NATURAL RECOVERY FROM ACUTE HEPATITIS C VIRUS INFECTION BY AGAMMAGLOBULINEMIC TWIN CHILDREN

Abstract

In February, 1994, 14 possible cases of acute hepatitis C virus (HCV) infection among recipients of an intravenous immunoglobulin (IVIG) preparation (Gammagard®; Baxter Healthcare Corporation) were reported to the United States Food and Drug Administration.1 Baxter withdrew Gammagard® distribution and recommended retrospective HCV infection surveillance for IVIG recipients of the suspected contaminated lots. This report documents acute HCV infection in X-linked agammaglobulinemic (XLA) twins (lack of HCV-specific antibody confirmed with a highly sensitive and specific enzyme-linked immunosorbent assay2) and describes an unusually mild HCV clinical course lacking progression to chronicity and accompanied by clearance of the virus. The benign outcome contrasts sharply with the frequent experience of chronic hepatitis3 and with the severe, rapidly progressive HCV disease recently described by Bjoro et al.4 in hypogammaglobulinemic patients receiving a Swedish HCV-contaminated IVIG (Gammonativ®) as well as by Jonas et al.5 Continuous clinical and virologic monitoring of disease progression in patients with a complete lack of antibody might provide valuable insights into the possible role of antibody in the progression of HCV infection. Methods and results. Nine-year-old XLA twin boys received monthly IVIG beginning soon after birth when their congenital absence of immunoglobulin-producing B lymphocytes was confirmed. They were known to be at risk for the disease because a deceased older brother had XLA. Routine liver function tests were performed in early March, 1994, because they had received Gammagard® IVIG suspected of being contaminated with HCV. Liver transaminase elevations reflected acute injury (Fig. 1). In addition to alanine aminotransferase (ALT), represented in the figure, aspartate aminotransferase, gamma-glutamyltranspeptidase, alkaline phosphatase and cholylglycine (bile acids) concentrations were also monitored, with similar results (data not shown). Serum iron, iron-binding capacity, ferritin and transferrin values were slightly low. XLA patients cannot synthesize antibody; therefore serologic markers of previous infection could not be used to identify the hepatitis etiology. Serum samples tested for hepatitis B DNA6 were negative. There were no reported cases of hepatitis caused by either hepatitis A virus (HAV) or hepatitis E virus in the vicinity and the patients had not traveled to places where HAV cases were prevalent. Within 2 months after peak elevation of the liver enzymes, HCV RNA was detected by reverse transcription (RT)-PCR (using either Primer Set 1 (antisense 5′-CCCACACTACTCGGCTA-3′; sense 5′-CACTCCACCATGAATCACTCCCC) or Primer Set 2 (antisense 5′-TCGCAAGCGCCCTATCAGGCAG-3′; sense 5′-GGCGACCACTCCACCATAGATC-3′)) by standard protocol,7 confirming HCV infections. Serial testing for HCV RNA indicates apparent clearance of the viremia. HCV RNA was not detected by the branched DNA assay. It is possible that during the peak elevation of ALT the serum samples were not properly collected and preserved, leading to possible false negative RTPCR for HCV RNA. Subsequently samples were carefully collected and preserved to ensure the stability of the RNA.Fig. 1: Serum levels of ALT of the twin boys postinfection and follow-up for >3 years, 1993 to 1996. Samples were drawn at regular intervals during their visits for IVIG infusion. Boxed area, period when the plasma samples were positive for HCV RNA by RT-PCR analysis. Samples drawn on July 7, 1993, November 15, 1993, March 11, 1994, May 13, 1994, and June 29, 1994, were tested for HCV RNA by bDNA analysis and were negative, implying an absence of HCV RNA or a very low level of viremia (<350 000 HCV eq/ml).Because Twin B had experienced enteroviral meningitis 9 months previously, he was receiving biweekly IVIG at the time of onset of hepatitis, rather than the customary monthly infusions his brother received. Although it was known that current IVIG preparations did not contain anti-HCV antibody, at that time both boys were given IVIG biweekly in hopes that it might confer some benefit. The only clinical indication of hepatitis was slight nontender liver enlargement. Liver function test results returned to normal (Fig. 1) and the children remained clinically normal. They have each maintained growth at the 50th percentile. They do have chronic sinusitis and other occasional pyogenic infections. Discussion. Asymptomatic XLA twins who had received Gammagard® IVIG were tested for hepatitis because other Gammagard® recipients had developed HCV