Objective: Patient- and virus-related factors influence the response of patients with chronic hepatitis C to interferon-based therapy. The purpose of this study was to model the probability of achieving a sustained virological response in individual patients, taking into consideration various predictive factors.
Material and methods: We combined data from two randomized, multinational trials in which patients received peginterferon alfa-2a (40KD) plus ribavirin. The logistic regression model for patients infected with hepatitis C virus genotype 1 included age, viral load, histology, alanine aminotransferase quotient, body mass index, treatment duration, ribavirin dose and adherence.
Results: In the genotype 1 model, varying baseline factors had a striking effect on the probability of sustained virological response. A dramatic difference in the probability of sustained virological response was seen in a series of hypothetical patients in whom five factors were varied to represent best and worst case scenarios. The best case scenario (age 20 years; no cirrhosis/bridging fibrosis; alanine aminotransferase quotient=7; body mass index 20 kg/m2; viral load 40,000 IU/mL) was associated with a 97% probability of sustained virological response, compared with 7% in the worst case scenario (age 60 years; cirrhosis/bridging fibrosis; alanine aminotransferase quotient=1; body mass index 30 kg/m2; viral load 9,000,000 IU/mL). Both adherence to treatment and achieving an early virological response increased the probability of sustained virological response.
Conclusions: In treatment-naïve patients with chronic hepatitis C, host factors play a major role in determining treatment outcome and the logistic regression model is useful for predicting the probability of sustained virological response in individual patients.