Changes from mid-1980s to late 1990s among clinical and demographic correlates of melanoma thickness.

Tumour thickness is the most relevant prognostic factor for cutaneous melanoma. Although the increasing incidence of melanoma is currently attributable to "thin" lesions, the incidence rates of "thick" melanomas have not declined. We want to identify the clinical and demographic characteristics of patients that are associated with diagnosis of thick (> 3 mm) cutaneous melanoma and whether they had changed from mid-1980s to late-1990s. Cutaneous malignant melanomas incidence in 1985-87 and in 1995-97 were retrieved from the Tuscany Cancer Registry, central Italy. Only cases with Breslow-thickness information (182/260 in 1985-87 and 387/490 in 1995-97) were included. Thickness was categorised in < = 1 mm, 1-3 mm and > 3 mm. Thickness was evaluated for each period of time according to gender, age, histological type, site and residence. For cases diagnosed in 1995-97 the effect of such variables in predicting the risk of a thick tumour (vs. a thin one) was analysed in a logistic model. In 1985-87 patients with thick melanoma were more likely to be - with a statistically significant difference - males (38.1 % of thick tumours) than females (19.4 %), over 70 (57.7 % of thick tumour), with nodular melanoma (62.1 %) and residents far from the city of Florence (30.3 %); no differences were evidenced according to site. From 1985-87 to 1995-97 there was a global shift towards thinner melanomas. In 1995-97 nodular type and old age were the only variables significantly associated with thick melanomas when other factors were taken into account in a multivariate analysis. According to most recent data, early detection activities should be focused on older patients and on nodular histotype. Male sex and residence was no longer found to be associated with late melanoma diagnosis.