Performance of second- and third-generation RIBAs for confirmation of third-generation HCV EIA-reactive blood donations. Retrovirus Epidemiology Donor Study.

BACKGROUND: Licensure of an enhanced HCV screening assay (HCV 3.0 EIA) without concurrent licensure of a complementary supplemental assay (i.e., RIBA HCV 3.0 strip immunoblot assay [RIBA-3]) decoupled screening and supplemental testing. In March 1998, the FDA Center for Biologics Evaluation and Research (CBER) recommended the use of RIBA-3 on RIBA HCV 2.0 strip immunoblot assay (RIBA-2)-indeterminate units screened with HCV EIA 3.0. STUDY DESIGN AND METHODS: The sensitivity of RIBA-2 and RIBA-3 was compared in tests on HCV 3.0 EIA-repeatably reactive (RR) units identified immediately after the implementation of HCV 3.0 EIA screening. Two protocols were evaluated: parallel testing of HCV 3.0 EIA-RR units by RIBA-2 and RIBA-3 and reflex testing of HCV 3.0 EIA-RR and RIBA-3-confirmed-positive units by RIBA-2. All specimens with discordant RIBA-2 and RIBA-3 results and a representative sampling with concordant RIBA results were tested by PCR. RESULTS: In the parallel testing protocol, 99,777 donations were screened, with 245 HCV 3.0 EIA-RR specimens included in the study. Of 166 RIBA-2-positive samples, 165 tested positive in RIBA-3 (1 sample reacted to the control superoxide dismutase antigen in RIBA-3). Thirty-two (74%) of 43 RIBA-2-indeterminate specimens and 4 (11%) of 36 RIBA-2-negative specimens tested positive in RIBA-3. HCV RNA was identified in 5 (16%) of 32 RIBA-2-indeterminate/RIBA-3-positive donations, as well as in 26 (70%) of 37 concordant RIBA-2/RIBA-3-positive donations. In the reflex testing protocol, 292,459 donations were screened, with 709 HCV 3.0 EIA-RR specimens included in the study. RIBA-3 testing yielded 517 (73%) positive specimens, of which 50 (9.7%) tested indeterminate and 15 (2.9%) tested negative in RIBA-2. Among the RIBA-discordant specimens, 10 (20%) RIBA-2-indeterminate specimens and 1 (7%) RIBA-2-negative specimens tested positive in PCR; in comparison, 60 (77%) of 78 concordant RIBA-2/RIBA-3-positive units tested positive in PCR. CONCLUSIONS: RIBA-3 is significantly more sensitive than RIBA-2 in testing of HCV 3.0 EIA-screened donations. During the review process of this manuscript, the FDA licensed the RIBA-3 test.

Evaluation of indeterminate c22-3 reactivity in volunteer blood donors.

BACKGROUND: Approximately 25 percent of blood donor sera that are repeatably reactive for hepatitis C virus (HCV) on second-generation enzyme immunoassay (EIA 2.0) are indeterminate on second-generation recombinant immunoblot assay (RIBA 2.0), and over 76 percent of these results are due to single reactivity to the HCV recombinant antigen c22-3. STUDY DESIGN AND METHODS: Data are presented on 46 volunteer allogeneic blood donors who were reactive on EIA2.0 and c22-3 indeterminate in RIBA 2.0. Index and follow-up samples were evaluated by using a panel of five synthetic peptide EIAs, a prototype strip immunoblot assay that uses synthetic peptides in addition to recombinant protein (RIBA 3.0), and polymerase chain reaction (PCR) for HCV RNA. RESULTS: All 46 donations had normal alanine aminotransferase values; only 2 (4.3%) reacted for antibody to hepatitis B core antigen. With a panel of 12 synthetic peptides spanning the entire sequence of the c22-3 recombinant antigen, 33 plasmas (72%) reacted to one peptide or none, including 19 plasmas with reactivity restricted entirely to the N-terminal peptide (1-15 amino acids) of c22-3. With RIBA 3.0, 28 donations (61%) were nonreactive, including 25 that reacted with one peptide or none in EIA. Of these 25 plasmas, 18 reacted with the N-terminal sequence only. All 46 index donations were tested by PCR; the single PCR-positive unit reacted with four HCV peptides, was positive by RIBA 3.0, and reacted for antibody to hepatitis B core antigen. Twenty-six index donors were successfully recalled 3 to 7 months after their index donation. None seroconverted to positivity in RIBA 2.0, 1 was nonreactive, and 25 remained positive for c22-3 only. The restricted epitope reactivity in peptide EIA and RIBA 3.0 was maintained over time in all cases. All 26 of the follow-up samples tested negative by PCR. CONCLUSION: On the basis of the restricted peptide reactivity and PCR negativity of index and follow-up samples, it is concluded that the majority of c22-3 RIBA 2.0-indeterminate results are due to nonspecific cross-reactivity to restricted (principally, N-terminal) regions of HCV core antigen.