ABSTRACT
Optical coherence tomography (OCT), a non-invasive diagnostic modality, may replace biopsy for diagnosing basal cell carcinoma (BCC) if a high-confidence BCC diagnosis can be established. In other cases, biopsy remains necessary to establish a histopathological diagnosis and treatment regimen. It is, therefore, essential that OCT assessors have a high specificity for differentiating BCC from non-BCC lesions. To establish high-confidence BCC diagnoses, specific morphological BCC characteristics on OCT are used. This study aimed to review several cases of non-BCC lesions that were misclassified as BCC by experienced OCT assessors, thereby providing insight into the causes of these misclassifications and how they may be prevented. The study population consisted of patients who had a histopathologically-verified non-BCC lesion. Patients from Maastricht University Medical Center+ from February 2021 to April 2021 were included in the study. Two independent OCT assessors assessed OCT scans. One OCT assessor recorded the presence or absence of validated morphological BCC characteristics. A false-positive OCT test result was defined as certainty of BCC presence in a non-BCC lesion. The frequency of misclassifications and the presence or absence of morphological BCC features are discussed. A total of 124 patients with non-BCC lesions were included. Six patients were misclassified by both OCT assessors and are discussed in more detail. Histopathological diagnoses were squamous cell carcinoma (n = 2/21), actinic keratosis (n = 2/29), squamous cell carcinoma in situ/Bowen's disease (n = 1/16), or interphase dermatitis (n = 1/4). In all misclassified cases, multiple, apparent morphological BCC characteristics on OCT were present. Most non-BCC lesions are recognized as such by OCT assessors. However, there remains a small risk that a high-confidence BCC diagnosis is established in non-BCC lesions wherein features mimicking validated BCC characteristics are present. Misclassification may be prevented by careful delineation of epidermal layers and good differentiation between dermal ovoid structures typical of BCC versus squamous cell carcinoma.
Subject(s)
Carcinoma, Basal Cell , Carcinoma, Squamous Cell , Skin Neoplasms , Humans , Skin Neoplasms/diagnostic imaging , Skin Neoplasms/pathology , Tomography, Optical Coherence/methods , Sensitivity and Specificity , Carcinoma, Basal Cell/diagnostic imaging , Carcinoma, Basal Cell/pathology , Carcinoma, Squamous Cell/diagnostic imagingABSTRACT
BACKGROUND: Recurrent/residual basal cell carcinoma (BCC) after topical treatment may not be visible during clinical and dermatoscopic examination (CDE). Optical coherence tomography (OCT) may detect these subclinical recurrences or residues. OBJECTIVE: To compare the diagnostic accuracy of CDE with that of CDE combined with OCT (CDE-OCT) for detecting recurrent/residual BCC after topical treatment of superficial BCC. METHODS: In this diagnostic cohort study, the suspicion level for recurrence or residue was recorded on a 5-point confidence scale. All patients with high suspicion of recurrence or residue based on CDE and/or CDE-OCT were referred for punch biopsy. Patients with a low suspicion on CDE and CDE-OCT were asked to (voluntarily) undergo a control biopsy. Histopathologic results of the biopsy were used for verification of CDE and CDE-OCT diagnoses (gold standard). RESULTS: This study included 100 patients. A histopathologic recurrent/residual BCC was found in 20 patients. For recurrence or residue detection, sensitivity was 100% (20 of 20) for CDE-OCT and 60% (12 of 20) for CDE (P = .005) and specificity was 95% for CDE-OCT and 96.3% for CDE (P = .317). The area under the curve for CDE-OCT (0.98) was significantly higher than that for CDE (0.77) (P = .001). LIMITATIONS: Results are based on 2 OCT assessors. CONCLUSION: Compared with CDE alone, CDE-OCT results in a significantly higher ability to detect recurrent/residual BCCs after topical treatment.
