Electrolyte prognosis scoring system can predict overall survival in patients with osteosarcoma

Abstract

Osteosarcoma stands as the most prevalent bone tumor, characterized by a heightened tendency for local recurrence and distant metastasis, resulting in a bleak prognosis. Presently, there exists a shortage of novel markers to effectively determine the prognosis of osteosarcoma patients. Recent research indicates that hematological markers partially mirror an individual's microenvironment, offering potential insights into predicting patient prognosis. However, prior studies predominantly focused on the prognostic significance of singular hematological indices, failing to comprehensively represent the tumor microenvironment of patients. In our investigation, we meticulously gathered data on 22 hematological and electrolyte markers, utilizing LASSO Cox regression analysis to devise an Electrolyte Prognostic Scoring System (EPSS). The EPSS encompasses various indicators, including immunity, inflammation, coagulation, and electrolyte levels. Our findings indicate that the EPSS stands as an independent prognostic factor for overall survival among osteosarcoma patients. It serves as a valuable addition to clinical characteristics, adept at discerning high-risk patients from those deemed clinically low-risk. Furthermore, EPSS-based nomograms demonstrate commendable predictive capabilities.

Keywords: OS; electrolyte; hematological markers; osteosarcoma; prognostic nomograms.

Figure 1

Work flow chart of this study.

Figure 2

Construction of EPSS and its comparison with individual hematological and electrolyte parameters. (A) Forest plot showing the results of univariate cox regression analysis of 22 hematological and electrolyte markers; (B) ROC curves showing the predictive power of EPSS in the training set versus a single hematology or electrolyte indicator; (C) ROC curves showing the predictive power of EPSS in the validation set versus a single hematology or electrolyte indicator.

Figure 3

There are significant differences between patients in EPSS risk groups. (A) High-risk patients in the training set had significantly lower overall survival than low-risk patients; (B) High-risk patients in the validation set had significantly lower overall survival than low-risk patients.

Figure 4

EPSS is an independent prognostic factor for overall survival in patients with osteosarcoma and has certain advantages compared with clinical characteristics. (A) Forest plot showing the results of univariate COX regression analysis of EPSS and clinical characteristics in the training set; (B) Forest plot showing the results of univariate COX regression analysis of EPSS and clinical characteristics in the validation set; (C) Forest plot showing the results of multivariate COX regression analysis of EPSS and clinical characteristics in the training set; (D) Forest plot showing the results of multivariate COX regression analysis of EPSS and clinical characteristics in the validation set; (F) Time-dependent ROC curves showing the predictive power of EPSS and clinical features in the training set; (F) Time-dependent ROC curves showing the predictive power of EPSS and clinical features in the training set; It can be seen that the predictive power of each variable varies over time.

Figure 5

A nomogram was constructed combining EPSS with clinical features and the predictive power of the nomogram was assessed. (A) The nomogram of the overall survival of patients with osteosarcoma shows that EPSS score and site are the two most important variables; (B) Calibration curves for nomogram predicting 36-Month and 60-Month survival of patients in the training set; (C) Calibration curves for nomogram predicting 36-Month and 60-Month survival of patients in the validation set.

Figure 6

The relationship between EPSS and clinical characteristics were assessed. (A) The relationship between EPSS and gender; (B) The relationship between EPSS and Tumor-site; (C) The relationship between EPSS and pathological fracture.