Changing Prognostic Factors in Osteosarcoma: Analysis of 381 Cases from Two Institutions
Deyin Xing, Shadi A Qasem, Kofi Owusu, Kui Zhang, Gene P Siegal, Shi Wei. University of Alabama at Birmingham, Birmingham, AL; Wake Forest University, Winston-Salem, NC
Background: Osteosarcoma (OSA) occurs most commonly in children/young adults, with a historic second incidence peak in the elderly. Most studies have focused on those occurring in adolescence. Detailed information on descriptive features and prognostic factors in patients of different age groups is lacking. The aim of this study was to identify clinicopathologic factors significantly associated with survival in young, middle, and older age groups.
Design: The tumor registries of the authors' institutions were searched to identify OSA cases between 1973 and 2012. The clinicopathologic factors were recorded, including age, sex, race, tumor site, type, grade and size, and % necrosis post-chemotherapy. Analyses of overall survival (OS) and recurrence free survival (RFS) were performed in each stratified age group (G1, <25, n=202; G2, 25-54, n=118; G3, >54, n=61) by utilizing the Cox proportional hazard model.
Results: A total of 381 OSA cases were retrieved. The peak incidence was in the second decade (41%), and 53% were younger than age 25. There was a steady incidence rate (10%/decade) thereafter until the 6th decade, when it started to decline. There were significant differences in tumor types (conventional vs. other) and sites (long bone vs. other, bone vs. extraskeletal) among different groups. Univariate analysis revealed that factors significant for OS were gender, tumor site, grade and size in G1, tumor type in G2, and tumor size in G3, whereas significant factors for RFS were tumor site and size in G1, tumor type in G2, and race and tumor size in G3. Moreover, % necrosis was a significant factor for RFS only when all age groups were combined, but not for OS. Upon multivariate analysis, factors significantly associated with a favorable OS were female gender (p<0.05), long bone location (p<0.01), low grade (p=0.01) and small tumor size (p<0.0001) in G1, parosteal OSA in G2 (p<0.01), and extraskeletal site in G3 (p=0.03); whereas factors significantly associated with favorable RFS were female gender (p<0.05) and small tumor size (p<0.001) in G1; and low grade in G2 (p=0.05).
Conclusions: Modern OSA did not have a second incidence peak. Rather, it showed a steady incidence rate in middle age individuals and gradual decline in the elderly. There were significant different distributions in tumor types and sites across age groups. Importantly, while %necrosis did not appear to be associated with OS, distinctive prognostic factors differed significantly among different age groups, thus providing a rationale for age-based management strategies.
Category: Bone & Soft Tissue
Tuesday, March 5, 2013 11:30 AM
Proffered Papers: Section G, Tuesday Morning