Purchase this article with an account.
Ketki Soin, Michael Andreoli, Felix Chau, Yannek Isaac Leiderman; Retinoblastoma: A Review of 1,452 Cases. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4473.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
This study aims to evaluate the epidemiological trends in retinoblastoma over 37 years.
Reported cases of retinoblastoma occurring between 1973 and 2009 in the Surveillance, Epidemiology, and End Results (SEER) Program tumor registry were identified for analyses. Overall survival was the primary outcome measure. Kaplan-Meier survival curves were generated for tumor grade, race, age at diagnosis, gender, year of diagnosis, and treatment modality. Cox proportional hazards regression was performed to assess the impact of patient and tumor characteristics on survival.
The SEER database contained 1,452 cases of retinoblastoma over 37 years. The mean patient age was 1.41 years (standard deviation 2.31 years, range 0 to 35 years). 51.9% of patients were boys. Cytological, histological, or microscopic confirmation of diagnosis was performed in 84.3% of cases. Tumors were unilateral in 72.4% (left-sided primary tumor 34.9%, right-sided 37.5%), and bilateral in 27.6% of patients. The mean follow-up interval was 127.8 months. Survival was not altered by patient gender or age at diagnosis. Increased survival was associated with more recent year of diagnosis in this study population. There was a steady decline in treatment with radiotherapy during the study interval. Patients with bilateral tumors were diagnosed earlier (0.46 years) than those with unilateral tumors (1.77 years; P < 0.0001). Unilateral tumors were associated with greater overall survival rates and a different subset of second malignancies than bilateral disease. Grade 1 tumors were diagnosed at a younger age (0.94 years) than grade 3 (2.24 years; P < 0.0001) or grade 4 tumors (2.14 years; P < 0.0001). Low-grade tumors were associated with improved survival rates. There was a decrease in high-grade tumors reported during the 37-year interval. A multivariate Cox proportional hazards analysis revealed that T stage and laterality were the only covariates that correlated with overall survival.
Evaluation of data from a large cohort of patients with retinoblastoma over a 37-year interval revealed several trends. Survival rates have improved during the interval analyzed, with lower T stage and unilateral disease correlating with improved survival. Data on tumor laterality, tumor grade, year of diagnosis, and secondary malignancies improve our ability to counsel patients with retinoblastoma and their families regarding prognosis, natural history, and future surveillance.
This PDF is available to Subscribers Only