Kaplan-Meier estimator and Cox proportional hazard model

Abstract

Lung cancer is one of the leading causes of death worldwide, accounting for an estimated 2.1 million cases in 2018. To analyze the risk factors behind the lung cancer survival, this paper employs two main models: Kaplan-Meier estimator and Cox proportional hazard model [1]. Also, log-rank test and wald test are utilized to test whether a correlation exists or not, which is discussed in detail in later parts of the paper. The aim is to find out the most influential factors for the survival probability of lung cancer patients. To summarize the results, stage of cancer is always a significant factor for lung cancer survival, and time has to be taken into account when analyzing the survival rate of patients in our data sample, which is from TCGA. Future study on lung cancer is also required to make improvement for the treatment of lung cancer, as our data sample might not represent the overall condition of patients diagnosed with lung cancer; also, more appropriate and advanced models should be employed in order to reflect factors that can affect survival rate of patients with lung cancer in detail.
Keywords
Lung Cancer, Survival Analysis, Kaplan-Meier Estimator, Cox Proportional Hazard Model
Survival Analysis of Lung Cancer Patients from TCGA Cohort.pdf