NAT screening of blood and plasma donations: evolution of technology and regulatory policy*

NAT for screening blood and plasma for viruses have been rapidly developed and implemented for the detection of HCV and HIV-1. The US FDA has had to develop regulatory policies under conditions not previously encountered, namely the development, improvement, and application of new test technologies occurring at such unprecedented speed that nearly all blood and plasma donations would need to be tested while the methods were still in clinical trials. This rapid pace of development can be clearly seen to extend into the future; NAT procedures for additional viruses and modifications of NAT for greater sensitivity and the rapid detection of many viruses simultaneously are currently under development. It had been clear for several years that the increased sensitivity for detecting HCV and HIV-1 by NAT would give an added level of safety to the nation's blood and plasma supply and that widespread use at the earliest possible date, even if conducted under investigational exemptions before approval of licenses for NAT systems, would be desirable. Furthermore, the expectation in the late 1990s that NAT screening of plasma for HCV would be required by European regulatory authorities, and the fact that substantial volumes of plasma collected in the United States were being exported to Europe, drove the efforts to initiate such testing in the United States. In the end, NAT for HCV was required in Europe for all plasma used for fractionation after July 1, 1999. The processing of blood and plasma collected in the United States includes screening for antibodies and antigens of HCV, HIV-1, and HBV by sensitive immunoassays (such as EIA). These immunoassays already had been in use for many years before NAT was introduced, and their use led to a high degree of safety. Nevertheless, some infectious units still escaped detection by these immunoassays. (However, plasma derivatives are subjected to procedures that would remove or inactivate all HCV, HIV-1, and HBV, if done appropriately.1) The development of the “minipool” concept made such NAT testing feasible; these minipools range from pools of 16 to 24 donations each for whole blood donations and 512 to 1200 donations each for source plasma donations. The use of minipools, accompanied by an algorithm for identifying an infected unit if a minipool tested positive, made the labor-intensive NAT screening affordable for the first time. Many of the few remaining infectious units can be identified by NAT and excluded from transfusion use. Today, NAT conducted on minipools can detect as few as 5 to 6 copies per mL of these viruses in the pool; depending on the minipool size, this is equivalent to 80 to 7200 copies of the virus per mL in the donor blood or plasma.2 Most of these infectious units are from donors in the “window period” of infection, when they are infectious but before they have serologic markers that can be detected by immunoassays. NAT screening can identify most of these donors, as well as the rarer cases of infectious atypical carriers of the viruses. For instance, NAT can detect HCV RNA in stored samples from anti-HCV-negative, NAT-positive donors who previously transmitted HCV to recipients over a long period of time, indicating that they were atypical carriers of HCV.3 Similarly, NAT can detect HIV-1 RNA in stored samples from anti-HIV-negative, p24 antigen-negative individuals who had previously transmitted HIV-1.1 Beginning in 1997, discussions were held between the US FDA and representatives of the fractionation community concerning the development of NAT screening to detect HCV and HIV-1. The risk of HCV or HIV-1 from plasma donations would be negated by the virus inactivation and removal procedures included in the manufacturing processes for plasma derivatives, if all procedures were properly conducted.1 Nevertheless, it was also thought that NAT improved the margin of safety because reduction in the overall viral load in the manufacturing pool of plasma would help ensure the effectiveness of the virus inactivation procedures. Although the initial discussions were about screening of plasma for use in fractionation, the real benefit was expected to come from screening whole blood donations, since no generally applicable procedures had yet been shown to be safe and effective for inactivation of viruses in blood components intended for transfusion. Subsequently, minipool algorithms were developed that made similar screening of blood donations possible. As a result of efforts by the FDA and the fractionation and testing communities, arrangements were made to permit the screening by NAT of plasma and whole blood in minipools under regulatory exemptions while large-scale clinical testing and laboratory validation of the assays were completed. These regulatory exemptions are officially called “investigational new drug applications” or “INDs,” even though they are regulated as biologic products rather than as new drugs. An investigator or company files an IND to obtain permission from FDA to conduct clinical trials of an unlicensed blood screening test. Existing data about the test are submitted, as well as a detailed study plan. The clinical trial is then carried out under FDA oversight. The data obtained from the study can then be used to support a biologics license application (BLA) to obtain FDA permission to market the test. The first BLAs for minipool NAT to detect HCV and HIV-1 in source plasma and whole blood donations were approved in 2001 and 2002, respectively (see later). The approval of the BLA for minipool NAT to detect HCV and HIV-1 in whole blood donations also included approval of NAT to detect these two viruses in individual whole blood donations without the use of minipools (single-unit NAT); however, the method was not sufficiently automated to be used for mass screening. Nevertheless, single-unit NAT systems that would permit large-volume screening are being developed to detect HCV, HIV-1, and HBV. Although single-unit NAT should be more sensitive