Constructing a Nomogram for Predicting Pelvic Osteosarcoma: A Retrospective Study Based on the SEER Database and a Chinese Cohort
Yefeng Xu,Qingying Yan,Jiewen Yang,Miao Cheng,Jialu Chen,Yongwei Yao,Yunxia Liu
DOI: https://doi.org/10.12968/hmed.2024.0339
2024-08-30
Abstract:Aims/Background Generally, pelvic osteosarcoma has a worse prognosis compared with limb osteosarcoma. This study aims to create and validate a new nomogram for predicting the prognosis of pelvic osteosarcoma. Methods Clinical data of 62 patients derived from the Surveillance, Epidemiology, and End Results (SEER) database and 31 Chinese patients diagnosed with pelvic osteosarcoma were gathered. Kaplan-Meier survival analysis was utilized to calculate the median survival time for all variables. Univariate and multivariate Cox regression models were employed to identify the prognostic factors of pelvic osteosarcoma. A nomogram was constructed using data gleaned from the SEER cohort and verified using the receiver operating characteristic (ROC) curve and calibration plot in the Chinese cohort. Results Kaplan-Meier analysis revealed that individuals of other races (Asians) (hazard ratio (HR) = 0.24, 95% confidence interval (CI): 0.1-0.57, p = 0.001), aged ≤51 years old (HR = 0.4, 95% CI: 0.22-0.73, p = 0.003), and with tumor size ≤160 mm (HR = 0.37, 95% CI: 0.2-0.71, p = 0.03) had better survival outcomes. Conversely, factors such as no primary surgery (HR = 3.6, 95% CI: 1.81-7.15, p < 0.001), lung metastasis (HR = 1.96, 95% CI: 1.17-3.28, p = 0.010), and radiotherapy (HR = 1.89, 95% CI: 1.10-3.25, p = 0.021) were associated with poorer survival. Multivariate Cox analysis indicated that lung metastasis (HR = 2.57, 95% CI: 1.29-5.13, p = 0.008), other races (Asians) (HR = 0.23, 95% CI: 0.07-0.75, p = 0.015), tumor size (HR = 0.28, 95% CI: 0.13-0.62, p = 0.001) and age (HR = 0.3, 95% CI: 0.16-0.59, p < 0.001) were independent prognostic factors for pelvic osteosarcoma. Univariate and multivariate Cox regression models identified three independent variables in the training cohort: age, lung metastasis, and tumor size. A predictive nomogram was developed based on the data from the SEER cohort and validated in the Chinese cohort. The areas under the curves (AUCs) that are used to predict 1-year, 2-year, and 3-year survival rates were 0.81 (95% CI: 0.68-0.94), 0.75 (95% CI: 0.63-0.86), and 0.80 (95% CI: 0.70-0.89) in the training cohort, and 0.67 (95% CI: 0.30-1.04), 0.66 (95% CI: 0.43-0.90) and 0.71 (95% CI: 0.50-0.93) in the validation cohort. Conclusion The predictive nomogram constructed in this study facilitates accurate and effective prediction of the overall survival of patients with pelvic osteosarcoma and helps enhance the clinical decision-making process.