Association of choroidal invasion with retinoblastoma survival rates

Purpose: To assess the impact of varying levels of choroidal invasion on survival from retinoblastoma. Methods: A retrospective nationwide analysis of retinoblastoma cases diagnosed between 2004-2016 using the Surveillance, Epidemiology, and End Results database was conducted. Overall survival, cause-specific survival, and all-cause mortality risk were assessed as primary outcomes. Results: A total of 393 retinoblastoma patients were included, of whom 268 (68.2%) had no choroidal invasion, 91 (23.2%) had focal choroidal invasion, and 34 (8.7%) had massive choroidal invasion on enucleation. A total of 6 deaths occurred throughout an average follow-up period of 72.2 ± 47.1 months: 4 deaths were cancer related. Adjusted Cox regression demonstrated higher all-cause mortality in patients with massive choroidal invasion (HR, 41.29; 95% CI, 4.05-420.49; P = 0.002) relative to those without choroidal invasion; however, those with focal choroidal invasion (HR, 2.69; 95% CI, 0.17-43.09; P = 0.484) demonstrated no difference in all-cause mortality. On further stratification by level of optic nerve invasion (ONI), all cancer-related deaths (4/4) were found to have occurred in patients with massive choroidal invasion and concomitant postlaminar ONI (PLONI). Patients with massive choroidal invasion without PLONI demonstrated 5-year overall and cause-specific survival of 100%, whereas patients with massive choroidal invasion and PLONI demonstrated 5-year overall and cause-specific survival of 80.2%. Conclusions: All retinoblastoma-related deaths occurred in patients with both massive choroidal invasion and PLONI. These findings could not establish that massive choroidal invasion is an independent risk factor for poor outcome.