Autofluorescence-based devices for oral cancer screening - A scoping review

Background Oral cancer (OSCC) and oral potentially malignant disorders (OPMDs) present persistent diagnostic challenges, with late detection driving poor prognosis and high treatment morbidity. Chairside autofluorescence-based devices such as GOCCLES and OralID have been developed to complement conventional oral examination (COE), but their diagnostic performance and clinical utility requires further exploration. This scoping review maps the available evidence on GOCCLES and OralID, characterising study populations, clinical contexts, and reported diagnostic outcomes. Methods This review followed the PRISMA-ScR guidelines and was developed using the Population-Concept-Context (PCC) framework. A systematic search of PubMed, Scopus, Web of Science, Embase, and Google Scholar (1 st January 1996 – June 2026) identified primary empirical studies evaluating GOCCLES or OralID in human subjects. Two independent reviewers performed title/abstract screening, full-text eligibility assessment, and data charting. Formal critical appraisal was not undertaken per JBI scoping review methodology. Synthesis was descriptive and narrative. Results Of 122 identified records, 13 studies met inclusion criteria (6 evaluating GOCCLES, 7 evaluating OralID), encompassing 577 participants across seven countries. Study designs included observational studies, clinical trials, case series, and case reports. Populations comprised mixed OPMD/OSCC cohorts (n=5), OPMD-only (n=2), OSCC-only (n=2), and other groups (supportive periodontal care, HIV-positive, autoimmune disease). GOCCLES showed low sensitivity (33–66%) and low to moderate specificity (40–90.6%) against histopathology, with better specificity (90.6%) in primary care screening. OralID demonstrated high sensitivity (74–100%) but low specificity (18–89%), depending on context. The strongest performance occurred in perioperative margin delineation (surgical margin accuracy improved from 52% to 74%; tumor-free margin rate 97% vs. 73% for conventional surgery). High false-positive rates and the ‘umbrella effect’ of hyperkeratosis limiting fluorescence penetration were consistently reported. Conclusion Evidence for GOCCLES and OralID is heterogeneous and context-dependent. Neither device consistently achieves high sensitivity and specificity concurrently. Their principal value appears adjunctive – guiding biopsy site selection or intraoperative margins – rather than as standalone screening tools. Future research requires standardised protocols, histopathological reference standards, and clearer definitions of appropriate clinical contexts, particularly in non-specialist primary care settings. Clinical Significance GOCCLES and OralID should be used as adjuncts to conventional oral examination rather than as standalone diagnostic or screening tools. Their clearest potential benefit is in focused clinical situations, particularly identifying the most representative biopsy site, defining the extent of heterogeneous or poorly demarcated lesions, and assisting intraoperative margin delineation in established oral cancer. They may also help highlight clinically occult mucosal abnormalities, support documentation, and prompt referral when findings are interpreted alongside lesion history, palpation, risk factors, and specialist judgement. Current evidence does not support routine population screening, use in asymptomatic low-risk individuals, exclusion of malignancy after a negative result, or replacement of biopsy and histopathology. False-positive findings may occur with inflammation, trauma, infection, scarring, and immune-mediated disease, while hyperkeratotic lesions may produce false-negative results because of reduced light penetration. Evidence is also insufficient to show that these devices improve referral accuracy, reduce diagnostic delay, predict malignant transformation, improve survival, or provide cost-effective routine surveillance in primary care.