Trilateral Retinoblastoma: Incidence and Outcomes

Overview

Patients with heritable retinoblastoma are at risk to develop an intracranial brain tumor, which is often fatal. The investigators intend to look at the incidence and survival of trilateral retinoblastoma (which is a brain tumor that can either be located in the pineal gland or elsewhere in the brain) in retinoblastoma patients globally. All retinoblastoma patients from participating centers will be included. The investigators hypothesize that the apparent incidence of trilateral retinoblastoma (especially the usually later diagnosed pineal trilateral retinoblastoma) in low-income countries will be lower because of low chances of surviving the ocular tumors at about 50% and also because of possible under-diagnosis. Therefore, as retinoblastoma care improves in low-income countries the incidence of (pineal) trilateral retinoblastoma might go up. Knowledge about incidence and survival can help improve health practices in parts of the world where this might be needed. Therefore this global study firstly aims 1) to evaluate survival after trilateral retinoblastoma and factors influencing survival and 2) to evaluate incidence of trilateral retinoblastoma by country income level. The study will run from 2024 through 2027.

This study will adhere to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for cohort studies and to the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) statement. The study is designed as a prospective cohort study of retinoblastoma patients diagnosed in one year from as many as possible worldwide treatment centers. The methods of patient recruitment are partly similar to previous global retinoblastoma studies. Data will be extracted from a the Global Retinoblastoma study follows up patients who will present with retinoblastoma in 2024-2027. All unique trilateral retinoblastoma patients (cases) and all other retinoblastoma patients (controls) with will be included in the incidence analysis. For survival analysis follow-up data will be required. After initial submission of case data by participating centers three years after the start of the study information on follow-up, metastasis, eye globe salvage, laterality, diagnosis of trilateral retinoblastoma and survival outcome will be available. At a later stage participating centers that reported trilateral retinoblastoma cases will be contacted again for additional information. All trilateral retinoblastoma patients will be included in the analysis, regardless if additional information is provided. The investigators will request the following additional data: * location (pineal, supra-/parasellar, other location (e.g., 3rd ventricle)), * treatment for trilateral retinoblastoma, * if not clear from initial report: cause of death (trilateral retinoblastoma, side effect of treatment, other), * details on (germline) RB1 mutation. * country of residence, * family history of retinoblastoma, * was patient symptomatic or asymptomatic when trilateral retinoblastoma was diagnosed (if symptomatic: what where the symptoms? If asymptomatic: how was the tumor diagnosed), * information on trilateral retinoblastoma confirmation (e.g., biopsy, CSF, treatment response, disease progression), * treatment for trilateral retinoblastoma, * was the treatment with intent to cure (active treatment) or was the intent palliative, * maximum diameter (mm) of tumor, * wat trilateral retinoblastoma metastasized at the time of diagnosis, * was this case published (if yes: where?), * also missing data during the initial submission phase will be requested again. If feasible additionally all centers that will also be asked if they can retrospectively provide data on trilateral retinoblastoma patients of whom retinoblastoma was not diagnosed in 2024 and were diagnosed with trilateral retinoblastoma in the years 2020 through 2027. Outcome measures Incidence and risk factors of developing trilateral retinoblastoma will be analyzed in the following manner (cases and controls who were diagnosed with retinoblastoma in 2024): * overall incidence of pineal and non-pineal trilateral retinoblastoma in all retinoblastoma patients, bilateral retinoblastoma patients and heritable retinoblastoma patients (all patients with wither bi- or trilateral retinoblastoma, familial retinoblastoma or a proven germline RB1 mutation will be considered heritable), * estimate possible under-diagnosis of trilateral retinoblastoma in lower-income countries, * potential modifiers of incidence (risk of developing trilateral retinoblastoma) will be analyzed: pineal versus non-pineal trilateral retinoblastoma, age at diagnosis of retinoblastoma (e.g., \< or ≥ 12 months old), country income level. Survival analysis of trilateral retinoblastoma alone (all cases who were diagnosed with trilateral retinoblastoma in the years 2024 through 2027): * overall and trilateral retinoblastoma specific survival will be analyzed, * potential predictors of survival (e.g., pineal versus non-pineal trilateral retinoblastoma, tumor size, previous treatment for retinoblastoma, treatment for trilateral retinoblastoma, symptomatic versus asymptomatic, RB1 mutation, active treatment, metastastic disease, country income level). Survival analysis of trilateral retinoblastoma in the entire retinoblastoma cohort (cases and controls who were diagnosed with retinoblastoma in 2024-2027): * trilateral retinoblastoma mortality versus other-cause mortality (also stratified by country income level and al versus non-pineal trilateral retinoblastoma), * event-free survival (with an event defined as trilateral retinoblastoma diagnosis) (also stratified by country income level and al versus non-pineal trilateral retinoblastoma), * event-free survival will be analyzed for potential predictors (age at diagnosis of retinoblastoma, previous systemic chemotherapy, previous external beam radiotherapy, country income level). Other analyses (all cases who were diagnosed with trilateral retinoblastoma in the years 2024 through 2027): * verify independence between the age at diagnosis of retinoblastoma and trilateral retinoblastoma, * verify the period that retinoblastoma patients are at risk to develop trilateral retinoblastoma, * whether trilateral retinoblastoma is usually diagnosed at a later date if the diagnosis of retinoblastoma was before the age of 12 months, * evaluate and more precisely estimate of the previously estimated lead time between symptomatic and asymptomatic trilateral retinoblastoma of about 1 year. The income level of a country will be determined by the at the time most recent World Population Prospects by the Department of Economic and Social Affairs of the United Nations. Statistical analysis Incidence will be reported as proportions with 95% exact binomial confidence intervals. Cumulative incidence of trilateral retinoblastoma curves will be created, accounting for death from other causes without diagnosis of trilateral retinoblastoma. Overall survival (all-cause mortality), disease specific survival and event free survival were performed. An event is defined as being diagnosed with trilateral retinoblastoma. The Kaplan-Meier product-limit method and the log rank test will be used to analyze survival data. A Cox proportional hazards model will be employed to analyze potential predictors factors. Schoenfeld residuals will be used to check the proportionality assumption of the model. The investigators each case with a unique identifier for stratification based on the center to account for heterogeneity between these centers. In case of subgroup analyses original P-values will presented, as well as adjusted p-values corrected for multiple hypothesis testing according to Bonferroni's method.