Salivary inflammatory and matrix remodeling biomarkers in oral submucous fibrosis: A systematic review of diagnostic accuracy

Abstract

To systematically synthesise evidence on salivary inflammatory, fibrotic, matrix remodelling, molecular signalling, and transcriptomic biomarkers in oral submucous fibrosis (OSMF), with emphasis on expression patterns and reported diagnostic utility. This systematic review was conducted according to PRISMA guidelines, with relevant Preferred reporting systems for systematic reviews and meta analysis- diagnostic test accurracy (PRISMA-DTA) items considered for diagnostic outcomes. PubMed/MEDLINE, Scopus, Cochrane Library, and Google Scholar were searched up to 31 May 2026. Human observational studies evaluating salivary biomarkers in OSMF were included if OSMF-specific data were extractable. Studies assessing only non-salivary biomarkers, broad salivary chemistry, oxidative stress markers, trace elements, or mixed oral potentially malignant disorders (OPMDs) without separate OSMF data were excluded. Risk of bias was assessed using domain-based appraisal, and findings were synthesised narratively because meta-analysis was not appropriate. Ten independent studies were included. The biomarkers assessed included MMP-12, MMP-8, IL-1 β , IL-8, IL-10, IL-17, TNF- α , TGF- β isoforms, epidermal growth factor receptor (EGFR), S100A7, S100P, OAZ1, SAT, and miRNA-21. MMP-12 showed the strongest individual diagnostic signal, with increased salivary levels in OSMF and reported receiver operating characteristic-based diagnostic utility. MMP-8 showed an inverse expression pattern. TGF- β isoforms and IL-17 demonstrated disease-associated changes, whereas IL-10, salivary IL-1 β , and TNF- α showed limited or inconsistent discriminatory value. EGFR, S100A7, miRNA-21, and transcriptomic panels showed promising but preliminary findings. Salivary biomarkers show potential as adjunctive tools for OSMF assessment, particularly MMP-12, TGF- β isoforms, and multi-marker transcriptomic panels. However, current evidence remains low to very low in certainty and requires external validation before clinical use.