This study was conducted to assess the diagnostic accuracy of long non-coding RNAs (lncRNAs) in differentiating patients with oral squamous cell carcinoma (OSCC), and to explore their potential role in early detection.

A literature search was conducted in PubMed, Embase, Web of Science, and the Cochrane Library for studies published updated to 1 December 2024. Studies that explored the diagnostic accuracy of lncRNAs in OSCC were included. The risk of bias was assessed using the QUADAS-2. Diagnostic accuracy was calculated using indicators such as sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) by random effects modeling.

A total of eight studies were included in this meta-analysis, comprising 532 patients diagnosed with OSCC and 446 controls. The pooled sensitivity was 0.87 (95 % CI: 0.61 to 0.97), and the specificity was 0.87 (95 % CI: 0.76 to 0.93). The pooled PLR was 6.75 (95 % CI: 3.12 to 14.58), and the NLR was 0.14 (95 % CI: 0.04 to 0.57). The overall DOR was 46.87 (95 % CI: 6.20 to 354.18). A higher diagnostic value was found for blood and/or saliva lncRNAs (DOR: 123.22; 95 % CI: 9.79 to 1550.8) than for tissue lncRNAs (DOR: 6.24; 95 % CI: 3.29 to 11.86).

This study has demonstrated that lncRNAs were found to achieve high diagnostic accuracy for OSCC. lncRNAs expressed in body fluids have higher diagnostic value than those in tissues, and can be effectively utilized as potential biomarkers for OSCC detection.