Plasma from 173 patients with HCV genotype 1 infection was analyzed for IP-10 levels prior to treatment with pegylated interferon-alpha-2a and ribavirin. Significantly lower IP-10 levels were observed in patients achieving a rapid viral response (RVR) (P < .0001), even in those with body mass index (BMI) > or = 25 kg/m2 (P = .004) and with baseline viral load > or = 2 million IU/mL (P = .001). Similarly, significantly lower IP-10 levels were observed in patients obtaining a sustained viral response (SVR) (P = .0002), including those having higher BMI (P < .05), higher viral load (P = .0005), and both higher BMI and viral load (P < .03). In multivariate logistic regression analyses, a low IP-10 value was independently predictive of both RVR and SVR. A baseline cutoff IP-10 value of 600 pg/mL yielded a negative predictive value (NPV) of 79% (19/24) for all genotype 1-infected patients, which was comparable with that observed using a reduction in HCV-RNA by at least 2 logs after 12 weeks of therapy (NPV 86%; 19/22); by combining the two, 30 of 38 patients (NPV 79%) potentially could have been spared unnecessary therapy. In patients having both higher BMI and viral load, cut-off levels of 150 and 600 pg/mL yielded a positive predictive value (PPV) of 71% and NPV of 100%, respectively. In conclusion, pretreatment IP-10 levels predict RVR and SVR in patients infected with HCV genotype 1, even in those with higher BMI and viral load. A substantial proportion of the latter patients may achieve SVR in spite of unfavorable baseline characteristics if their pretreatment IP-10 level is low. Thus, pretreatment IP-10 analysis may prove helpful in decision-making regarding pharmaceutical intervention.