Cutaneous malignant melanoma aspects of prognostic factors and time : trends in a Swedish population

Abstract

One of the aims of this thesis was to assess the possible impact of primary and secondary preventive activities in the population of Stockholm-Gotland with the objective to reduce the incidence and mortality from cutaneous melanoma. Another aim was to investigate whether thin melanomas and hereditary melanomas represent separate biologic entities.

The study base was cases with cutaneous melanoma reported to the Swedish Cancer Registry 1976 - 1994, patients registered in the Stockholm-Gotland Regional Melanoma Registry 1976-1994, individuals reported with malignant melanoma as an underlying cause of death to the Swedish Cause- of-Death Registry 1970-1996, and patients reported to the National Familial Melanoma data base 1987-1994.

The average annual increase of the age-standardised incidence was about 5 % in both genders. The increase mainly concerned thin turnouts and melanoma in situ. During the 1990s the incidence in males levelled off. This apparent change in trend was not observed outside the Stockholm-Gotland region. The estimated five-year melanoma-specific survival rate in patients diagnosed with localised cutaneous melanoma increased from 84 % to 92 % during 1976-1994. No time-trend for melanoma-specific survival was observed among patients with regional or distant metastases.

The increased survival during 1976-1989 could be fully explained by a trend towards more favourable tumour characteristics. During 1990-1994, screening activities may have contributed to the observed survival improvement. An upward trend in melanoma mortality appeared to level off in the mid 1980s. In females a slight decrease was observed during 1987-1996. The change in trend was most pronounced in the Stockholm-Gotland region. The results indicate that the interventional activities in Stockholm-Gotland may have influenced both melanoma incidence and mortality.

Of patients diagnosed with thin cutaneous melanoma (< 0.8 nun), four percent developed recurrent disease after a median follow-up of 50 months. Anatomic site, tumour thickness, level of invasion and turnout regression were significant prognostic factors. No subgroup of patients could be identified that was without risk of recurrent disease. Thin melanomas did not appear to constitute a separate biologic entity. A family history of melanoma was reported in five percent of patients diagnosed with cutaneous melanoma. Familial melanoma was associated with young age at diagnosis, the presence of dysplastic nevi, multiple melanoma and melanomas detected at an early stage. The prognosis of the familial cases did not differ from that of sporadic melanomas when established prognostic factors were taken into account.