Applying Cox proportional hazards model to identifying the factors affecting the survival of patients with brain metastases

Abstract

Abstract Background : The aim of this study is to determine of risk factors in patients with brain metastases, and prognostic factors affecting survival of patients by using the Cox proportional hazards model. Methods : This descriptive - analytic retrospective study was performed on 197 patients with brain metastases who Referred to Shahid Ramezanzadeh Radiation Center, Yazd, Iran. Several risk factors in This patients were analyzed by the Cox model. Survival times were estimated by Kaplan - Meier and significant difference between the survival distributions of sub groups were compared with the log-rank test. The data were analyzed by R and SPSS softwares. Results : The Kaplan-Meier estimate of patients overall survival were 27% at 12 months and 12% at 24 months. by the log - rank test we founded that there was significant difference between the patients survival in subgroups of variables, such as: age at diagnosis, outside brain diseases (primary tumor), type of treatment, response to radiotherapy and re-radiation therapy (P<0.05). Cox model analysis indicated that the survival of patients with brain metastases influenced by the outside brain diseases (primary tumor) and response to radiotherapy variables. Conclusion : According to this result, control of the primary tumor and appropriate by radiation therapy seems to be effective to increase patients Lifetime. Also, it is more appropriate to use the Cox model in medical studies because of its easy for interpretation and not limited to use the type of covariates.

Keywords

• Cox proportional hazards model
• Kaplan-Meier
• survival
• brain metastasis