To update the current evidence on the malignant transformation of oral leukoplakia (OL), including all studies published worldwide on the subject, selected with the maximum rigor regarding eligibility. MEDLINE, Embase, Web of Science and Scopus were searched for studies published before June-2024, with no lower date limit. The risk of bias was analyzed using the Joanna Briggs Institute tool for meta-analyses of proportions. We carried out meta-analyses, explored heterogeneity across subgroups and identified risk factors with potential prognostic value. Fifty-five studies (41,231 with OL) were included. The pooled malignant transformation proportion for OL was 6.64% (95% CI = 5.21-8.21). The malignant transformation did not significantly vary by time periods (p = 0.75), 5.35% prior to 1978, 7.06% from 1979 to 2007 and 6.97% during more recent times. The risk factors that significantly had a higher impact on malignant transformation were the non-homogeneous leukoplakias (RR = 4.23, 95% CI = 3.31-5.39, p < 0.001), the larger size (RR = 2.08, 1.45-2.96, p < 0.001), leukoplakia located on the lateral border of tongue (malignant transformation = 12.71%; RR = 2.09, 95% CI = 1.48-2.95, p < 0.001), smoking (RR = 1.64, 95% CI = 1.25-2.15, p < 0.001), and the presence of epithelial dysplasia (RR = 2.75, 95% CI = 2.26-3.35, p < 0.001). OL presents a considerable malignant transformation probability that is especially increased in large non-homogeneous lesions in smokers, located on the lateral border of the tongue, with epithelial dysplasia.