Incidence and outcomes of pediatric extremity melanoma: A propensity score matched SEER study

Abstract

BackgroundThere is a paucity of literature on treatment of melanoma in children with surgical management extrapolated from adult experience. The incidence and clinical outcomes of pediatric extremity melanoma were studied. MethodsSEER registry was analyzed between 1973 and 2010 for patients <20years old with extremity melanoma. Multivariate and propensity-score matched analyses were performed to identify independent predictors of survival. ResultsOverall, 917 patients were identified with an age-adjusted incidence of 0.2/100,000 persons, annual percent change 0.96. Most had localized disease (77%), histology revealing melanoma-not otherwise specified (52%). Surgical procedures performed included wide local excision (50%), excisional biopsy (32%), lymphadenectomy (LA) (28%), and sentinel lymph node biopsy (SLNB) (15%). Overall, 30-year disease specific mortality was 7% with lower survival for extremity melanoma (90%), males (89%), nodular histology (69%), and distant disease (36%) (all P<0.05). Post-treatment multivariate analysis revealed localized disease (HR 9.76; P=0.006) as an independent prognosticator of survival; earlier diagnostic years 1988–1999 (HR 2.606; P=0.017) were a negative prognosticator of survival. Propensity-score matched analysis found no difference in survival between SLNB/LA vs no sampling for regional/distant disease. ConclusionsPediatric extremity melanoma in SEER demonstrate no survival advantage between children undergoing sampling procedures vs no sampling for regional/distant disease. Type of studyRetrospective, prognostic study. Level of evidenceIII.