Routinely collected incidence data have often lacked specific identification of ocular melanoma in the past and with increasing diagnosis and management of this disease by noninvasive techniques may now underestimate the true incidence. We attempted to accurately measure the incidence of ocular melanoma in Australia from 1990 to 1998 using 2 population-based sources, cancer registries and ophthalmologists. We examined the distribution across the continent, by latitude and in subpopulations, and evaluated the extent of nonnotification to cancer registries. One-half (51%) of the incident cases from 1996 to 1998 were diagnosed clinically and had no tissue diagnosis. An estimated 20% of melanomas, mainly those lacking a tissue diagnosis, were not notified to Australian cancer registries, but only 1.3% were not notified by ophthalmologists. Expert reviewers agreed that a high proportion (95%) of clinically diagnosed lesions were probable or possible melanomas. Incidence was significantly higher in men than in women, especially at older ages, relatively uniform across the Australian states with only weak evidence of a latitude gradient, and higher in rural than in urban areas. The incidence of ocular melanoma in people born in Southern Europe was half and in those born in Asia only 20% of the rate in people born in Australia and New Zealand. We concluded that it was possible to identify a high proportion of cases by surveying relevant sources additional to cancer registries and to be reasonably confident of the accuracy of the clinical diagnoses. The higher incidence in older men, who probably have higher sun exposure, than in older women and in residents of rural areas, where outdoor work is more prevalent, than in urban areas; the low incidence in dark-eyed populations, who may have a lower sensitivity to or less transmission of solar radiation to the choroid; and the preferential location of ocular melanomas in ocular sites known to receive the highest exposure to solar radiation all support a role for solar radiation in the aetiology of ocular melanoma. The absence of a strong latitude gradient in incidence does not argue against such a role because exposure of the eye to solar UV is probably determined most by the horizon sky, where ambient solar UV is less affected by latitude.
Copyright 2003 Wiley-Liss, Inc.