468A AASLD ABSTRACTS HEPATOLOGY, October 2003 638 ANALYSIS OF THE INVOLVEMENT OF PKR PROTEIN IN RESISTANCE TO THERAPY FOR CHRONIC HEPATITIS C VIRUS INFECTION. Gerry C MacQuillan, University Hospital Birmingham, Birmingham, UK; Paul Caterina, W Bash'aan de Boer, Michael Platten, PathCentre, Nedlands, Australia; William D Reed, Hollywood Pm'vate Hospital' Nedlands, Australia; Gary P Jeffrey, University of Western Australia / Sir Charles Gairdner Hospital' Nedlands, Australia Background: The frequent failure of IFN based therapies to clear HCV in patients chronically infected with the genotype I strain has been attributed to inhibition of PKR function by HCV pro- teins, particularly NS5A and E2. Functional PKR phosphorylates downstream eIF-2 alpha shutting off cellular and viral translation. Recently a SNP has been identified in the regulatory region of the eIF-2 alpha gene which may influence response to IFN treatment in chronic HBV infection (King et ai, Hepatology, 2003). Whether this is relevant in chronic HCV disease is unknown. Method: Since functional PKR is required for phosphorylation of eIF-2 alpha, alteration in the activity of PKR during chronic HCV infection was assessed using immunohistochemistry to stain for the presence of phosphorylated eIF-2 alpha protein. Evidence of a change in PKR function was also searched for through alteration in the intracel- lular in vivo location of PKR through immunogold electron-mi- croscopy. Results (1): Phosphorylated eIF-2 alpha protein expression was largely confined to hepatocyte nuclei in liver tis- sue, although a weak cytoplasmic blush was observed which was not readily quantifiable. The number of positive staining hepato- cyte nuclei per 400 hepatocytes counted (-+ S.D) was not signifi- canfiy different between HCV genotype I (60% nuclei positive -+ 5.4%, n - 12), genotype 3 infected patients (55% nuclei positive -+ 6.2%, n - 16) and normal controls (59% nuclei positive -+ 3.5%, n - 4). In addition, there was no significant difference in pre-treat- ment phosphorylated eIF-2 alpha protein expression between HCV genotype I infected sustained responders (60% nuclei pos- itive -+ 4.3%, n - 5) and non-responders (61% nuclei positive -+ 3.9%, n - 7) to standard IFN-alpha/ribavirin combination therapy. Results (2): Quantification of immunogold labelling was expressed as the mean number of gold particles//±m 2 per nuclear or cyto- plasmic region of each cell -+ S.E.M. Immunogold staining of PKR protein in IFN-alpha treated PBMC normal controls demonstrated a significant increase in nuclear (p - 0.03) and cytoplasmic vesic- ular (rough endoplasmic reticulum) labelling (p - 0.004) for PKR (36 -+ 4 gold particles//±m 2 and 45 _+ 7 gold particles//±m 2 respec- tively), compared to the untreated PBMC (18 -+ 5 gold particles/ p~m 2 and 3 _+ I gold particles//±m 2 respectively). A similar distribution of PKR protein in the nucleus and cytoplasm of HCV positive (n - 4) and HCV negative liver biopsies (n - 1) was observed (mean nuclear and cytoplasmic count for PKR protein for the HCV group was 21 _+ 4 and 18 _+ 3 gold particles//±m 2 respectively). These finding confirmed previous in vitro studies of PKR ultrastructural localisation. Conclusion: Differences in re- sponse to IFN-alpha/ribavirin therapy in an Australian cohort of HCV genotype I patients could not be attributed to differences in the extent of pre-treatment activation of PKR protein. These find- ing suggest that PKR function and location in cellular translation is uninterrupted despite the presence of the potentially inhibitory HCV proteins. Disclosures: Paul Caterina - No relationships to disclose W. Bastiaan de Boer - No relationships to disclose Gary