfusion samples from recipients and in the range of 8- to 10-fold higher than in allo donors, so we are using auto donors as a surrogate to get a background rate in recipients, and these other donors, the samples will be tested in an anonymized context, non-linked, so we don't need to recall those people, and the samples will actually be collected, the samples that we will test will be samples that will have been collected around the same time as we are testing the recipients. The next issue, and the critical really outcome of the study, in addition to donor viremia, is transmission rate to recipients. We really don't understand that particularly, the relationship if we do find a population of low level viremics, you know, are they infecting recipients, if IgM is present in a subset of viremics, does that influence transmission rate. So, we will be recalling the recipients of the viremic donation, testing them for both IgM and RNA to determine did they become infected, looking then at the donor and recipient factors that may influence whether transmission occurred, so viral load in the unit, IgM status, infectivity in tissue culture, and the recipient sort of underlying disease considerations. Also, then, recipients will be administered a symptom questionnaire to try to determine what proportion of recipients of viremic units who became infected actually manifested West Nile related disease. Now, to show you the kinds of samples that we have, there are two repositories I alluded to, one RADAR, the other TRIPS. Now, RADAR is an NHLBI-sponsored linked prospective donor recipient study. It began about two years ago at seven regional sites. The goal of the study is to enroll 4,000 recipients having pre- and post-transfusion, about six months post-transfusion samples from these recipients. We are about almost at 3,000 at this point, so this is a study that is going on at the opportune time. Now, these 4,000 recipients will have received about 15,000 donations, so we will have 15,000 donor exposures over this period of time in these regions. In order to be sure that the maximum number of donor units go into the recipients, as you will see, the blood centers actually over-collect donations and designate them for RADAR recipients. So, we have about 8-fold the number of donations in the repository than actually went in to the recipients who enrolled in the study, and that becomes a factor in a few minutes. The strategy of this, it is a study that is really intended to test donation samples and have the specimens from the pre- and post-transfusion samples and the recipients in the freezer, so that as a new agent like West Nile now comes along, we can very quickly go to these samples and simply test them. The recipients have consented to and the donors to storage of their specimens and to subsequent testing for new agents although we do have to go back and get IRB clearance that all the pieces are in place, but we do not have to reconsent the donors and recipients. We have both frozen whole blood aliquots and plasma aliquots from donations and recipient samples. The general strategy is to test the recipient 6- to 12-month sample, and then if the recipient is positive, then, we test the 3 to discriminate a transmission event from a pre-existing infection in the recipients. The TRIPS study stands for Transfusion Related Infections Prospectively Study. It is Harvey Alter-ism. This again is an ongoing study that essentially carries on what Harvey has done for several decades, focused in the D.C. area. In addition to the NIH Clinical Center, the study actually includes a site at the Children's Hospital D.C. site. This is somewhat similar although Harvey is really endeavoring to get all donation samples for all enrolled recipients frozen away. He is focused on highly transfused patients whereas the RADAR is particularly looking at orthopedic type in cardiac bypass. Now, Harvey's, the big difference again is the RADAR only has pre-transfusion 6 months, whereas Harvey's collections include essentially weekly samples for the first month, then monthly, so it allows much more careful characterization of the time course of events in infected recipients. In fact, in a prospective context, these recipient serial samples are being monitored for all the viruses shown here by both nucleic acid, as well as serologic methods. So, these are, you know, wonderful resources and a large investment of both the CDC and the NIH to build these repositories, but as we have come to learn, when you have got something as temporal and focal as West Nile, this repository is not the whole answer. So, what you see here are the samples in the repositories from the participating centers by month, so it includes both the RADAR and TRIPS samples. So, we have a total of about 15,000 donation samples of over this 4-month period of interest, but when we look then at the West Nile activity within the regions represented in the study, it turns out that only 3,000 or so of the 15,000 samples are from regions that have proven to be relatively high in terms of West Nile activity. In addition, if we look at the curves