Objectives: We performed a systematic review to evaluate the published biomarkers related to oral leukoplakia (OL), aiming to identify the biomarkers that indicate any future risk of cancer in patients with oral leukoplakia.

Methods: A search strategy was developed for three main electronic databases: PubMed, Cochrane Library, and EBSCO, and also for Google Scholar, until February 28, 2020. The study selection was performed in a two-phase process aiming at studies assessing tissue biomarkers for "malignant transformation of OL." Risk of bias analysis of included studies was performed using the Quality in Prognosis Studies Tool.

Results: From 3,130 articles initially identified by searching databases, a total of 46 studies were included in this systematic review, with a combined sample of 3,783 patients, of whom 1,047 presented with malignant transformation of a previously diagnosed OL as reported by the authors. The cancer incidence in the whole group was 27.6% (range: 5.4% to 54.1%). The studies were derived from different geographic areas, including Asia (n = 21), Europe (n = 15), North America (n = 9), and Oceania (n = 1). There were 49 different molecular biomarkers evaluated in the 46 included studies: p53 and podoplanin proteins were the most frequently reported, followed by abnormalities at particular chromosomal loci (e.g., LOH). Risk of bias analysis revealed concerns associated with "measurement of prognostic factor," "study confounding" and "statistical analysis and reporting."

Conclusions: Substantial heterogeneity and lack of standardized reporting of data among the studies were identified. The most promising biomarkers reported to have a significant association with the malignant transformation in OL included podoplanin and chromosomal loci abnormalities. A critical examination of the follow-up studies on OL published so far indicated that tissue biomarkers that could predict the risk of oral cancer in patients with OL are still in a discovery phase.