Addison M Demer1,2, Jamie L Hanson1, Ian A Maher1, Walter Liszewski3. 1. Department of Dermatology, University of Minnesota, Minneapolis. 2. Now with Department of Dermatology, Mayo Clinic, Rochester, Minnesota. 3. Department of Dermatology, Northwestern University, Chicago, Illinois.
Abstract
Importance: Although previous database studies suggest that Mohs micrographic surgery (MMS) treatment is associated with improved overall survival (OS) for head and neck melanomas, outcomes for trunk and extremity (T&E) tumors have not been adequately evaluated. Objective: To assess survival outcomes for patients with melanomas of the T&E treated with MMS vs wide local excision (WLE). Design, Setting, and Participants: This retrospective cohort study examined deidentified data from the National Cancer Database between 2004 and 2015. Inclusion criteria for the analysis included diagnosis of trunk, upper extremity, or lower extremity melanoma; known Breslow depth; removal by MMS or WLE; and known last date of survival status. Main Outcomes and Measures: Five-year all-cause mortality (ACM) rates. Results: A total of 188 862 in situ and invasive melanomas were included in the analysis (MMS, 2.3%; WLE, 97.7%); the mean (SD) age of patients included was 58.8 (16.0) years, and 52.7% were male. Multivariate analysis demonstrated no OS difference among trunk (WLE hazard ratio [HR], 1.097; 95% CI, 0.950-1.267; P = .21), upper extremity (WLE HR, 1.013; 95% CI, 0.872-1.176; P = .87), lower extremity (WLE HR, 0.934; 95% CI, 0.770-1.134; P = .49), or combined T&E (WLE HR, 1.031; 95% CI, 0.941-1.130; P = .51) tumors. Factors associated with increased risk of ACM on multivariate analysis of all tumors included increasing age (HR, 1.043; 95% CI, 1.042-1.044; P < .001), no insurance or nonprivate insurance (none: HR, 1.921 [95% CI, 1.782-2.071]; Medicaid: HR, 2.410 [95% CI, 2.242-2.591]; Medicare: HR, 1.237 [95% CI, 1.194-1.281]; other government insurance: HR, 1.279 [95% CI, 1.117-1.465]; P < .001 for all), positive surgical margins (HR, 1.609; 95% CI, 1.512-1.712; P < .001), a Charlson-Deyo comorbidity score greater than 0 (Charlson-Deyo score of 1: HR, 1.340; 95% CI, 1.295-1.385; P < .001; Charlson-Deyo score of ≥2: HR, 2.044; 95% CI, 1.934-2.159; P < .001), tumor ulceration (HR, 2.175; 95% CI, 2.114-2.238; P < .001), and increasing Breslow depth (HR, 1.002 [per 0.1 mm]; P < .001). Female sex (HR, 0.698; 95% CI, 0.680-0.716; P < .001) and nonnodular subtype (lentigo maligna/lentigo maligna melanoma: HR, 0.743; 95% CI, 0.686-0.805; P < .001; superficial spreading: HR, 0.739; 95% CI, 0.710-0.769; P < .001; other subtype: HR, 0.817; 95% CI, 0.790-0.845; P < .001; nodular: HR, 1 [reference]) were associated with improved OS. Conclusions and Relevance: This cohort study of patients surgically treated for melanomas of the trunk and/or extremities found that, compared with WLE, MMS was not associated with significantly different OS for T&E melanomas.
Importance: Although previous database studies suggest that Mohs micrographic surgery (MMS) treatment is associated with improved overall survival (OS) for head and neck melanomas, outcomes for trunk and extremity (T&E) tumors have not been adequately evaluated. Objective: To assess survival outcomes for patients with melanomas of the T&E treated with MMS vs wide local excision (WLE). Design, Setting, and Participants: This retrospective cohort study examined deidentified data from the National Cancer Database between 2004 and 2015. Inclusion criteria for the analysis included diagnosis of trunk, upper extremity, or lower extremity melanoma; known Breslow depth; removal by MMS or WLE; and known last date of survival status. Main Outcomes and Measures: Five-year all-cause mortality (ACM) rates. Results: A total of 188 862 in situ and invasive melanomas were included in the analysis (MMS, 2.3%; WLE, 97.7%); the mean (SD) age of patients included was 58.8 (16.0) years, and 52.7% were male. Multivariate analysis demonstrated no OS difference among trunk (WLE hazard ratio [HR], 1.097; 95% CI, 0.950-1.267; P = .21), upper extremity (WLE HR, 1.013; 95% CI, 0.872-1.176; P = .87), lower extremity (WLE HR, 0.934; 95% CI, 0.770-1.134; P = .49), or combined T&E (WLE HR, 1.031; 95% CI, 0.941-1.130; P = .51) tumors. Factors associated with increased risk of ACM on multivariate analysis of all tumors included increasing age (HR, 1.043; 95% CI, 1.042-1.044; P < .001), no insurance or nonprivate insurance (none: HR, 1.921 [95% CI, 1.782-2.071]; Medicaid: HR, 2.410 [95% CI, 2.242-2.591]; Medicare: HR, 1.237 [95% CI, 1.194-1.281]; other government insurance: HR, 1.279 [95% CI, 1.117-1.465]; P < .001 for all), positive surgical margins (HR, 1.609; 95% CI, 1.512-1.712; P < .001), a Charlson-Deyo comorbidity score greater than 0 (Charlson-Deyo score of 1: HR, 1.340; 95% CI, 1.295-1.385; P < .001; Charlson-Deyo score of ≥2: HR, 2.044; 95% CI, 1.934-2.159; P < .001), tumor ulceration (HR, 2.175; 95% CI, 2.114-2.238; P < .001), and increasing Breslow depth (HR, 1.002 [per 0.1 mm]; P < .001). Female sex (HR, 0.698; 95% CI, 0.680-0.716; P < .001) and nonnodular subtype (lentigo maligna/lentigo maligna melanoma: HR, 0.743; 95% CI, 0.686-0.805; P < .001; superficial spreading: HR, 0.739; 95% CI, 0.710-0.769; P < .001; other subtype: HR, 0.817; 95% CI, 0.790-0.845; P < .001; nodular: HR, 1 [reference]) were associated with improved OS. Conclusions and Relevance: This cohort study of patients surgically treated for melanomas of the trunk and/or extremities found that, compared with WLE, MMS was not associated with significantly different OS for T&E melanomas.