In long-term survivors of osteosarcoma and Ewing sarcoma treated with the addition of radio- and chemotherapy, low bone mineral density (BMD) and fractures have been observed, presumably resulting from these adjuvants. Because patients with chondrosarcoma usually are not treated with conventional adjuvant treatment, observation of low BMD in patients with chondrosarcoma presumably would be the result of other mechanisms. However, BMD in patients with a history of chondrosarcoma has not been well characterized.Questions/Purposes
The aim of our study was to address the following questions: (1) Do long-term survivors of chondrosarcoma have normal BMD and, if not, which factors contribute to low BMD? (2) Is there a greater risk of fracture and does the Fracture Risk Assessment Tool (FRAX®) score reflect fracture likelihood?Methods
All known patients with a history of chondrosarcoma treated at our institution before 2006 were identified. Of 127 patients believed to be alive at the time of this study, 30 agreed to participate in this study (11 females, 19 males; mean age at surgery, 39 ± 12 years; mean followup, 12 ± 5 years). With the data available, the 30 participants were not different from the 97 nonparticipants in terms of age, sex, BMI, tumor grade, tumor location (axial versus appendicular, lower extremity versus elsewhere), and use of any treatment known to influence osteopenia (chemotherapy, lower extremity surgery). BMD was measured and history of fractures was assessed using a questionnaire. The patients' BMD measurements in this study were sex- and age-matched with a normative sex- and age-categorized reference population reported by Kudlacek et al. Associations were tested by univariate regressions and ANOVAs of all measures of BMD and eligible oncologic and demographic factors.Results
Eighteen of 30 (60%) patients had a pathologic BMD according to the WHO dual-energy x-ray absorptiometry definition, 15 (50%) had osteopenia, and three (10%) had osteoporosis. T-scores in the study cohort were lower than reference values for the femur neck (mean difference, 0.64; 95% CI, 0.27-1.01; p < 0.0015), but not for the spine (mean difference, 0.39; 95% CI, −0.06 to 0.84; p = 0.09). Thirteen patients (45%) reported a history of fractures not distinguishing between low and high impact. The incidence of fractures was 2.8 greater than expected from a comparison with a published microcensus survey of the Austrian population. No effect of the FRAX® score on fracture risk could be identified (p = 0.057).Conclusions
Long-term survivors of chondrosarcoma appear to be at greater risk for having low BMD develop than the healthy population. Although these results are preliminary and based on a very small sampling of patients, if they can be confirmed in larger studies, BMD assessment by dual-energy x-ray absorptiometry might be considered as these patients are followed posttreatment by sarcoma care units. The reasons for low BMD still must be elucidated.Level of Evidence
Level IV, prognostic study.