Sex differences in prognosis factors in patients with lung cancer: A nationwide retrospective cohort study in Korea.

Abstract

Large-scale studies elucidating sex differences in factors impacting prognosis and sex-specific prognossis factors scoring in patients with lung cancer are insufficient. The present study aimed to develop a model to predict sex-specific prognosis factors in Korean patients with lung cancer. This nationwide cohort study included 96,255 patients aged ≥19 years diagnosed with lung cancer and underwent Korean National Health Insurance Service health examinations between January 1, 2005 and December 31, 2015 and followed until 2020. Factors associated with prognosis were estimated using multivariable Cox proportional hazards regression analyses, and separate prognosis scores were calculated for male and female patients. The sex-specific risk scoring models were validated with Kaplan-Meier survival curves and c-statistic. During a mean follow-up of 2.8 years, 60.5% of the patients died. In male patients with lung cancer, age ≥ 65 years (24 points) had the highest mortality risk score, followed by chemotherapy in combination with radiotherapy (16 points), chemotherapy (14 points), and radiotherapy (11 points). In female patients with lung cancer, chemotherapy in combination with radiotherapy (19 points) had the highest mortality risk score, followed by chemotherapy (16 points), age ≥ 65 years (13 points), and radiotherapy (13 points). The analysis of patients categorized into three risk groups based on risk scores revealed that the fatality rates within 5 years were 7%, 54%, and 89% in the low-, intermediate-, and high-risk groups for male patients and 3%, 46%, 85% in the low-, intermediate-, and high-risk groups for female patients, respectively. The c-statistic was 0.86 for male patients and 0.85 for female patients. The strongest fatality risk factors in lung cancer were age ≥ 65 years in male patients and chemotherapy in female patients. The present study developed sex-specific prognosis scoring models to predict fatality risk in patients with lung cancer.