Osteosarcomas in retinoblastoma-survivors. A report of 28 affected patients from the Cooperative Osteosarcoma Study Group (COSS)

Abstract

PurposeMany publications address the epidemiology of secondary osteosarcomas following retinoblastoma. Treatment- and outcome-related information is, however, scarce. We used a large cooperative group’s database to study this issue. Patients and methodsThe database Cooperative Osteosarcoma Study Group COSS was searched for patients with osteosarcoma following a previous retinoblastoma. Patients were then analyzed for demographic factors, local/systemic treatments received, and outcomes. Results28 eligible patients were identified. Median age at retinoblastoma was .5 years, 89% occurred bilaterally. Retinoblastoma-therapy was by surgery in 26/27, radiotherapy in 26/27, chemotherapy in 10/26 (rest unknown). Osteosarcoma was diagnosed after 13.7 (3.1 – 36.1) years, extremities and head/neck affected in 14/28, each. Four/28 patients had primary metastases. Osteosarcoma treatment included chemotherapy in all cases, local therapy surgery in 27/28. Histological response was good in 9/16 evaluable cases. Surgery of all sites was macroscopically complete in 23/27 operated tumors and microscopically complete in 17/26 (1 unknown). Radiotherapy was administered to 3 craniofacial tumors. Median follow-up was 3.5 (.4 – 30.1) years from osteosarcoma diagnosis, 8/28 patients remaining event-free. Altogether, five patients suffered further secondary malignancies. Actuarial overall and event-free survival at 2 and 5 years from osteosarcoma were 73% (standard error: 8%) / 50% (10%) and 47% (10%) / 22% (9%), respectively. ConclusionThis comparatively large cohort of osteosarcomas after retinoblastoma proves that the latter may be treated curatively. While their prognosis is far worse than that of primary osteosarcomas, partly due to a predilection for craniofacial involvement, selected patients may still become long-term survivors with appropriate therapies.