HISTOLOGIC STAGING IN MALIGNANT-MELANOMA - CROSS-SECTIONAL AREA REVISITED

Background and Objectives: Tumor thickness is considered the single mo st important predictor of survival in clinically localized malignant m elanoma. A recent study found tumor volume a more sensitive predictor of survival than thickness. Volume measurement, however, is complicate d, time consuming, and based on biologically imprecise mathematical mo dels of tumor configuration. This report compares the prognostic power of cross-sectional area (CSA), a simpler measurement than volume, wit h tumor thickness. Methods: Forty-five patients with clinically locali zed malignant melanoma and a minimum 5-year follow-up post excision wi th negative resection margins were retrospectively followed for diseas e recurrence or death. Digitalized histologic images of each tumor wer e made from the original pathology slides and stored on a compact disc . Maximum tumor thickness and CSA were calculated for each primary mel anoma using an image analysis program and compared for predictive accu racy of 5-year survival. Results: CSA was positively correlated with m aximum tumor thickness (r = 0.76). Both measures had a similar predict ive accuracy for survival. Patients with melanomas less than 8 mm(2) h ad superior 5-year (94%) and disease-free survival rates (78%) compare d to patients with melanomas exceeding 8 mm(2) (5-year survival, 62%; 5-year disease-free survival, 23%). Conclusions: CSA is an easily calc ulated measurement that is as predictive for 5-year survival as is Bre slow's thickness. Prospective assessment Of CSA is warranted. (C) 1998 Wiley-Liss, Inc.