hepatitis. They had no evidence of autoimmune disease or other systemic illness. Diagnosing an infectious etiology of hepatitis in XLA patients presents difficulty because they do not mount an antibody response; therefore serologic markers of infection by usual hepatitis viruses are not available, and it is necessary to search directly for the agents' antigens or genomes. In these cases hepatitis B surface antigen and HBV DNA test results were negative. We were not able to perform HAV antigenic or genomic tests, but there was no clinical or epidemiologic evidence of HAV infection. HCV RNA was identified in serum samples by two independent laboratories, suggesting that the boys had HCV hepatitis. Liver biopsy was not clinically indicated because of the patients' benign courses; therefore definitive demonstration of HCV-RNA in affected liver tissue was not available. The lack of detection of HCV-RNA by bDNA is a likely indication of a low level of viremia. HCV-PCR testing of IVIG lots associated with postinfusion hepatitis in Gammagard® recipients reveals that each of our patients received IVIG from at least four HCV RNA-contaminated lots (Table 1).8TABLE 1: Dates of IVIG infusion and the corresponding lot numbers that were received by each of the twins Comparison of infusion dates with illness onset dates suggests that infusions on October 1, 1993, or February 4, 1994, were most likely to have been responsible for the infection. Lot 2317A 166 infused on October 1, 1993, had a higher HCV concentration than the other lots. Although patient-to-patient HCV transmission has been reported in a hematology ward,9 nosocomial infection is very unlikely in our patients. Perhaps the most interesting HCV infection feature in these boys is the clearance of virus from the serum in the absence of specific antibody associated with complete recovery, sustained for 30 months as of this writing. This recovery contrasts sharply with the rapidly progressive, severe HCV disease of hypogammaglobulinemic patients receiving Swedish IVIG, Gammonativ®.4 The study by Jonas et al.5 included one patient with XLA and although the follow-up of that study is recorded up to only 1 year, this single patient had normal serum transaminase concentrations, further confirming our unique observation of a mild course of HCV in infected XLA patients. To date there have been 11 reported cases of XLA patients with HCV infection; of these 9 had a mild course, 1 has chronic HCV disease and 1 is deceased. In contrast there are 24 reported cases in patients with common variable immunodeficiency; 6 had a mild course, 11 had chronic HCV infection and 7 patients are deceased.10 It is possible that our patients have a latent infection that will be manifested later. It is also possible from their experience that specific antibody is not required for HCV clearance and recovery from HCV hepatitis. If true, this might imply that HCV antibody plays a necessary role in the usual progression of HCV infection to chronic hepatitis. It is also possible that in some such XLA cases the lack of mature B lymphocytes may account for the unusual progression of HCV infection; if B cell infection is a prerequisite to establish an HCV productive infection in the host, lack of B cells may hamper the establishment of HCV persistency. Another possibility is whether antibody-dependent enhancement, a phenomenon not described for HCV but occurring in many flaviviridae infections, accounts for the abortive infection in some agammaglobulinemic patients. This study also supports a plausible hypothesis that HCV elicits an antibody-dependent autoimmune response as well as an antibody-dependent cell-mediated cytotoxic response which contributes to the liver damage. If this hypothesis is proved to be correct then suppression of antibody production with immunosuppressive agents might prevent the development of liver damage and possibly progression to chronic infection. Acknowledgments. The authors gratefully acknowledge the assistance of Drs. Peter J. Dailey and Judith Wilbur from Chiron Corporation with bDNA analysis. The RT-PCR analysis data on the IVIG lots was kindly provided by Dr. Mei-Ying W. Yu of the Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration. Ms. Cathie G. Miller assisted in preparation of the figure. This work was supported by funds from Humana, provided to GJK. Garrett Adams, M.D., M.P.H. Sean Kuntz, B.S. Gerard Rabalais, M.D. Denise Bratcher, D.O. Carlo H. Tamburro, M.D., M.P.H. Girish J. Kotwal, Ph.D. University of Louisville School of Medicine; Department of Pediatrics (GA, GR, DB), Microbiology and Immunology (SK, GJK); and Medicine (CHT); Louisville, KY