Prognostic factors in patients with melanoma metastatic to axillary or inguinal lymph nodes. A multivariate analysis.
Academic Article
Overview
abstract
Although pathologic nodal status is a major determinant of outcome in melanoma, there is substantial prognostic heterogeneity among node-positive patients. This study was undertaken to further clarify significant variables predicting survival in patients with melanoma metastatic to axillary or groin nodes. From 1019 patients with melanoma undergoing axillary or groin dissection between 1974 and 1984, the authors identified 449 patients with histologically positive nodes. Both univariate and multivariate analyses were performed using the Kaplan-Meier product limit method and the Cox model of proportional hazard regression. The major determinant of survival was pathologic stage (PS) according to the 1983 AJCC staging system. Three hundred fifty patients (78%) were classified PS-III (one nodal group involved), with a survival of 39% at 5 years and 32% at 10 years. Factors independently predictive of a favorable outcome in these patients were nontruncal primary site (p = 0.0002), microscopic nodal involvement (p = 0.001), number of positive nodes less than three (p = 0.003), and absence of extranodal disease (p = 0.01). Ninety-nine patients (22%) were classified PS-IV, 51 with two nodal stations involved (N2), 25 with intransit disease and one nodal station involved (N2), 7 with extraregional soft tissue metastases (M1), and 16 with visceral metastases (M2). Survival for PS-IV patients was 9% at 5 and 10 years, respectively. Within PS-IV, factors independently predictive of a more favorable outcome were the absence of extranodal disease (p = 0.0001), female sex (p = 0.03), and a long interval from diagnosis to lymph node dissection (p = 0.04). These factors were incorporated into a model predicting relative risk of death from disease for both PS-III and PS-IV patients, separating patients into groups at high, intermediate, and low risk of recurrence after lymphadenectomy.
