Bridging the gap: meta-epidemiological analysis on the clinical translation of stem cell-based therapies in women's reproductive diseases.

study questionWhat is the current landscape of randomized controlled trials (RCTs) evaluating stem cell-based therapies for women's reproductive diseases, and how effectively has preclinical research informed their clinical translation? SUMMARY ANSWER: The current clinical trial landscape for stem cell-based therapies in women's reproductive diseases is characterized by rapid growth in trial registration, particularly for premature ovarian failure (POF) and intrauterine adhesions (IUA), yet remains predominantly in early-phase, single-center, single-country studies with significant heterogeneity in design and outcome reporting. WHAT IS KNOWN ALREADY: Stem cell-based therapies show promise in preclinical models for conditions such as POF, IUA, and endometriosis. However, despite increasing clinical interest, the extent to which these therapies have been rigorously evaluated in RCTs and how well clinical trials reflect preclinical evidence remains unclear. STUDY DESIGN SIZE DURATION: This translational analysis integrated a meta-epidemiological approach with bibliometric methods. We searched the International Clinical Trials Registry Platform (ICTRP), ClinicalTrials.gov, EudraCT, and ChiCTR from inception to 1 December 2025, using terms related to 'stem cells', 'reproductive diseases', and 'randomized controlled trials'. Only English and Chinese registration records were included. PARTICIPANTS/MATERIALS SETTING

methodsTwo reviewers independently screened registration records for RCTs assessing stem cell or derivative therapies in women's reproductive diseases. Eligible criteria included RCTs that reported the use of stem cells or their derivatives in female patients with reproductive conditions. Data on trial design, disease focus, stem cell type, source, phase, sample size, outcomes, and results were extracted. A hybrid approach combined traditional systematic data extraction with machine learning-assisted screening (ASReview) for basic research. Bibliometric analysis using VoSviewer and CiteSpace mapped research trends, while Health Research Classification System (HRCS) scores assessed translational potential. Consistency in screening and extraction was evaluated using kappa and Cronbach's alpha. MAIN RESULTS AND THE ROLE OF CHANCE: From 804 records, 38 RCTs met inclusion criteria. The majority were early-phase (Phase 1/2, 76.3%), single-center (81.6%), and conducted in one country (97.4%), with China (50.0%) and the USA (18.9%) leading. POF (34.2%), ovarian cancer (23.7%), and IUA (18.4%) were the most studied conditions. Umbilical cord-derived mesenchymal stem cells (28.9%) were the most commonly used. While preclinical research shows strong mechanistic support, especially for POF and IUA, there was minimal overlap in primary outcome measures across trials, limiting comparability. Only 36.8% of RCTs included safety as a co-primary endpoint, and 49% had short-term follow-up, raising concerns about long-term risks such as tumorigenicity. LIMITATIONS REASONS FOR CAUTION: The analysis may be subject to language bias (English and Chinese only), and some included trials were incomplete, relying on interim reports or press releases not peer-reviewed. The quality of individual RCTs was not formally assessed, and publication bias in basic research may influence bibliometric trends. WIDER IMPLICATIONS OF THE

findingsThis study highlights a significant gap between robust preclinical evidence and fragmented clinical evaluation, with over-concentration on POF and underrepresentation of other high-potential conditions like endometriosis and thin endometrium. The findings underscore the need for international collaboration, standardized outcome sets, and harmonized regulatory frameworks to enable large-scale, high-quality trials that can translate regenerative potential into clinical reality. STUDY FUNDING/COMPETING INTERESTS: This study was supported by the National Natural Science Foundation of China (82301835 and 82371682), the Science and Technology Program of Hunan Province (2024JJ4091 and 2023JJ40956), the Youth Fund of Xiangya Hospital, Central South University, China (2021Q03 and 2209090550121), the Postdoctoral Science Foundation of China (2022M713522 and 2021TQ0372), and the Innovation and Entrepreneurship Training Program for College Students (202510533012, S202410533307, and CXPY2025224). The authors declare no competing interests. REGISTRATION NUMBER: https://doi.org/10.17605/OSF.IO/JM9EK.