A single-institution validation of the AJCC staging system for stage IV melanoma

Abstract

8517 Background: We sought to validate the AJCC staging system for stage IV melanoma in a contemporary, prospective, single- institution cohort and explore additional clinical factors that may influence prognosis. Methods: A prospective institutional database identified 1319 patients with stage IV melanoma. To minimize tertiary center referral bias, only patients seen prior to a stage IV diagnosis were included (n=687). Variables were dichotomized: lactate dehydrogenase (LDH) (=200, >200), number of metastases (1, >1), and number of involved organs (1, >1). Kaplan-Meier curves were generated and Cox regression was used to identify factors independently predictive of survival. Results: Demographics are provided in the table . The median age at diagnosis of stage IV was 55 years (range 16–94) and the median disease free interval (DFI) was 12 months (0–181). The overall median survival was 10 months (5–21) with a median follow-up for survivors of 31 months (9–68); 569 deaths were observed. Cox regression analysis ( table ) identified younger age at stage IV diagnosis, a longer DFI, and a normal LDH to be predictive of improved survival. Patients with either distant skin/subcutaneous/nodal or pulmonary disease experienced prolonged survival when compared to patients with metastases to other visceral sites. Survival was improved in patients with a single metastatic site at diagnosis of stage IV. Gender, antecedent stage, and number of involved organs were not associated with outcome. Conclusions: In this single institution cohort of patients with stage IV melanoma, poorer survival in patients with non-pulmonary visceral metastases and/or abnormal LDH levels as described by the AJCC staging system was confirmed. Additionally, the number of metastases at the time of diagnosis of stage IV was the most powerful predictor of poorer survival and may be a variable to consider in future staging systems. No significant financial relationships to disclose. [Table: see text]