Age as a prognostic factor for patients with osteosarcoma: an analysis of 438 patients

Abstract

Background: It is unclear whether age at diagnosis is an important prognostic factor in patients with osteosarcoma. Understanding this relationship could yield valuable insight into therapeutic rationale, focus patient selection for clinical trials, advance molecular concepts and theories, and expand current principles guiding prognosis. Our aim was to understand if age at diagnosis is a prognostic indicator for eventual outcome, as measured by disease-free survival and overall survival in patients with osteosarcoma.

Methods: Our cohort consisted of 438 patients of all ages who were diagnosed with osteosarcoma between 1 January 1980 and 31 December 2000 and who underwent the majority of their treatment at M.D. Anderson Cancer Center (MDACC). Patient and tumor specific variables were collected including patient demographics, patient history, primary tumor information (i.e., location, size, histology, extension, necrosis, etc.), treatment strategy (i.e., surgery, chemotherapy, and/or radiotherapy), metastatic disease information, long-term follow-up, and eventual outcome. Statistical analyses, including univariate and multivariate analyses were performed, with overall survival and disease-free survival as the primary outcome measures.

Results: The median age at diagnosis was 18.1 years (range 2 months to 78.8 years). Median follow-up was 4.2 years (range 5 days to 22.8 years) for all patients and 12.3 years (range 1 month to 22.8 years) for 209 surviving patients. Survival rates at 5, 10, and 15 years were 54.1, 47.2, and 45.2%, respectively. On univariate analyses, age >or= 40 was found to be a poor prognostic factor. Other prognostic factors included tumor size, metastasis at diagnosis, soft-tissue tumor extension, surgery type, chemotherapy group, and tumor necrosis. Age was not identified a statistically significant prognostic variable on multivariate analysis.

Conclusions: Age at diagnosis does not appear to be a significant independent prognostic variable for overall survival or disease-free survival in patients with osteosarcoma. Although our data indicate that patients in the fifth decade and older fare worse than younger patients, other variables such as tumor necrosis, tumor extension, and tumor location are likely responsible for the observed decline in overall survival and disease-free survival.

Fig. 1

Kaplan–Meier curves of overall and disease-free survival

Fig. 2

Kaplan–Meier curves of overall survival grouped by age. (log-rank, p = 0.869)

Fig. 3

Kaplan–Meier curves of overall survival grouped by age. (log-rank, p = 0.051)

Fig. 4

Graphs of the log likelihoods (a) and p values (b) at ages of dichotomization, generated by repeated Cox regression analyses. (red line signifies p = 0.05)