Racial and Ethnic Differences in Sarcoma Incidence Are Independent of Census-Tract Socioeconomic Status
Brandon J. Diessner, Brenda J. Weigel, Paari Murugan, Lin Zhang, Jenny N. Poynter, Logan G. Spector
Abstract
Abstract Background: Epidemiologic analyses of sarcoma are limited by the heterogeneity and rarity of the disease. Utilizing population-based surveillance data enabled us to evaluate the contribution of census tract-level socioeconomic status (CT-SES) and race/ethnicity on sarcoma incidence rates. Methods: We utilized the Surveillance, Epidemiology, and End Results program to evaluate associations between CT-SES and race/ethnicity on the incidence rates of sarcoma. Incidence rate ratios and 99% confidence intervals were estimated from quasi-Poisson models. All models were stratified by broad age groups (pediatric: <20 years, adult: 20–65 years, older adult: 65+ years) and adjusted for sex, age, and year of diagnosis. Within each age group, we conducted analyses stratified by somatic genome (fusion-positive and fusion-negative sarcomas) and for subtypes with >200 total cases. A P value less than 0.01 was considered statistically significant. Results: We included 55,415 sarcoma cases in 35 sarcoma subtype–age group combinations. Increasing CT-SES was statistically significantly associated with 11 subtype–age group combinations, primarily in the older age group strata (8 subtypes), whereas malignant peripheral nerve sheath tumors in adults were associated with decreasing CT-SES. Nearly every sarcoma subtype–age group combination displayed racial/ethnic disparities in incidence that were independent of CT-SES. Conclusions: We found race/ethnicity to be more frequently associated with sarcoma incidence than CT-SES. Our findings suggest that genetic variation associated with ancestry may play a stronger role than area-level SES-related factors in the etiology of sarcoma. Impact: These findings provide direction for future etiologic studies of sarcomas.