Global Retinoblastoma Treatment Outcomes: Association with National Income Level.

PURPOSE: To compare metastasis-related mortality, local treatment failure, and globe salvage after retinoblastoma in countries with different national income levels. DESIGN: International, multicenter, registry-based retrospective case series. PARTICIPANTS: Two thousand one hundred ninety patients, 18 ophthalmic oncology centers, and 13 countries on 6 continents. METHODS: Multicenter registry-based data were pooled from retinoblastoma patients enrolled between January 2001 and December 2013. Adequate data to allow American Joint Committee on Cancer staging, eighth edition, and analysis for the main outcome measures were available for 2085 patients. Each country was classified by national income level, as defined by the 2017 United Nations World Population Prospects, and included high-income countries (HICs), upper middle-income countries (UMICs), and lower middle-income countries (LMICs). Patient survival was estimated with the Kaplan-Meier method. Logistic and Cox proportional hazards regression models were used to determine associations between national income and treatment outcomes. MAIN OUTCOME MEASURES: Metastasis-related mortality and local treatment failure (defined as use of secondary enucleation or external beam radiation therapy). RESULTS: Most (60%) study patients resided in UMICs and LMICs. The global median age at diagnosis was 17.0 months and higher in UMICs (20.0 months) and LMICs (20.0 months) than HICs (14.0 months; P < 0.001). Patients in UMICs and LMICs reported higher rates of disease-specific metastasis-related mortality and local treatment failure. As compared with HICs, metastasis-related mortality was 10.3-fold higher for UMICs and 9.3-fold higher for LMICs, and the risk for local treatment failure was 2.2-fold and 1.6-fold higher, respectively (all P < 0.001). CONCLUSIONS: This international, multicenter, registry-based analysis of retinoblastoma management revealed that lower national income levels were associated with significantly higher rates of metastasis-related mortality, local treatment failure, and lower globe salvage.

A Multicenter, International Collaborative Study for American Joint Committee on Cancer Staging of Retinoblastoma: Part II: Treatment Success and Globe Salvage.

PURPOSE: To evaluate the ability of the American Joint Committee on Cancer (AJCC) 8th edition to predict local tumor control and globe salvage for children with retinoblastoma (RB). DESIGN: International, multicenter, registry-based retrospective case series. PARTICIPANTS: A total of 2854 eyes of 2097 patients from 18 ophthalmic oncology centers from 13 countries over 6 continents. METHODS: International, multicenter, registry-based data were pooled from patients enrolled between January 2001 and December 2013. All RB eyes with adequate records to allow tumor staging by the AJCC 8th edition criteria and follow-up to ascertain treatment outcomes were included. MAIN OUTCOME MEASURES: Globe-salvage rates were estimated by AJCC clinical (cTNMH) categories and tumor laterality. Local treatment failure was defined as use of enucleation or external beam radiation therapy (EBRT), with or without plaque brachytherapy or intra-arterial chemotherapy (IAC). RESULTS: Unilateral RB occurred in 1340 eyes (47%). Among the 2854 eyes, tumor categories were cT1 to cT4 in 696 eyes (24%), 1334 eyes (47%), 802 eyes (28%), and 22 eyes (1%), respectively. Of these, 1275 eyes (45%) were salvaged, and 1179 eyes (41%) and 400 eyes (14%) underwent primary and secondary enucleation, respectively. The 2- and 5-year Kaplan-Meier cumulative globe-salvage rates without the use of EBRT by cTNMH categories were 97% and 96% for category cT1a tumors, 94% and 88% for cT1b tumors, 68% and 60% for cT2a tumors, 66% and 57% for cT2b tumors, and 32% and 25% for cT3 tumors, respectively. Risk of local treatment failure increased with increasing cT category (P < 0.001). Cox proportional hazards regression analysis confirmed a higher risk of local treatment failure in categories cT1b (hazard ratio [HR], 3.5; P = 0.004), cT2a (HR, 15.1; P < 0.001), cT2b (HR, 16.4; P < 0.001), and cT3 (HR, 45.0; P < 0.001) compared with category cT1a. Use of plaque brachytherapy and IAC improved local tumor control in categories cT1a (P = 0.031) and cT1b (P < 0.001). CONCLUSIONS: Multicenter, international, internet-based data sharing validated the 8th edition AJCC RB staging to predict globe-salvage in a large, heterogeneous, real-world patient population with RB.