EDITORIAL

Volume 40, January 2000 TRANSFUSION 3

E D I T O R I A L

Hepatitis C virus (HCV) lookback is the effort totrace and notify persons who previously re-ceived blood or blood components from do-nors who later were found to have possible

HCV infection on the basis of a repeatedly reactive licensedscreening test for anti-HCV and results of further testing.The Food and Drug Administration (FDA) first issued guid-ance for industry on HCV lookback in March 1998.1 Twoarticles in the current issue of TRANSFUSION report scien-tific data that strongly influenced current policies of thePublic Health Service (PHS) regarding HCV lookback.2,3 Theevents leading to these publications illustrate the depen-dence of public health policy on a foundation of good sci-ence and the benefits of a constructive, ongoing interactionbetween policymakers and scientists. The lessons learnedfrom these events may help us to face future challenges inblood safety.

HCV lookback is complicated by the fact that many dif-ferent HCV tests have been in use over time. Routine blooddonor screening for infection with HCV was implementedin the United States in May 1990 with the use of an FDA-licensed, single-antigen-based, enzyme-linked immunoas-say (EIA), HCV EIA 1.0. A more sensitive EIA, based on mul-tiple viral antigens, became commercially available inMarch 1992 and quickly replaced the HCV EIA 1.0. A sec-ond-generation HCV recombinant immunoblot assay(RIBA 2.0), the first commercial HCV supplemental test toconfirm antibody-positive screening test results, becameavailable in June 1993, although some blood organizationshad earlier access to experimental supplemental tests. Later(third-generation) versions of screening and supplementaltests, namely HCV EIA 3.0 and HCV RIBA 3.0, were licensedin May 1996 and February 1999, respectively. These tests co-exist in the marketplace with HCV EIA 2.0 and HCV RIBA2.0. In developing recommendations for HCV lookback, theFDA sought to minimize recipient notifications based onfalse-positive test results in the donor while at the sametime identifying recipients who were placed at increasedrisk by transfusion. In this regard, two problems were ofmajor significance, namely, the clarification of indetermi-nate results of HCV RIBA 2.0 and the resolution of repeat-edly reactive HCV EIA 1.0 results in the absence of supple-mentary test results.

In this issue of TRANSFUSION, Busch et al.2 provideoriginal data and summarize previous literature reports onHCV EIA 3.0 testing of blood donor samples previouslyidentified as repeatedly reactive by HCV EIA 2.0 and as in-determinate by HCV RIBA 2.0. They found that negativeHCV EIA 3.0 results in such samples are strongly predictiveof a negative HCV RIBA 3.0 (over 99% of cases) and a nega-tive test result for HCV RNA. Conversely, repeatedly reac-tive EIA 3.0 results in such samples often are associated witha positive HCV RIBA 3.0 (20% of cases), and positive HCVRNA tests may be found in the latter group (20% of HCVRIBA 3.0-positive samples). On the basis of these results,which were provided confidentially to the FDA before pub-lication, the authors suggested the possibility of clarifyingprior HCV EIA 2.0-repeatedly reactive/HCV RIBA 2.0-inde-terminate donor test results by the use of HCV EIA 3.0. Morespecifically, they proposed that one could use HCV EIA 3.0rather than (the then investigational) HCV RIBA 3.0 to re-test stored, repeatedly reactive donor samples. This strat-egy, which the FDA adopted in a revised guidance docu-ment for industry on HCV lookback issued in September1998,4 substantially reduced the need for additional testingwith HCV RIBA 3.0. In addition, it averted a large numberof unnecessary patient notifications, due to false-positivescreening test results, that would have occurred in the ab-sence of further testing. Clearly, the revised policy had ben-efits both for the blood industry (lower costs of testing andnotification) and for the public (fewer notifications basedon false-positive screening tests).

