Abstract Background and aims Although the CHAIN trial showed neutral overall effects of FYTF-919 on ICH, treatment responses appeared to vary across patient profiles. We aimed to characterise treatment-effect heterogeneity and identify potential subgroups that derive benefit from FYTF-919. Methods CHAIN was a prospective, randomised, double-blinded, placebo-controlled trial at 26 hospitals in China. Adults with acute moderate-to-severe ICH were randomly assigned to FYTF-919 or placebo for 28 days. The primary outcome was uw-mRS at 90 days. Baseline characteristics were assessed for effect modification using causal forest and non-linear regression, and further evaluated in categorised subgroups analyses with interaction terms added. This trial is registered at ClinicalTrials.gov, NCT05066620. Results Between Nov 24, 2021 and Dec 28, 2023, 1619 participants with available 90-day uw-mRS were included (801 FYTF-919; 818 placebo). Hematoma volume, age and time to randomisation were the strongest contributors to treatment-effect heterogeneity. Non-linear interaction was observed for hematoma volume (p=0·008). Patients with moderate hematoma (16-56mL) showed more favourable 90-day uw-mRS with FYTF-919 (difference 0·07, 95%CI 0·02 to 0·11), whereas no benefit was observed among small or large hematoma. Treatment effects also differed by time to randomisation (15h vs. ≥15h, p=0·029 for interaction), with more favourable outcomes among participants randomised later. No heterogeneity was observed for other characteristics including age, GCS and hematoma location. Conclusions Treatment effects of FYTF-919 in acute ICH varied across hematoma volume and timing of randomisation. These findings suggest potential benefit in selected subgroups and warrant validation in future trials. Conflict of interest Weiyi Zhong, Dou Wang, Liling Zeng, Jianwen Guo, Craig Anderson, Lili Song: nothing to disclose Figure 1 - belongs to Results Figure 2 - belongs to Results
Zhong et al. (2026) studied this question.