Vidal, Nahid Y.

Background: Mohs micrographic surgery (MMS) is a promising treatment modality for melanoma in situ (MIS). However, variations in surgical technique limit the generalizability of existing data and may impede future study of MMS in clinical trials. Methods: A modified Delphi method was selected to establish consensus on optimal MMS techniques for treating MIS in future clinical trials. The Delphi method was selected due to the limited current data, the wide range of techniques used in the field, and the intention to establish a standardized technique for future clinical trials. A literature review and interviews with experienced MMS surgeons were performed to identify dimensions of the MMS technique for MIS that (1) likely impacted costs or outcomes of the procedure, and (2) showed significant variability between surgeons. A total of 8 dimensions of technical variation were selected. The Delphi process consisted of 2 rounds of voting and commentary, during which 44 expert Mohs surgeons across the United States rated their agreement with specific recommendations using a Likert scale. Results: Five of eight recommendations achieved consensus in Round 1. All 3 of the remaining recommendations achieved consensus in Round 2. Techniques achieving consensus in Round 1 included the use of a starting peripheral margin of ≤5 mm, application of immunohistochemistry, frozen tissue processing, and resecting to the depth of subcutaneous fat. Consensus on the use of Wood’s lamp, dermatoscope, and negative tissue controls was established in Round 2. Conclusions: This study generated 8 consensus recommendations intended to offer guidance for Mohs surgeons treating MIS. The adoption of these recommendations will promote standardization to facilitate comparisons of aggregate data in multicenter clinical trials.

BACKGROUND While increasing evidence supports the safety and effectiveness of immunohistochemistry-assisted Mohs micrographic surgery (MMS) for superficially invasive melanoma, there is a paucity of the literature investigating its effect on final defect size. OBJECTIVE To evaluate the tissue sparing effect of MMS for melanoma. MATERIALS AND METHODS Three hundred and twenty-eight patients with early-stage (T1a/T1b) cutaneous melanomas treated with MMS from January 2008 to December 2018 were evaluated. Measured defect sizes after Mohs tumor extirpation were compared with anticipated defect size that would result from standard-margin wide local excision (WLE). Average actual versus anticipated defect areas were compared using a paired t-test (95% confidence intervals). RESULTS The following groups demonstrated a significantly smaller defect area for MMS-treated tumors when compared with anticipated standard-margin WLE defect: All tumors combined (13.8 cm2 vs 10.4 cm2, p < .001), tumors requiring 1 stage (13.6 cm2 vs 10.1 cm2, p < .001), and tumors requiring 2 stages (13.2 cm2 vs 10.5 cm2, p = .004). The majority of patients (83.5%, n = 274) achieved clear margins with 1 stage. CONCLUSION Immunohistochemistry-assisted MMS for early-stage invasive melanoma is associated with smaller final defect size and overall tissue sparing effect compared with standard WLE margins.

AbstractFrozen sections are a useful pathologic tool, but variable image quality may impede the use of artificial intelligence and machine learning in their interpretation. We aimed to identify the current research on machine learning models trained or tested on frozen section images. We searched PubMed and Web of Science for articles presenting new machine learning models published in any year. Eighteen papers met all inclusion criteria. All papers presented at least one novel model trained or tested on frozen section images. Overall, convolutional neural networks tended to have the best performance. When physicians were able to view the output of the model, they tended to perform better than either the model or physicians alone at the tested task. Models trained on frozen sections performed well when tested on other slide preparations, but models trained on only formalin‐fixed tissue performed significantly worse across other modalities. This suggests not only that machine learning can be applied to frozen section image processing, but also use of frozen section images may increase model generalizability. Additionally, expert physicians working in concert with artificial intelligence may be the future of frozen section histopathology.

Abstract Evaluation of basal cell carcinoma (BCC) involves tangential biopsies of a suspicious lesion that is sent for frozen sections and evaluated by a Mohs micrographic surgeon. Advances in artificial intelligence (AI) have made possible the development of sophisticated clinical decision support systems to provide real-time feedback to clinicians that could have a role in optimizing the diagnostic workup of BCC. There were 287 annotated whole-slide images of frozen sections from tangential biopsies, of which 121 contained BCC, that were used to train and test an AI pipeline to recognize BCC. Regions of interest were annotated by a senior dermatology resident, an experienced dermatopathologist and an experienced Mohs surgeon, with concordance of annotations noted on final review. Final performance metrics included a sensitivity and specificity of 0.73 and 0.88, respectively. Our results on a relatively small dataset suggest the feasibility of developing an AI system to aid in the workup and management of BCC.