than minipool NAT, the ultimate relative sensitivities will depend on the outcome of continuing improvements in both types of NAT technology and might differ depending on the virus being screened. The relative sensitivities will have to be determined by studies conducted in actual blood and plasma collection and testing centers. However, sensitive, specific, and affordable single-unit NAT screening might not be available for at least several years. A license for a NAT system for screening source plasma for HCV and HIV-1 on minipools of 512 donations was approved by the FDA on September 18, 2001. This license was approved for the National Genetics Institute, a large national commercial testing facility in Los Angeles, California. A supplement to the product license of one of the major fractionators, Alpha Therapeutic Corporation (Los Angeles, CA), was simultaneously approved for use of this NAT system. Following these approvals, the other three major fractionators continued to screen donations for HCV and HIV-1 by minipool NAT under INDs. On December 4, 2001, the FDA posted on its Web site a draft guidance document (published as 67 FR4719, January 31, 2002) that stated that all collectors of source plasma would have to submit license amendments for the use of a licensed NAT for HCV and HIV-1 within 6 months of the publication of the final guidance document (which is still pending). All of the major fractionators are currently developing strategies to use licensed NAT systems to screen source plasma for HCV and HIV-1. NAT screening of nearly all whole blood donations is continuing to occur. A license for a NAT system for screening whole blood donations for HCV and HIV-1 on minipools of 16 donations was approved by FDA on February 27, 2002. This license was approved for Gen-Probe, Inc. (San Diego, CA), a test manufacturer that developed the test method with support from the National Heart Lung and Blood Institute, National Institutes of Health (Bethesda, MD). Unlike the license for source plasma NAT, which is a license for a national testing laboratory, this license for whole blood NAT is a license for a test kit that can be used at blood banks throughout the country. It is expected that more than half of the blood collections in the United States will switch their NAT for HCV and HIV-1 from testing under an IND to testing with the newly licensed assay by June 2002. On March 13, 2002, the FDA posted on its Web site a draft guidance document (published as 67 FR17077, April 9, 2002) that stated that all blood establishments must implement a licensed NAT for HCV and HIV-1 within 6 months of the publication of the final guidance document (which is still pending). Screening by NAT for HCV was developed more rapidly than screening for other viruses because the residual prevalence was greatest for HCV and because the high concentration of virus in plasma1 during the relatively long window period (approximately 70-80 days from exposure to seroconversion4) was expected to make minipool detection of the virus very effective. (Although virus removal and inactivation procedures included in the manufacture of plasma derivatives will remove any risk of transmission of HCV by those derivatives if the procedures are carried out correctly, NAT screening was believed to provide an additional safeguard.) In the United States, NAT screening of source plasma for HCV was common by the end of 1998 and reached greater than 99 percent by the end of 1999. NAT screening of whole blood for HCV under INDs was common by the end of 1999 and reached greater than 99 percent by the middle of the year 2000. Logistically, completing NAT screening and reporting the results to the collection facility is not difficult to complete because the 42-day shelf-life of RBCs is long enough to permit testing before release. In addition, all facilities performing NAT for HCV on whole blood are also completing the tests in time to report the results before platelets are released (i.e., within the 5-day dating period of platelets). The only facilities not performing NAT screening for HCV are a few small hospital blood banks, representing fewer than 1 percent of blood collected in the United States; these blood banks will also be performing minipool NAT for HCV with a licensed test in the near future. A WHO international standard reagent for HCV RNA has been established and an FDA HCV RNA test has been to the of a level of sensitivity for all NAT all minipool testing of pool The level of sensitivity to be required has not yet been in but it will be equivalent to copies per mL in the NAT of studies conducted to NAT sensitivity have results in viral copies per In the WHO has developed that NAT results for HCV, and HBV can be in per The to copies per mL to per mL from one laboratory to the because of in the methods For this results that have been in copies per mL have been in this in this the is in the range of to copies 1 depending on which virus is being and which laboratory is the A sensitivity of per mL for detection of HCV in the donor plasma has been in some in Europe and is also the for NAT systems in the United States. The FDA Blood after of a NAT system for HCV, results of those tests be used to donors who had been or would be because of an test for A draft algorithm to be used as a for these donors was at that and the FDA is developing a guidance document for on this in developing and NAT for detecting HIV-1 has that for of NAT to detect is at an of development, but the risk of from transfusion has very in the United States. Although the prevalence of in US blood donors has not been serologic testing of blood donations between and not any donors infected with this of NAT screening for HIV-1 in the United States reached greater than 99 percent of source plasma by the end of 1999 and greater than 99 percent of whole blood donations by the middle of 2000. for HIV-1 NAT screening of whole blood donations are being before the platelets are As with HCV, the only facilities not performing NAT screening for HIV-1 are a few small hospital blood these blood banks