Prospective surgical trials provide the cornerstones of current practice guidelines for cutaneous melanoma (JW Dermatol Oct 28 2011). In two prior assessments of optimal margins of excision for intermediate-thickness cutaneous melanomas (Ann Surg Oncol 2001; 8:101, and N Engl J Med 2004; 350:757), results were mixed: In the first study, survival rates did not differ significantly with 2-cm vs. 4-cm margins; and in the second, a 1-cm margin was associated with a significantly greater risk for regional recurrence than a 3-cm margin, but with similar overall survival. Thus, the optimal excision margin for cutaneous melanomas thicker than 2 mm remained unresolved.

The Swedish Melanoma Study Group, in cooperation with the Danish Melanoma Group, assigned 936 patients with cutaneous melanomas thicker than 2 mm to local excision margins of 2 cm or 4 cm. The primary endpoint was overall survival. After a median follow-up of 6.7 years, 181 and 177 patients died in the 2-cm and 4-cm groups, respectively, a nonsignificant difference. Five-year overall survival was 65% in both groups. Although rare, local recurrences were more likely in the 2-cm group (4.3%) than in the 4-cm group (1.9%). The authors conclude that a 2-cm resection margin is sufficient and safe for cutaneous melanomas thicker than 2 mm.

Comment: Many centers now take 2-cm margins regardless of tumor thickness, but for thick melanomas, this margin width had never been fully tested. With this trial, it is clear that 2 cm is sufficient even for high-risk melanomas. A few more local recurrences developed in the narrow-margin group but without an impact on overall survival. On a statistical note, the original intended accrual was 2000 patients, but fewer than 1000 ultimately enrolled, reducing the statistical power to show equivalence to 87%. The accompanying editorial by melanoma surgeons John Thompson and David Ollila raises this issue as a subtle but important point: This study does not have the power to prove equivalence; it can only demonstrate noninferiority of the 2-cm margin. I call this type of statistical phenomenon "power failure." Nevertheless, I would not recommend margins greater than 2 cm for primary melanomas, regardless of thickness.

— Hensin Tsao, MD, PhD

Published in Journal Watch Dermatology November 4, 2011

=======================================================