Background & aims: We evaluated the impact of hepatitis C virus (HCV) treatment adherence support among people who inject drugs (PWID) in India using a precision trial design.

Methods: Treatment-naïve participants with a history of injection drug use were recruited from community-based clinics across seven cities. All received sofosbuvir/velpatasvir once daily for 12 weeks. Failure risk was defined a priori using a prognostic score including age, sex, income, homelessness, injection frequency, depressive symptoms, quality of life indicators and sexual partners. Participants at elevated risk were randomized 3:2:1 to high- (patient navigator [PN]+flexible directly observed therapy [≥1 dose/week observed]), medium- (PN contact ≥every 2 weeks) or low-intensity support. Those at minimal risk were randomized 1:2:3 to high-, medium-, and low-intensity support. The primary outcome was sustained virologic response (SVR; HCV RNA <LLOQ 24 weeks post-randomization; intention to treat).

Results: A total of 3,000 participants were randomized between January 2021 and December 2022 (2,048 minimal risk and 952 elevated risk), and 2,798 (93.3%) completed SVR assessment. Among minimal-risk participants, SVR was 62.6%, 60.8%, and 68.3% with low-, medium-, and high-intensity support, respectively. Among elevated-risk participants, SVR was 48.4%, 45.5%, and 50.8%, respectively. Fifty-one participants experienced a serious adverse event, including 35 deaths. Among minimal-risk participants, high-intensity support was superior to low-intensity support (adjusted relative risk [aRR] 1.09; 95% CI 1.00-1.19; p = 0.04), whereas medium-intensity support was not (aRR 0.97; 95% CI 0.90-1.05). Among elevated-risk participants, neither low-intensity (aRR 0.96; 95% CI 0.80-1.15) nor medium-intensity (aRR 0.89; 95% CI 0.77-1.04) support differed from high-intensity support. Each 10% decrease in prognostic score was associated with a 6% increase in the likelihood of SVR (aRR 1.06; 95% CI 1.04-1.08).

Conclusions: Prognostic scores could help target adherence interventions more efficiently; however, more intensive adherence support and/or novel interventions are needed to improve SVR among those at highest predicted risk.