will also be performing minipool NAT for HIV-1 with a licensed test in the near future. A WHO international standard for HIV-1 RNA has been with plasma by and an FDA HIV-1 RNA test has been and As for HCV NAT, the for HIV-1 NAT sensitivity has not yet been in but it will be equivalent to copies per mL in the NAT A sensitivity of per mL for detection of HIV-1 in the donor plasma has been in at least one European and is also the for NAT systems in the United States. for the use of NAT for HIV-1 are being developed by the FDA simultaneously with for HCV screening. The FDA Blood after NAT for HIV-1 be used to donors who had been or would be because of an assay for The FDA is developing a guidance document on this NAT screening for HIV-1 permit of the screening test for HIV-1 p24 a test that was as an in data will be by FDA for each application to the p24 test. a FDA on NAT it was that since the test for p24 has only detected 6 units to the and additional units Blood that would not have been detected by tests for This to a detection of fewer than 1 unit 6 units In all of the data at the FDA HIV-1 NAT screening of minipools was more sensitive than testing for p24 in individual in no was an that would not have been detected by that used HIV-1 NAT on minipools of 512 units detected that was for of or 16 units detected samples that would not have been detected without These data that licensed NAT screening might p24 screening of both plasma and blood in the future. The FDA is to permit individual to NAT for p24 screening if they submit data to support equivalent or greater sensitivity of a NAT system to the p24 test. the FDA for p24 screening be if the of the data were to to all licensed NAT of the currently licensed NAT systems for detecting HIV-1 have been for sensitivity to or greater than p24 and this fact has been included in their Blood and plasma establishments that implement the currently licensed NAT systems for HIV-1 can their licenses to p24 one license approved September 18, 2001, the of HIV-1 minipool NAT for p24 and that p24 screening in September NAT screening for HCV and HIV-1 has the of residual transmitted by transfusion NAT it was that was a residual prevalence of to units HCV per 1 donations of and units per 1 donations of whole the of NAT, the residual prevalence for HCV is thought to be from units for to to per 1 donations of whole The residual prevalence for HCV in source plasma after NAT has not been NAT the residual prevalence of HIV-1 after screening by and tests was about per 1 donations for source and per 1 donations for whole The residual prevalence for HIV-1 after NAT is to be from units for to to per 1 donations of whole The residual prevalence for HIV-1 in source plasma after NAT has not been the year minipool NAT screening for HBV under has been to all source plasma collected in the United States, in by the expectation that regulatory in other might HBV is in units per 1 plasma donations even after application of procedures that can remove or inactivate HBV during the manufacture of plasma derivatives would any residual HBV the viral load in the fractionation pool is seen as a to help ensure the effectiveness of these in the of virus inactivation to a viral load might still be by the fractionation or by the antibodies to HBV in the manufacturing the time of HBV, to for HCV and for and the of HBV in blood during the window period 1 and per that NAT screening of minipools for HBV would not be very effective. However, results of screening source plasma for HBV with minipool NAT than expected of units plasma donations that had already been tested by licensed the for this are not clear at company performing such screening detecting HBV in of company detecting HBV in of The testing in the of units as a result of an of an additional of a new sensitive assay such as HBV In minipool NAT for HBV has not been for whole blood donations in the United States to minipool NAT from to units per blood donations are to HBV not detected by screening for and for minipool NAT for HBV were introduced, the residual risk of transmission of HBV by whole blood would be units per donations for The expected of minipool NAT screening to detect HBV in whole blood is a this screening has not yet been However, the minipool NAT to detect HBV in whole blood donations has not been in the United States is the expectation that new immunoassays that have been approved or are currently under development are of or greater sensitivity than minipool These new immunoassays are more sensitive than most of the tests currently licensed in the United The FDA will the sensitivity of licensed assays by these assays to be to detect a concentration of on the A draft FDA guidance document to this was posted on the FDA Web site on April (published as 67 April These screening tests would detect any infected unit about HBV per mL in some cases can detect even NAT screening for HBV would have to be very sensitive to be more than the more sensitive use of a minipool as small as minipool NAT with a sensitivity of per mL would be only equivalent in sensitivity to the types of assays are continuing to in but the relative sensitivity of minipool NAT for HBV only than the immunoassays for However, single-unit NAT can detect the of HBV to days than the currently licensed tests and is more sensitive than the the technology for single-unit NAT to detect HBV any is not yet The FDA will a guidance document on HBV NAT near the time of the approval of an HBV NAT A technology might be in which a sensitivity level for detection of HBV would be a would be and the standard be by use of an HBV NAT minipool a sensitive or some other technology yet to be For instance, a sensitivity might be that is the detection of the tests equivalent to of HBV In this a WHO HBV international standard has been developed for use in HBV and the FDA is being to be more for NAT HBV NAT on minipools would not permit the of the currently screening of whole blood donations or the of the required testing of blood and plasma donations, to data at a FDA is that a very small of donor