Prognostic value of risk factors in Oral Potentially Malignant disorders and Oral Squamous Cell Carcinoma: an Umbrella Review

Background Oral Squamous Cell Carcinoma ranks the 7th most common cancer worldwide. Numerous systematic reviews and meta-analyses have been published over the past decade with many identified risk factors. This Umbrella aims to synthesize from the body of available evidence, which features of oral potentially malignant disorders (OPMDs) most contribute to malignant transformation, while reporting risk prognostic/prediction models and methodological gaps. Methods This umbrella review followed the PRISMA 2020 checklist and was registered with PROSPERO (CRD42024584847). A Comprehensive search of PubMed, Scopus, and Web of Science (up to February 2025) was done to identify systematic reviews and meta-analyses evaluating risk and prognostic factors in OPMDs and OSCC. Screening was done by two independent reviewers. Eligible studies were appraised using AMSTAR-2, with data synthesized descriptively. Results Ten systematic reviews and meta-analyses published between 2019 and 2024 were included, encompassing diverse populations and study designs. Frequently assessed prognostic factors included age, sex, tobacco and alcohol use, lesion site, histopathological grade, and molecular markers such as podoplanin, ALDH1A1, PROM1, and cancer stem cell–related biomarkers. Pooled malignant transformation rates varied across lesion types—1.16% for oral lichen planus and up to 19.9% for Erythroplakia. Strong associations were found between tobacco exposure and oral cancer risk (OR up to 15.3). Biomarkers like podoplanin and ALDH1A1 exhibited significant prognostic potential (HR: 2.9–8.9). Conclusion This umbrella review underscores that both molecular and clinical parameters, especially epithelial dysplasia, lesion site, tobacco exposure, and biomarkers that hold prognostic value for malignant transformation in OPMDs and survival in OSCC. Future research should prioritize standardized biomarker validation, multicentric prospective designs, and integrated prognostic models to improve personalized risk prediction and management.