Skin Cancer 2019: Detecting Non-Melanoma Skin Cancer Through Multimodal Optical Imaging
Posted: Monday, April 15, 2019
According to research presented at the 2019 World Congress on Advanced Treatments and Technologies in Skin Cancer (Skin Cancer 2019, poster 4), multimodal optical imaging may be efficient in detecting and diagnosing non-melanoma skin cancers. The researchers sought to determine whether technology such as reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) could be combined within the same tool to provide high-resolution three-dimensional imaging of potentially cancerous lesions.
“RCM-OCT imaging may be prospectively used to comprehensively diagnose suspicious [basal cell carcinoma] lesions, determine subtype, and triage for treatment,” noted Chih-Shan J. Chen, MD, PhD, of Memorial Sloan Kettering Cancer Center, New York, and colleagues.
A total of 85 patients with either suspicious (n = 60) or biopsy-confirmed (n = 25) basal cell carcinomas were included in the study. The investigators correlated primary features in RCM and OCT images with histopathology, including tumor nests of varying sizes, dilated blood vessels, dark silhouettes, tumor extensions from the basal cell layer, in dermis, horn cyst, and bright peritumoral stroma. In addition, the diagnostic accuracy was calculated, and OCT-predicted depth was correlated with histopathologically measured depth. Some features, including necrosis and mucin pools within tumors, were correlated on histology.
The RCM-OCT tool produced images for all lesions with higher sensitivity and negative predictive value (100%), as well as comparable specificity (48% vs. 56% on RCM) and positive predictive value (82.19% vs. 84.59% on RCM). In suspicious lesions alone, relatively higher specificity (94.1%) and positive predictive value (75%) were observed. In addition, high correlation (R = .86) between the depth as predicted by OCT and as measured histopathologically was observed.
