Survival from uveal melanoma in Western Australia 1981-2005.

BACKGROUND

The survival rates for patients diagnosed with uveal melanoma in Australia are unknown. Few long-term studies of uveal melanoma are available, and it is unclear whether their results are applicable to the Australian population.

DESIGN

Retrospective population-based study.

PARTICIPANTS

Patients diagnosed with uveal melanoma between 1981 and 2005 in Western Australia.

METHODS

Three hundred eight cases were included. Relative survival and Cox regression were performed. Variables tested for their predictive ability included patient age and sex, tumour-specific variables, and treatment modality.

MAIN OUTCOME MEASURES: All-cause survival rates and relative survival rates of patients with diagnosed uveal melanoma.

RESULTS

Relative survival rates for the entire cohort were 88.2%, 81.4% and 71.4% at 3, 5 and 10 years, respectively. Predictors of worse survival included mixed-cell tumour morphology (hazard ratio [HR] = 2.1; P-value = 0.002), tumour location at the ciliary body (HR = 1.7; P-value = 0.029) and tumour apical height more than 5 mm (HR 1.9, P-value = 0.026). Of all patients who underwent enucleation, those diagnosed in 1998-2005 died twice as fast (HR = 2.3; P-value = 0.004). In the 17 patients with metastasis, the median survival time from date of diagnosis of metastasis was 3.1 months.

CONCLUSIONS

These survival estimates are comparable to those reported for the USA, and more optimistic than those reported for most European-based studies. Tumour apical height, tumour site, tumour morphology and having an enucleation in certain calendar periods of diagnosis were independent predictors of survival. Survival prognosis for patients with diagnosed metastatic uveal melanoma is very poor.