In their article, Tobler et al.3 report original studiesdemonstrating that a signal-to-cutoff (S:CO) ratio >2.5 inthe HCV EIA 1.0 had an 89-percent sensitivity for detectingHCV RIBA 2.0-positive samples and an 87-percent specific-ity for correctly identifying HCV RIBA 2.0-negative samples.These studies provided a scientific basis for limiting HCVlookback, if it were extended to include HCV EIA 1.0 test-ing of donors since 1990. The strategy of including consid-eration of the S:CO ratio would avoid false notifications thatotherwise would occur in the absence of additional testingto clarify the significance of historical, repeatedly reactiveHCV EIA 1.0 screening test results for a donor. The datacontained in the article by Tobler et al. were presented at apublic meeting of the PHS Advisory Committee on BloodSafety and Availability (ACBSA) in January 1999. Discussionof these data led to recommendations by the PHS ACBSA

Hepatitis C virus lookback:emerging science and public policy

TRANSFUSION 2000;40:3-5.

EDITORIAL

4 TRANSFUSION Volume 40, January 2000

that were adopted later by the Department of Health andHuman Services (DHHS) and then incorporated in a draftcomprehensive revised guidance published by the FDA inJune 1999.5

The public health value of these scientific studies maybest be understood by reviewing the evolution of publicpolicy on HCV lookback. Debate on the benefits of HCVlookback began in 1989, before the approval of the firstdonor-screening test for anti-HCV. Lookback was not rec-ommended early on because of the absence of a provenmedical benefit and concern over false-positive tests indonors that would lead to inappropriate notification ofprior blood recipients. HCV lookback later became a mat-ter of concern to the Congressional Committee on Govern-ment Reform and Oversight of the 104th Congress, becauseof the increased recognition of an emerging epidemic ofliver disease due to chronic HCV infections.6 This commit-tee strongly advocated for HCV lookback as part of a broadstrategy to find persons with HCV infection who might ben-efit from medical intervention. In the United States, illegaldrug use and high-risk sexual behavior cause most HCVinfections, whereas prior blood transfusion accounts forapproximately 7 percent of the 3.9 million living personswith serologic evidence of HCV.7

Since 1985, the risk of transfusion-related HCV infec-tion has decreased significantly as a result of effective do-nor screening,8 and transfusion transmission of HCV nolonger represents a major cause of new infections. Consis-tent with a commitment made to the congressional com-mittee by DHHS, the issue of HCV lookback was broughtbefore the PHS ACBSA in April 1997 and August 1997. Fac-tors affecting the PHS ACBSA recommendations were thepublication for the first time in March 1997 of consensusrecommendations on the treatment of chronic HCV, theavailability since 1992 of accurate screening tests and ofsupplemental tests for anti-HCV, and cost-effectivenessanalyses of HCV lookback.9 After a lengthy debate and rec-ognizing the value of supplemental tests to avert notifica-tions based on false-positive screening test results, the PHSACBSA recommended HCV lookback based on donorscreening with the more accurate “multi-antigen” tests(HCV EIA 2.0 and 3.0). After the acceptance of these recom-mendations by the DHHS in January 1998, the FDA issuedits first guidance on targeted HCV lookback in March 1998.1

The FDA sought to minimize false notifications by rec-ommending appropriate additional testing of the repeat-edly reactive donor samples or fresh donor samples. Therecommendations for such additional testing necessarilywere limited to strategies of known validity. The FDA’s re-vised guidance issued in September 19984 included a pro-vision for the retesting of stored repeatedly reactive donorsamples by EIA 3.0 in lieu of RIBA 3.0. This modification,made in response to an advance review of the data andanalysis of Busch et al.2 in this issue of TRANSFUSION,

avoided many false notifications that would have occurredfor lack of availability of RIBA 3.0 or fresh donor samples foradditional testing. Even with the subsequent commercialavailability of RIBA 3.0, the prior use of EIA 3.0 to clarifythese cases allowed the implementation of a more cost-ef-fective strategy with minimal loss of sensitivity. The timelydevelopment of an improved testing strategy created a win-win situation for both the blood banks and public health.

However, the HCV lookback policy as defined by theFDA in March 1998 and September 1998 did not addressdonor testing between May 1990 and March 1992 as alookback “trigger,” and, in October 1998, the CongressionalCommittee on Government Reform and Oversight criti-cized the PHS effort on HCV lookback as too limited.10 Forthis reason, the PHS ACBSA reexamined its prior recom-mendations. At public meetings of the PHS ACBSA held onthis issue in November 1998 and January 1999, a centralconcern was the possibility of false notifications due tounconfirmed donor-screening tests performed with HCVEIA 1.0.11 The cited studies by Tobler et al.3 showing that theuse of an S:CO ratio <2.5 could be an accurate method ofidentifying false-positive HCV EIA 1.0 screening tests werecritical to the discussion. Use of the test ratio averted mostof the notifications based on false-positive screening teststhat would have occurred in the absence of additional test-ing of the donor samples.

