WASHINGTON _ The FDA this week challenged biotech firms to"close the HIV window" in blood-donor screening by developingmass screening methods that detect the AIDS virus itself, not justantibodies to the virus.Without such a test, blood banks cannot detect AIDS-tainted bloodbetween the instant of infection and the time the immune systembegins making antibodies to the HIV, a "window" which can lastseveral weeks.Consequently, more than a decade into the AIDS epidemic, bloodrecipients still face a small risk _ approximately one contaminatedunit of blood in every 420,000 _ of contracting AIDS from a donorwith a serologically silent infection, according to the federal Centersfor Disease Control and Prevention (CDC) in Atlanta.Closing the window could prove exceedingly profitable for thecompany that manages to surmount the scientific, manufacturing andregulatory hurdles to the development of a test.Each year, blood banks draw 12 million units of blood from eightmillion volunteers. The FDA requires that each unit be tested for avariety of infectious agents, including syphilis, two strains of HIV,two strains of human T Cell lymphotrophic virus,hepatitis B, andhepatitis C.Roger Dodd, head of transmissible diseases at the American RedCross Holland Laboratory in Rockville, Md., said that screeningmethods now in use, although imperfect, have protected thousands ofseverely ill transfusion recipients from AIDS infection.As of December 1993, he said, the CDC had reported 6,291transfusion-related AIDS cases, out of a nationwide total of 361,164.The overwhelming majority of those cases were reported before 1985,when the first AIDS tests were introduced. Since then, just 29transfusion-linked cases have been reported."Of course," Dodd said, "it is specious to argue that these are the onlypeople who have been infected through transfusions because weknow that AIDS is only the terminal end of HIV illness."The FDA estimates that from 80 to 450 people are actually infectedeach year, but that just a few of them live long enough to developclinical symptoms of AIDS because most are aged or in the advancedstages of disease.New generations of serological antibody-screening and confirmatorytests may have already begun to reduce the number of people who areinfected even further by narrowing the window without resorting togenetic technology.Third generation serologic tests now detect HIV antibodies within 22days of infection, down from 45 days in 1990, said Michael Busch, ofIrwin Memorial Blood Centers and the University of California, SanFrancisco.All told, Busch said, the serologic tests have become so sophisticatedthat blood-bankers expect that gene-based tests would bring onlyincremental improvements. Screening for cell-free HIV RNA, hesaid, would narrow the window by just 11 days, while screening forDNA would reduce it by about six days.Blood Banks Question The EffortIn light of this evidence, some in the blood banking community havequestioned the expense and effectiveness of mounting a campaign toclose the HIV window.Peter Gill, director of the National Testing Program for the CanadianRed Cross, said donor-screening in Canada turns up just 20 units ofAIDS-contaminated blood a year. "An extraordinary effort wouldhave to be put in to reliably close that window down," Gill said. "Is itworth it?"FDA Commissioner David Kessler told participants that even thelowest estimated risk is unacceptable."We want to eliminate the HIV window," Kessler said in his openingaddress to the three-day meeting, which ended today. "We want toencourage the development and commercialization of usefultechniques for direct viral detection." He said the lessons learnedfrom this process will not only assure a safer blood supply, but alsoenhance the nation's "ability to deal with infectious agents that maybe discovered in the future."Kessler said he has another agenda as well. He would like to changethe FDA's role, and its reputation, from an agency that is so focusedon safety that it appears to hinder progress to one that "fosters thedevelopment of new technologies."But researchers have not waited for the FDA to attempt to developnew gene-based tests for viruses.Among the most promising are polymerase chain reaction (PCR) andother gene amplification techniques, which are so powerful they candetect a single molecule of a known gene sequence. Withmodifications, these techniques also enable researchers to preciselydetect RNA molecules.PCR May Be The Best OptionPerhaps the best known of the amplification techniques involvesusing PCR to copy a minute quantity of DNA millions of times, sothat mere molecules can be detected using chemical tests.The technique, discovered in 1985, has been steadily improved sothat results can be obtained in hours, rather than days, and tests canbe primed to react with more than one segment of DNA. Such aproperty could prove to be extremely useful to blood banks, because itwould permit them to use a single test to screen for several of thepathogens now tested for individually.Thomas White, of Roche Molecular Systems, which has developed aPCR-based detection kit for diagnostic use, said, however, "theformat of the current tests is not amenable for high volume screeningof the blood supply."But John Sninsky, senior director of research for Roche MolecularSystems in Alameda, Calif., said PCR, which is highly accurate,might nevertheless be effectively phased in as an HIV screeningtechnique.For instance, he said, "PCR may be considered first to resolveambiguous or indeterminate Western Blots before being called uponfor increasingly higher volume uses." Western blots are nowprimarily used to guard against false positive antibody tests.But there are other possible approaches as well. According to theFDA, they include:- Ligase chain reaction uses a thermostable enzyme to ligate thephosphate backbone of complementary pairs of nucleotide probes thatare bound to the target DNA sequence. In the presence of the secondset of probes, the first set acts as a template for ligation of the secondset, which generates a chain reaction that multiplies the DNAsequence. The gene can then be detected using electrophoretic gelanalysis.- Branched DNA oligonucleotides provide a means of amplifyingthe signal generated by a limited number of targets rather than bymultiplying the targets themselves. In this technique, scientistssynthesize branched DNA, containing primary and secondarysegments. The first segment hybridizes with a target sequence; thesecond hybridizes to a small oligonucleotide that is labeled withalkaline phosphatase for detection.- Transcription mediated amplification uses acridinium ester-labeled oligonucleotide probes for hybridization. Detection isaccomplished by chemiluminescence.Jay Epstein, acting director of the FDA's Office of Blood Researchand Review, said "assays focusing on RNA detection seem to offerthe greatest promise at this time for narrowing the HIV window.Branched DNA testing offers potential as well."Overcoming The ObstaclesHowever, to adapt tests that were developed for diagnostic use tomass blood-donor screening programs, researchers will have toovercome formidable obstacles.Among other things, they must learn how genetic variation in geneamplification and hybridization techniques affects the sensitivity of aprospective test; they must create tests that can detect manycirculating strains of rapidly mutating HIV; and they must find waysto control for the sample cross-contamination that can producedeceptive, false-positive results.In addition, commercial laboratories that use currently available PCRtests must invest millions in building separate work areas withsophisticated ventilation to minimize the likelihood of cross-contamination. They must also train their lab technicians in carefuland time-consuming sample preparation.The ideal would be to develop a test that could be adopted byscreening labs as easily as serological testing, needing no specialhandling or work areas. Participants said "walk-away" automationwould be a crucial feature, since large blood banks may draw over athousand units of blood a day. n092994

-- Steve Sternberg Special To BioWorld Today

(c) 1997 American Health Consultants. All rights reserved.