Melbourne Dental School, The University of Melbourne, Melbourne, Vic., Australia.
Center for Oral and Systemic Health, Marshfield Clinic Research Institute, Marshfield, WI, USA.
Oral Dis. 2021 May;27(4):848-880. doi: 10.1111/odi.13363. Epub 2020 May 12.
To systematically review retrospective studies examining prognostic potentials of candidate biomarkers to stratify malignant progression of oral leukoplakia (OL) and proliferative verrucous leukoplakia (PVL).
A systematic literature search of PubMed, EMBASE, Evidence-Based Medicine and Web of Science databases targeted literature published through 29 March 2018. Inter-rater agreement was ascertained during title, abstract and full-text reviews. Eligibility evaluation and data abstraction from eligible studies were guided by predefined PICO questions and bias assessment by the Quality in Prognosis Studies tool. Reporting followed Preferred Reporting Items for Systematic Review and Meta-Analysis criteria. Biomarkers were stratified based on cancer hallmarks.
Eligible studies (n = 54/3,415) evaluated 109 unique biomarkers in tissue specimens from 2,762 cases (2,713 OL, 49 PVL). No biomarker achieved benchmarks for clinical application to detect malignant transformation. Inter-rater reliability was high, but 65% of included studies had high "Study Confounding" bias risk.
There was no evidence to support translation of candidate biomarkers predictive of malignant transformation of OL and PVL. Systematically designed, large, optimally controlled, collaborative, prospective and longitudinal studies with a priori-specified methods to identify, recruit, prospectively follow and test for malignant transformation are needed to enhance feasibility of prognostic biomarkers predicting malignant OL or PVL transformation.
系统回顾回顾性研究，以评估候选生物标志物在口腔白斑（OL）和增殖性疣状白斑（PVL）恶性进展分层中的预后潜力。
通过 3 月 29 日检索 PubMed、EMBASE、循证医学和 Web of Science 数据库，对目标文献进行检索。在标题、摘要和全文审查期间确定了评分者间的一致性。根据预设的 PICO 问题和预后研究质量工具对合格研究进行了评估和数据提取。报告遵循系统评价和荟萃分析的首选报告项目标准。根据癌症特征对生物标志物进行分层。
合格研究（n=54/3415）评估了 2762 例（2713 例 OL，49 例 PVL）组织标本中的 109 种独特的生物标志物。没有生物标志物达到检测恶性转化的临床应用基准。评分者间可靠性较高，但 65%的纳入研究存在高“研究混杂”偏倚风险。
没有证据支持候选生物标志物预测 OL 和 PVL 恶性转化的转化。需要设计系统、大型、最佳控制、协作、前瞻性和纵向研究，使用预先指定的方法来识别、招募、前瞻性随访和检测恶性转化，以提高预测恶性 OL 或 PVL 转化的预后生物标志物的可行性。