that Lyle showed yesterday, and the fact that the epidemic really peaked in a very brief, you know, one- to two-month period, only a subset of these samples are really very high yield samples. In addition, as I mentioned, we over-collect in order to have enough donations to support the enrolled recipients. We enroll and freeze away a large number of additional units. What this slide shows is the percentage of the RADAR 15,000 donations in the repository during the period of interest that are actually from recipients, the corresponding units that went into enrolled recipients, that were linked to enrolled recipients. You can see that actually only 13 percent of these units have recipients for whom we already have pre-sample and post-sample in progress. So, in fact, in this study, we are not only intending to test the link, but we will also test the donations that went into the recipients who did not enroll, and these recipients will be traced through lookback. We won't have a pre-transfusion sample, but these are important samples to contribute to a prevalence and estimate, and through lookback, will determine transmission rate. Now, again realizing that the numbers from these formal studies were not sufficient, we continued to closely work with Sue and the Red Cross folks. Sue immediately on, you know, the appreciation this was going to be a problem beginning of September, initiated retrieving and freezing away plasma, actually freezing down the residual plasma in the PPT tubes, which are used at Red Cross for NAT testing. This slide shows the number of specimens that the different sites have frozen from sites that, in conjunction with CDC, were determined to be potentially important sites for prevalence. You see that she has got 27,000 donations from St. Louis, 23,000 from Detroit, 21,000 from Cleveland. These are collected over about a six-week period, and we are working now with Lyle to determine which of these samples are the highest probability given the temporal epidemic in these regions, and just continuing three additional sites of Gulf Coast Red Cross collection, Chicago, and Memphis, the numbers you see here. But again if you focus on the periods of interest, we will probably end up focusing or prioritizing testing a subset of these specimens. Actually, this then summarizes the available specimens and the subset that are likely to be of reasonable probability of yielding viremic donations based on the epidemic, so these are the numbers we just talked to, only 3,000 of the 14,000 or so in RADAR and TRIPS are likely to be informative for yield. Potentially as many as 45,000 at the Red Cross, 90,000. We have also supplemented with 3,000 samples from both the Mississippi UVS collection site and the Life Source Chicago collection site, separate aliquots frozen away for the link study, and there is close discussion with Roche. The Roche Net System creates an archive plate, which is a 1.5 MLD bottom microplate archive, and our hope is to work to use some of these samples in conjunction with Roche at either in an initial phase, perhaps at the CDC Fort Collins test lab, but subsequently, as Roche brings up their assay in the spring, we hope to work with them to employ the same protocol to expand the data on prevalence and transmission. So, a total of over 100,000 specimens available of which potentially, more than 50,000 may be informative. Again, just to step back, in terms of the study, we sort of defined two phases, a Phase I, which is really the period where we are now working to establish the performance of candidate assays, benchmarking these candidate assays that are being developed to get program Roche assays against the Fort Collins' assays, so really creating panels that can, head to head, define the relative sensitivity and absolute sensitivity of these tests, also developing larger performance panels that would be available to anyone who would be interested to assess the performance of the RNA, as well as IDM tests. Then, the major Phase II component, which will involve testing. The plan is to test at least 50,000 specimens over time, perhaps in a phase mode, prioritizing these samples that have the highest probability of viremia, again over this past summer/fall. This will allow us to define the prevalence of viremia in the donors and to characterize the viremic samples, recall the recipients and testing of the recipient follow-up samples that are already in the RADAR/TRIPS repository will allow definition of transmission rate and lookback will be required, though, for the majority of samples because they are not enrolled in that study, and we will be able to look at correlates of transmission. I mentioned that we will be getting background rates both in allogeneic donors to model the probability of window phase donations, as well as in autologous donors to assess the background rate of recipients for definition of a transmission rate from the viremic donations by subtracting away the background rate. Timing sort of summary. Again, fortunately, these RADAR/TRIPS samples have been collected, you know, for the last few years, and continue to |