Usefulness of optical coherence tomography as an adjunct diagnostic tool in the detection of oral squamous cell carcinoma: An in vitro study / Ali Yassen Obade

Objective: To correlate OCT imaging with histopathological diagnosis of biopsy specimen; To identify the structural change in oral cancers biopsy specimens by OCT; To evaluate OCT capability of defining the indication for surgical biopsy and measure the accuracy of the OCT findings with the final diagnosis. Material and method: This prospective study included 52 oral lesions from 44 patients (male = 21, female = 23). Surgical biopsies (excisional = 23, incisional = 29) were obtained under local anesthesia. Two assessors (surgeon and pathologist) examined OCT images of 52 samples in two separate sessions. In first session, they were asked to evaluate the mucosal architectural status including keratin layer (KL), epithelial layer (EP), lamina propria (LP), basement membrane (BM) and the reflection’s degree of epithelial layer (EP Re.) and report biopsy need. In the second session, the same two assessors reassessed the OCT images in the same manner of the first session. Firstly, they were asked to give a differential diagnosis for that lesion based only on the brief clinical history and secondly, they were required to score on the five OCT variables and based on that they were asked to report the OCT agreement (agree, disagree) with their clinical differential diagnosis that they had made earlier. The inter-observer agreement (first and second session) and intra-observer agreement were calculated using Kappa scores. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the accuracy of COE, OCT and combined (COE + OCT) were also calculated. Results: The histopathological results showed there were SCC (14), verrucous carcinoma (1), epithelial dysplasia (7), epithelial hyperkeratosis (1), epithelial hyperplasia (1), and OLP (5). Other lesions included were mucoceles and pyogenic iii granuloma (4 each), traumatic eosinophilic ulcer, fibroepithelial polyps and traumatic neuroma (3 each), irritation fibroma, squamous cell papilloma, oral melanotic macule, lymphedema, hemangiolymphangioma and inflamed mucosa (1 each). Sn, Sp, NPV, PPV and Ac for 1st assessor were 86%, 77%, 88%, 73% and 81% (COE), 95%, 80%, 96%, 78% and 87% (OCT), 95%, 77%, 96%, 75% and 85% (combined), while for the 2nd assessor 91%, 70%, 91%, 69% and 79% (COE), 95%, 71%, 96%, 69% and 81% (OCT), 96%, 72%, 95%, 73% and 83% (combined). Kappa score of inter-observer agreement was 0.92 for biopsy need. Conclusion: Optical coherence tomography (OCT) is a non-invasive technique with a high-resolution capability that is considered a useful tool for obtaining cross-sectional real-time images for different parts of the human body. Basement membrane is considered a key parameter in the detection of oral cancer and for differentiating it from other oral pathological conditions. The EP thickness and its degree of reflection are valuable parameters for recognizing healthy tissues from most pathological conditions, but, they cannot be considered as accurate diagnostic criteria to discriminate between the dysplastic epithelium and benign oral tissues. Overall OCT is a promising optical technique that might be able to define the grades of oral dysplasia in the near future with the continuous increase in its resolution capability