There is a moral to this story, namely, that sound pub-lic health policy related to blood safety should be basedupon and influenced by current science. The public healthdebate on HCV lookback evolved with time and so will thedebates on other issues of concern. The FDA has under-taken the reexamination of its existing blood safety policies,which were implemented over the last two decades. Thistask is part of a broad plan to address blood safety regula-tion and the risks related to blood components.12 Althoughmany factors may affect future policy decisions, the factorwith the greatest influence should be current scientific data.Hopefully, the model of a constructive interplay betweenscientists and policymakers regarding policy on HCVlookback that is illustrated by the cited articles in this issueof TRANSFUSION will serve as a useful precedent.

Jay Epstein, MDOffice of Blood Research and Review

Center for Biologics Evaluation and Research

Food and Drug Administration

Department of Health and Human Services

Rockville, MD 20852-1448

REFERENCES01. Guidance for industry: supplemental testing and the notifi-

cation of consignees of donor test results for antibody to

hepatitis C virus (anti-HCV). Fed Regist 1998;63(54):13675-6.

EDITORIAL

Volume 40, January 2000 TRANSFUSION 5

02. Busch MP, Tobler LH, Tegtmeier G, et al. Use of third-gen-

eration hepatitis C virus (HCV) enzyme immunoassay (EIA)

to resolve second-generation HCV EIA-reactive and sec-

ond-generation recombinant immunoblot assay-indeter-

minate blood samples: data to support current Food and

Drug Administration Guidance on HCV Lookback. Transfu-

sion 2000;40:10-14.

03. Tobler LH, Tegtmeier G, Stramer SL, et al. Lookback on do-

nors who are repeatedly reactive on first-generation

hepatits C virus assays: justification and rational imple-

mentation. Transfusion 2000;40:15-24.

04. Guidance for industry: current good manufacturing prac-

tice for blood and blood components: (1) quarantine and

disposition of units from prior collections from donors with

repeatedly reactive screening test for antibody to hepatitis

C virus (anti-HCV); (2) supplemental testing, and the notifi-

cation of consignees and blood recipients of donor test re-

sults for anti-HCV. Fed Regist 1998;63(203):56198-9.

05. Draft guidance for industry: current good manufacturing

practice for blood and blood components: (1) quarantine

and disposition of prior collections from donors with re-

peatedly reactive screening tests for hepatitis C virus (HCV);

(2) supplemental testing, and the notification of consignees

and transfusion recipients of donor test results for antibody

to HCV (anti-HCV). Fed Regist 1999;64(119):33309-13.

06. Protecting the nation’s blood supply from infectious

agents: the need for new standards to meet new threats.

Tenth Report of the Committee on Government Reform

and Oversight. 104th Congress, second session. House Rep

104-746 (August 2, 1996).

07. Alter MJ, Kruszon-Moran D, Nainan OV, et al. The preva-

lence of hepatitis C virus infection in the United States 1988

through 1994. N Engl J Med 1999;341:556-62.

08. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP.

Medical progress: transfusion medicine (first of two

parts)—blood transfusion. N Engl J Med 1999;340:438-47.

09. Department of Health and Human Services Advisory Com-

mittee on Blood Safety and Availability, April 24-25, 1997,

and August 11-12, 1997. Available at http://www.hhs.gov/

partner/bloodsafety/.

10. Hepatitis C: silent epidemic, mute public health response.

Seventh Report by the Committee on Government Reform

and Oversight. 105th Congress, second session. House Rep

105-820 (October 15, 1998).

11. Department of Health and Human Services Advisory Com-

mittee on Blood Safety and Availability, November 24,

1998, and January 28, 1999. Available at http://

www.hhs.gov/partner/bloodsafety/.

12. Center for Biologics Evaluation and Research. Blood action

plan. http://www.fda.gov/cber/blood/bap.htm.