Background: Genotype 1 HCV patients treated withpegIFN/RBV who carry the good response IL28B variant (C/C,rs12979860) have dramatically improved early viral kinetics.Current strategies for the clinical development of directly-actinganti-HCV agents involve combination with pegIFN/RBV torestrict the development of antiviral resistance. Given the populationfrequency of the good response IL28B variant it is possiblefor small early phase efficacy trials to be confounded byimbalance for IL28B genotype across treatment arms. Methods:We statistically modeled the probability of an imbalance for theC/C IL28B genotype between treatment arms for 3 hypotheticalsituations - a phase 1 (n=60), 2a (n=120) and 2b (n=240)trial, each involving 3 randomized arms. We then modeled theimplication of such an imbalance for the primary outcome ofviral load reduction at week 4 in studies that combine a directantiviral with pegIFN plus RBV. We assumed i) a population frequencyof C/C = 33%; and ii) median viral load reduction at4 weeks of treatment of 3.8 logIU/mL vs 1.4 logIU/mL in C/Cvs non-C/C patients due to pegIFN alone, as previouslyreported for North American Caucasians (Thompson,Gastro,Apr 15, Epub). Results: The probability of an imbalance inone treatment arm of ± 10% (<23% or > 43%) was 31%, 18%and 6% for the phase 1, 2a and 2b trials, and for an imbalance± 20% was 10%, 0.4% and <0.01%. In a phase 1 trial,an imbalance for the good response C/C genotype of 10% -20% could lead to differences of 0.2 - 0.5 log between treatmentarms at week 4, due to pegIFN alone (Table 1). Conclusion:Knowledge of IL28B genotype distribution is important forinterpreting early phase clinical trial results, particularly dosefindingstudies where dose-related antiviral potency must beweighed against toxicity. Confounding by IL28B genotypeimbalance between treatment arms might affect the decision toadvance a compound from proof-of-concept to the next stage ofclinical development.