Conditional survival of patients with nonmetastatic bone osteosarcoma.

Abstract

e23511 Background: Conditional survival provides a dynamic prediction of prognosis for patients surviving a defined period of time after diagnosis. This study aimed to determine the conditional survival and prognostic factors over time among patients with non-metastatic bone osteosarcoma. Methods: We reviewed 714 bone osteosarcoma patients treated from 1985 to 2016. Patients with metastatic disease at diagnosis or limited follow up were excluded, resulting in 587 cases for analysis. Clinical and pathological variables were recorded. Predictive variables included age at diagnosis, gender, previous radiation history, tumor site, tumor size, histologic subtype, histologic grade, resection margin, chemotherapy, and radiation therapy. The multivariate Cox proportional hazards regression was used to analyze prognostic factors of conditional overall survival and progression-free survival at baseline and 5 years after diagnosis. Results: The estimated 5-year conditional overall survival increased from 71.0% (95% CI: 67.5%-75.0%) at baseline to 86.9% (95% CI: 82.6%-90.5%) at 5 years, which means if a patient with non-metastatic bone osteosarcoma survived 5 years, the chance of surviving another 5 years was 86.9%. If the patient was progression-free for 5 years, the 5-year conditional overall survival was even higher, 93.2% (95% CI: 89.5%-96.4%), and the 5-year conditional progression-free survival improved from 57.1% (95% CI: 53.3%-61.0%) at baseline to 91.2% (95% CI: 87.5%-94.6%) at 5 years. Prognostic factors for mortality and disease progression change as survival time increases. At baseline, age (p < 0.001 and p = 0.003), histologic subtype (p < 0.001 and p = 0.001), grade (p < 0.001 and p < 0.001), tumor size (p = 0.002 and p = 0.002), resection margin (p < 0.001 and p < 0.001) and chemotherapy (p = 0.001 and p = 0.001) were predictive of both overall survival and progression-free survival. However, only age (p < 0.001) and histologic subtype (p = 0.015) remained significant for mortality and resection margin (p = 0.001) for disease progression at 5 years. Conclusions: The survival probability of osteosarcoma improves as survival time increases. Estimates of conditional survival can provide useful information for individualized surveillance strategies, risk evaluation, patient counseling, and making clinical decisions.