Unveiling the therapeutic profile of pomiferin: a meta-analysis of cytotoxicity and oxidative stress modulation in preclinical models

Objective Pomiferin, a prenylated isoflavonoid from Maclura pomifera , shows anticancer and antioxidant potential. However, data vary across cell lines and models. This meta-analysis aims to quantify its IC 50 in different cancer types and systematically synthesise its effects on oxidative stress biomarkers, thereby establishing a quantitative synthesis for its therapeutic profile. Methods A systematic search in PubMed, Scopus, and Web of Science was conducted following PRISMA guidelines. A “mean-only” approach was used to manage within-study dependency and inconsistent variance reporting. Data were analysed with multilevel random-effects models and Cluster-Robust Variance Estimation (CR2). Results The pooled geometric mean IC 50 for Pomiferin was 12.6 μM ( logIC 50 = 2.53). Meta-regression showed that efficacy varied by tumour type, with Glioma/neurological models having higher IC 50 values (lower efficacy) than others ( p = 0.012). Pomiferin increased Catalase activity ( p = 0.044) and protein levels ( p = 0.033 ). No statistically significant difference in efficacy was detected between in vitro and in vivo models ( p = 0.190 ); however, current data limitations preclude establishing definitive biological equivalence. Conclusion This meta-analysis demonstrates that Pomiferin is a context-dependent agent that modulates enzymatic defences, such as CAT, confirming its potential as a therapeutic candidate. It highlights the efficacy limits in resistant types, such as Glioma, and provides a quantitative basis for future research. This study consolidates diverse findings into a robust evidence base, serving as the first comprehensive meta-analysis of Pomiferin’s biological activity and guiding future studies.