Factors predictive of radiation retinopathy post (125)Iodine brachytherapy for uveal melanoma.

OBJECTIVE To report the host, tumour, and radiation-related predictive factors for developing radiation retinopathy post (125)Iodine brachytherapy for uveal melanoma. DESIGN A retrospective clinical case series. PARTICIPANTS Three hundred consecutive patients with uveal melanoma treated with (125)Iodine brachytherapy. METHODS Electronic chart review of demographic, clinical, treatment, and follow-up data. Proliferative and (or) nonproliferative radiation retinopathy patients were included. Cumulative incidence rates were calculated using Kaplan-Meier estimates. Univariate and multivariate statistical regression analyses were performed to identify factors predictive of radiation retinopathy. RESULTS The mean follow-up period was 48 months. Radiation retinopathy occurred in 107 patients (36%). Actuarial incidence of radiation retinopathy was 30% (CI 24%-36%) at 24 months post-treatment. In the multivariate model, the predictive factors were a younger age (hazard ratio [HR] 0.98, p < 0.03), diabetes (HR 2.17, p < 0.007), and hypertension (HR 2.17, p < 0.004). Tumour-related factors were proximity to optic disc (HR 0.95, p < 0.02) and proximity to foveola (HR 0.96, p < 0.02). Pretreatment tumour dimensions, other tumour characteristics, and total radiation dose did not demonstrate statistically significant risks for developing radiation retinopathy. CONCLUSIONS Radiation retinopathy affects 30% of patients 2 years following (125)Iodine brachytherapy for uveal melanoma. The predictive factors for radiation retinopathy are a younger age, comorbidity with diabetes or hypertension, and proximity of the tumour margin to the optic disc or foveola. Identifying the predictive factors for developing radiation retinopathy can modify follow-up for patients at risk, which may permit earlier management of the developing radiation-induced ischemic retinal changes.