Background: Large, prospective, randomized trials with long term follow-up are required to obtain an unbiased evaluation of the significance of resection margins in patients with cutaneous melanoma.
Methods: The Swedish Melanoma Study Group performed a prospective, randomized, multicenter study of patients with primary melanoma located on trunk or extremities and with a tumor thickness > 0.8 mm and </= 2 mm. Patients were allocated randomly to a 2-cm excision margin or a 5-cm excision margin. In total, 989 patients were recruited during the period 1982-1991. The median follow-up was 11 years (range, 7-17 years) for estimation of survival and 8 years (range, 0-17 years) for evaluation of recurrent disease.
Results: The crude rate of local recurrence, defined as a recurrence in the scar or transplant, was < 1% (8 of 989 patients). Twenty percent of the patients (194 of 989 patients) experienced any disease recurrence, and 15% (146 of 989 patients) died of melanoma. There were no statistically significant differences between the two treatment arms. In a multivariate Cox analysis with patients allocated to wide excision as the reference group, the estimated relative hazards for overall survival and recurrence free survival among those allocated to a 2-cm resection margin were 0.96 (95% confidence interval, 0.75-1.24), and 1.02 (95% confidence interval, 0.80-1.30), respectively.
Conclusions: In this long term follow-up study, local recurrences were found to be rare among patients with tumors > 0.8 mm thick and </= 2.0 mm thick. No difference in recurrence rate or survival between the two treatment groups was found. Patients in this category can be treated with a resection margin of 2 cm as safely as with a resection margin of 5 cm.