OCT in the diagnosis of head and neck pre-cancerous and cancerous cutaneous lesions: An immediate ex vivo study

Background Optical coherence tomography (OCT) has been shown to reliably identify skin changes with the combined advantage of resolution and penetration depth. The objective of this study was to assess the accuracy of OCT in diagnosing various skin pathologies, using pre-set diagnostic criteria, by two reviewers blinded to the actual diagnosis. Patients and methods This immediate ex-vivo study included skin specimens from 103 patients with suspicious skin lesions. In total, 110 lesions were scanned using OCT immediately following surgical resection. Two experienced OCT reviewers assessed these OCT diagnostic parameters blindly at two different intervals based on the parameters of the pre-set criteria. Furthermore, based on the diagnostic suggestions of the two reviewers, the sensitivity and specificity were calculated along with the kappa agreement statistic. Results A variety of lesions were diagnosed, as part of this study, including actinic keratosis (AK), basal cell carcinoma (BCC), cutaneous squamous cell carcinoma (SCC), lentigo maligna (LM) and malignant melanoma (MM). Using the previously set diagnostic criteria, a sensitivity and specificity for AK was reported as 97.5% and 100%, respectively. For BCC it was 98% and 95.3%, respectively; while for cutaneous SCC was 83.3% and 94.9%, respectively. The sensitivity and specificity for LM was reported as 55% and 89.5%, respectively; while for MM it was 43.8% and 76.2%, respectively. High overall levels of intra- and inter-observer agreements were shown for most of the evaluated diagnostic parameters. Conclusion This study indicates that OCT shows good promise as an optical diagnostic technique in diagnosing skin pathologies. The sensitivity in diagnosing actinic keratosis and basal cell carcinomas was impressive, with satisfactory results in diagnosing cutaneous squamous cell carcinoma. The sensitivity in diagnosing lentigo maligna and malignant melanoma remains poor and require further studies. Experienced OCT reviewers are required to improve diagnostic accuracy.