Because of the recent Gammagard incident (see Z01 BQ 04007-99), a retrospective cohort study was initiated to determine the risk of and risk factors for HCV infection among patients with immune deficiencies who had received IGIV between March 1993 and February 1994. Serum from each patient was tested for evidence of HCV infection. Gammagard lots administered to at least 10 patients in the cohort were tested for HCV RNA by our RT-PCR assay and the titer of HCV RNA in positive lots was determined by limiting dilution. The RNA was directly extracted from 4 ml of the 10% reconstituted IG solutions without ultracentrifugation because of the poor pelleting efficiency observed especially for those prepared from multiantigen screened plasma. HCV RNA was determined in IGIV lots made by Baxter which include 37 lots of Gammagard (made mainly from Source Plasma) and 6 lots of Polygam (made from Recovered Plasma from the American Red Cross) used in the cohort. HCV RNA was not detected in 9 lots of Gammagard prepared from a mixture of both EIA-1 and EIA-2 screened plasma but detected in 18 of 28 lots prepared solely from EIA-2 screened plasma. Among Polygam lots tested, one of the 5 lots prepared mostly from EIA-1 screened plasma was HCV RNA-positive albeit with a low titer while one lot solely from EIA-2 screened plasma was negative. Other IGIV products used in the cohort which included Ivegam and Gamimune-N had no detectable HCV RNA. Of 210 patients who received Gammagard, 23 became infected compared to none of 52 patients who received exclusively other IGIV products (including Polygam). HCV infection was associated only with Gammagard produced solely from plasma screened by EIA-2. There was a dose-response relationship between HCV infection and the quantity of HCV RNA infused in patients. As an interim measure to provide additional assurance of safety, we in the FDA began (since January 1995) the lot-release testing of those immunoglobulin products which have no viral inactivation procedures in their manufacture. About 90% of the lots had no detectable HCV RNA and were subsequently released. Since March 1995, those previously released in-date lots were also tested for HCV RNA. Our preliminary data also indicated that (1) the buoyant density profile of HCV present in an anti-HCV negative Gammagard lot was different from that observed in an anti-HCV positive IG lot, and (2) in the absence of anti-HCV, more HCV partitions into immunoglobulin during fractionation. The study is still in progress to determine reasons for the infectivity of Gammagard.
