Ulceration as a predictive marker for response to adjuvant interferon therapy in melanoma

• Published in: Annals of surgery Vol. 252; no. 3; pp. 460 - 466
• Main Authors: McMasters, Kelly M, Edwards, Michael J, Ross, Merrick I, Reintgen, Douglas S, Martin, 2nd, Robert C G, Urist, Marshall M, Noyes, R Dirk, Sussman, Jeffrey J, Stromberg, Arnold J, Scoggins, Charles R
• Format: Journal Article
• Language: English
• Published: United States 01-09-2010
• Subjects: Adolescent; Adult; Aged; Antineoplastic Agents - therapeutic use; Biomarkers, Tumor - analysis; Chemotherapy, Adjuvant; Female; Humans; Immunohistochemistry; Interferons - adverse effects; Interferons - therapeutic use; Lymph Node Excision; Lymphatic Metastasis; Male; Melanoma - drug therapy; Melanoma - pathology; Melanoma - surgery; Middle Aged; North America; Prognosis; Proportional Hazards Models; Prospective Studies; Reverse Transcriptase Polymerase Chain Reaction; Sentinel Lymph Node Biopsy; Skin Neoplasms - drug therapy; Skin Neoplasms - pathology; Skin Neoplasms - surgery; Skin Ulcer - chemically induced; Survival Rate; North America
• ISSN: 0003-4932; 1528-1140
• DOI: 10.1097/sla.0b013e3181f20bb1

This analysis was performed to investigate the hypothesis that ulceration predicts improved response to adjuvant interferon (IFN) therapy.
Several studies have demonstrated that adjuvant therapy for high-risk melanoma patients with IFN alfa-2b improves disease-free survival (DFS), although the impact on overall survival (OS) is controversial. Recent data have suggested that IFN therapy may preferentially benefit patients with ulcerated primary melanomas.
Post hoc analysis was performed by a prospective multi-institutional randomized study of observation versus adjuvant IFN therapy for melanoma. All patients underwent sentinel lymph node biopsy; completion lymphadenectomy was performed for patients with sentinel lymph node metastasis. Patients were stratified by Breslow thickness, ulceration, and nodal status. Kaplan-Meier analysis of DFS and OS was performed and included univariate and multivariate analyses.
A total of 1769 patients were analyzed (1311 without ulceration, 458 with ulceration) with a median follow-up of 71 months. Ulceration was associated with significantly worse DFS and OS in both node-negative and node-positive patients. Kaplan-Meier analysis of node-negative and node-positive patients by ulceration status revealed that the only significant impact of interferon was improved DFS in the ulcerated node-positive patients (P = 0.0169). IFN therapy had no significant impact on OS regardless of ulceration status, however. On multivariate analysis, IFN treatment was a significant independent predictor of DFS among ulcerated patients (odds ratio, 0.51; 95% confidence interval, 0.30-0.83; P = 0.0053), but not among patients without ulceration.
These data support the conclusion that ulceration is a predictive marker for response to adjuvant IFN therapy. Future studies to evaluate specifically the differential effect of IFN on patients with ulcerated melanomas may allow us to focus this therapy on patients most likely to benefit from it.