# Optical Coherence Tomography for Invasive Oral Squamous Cell Carcinoma: Diagnostic Accuracy and Grade- and Subsite-Associated Imaging Features

## TL;DR

Optical coherence tomography (OCT) is highly accurate for diagnosing invasive oral cancer, with features varying by tumor grade and location.

## Contribution

The study demonstrates OCT's diagnostic accuracy and identifies subsite-specific imaging patterns in oral squamous cell carcinoma.

## Key findings

- OCT achieved 98.5% sensitivity and 96.7% specificity for detecting invasive OSCC.
- OCT biomarkers varied systematically with tumor differentiation grade.
- Performance was highest in the tongue and floor of the mouth but lower in posterior/keratinized subsites.

## Abstract

Background: Early and accurate diagnosis remains crucial to improving outcomes in oral cancer. Optical coherence tomography (OCT) offers real-time, high-resolution imaging that may support diagnosis and treatment planning in oral squamous cell carcinoma (OSCC). Methods: In this prospective study, preoperative OCT scans were obtained from 68 histologically confirmed OSCC lesions, with 30 paired adjacent mucosa samples from the same patients as histologically negative comparators (diagnostic dataset: 98 lesions). OCT findings were compared with histopathology for diagnostic performance, OCT biomarker patterns by tumour grade, tumour depth measurement, margin assessment, and subsite-specific performance. Results: OCT demonstrated 98.5% sensitivity, 96.7% specificity, and an AUC of 0.98 for detection of invasive OSCC. OCT biomarkers—including abnormal epithelial architecture with variable epithelial thickness, stratification loss, basement membrane disruption, and increased subepithelial reflectivity—varied systematically with tumour differentiation grade. Tumour depth measurements showed acceptable agreement with histology, while margin definition was correct in 80% of cases. Performance was highest in the tongue and the floor of the mouth, with reduced performance in posterior/keratinised subsites. Image artefacts occurred in 5.1% of scans. Conclusions: OCT provides a reproducible, real-time adjunct for diagnosis, margin planning, and lesion stratification in OSCC, with recognised limitations related to light attenuation and operator-dependent factors. Multicentre validation and integration with digital interpretation platforms are warranted.