Poster Presentations

Session Title: Category 5g. VIRAL HEPATITIS: g. HEPATITIS C - CLINICAL (THERAPY)
Presentation Date: Apr 15, 2010

EVALUATING VIROLOGIC RESPONSE WITHIN 2-4 WEEKS HAS HIGH POSITIVE/NEGATIVE PREDICTIVE VALUES FOR SUSTAINED VIROLOGIC RESPONSE IN PATIENTS WITH GENOTYPE 1 CHRONIC HEPATITIS C DURING TREATMENT

A. Neumann^1,2*, M. Shiffman³, P. Ferenci⁴, G. Foster⁵, E. Yoshida⁶, S. Pianko⁷, M. Sulkowski⁸, S. Zeuzem⁹, Y. Benhamou¹⁰, D. Nelson¹¹, J. McHutchison¹², E. Pulkstenis¹³, G. Subramanian¹³, for the ACHIEVE Study Group
¹Faculty of Life Sciences, Bar-Ilan University, Ramat-Gan, Israel, ²Institute for Theoretical Biology, Humboldt University, Berlin, Germany, ³Virginia Commonwealth University Medical Center, Richmond, VA, USA, ⁴Medical University of Vienna, Vienna, Austria, ⁵The Royal London Hospital, London, UK, ⁶University of British Columbia, Vancouver, BC, Canada, ⁷Monash Medical Centre, Clayton, VIC, Australia, ⁸Johns Hopkins Center for Viral Hepatitis, Baltimore, MD, USA, ⁹J.W. Goethe-University Hospital, Frankfurt/Main, Germany, ¹⁰Hôpital Pitié-Salpêtrière, Paris, France, ¹¹University of Florida, Gainesville, FL, ¹²Duke Clinical Research Institute, Durham, NC, ¹³Human Genome Sciences, Inc., Rockville, MD, USA. *neumann@mail.biu.ac.il

Background and aims: Knowing the positive/negative predictive values (PPV/NPV) of SVR as early as possible during CHC treatment allows for better patient management. Albinterferon alfa-2b (albIFN) is a novel, long-acting, fusion polypeptide that is dosed every 2 weeks. This study validated a previously identified prediction algorithm that allows high positive prediction as early as 2 weeks and high negative prediction as early as 4 weeks after starting therapy.
Methods: 1323 patients were randomized to peginterferon alfa-2a 180 µg/wk, or albIFN 900 or 1200 µg q2wk, all with weight-based ribavirin 1000-1200 mg/d, for 48 weeks. HCV RNA was measured by real-time PCR (lower limit of detection 15 IU/mL; limit of quantitation 43 IU/mL). A prospectively defined positive and negative prediction algorithm at treatment weeks 2-4 was compared with rapid virologic response at week 4 and no early virologic response at week 12 in the per-protocol population (n=747). Since noninferiority in SVR rates (56%-59%) and similarity of early predictive parameters among all arms were established, data were pooled for the prediction analysis (n=747). Results were similar for the ITT population.
Results: Rapid initial virologic response at week 2 (viral decline >2 log₁₀IU/mL) was observed in 34% of patients, with a high PPV (92%) for SVR. This early week-2 algorithm applied to more patients and compared favorably with rapid virologic response found in 26% of patients (PPV 98%). No initial virologic response at week 4 (viral decline < 2 log₁₀IU/mL; viral load >5.5 log₁₀IU/mL), found in 14% of patients, demonstrated a high NPV (96%) with 33% specificity. This early endpoint compared favorably with no early virologic response observed 8 weeks later at week 12 (NPV 98% in 14% of patients). A sequential prediction algorithm based on viral kinetics at weeks 2 and 4 reliably predicted treatment outcome in 67.5% of patients significantly earlier than week 12.
Conclusions: Using early viral kinetics information, it is possible to achieve both a high PPV and NPV for SVR as early as treatment weeks 2-4. Identifying early predictors of on-treatment response is important for tailoring response-guided therapy in the future with direct antiviral CHC therapy.