Patterns of bone sarcomas as a second malignancy in relation to radiotherapy in adulthood and histologic type.

Abstract

Radiotherapy decreases cancer mortality, but is associated with an increased incidence of second primary cancers, including osteosarcomas, especially after exposure in childhood. It remains uncertain whether radiation is related to other histologic types of bone sarcomas such as chondrosarcomas that are more common in adulthood. Using data from 1973 to 2008 Surveillance Epidemiology and End Results registries, we evaluated long-term risk of bone cancer in 1,284,537 adult 5-year cancer survivors. We used standardized incidence ratios (SIR) to compare second bone sarcoma rates to the general population for each histologic type. We also used multivariate Poisson regression to estimate the relative risk (RR) associated with radiotherapy for the most common subtypes, osteosarcoma and chondrosarcoma. By the end of 2008, 159 second bone sarcomas were reported. Compared with the general population, the risk of developing any bone sarcoma was increased by 25% in patients with no history of radiotherapy [Observed (O) = 89, SIR = 1.25 (1.00-1.54)] and by 257% in patients with a history of radiotherapy [O = 70, SIR = 3.57 (2.78-4.50)]. For each histologic subtype, SIRs were higher among patients who had previously received radiotherapy than among those who had not. The RR for radiotherapy for osteosarcoma (n = 63) was 5.08 (3.05-8.59) and for chondrosarcoma (n = 69) was 1.54 (0.88-2.59), and these risks were even greater for second sarcomas that arose in the radiotherapy field used to treat the first cancer [osteosarcoma, RR = 10.35 (4.96-23.66); chondrosarcoma RR = 8.21 (2.09-39.89)]. Our findings provide the first evidence of a likely association between radiation exposure and chondrosarcoma. These results further our understanding of radiotherapy-related cancer risks and will potentially direct practices in long-term surveillance of cancer survivors.