Characteristics and prognostic factors of adult patients with osteosarcoma from the SEER database

Abstract

Osteosarcoma is the most common bone malignancy. There are many studies on the prognostic factors of children and adolescents, but the characteristics and prognostic factors of adult osteosarcoma are rarely studied. The aim of this study was to construct a nomogram for predicting the prognosis of adult osteosarcoma. Information on all osteosarcoma patients aged ≥ 18 years from 2004 to 2015 was downloaded from the surveillance, epidemiology and end results database. A total of 70% of the patients were included in the training set and 30% of the patients were included in the validation set. Univariate log-rank analysis and multivariate cox regression analysis were used to screen independent risk factors affecting the prognosis of adult osteosarcoma. These risk factors were used to construct a nomogram to predict 3-year and 5-year prognosis in adult osteosarcoma. Multivariate cox regression analysis yielded 6 clinicopathological features (age, primary site, tumor size, grade, American Joint Committee on Cancer stage, and surgery) for the prognosis of adult osteosarcoma patients in the training cohort. A nomogram was constructed based on these predictors to assess the prognosis of adult patients with osteosarcoma. Concordance index, receiver operating characteristic and calibration curves analyses also showed satisfactory performance of the nomogram in predicting prognosis. The constructed nomogram is a helpful tool for exactly predicting the prognosis of adult patients with osteosarcoma, which could enable patients to be more accurately managed in clinical practice.

Figure 1.

Flowchart of sample selection.

Figure 2.

Kaplan–Meier estimated survival in patients with osteosarcoma stratified by age (A), sex (B), race (C), marital (D), primary site (E), tumor size (F), tumor number (G), grade (H), AJCC Stage (I), tumor stage (J), surgery (K), radiation (L), and chemotherapy (M). AJCC = American Joint Committee on Cancer.

Figure 3.

Nomogram to predict 3- and 5-year overall survival for osteosarcoma patients.

Figure 4.

Predictive performance of the survival nomogram reflected by ROC curves. ROC curves for the 3-year and 5-year osteosarcoma in patients in the training cohort (A–B) and in the SEER validation cohort (C–D). ROC = receiver operating characteristic, SEER = surveillance, epidemiology, and end results.

Figure 5.

The calibration curves for predicting osteosarcoma in the training and validation sets. Calibration plots of 3-year and 5-year osteosarcoma in the training cohort (A–B) and in the SEER validation cohort (C–D). SEER = surveillance, epidemiology, and end results.