The accuracy of detecting melanoma on frozen section melanoma antigen recognized by T cells 1 (MART-1) stains and on permanent sections of previously frozen tissue: A prospective cohort study

To the Editor: Mohs micrographic surgery (MMS) for melanoma is controversial because of concerns about accuracy detecting on frozen sections. This study's primary analysis examined sensitivity and specificity using hematoxylin eosin (H&E) melanoma-associated antigen recognized by T cells (MART-1) sections compared with permanent The secondary evaluated whether section specimens can be subsequently processed without diminishing diagnostic accuracy. Institutional Review Board-approved prospective study enrolled consecutive consenting adults (aged >18 years) biopsy-proven in situ or invasive treated MMS MART-1 immunohistochemical stains (MMS-IHC) from 2016 to 2018. All melanomas were excised a previously published MMS-IHC protocol.1Etzkorn J.R. Sobanko J.F. Elenitsas R. et al.Low recurrence rates 1 immunostaining: tissue processing methodology optimize pathologic staging margin assessment.J Am Acad Dermatol. 2015; 72: 840-850Abstract Full Text PDF PubMed Scopus (86) Google Scholar Two opposing vertical isolated debulking excision (Fig 1) . One specimen (frozen-to-permanent) was sections, then thawed (direct-to-permanent) immediately fixed formalin stained both H&E single surgeon. H&E, but ordered at discretion dermatopathologist. Permanent 5 dermatopathologists, who blinded interpretation surgeon not had been frozen. Pathologic interpretations categorized into 4 groups: (1) malignant; (2) benign; (3) qualified (4) benign (see Supplemental Table I definitions, available via Mendeley https://data.mendeley.com/datasets/hrjjcr43nx/1). A positive result (melanoma detected) defined as diagnosis “malignant” “qualified malignant.” negative “benign” benign.” Tissue samples 169 patients, comprising 119 (70.4%) 50 (29.6%) (Table I). showed that detected (test specimen) 95.3% (95% confidence interval [CI], 89.6%-98.1%) 95.1% CI, 82.2%-99.2%) dermatopathologist's same (frozen-to-permanent; gold standard) 1).Table ICharacteristics cohortVariable∗Data are presented number patients (%) mean (range).Number Patients (N = 169)Sex Male104 (62) Female65 (38)Age, y67.7 (37-101)Preoperative Melanoma situ119 (70)Lentigo maligna type65Superficial spreading1Acral lentiginous1Not specified52 Invasive melanoma50 (30)Lentigo maligna30Superficial spreading14Nodular3Not specified3Depth invasion, mm <141 1-27 >2-41 Not specified1Anatomic location Head neck147 (87) Hands feet5 Pretibial leg4 (2.3) Trunk extremity (nonacral, non-pretibial)13 (7.7)∗ Data (range). Open table new tab dermatopathologists frozen-to-permanent 98.3% 93.6%-99.7%) 89.6% 76.6%-96.1%), their own direct-to-permanent (gold 1). Of 15 total discordant diagnoses analyses, 12 involved diagnosis, indicating challenge diagnosing these (Supplemental Fig shows surgeons detect high dermatopathologists' It also sent directly Therefore, thaw challenging second opinions diagnostically accurate Diagnostic discordances expected melanocytic tumors.2Piepkorn M.W. Longton G.M. Reisch L.M. al.Assessment second-opinion strategies cutaneous lesions.JAMA Netw Open. 2019; 2: e1912597Crossref (15) Scholar, 3Santillan A.A. Messina J.L. Marzban S.S. Crespo G. Sondak V.K. Zager J.S. Pathology review thin multidisciplinary clinic: impact treatment decisions.J Clin Oncol. 2010; 28: 481-486Crossref (57) 4Gonzalez M.L. Young E.D. Bush J. al.Histopathologic features difficult-to-diagnose lesions: case-control study.J 2017; 77: 543-548.e1Abstract (6) Although this limited its single-center design, it demonstrates reliable technique melanoma. supports use precise microscopic margin-directed None disclosed.