Chenchen Wu
We commend Huang et al. for separating treatment from postoperative prophylaxis in CAT/VTE. However, the pooled analysis still combines clinically distinct exposure patterns, heterogeneous LMWH comparators, and studies with different follow-up windows, which limits interpretability of the reported risk ratios. Observational before-after and quality-improvement studies are also vulnerable to secular trend bias, and acute treatment should not be merged with extended therapy. Future meta-analyses should use drug-specific stratification, time-standardized or hazard-based methods, concurrent controls, and phase-specific analyses to better estimate apixaban's effect while preserving clinically meaningful differences between indications and treatment phases in cancer-associated thrombosis across settings and improving decision-making in practice.