P Jeffrey - No relationships to disclose Gerry C MacQuillan - No relationships to disclose Michael Platten - No relationships to disclose William D Reed - No relationships to disclose 639 THE HCV CORE AG/HCV RNA RATIO PREDICTS THE COURSE OF HCV REPLICATION IN TREATED AN UNTREATED LIVER TRANSPLANTED PATIENTS. Cyrille Feray, Paul Brousse Hospital, Villejuif, France; Pascale Berthillon, Hotel-Dieu Hospital, Lyon, France; Delphine Ducoulombier, Paul Brousse Hospital, Villejuif, France; Marianne Maynard, Liana Codes, Hotel-Dieu Hospital, Lyon, France; Michelle Gigou, Paul Brousse Hospital, Villejuif, France; Thierry Bizollon, Hotel-Dieu Hospital, Lyon, France; Didier Samuel, Paul Brousse Hospital, Villejuif, France; Christian Tr~po, Hotel-Dieu Hospital, Lyon, France Background : Accurate quantification of HCV core Ag in plasma is now available. The ratio HCVcoreAg/HCV RNA may theoretically reflect composition of circulating viral particles and the presence of defective interferent particles. Patients and methods : Liver transplanted patients with chronic hepatitis C (mean: 5 years after transplantation) were randomized to received either no therapy or interferon alpha-b and ribavirin for 48 weeks (Gastroenterology 2003;124:642-65). Among these 52 patients, 46 were followed through longitudinal measurements of serum HCVag (HCV Track(r)) and viral load each 3 months from randomization for up to 24 months. Results : The ratio HCVcoreAg/HCV RNA (R) measured at entry (R0) had a gaussian distribution. In non-treated patients, high levels of HCV core Ag and R at entry were correlated to a decrease of viral load in the 2 following years (p < 0.01 ; p<0.04). In treated patients, prolonged viral response was positively related to high R0 (t-test ; p-0.03 ; p-0.04) but not to initial HCV RNA or HCV core Ag. In non-treated patients, R0, R 1 year and R 2 years were not correlated. In contrast, when R was compared at 3 months interval, a correlative pattern emerged (R1 month vs R3 months : p-0.01 ; R3 months vs R6 months : p : 0.01 ; R 6 months vs R 1year : p : 0.04). Progression of viremia or response to therapy were not related to other factors (genotype, initial viral load, histological scores, gender or age). Conclusion : In this prospective study, the ratio HCVcoreAg/ HCVRNA fluctuated during 2-year observation and was a predic- tive marker of interferon response in treated patients and of spontaneous decrease of viral replication in others. Increase ratio could reflect defective interfering particles without HCV RNA and could help to determine the optimal timing for initiating inter- feron. Disclosures: Pascale Berthillon - No relationships to disclose Thierry Bizollon - No relationships to disclose Liana Codes - No relationships to disclose Delphine Ducoulombier - No relationships to disclose Cyrille Feray - No relationships to disclose Michelle Gigou - No relationships to disclose Marianne Maynard - No relationships to disclose Didier Samuel - No relationships to disclose Christian Tr6po - No relationships to disclose 64O HCV ALTERNATE READING FRAME PROTEINS (ARFPS) MAY BE VIRULENCE FACTORS THAT HELP THE VIRUS SURVIVE ADVERSE CONDITIONS. Andrea D Branch, J L Walewski, J A Guh'errez,~ J Fernandez, G Y Im, D T Dieterich, T D Schiano, S H Sigal' M L Schilsky, M~ Schwartz, Mount Sinai School of Medicine, New York, NY; C R Kyrk, A S Muerhoff, T P Leafy, J A Stewart, G J Dawson, S M Desai, Abbott Laboratories, Abbott Park, IL Background: HCV expresses alternate reading frame proteins (ARFPs) that stimulate immune responses (see Walewski et al., RNA 2001; Gosert and Moradpour, Hepatology 2002). We hypoth- esize that expression of ARFPs is modulated by RNA signals in the