samples are and and have HBV but the HBV is very than copies per mL in some These samples are infectious but would not be detected by HBV NAT conducted on minipools, the only the development of very sensitive, single-unit NAT screening might permit the of screening of whole blood donations. Furthermore, of donations that were but were to be NAT-positive when volumes of plasma were with the long of the effectiveness of screening of blood donations, that it would be to the level of sensitivity of NAT minipool screening. of this will data from a large study with sensitive assays to detect HBV and to HBV viral in which samples from the donors are also obtained and tested and in which the is on the to which single-unit NAT greater assay NAT screening for HCV, HIV-1, and HBV has been by the FDA to be donor screening rather than a in the processing or of a product as in the of a minipool a unit for one of these viruses a and to identify and the donor to donations by that to the donor to to the donor to to and some to conduct of donations by that donor and to identify those who had components made from donations by that A has been for the use of minipool NAT to detect and would be by the virus inactivation procedures that are used for HCV, HIV-1, and HBV during the manufacture of plasma derivatives, since and are viruses. NAT systems for these two viruses are because of similar regulatory even though is no These tests for screening source plasma donations have been by the FDA as testing than donor because of the donor would in most cases after was and because would not a is no need to screen all units for the of these In the of the virus is that only benefit can be by the donor or by at large as a result of the donor of the test. For as a result of donor would be only during a period of before the of individuals with would be excluded by or by the donor and the benefit of other to the donor would only for a few The FDA not clinical studies to the of tests that are however, the FDA validation of the tests as with the data to FDA to under a supplement to the BLA for the final In addition, if the testing results in of the individual unit the minipool NAT it is at FDA that such testing might testing and would have to be developed under an investigational for is in the with new occurring at a of about percent per many of the procedures to remove or inactivate viruses in plasma derivatives are not effective because it is a NAT to detect in source plasma has been by all major fractionators as an to the level of per mL in the manufacturing level was on the results of a study for plasma that product with a high concentration of virus in the manufacturing with of are not because would the level of antibodies to in the manufacturing As a plasma derivatives, which is to should be for use in this and other that are to from of of viral transmission by NAT to detect has been implemented both on minipools and on the final for that NAT assays to detect only high of detect about 1 unit per to donations of source plasma with assays with between and viral copies per of the fractionators, but not are screening source plasma for by minipool The of units is expected to be between 1 in and 1 in source plasma Although it is expected that most whole blood donations will be for and in minipools in the near the regulatory for these tests is still being Blood and the have stated that they are to such testing of this testing will to screening only plasma because in some cases components will be at the time of this not be the and in test methods permit the results It is expected that will be at high in 1 in to 1 in whole blood It is at the FDA that the results of minipool NAT to detect should be to identify the before of blood this testing the individual it be to be that it should be conducted under an investigational before approval of a license It is at FDA that minipool NAT for need not be required for whole blood donors because of the of It is expected that screening of whole blood donations by minipool NAT would detect 1 unit per that percent of are and that percent of transfusion recipients have to this would that such screening would one years. Nevertheless, if minipool NAT for is being on whole blood donations, it is at the FDA that the individual donor should be identified before the of blood components for if that these components can be This might in which the test should be under an investigational of a whole blood donor of a NAT for benefit the donor or or is still under that permit the of many such as are to to increased test and blood safety. methods are not for use by because of a of that can with application of this technology for blood screening is believed to be still 5 to years However, the technology is and is that rapid and screening of blood and plasma donations for on one will be minipool NAT with single-unit NAT can be expected to the sensitivity of screening. NAT might the few remaining not currently detected by serologic screening or minipool of single-unit NAT for HCV and HIV-1 was conducted by at least one in FDA approval of that system. In addition, the and blood banks in the United States have been testing all donations for HCV and HIV-1 with single-unit NAT under an IND for NAT, though not yet generally available because of its is believed by many to be the ultimate of NAT systems development, its will depend on other in testing have and its for the Although safe and effective methods have not yet been approved to inactivate viruses in whole blood without the such methods are under development. inactivation of whole blood will be in inactivation for detection by immunoassays and Although it is not yet this will be a similar source plasma provide NAT screening of source plasma for HCV, HIV-1, and HBV was implemented the of very effective methods to inactivate those viruses during the of manufacturing plasma In it was that by the viral load of the manufacturing the use of donor screening of the effectiveness of the inactivation It is not a similar will be when inactivation of whole blood is but this is an for and