Salivary Genetic Biomarkers of Lung Cancer: A Systematic Review and Meta-Analysis of the Diagnostic Accuracy Publisher



Koopaie M ; Fatahzadeh M ; Kolahdooz S
Source: Cancer Cell International Published:2025

Abstract

Background: Considering the elevated mortality linked to lung cancer and the constraints of diagnostic and screening techniques, non-invasive biomarkers offer a promising alternative for early diagnosis. This systematic review and meta-analysis investigated the diagnostic accuracy of salivary genetic biomarkers for lung cancer. Methods: We searched for pertinent keywords in the Cochrane Library, Embase, LIVIVO, MEDLINE, Web of Science, Scopus, and Google Scholar for relevant studies. A systematic review was performed, and data were extracted from the eligible studies. The methodological quality of the studies included in this review was assessed using the Quality Assessment tool for Diagnostic Accuracy Studies-2 (QUADAS-2). The split component synthesis method was used to determine sensitivity, specificity, likelihood ratios, and diagnostic odds ratios. Studies were classified into three categories based on the type of biomarkers used: the single salivary biomarker model, the multiple salivary biomarker model (using multiple salivary biomarkers), and the mixed biomarker model (combining salivary biomarkers with other data, such as blood biomarkers or demographic data). By synthesizing data from six studies, including single, multiple, and mixed biomarker models, the study aimed to evaluate the diagnostic accuracy of these models. Results: The findings indicate that single-salivary biomarker models exhibit moderate diagnostic performance, with sensitivity and specificity of 0.72 and 0.73, respectively. However, multiple salivary biomarker models demonstrate improved diagnostic accuracy (sensitivity: 0.88, specificity: 0.75), supporting the use of multiple salivary biomarkers to increase accuracy and reduce false-positive results. Mixed models using a combination of salivary and blood biomarkers outperformed other models, achieving a diagnostic odds ratio of 115.66, with sensitivity at 0.92 and specificity at 0.91. Conclusions: Our findings show that the mixed biomarker model (model using salivary and blood genetic biomarkers) yields the highest diagnostic odds ratio and area under the curve, making it the most efficacious diagnostic approach among the models analyzed. © 2025 Elsevier B.V., All rights reserved.