Biostatistical basis of elective node dissection for malignant melanoma.

During the years 1954 through 1964, 259 individuals with primary malignant melanoma had an elective node dissection. Microscopic metastases were found in 15% of these patients. The presence of only a microscopic focus of involvement gave a 10-year cure rate of 67%; metastasis larger than a microscopic focus in a single node, 50%; and more than one node, 15%. One hundred forty-five individuals were treated by wide excision alone with 18% subsequently requiring a therapeutic lymphadenectomy with a ten-year cure of only 6%. A prospective study was then initiated which was concerned with efficacy of selection of patients for elective node dissection. Clark's level of invasion was determined for 258 patients treated since January 1972. The depth of invasion of the primary lesion was found to correlate directly with the absence of lymph node metastases, extent of nodal involvement, and rate of recurrence. It is concluded that the concept of elective node dissection is valid.