Lung cancer risk and exposure to air pollution: a multicenter North China case-control study involving 14604 subjects

Abstract

Background: For North Chinese lung cancer patients, there is limited study on the distribution of air pollution and smoking related features based on analyses of large-scale, high-quality population datasets. The aim of the study was to fully analyze risk factors for 14604 Subjects.

Methods: Participants and controls were recruited in 11 cities of North China. Participants' basic information (sex, age, marital status, occupation, height, and weight), blood type, smoking history, alcohol consumption, history of lung-related diseases and family history of cancer were collected. PM2.5 concentration data for each year in each city of the study area from 2005 to 2018 were extracted based on geocoding of each person's residential address at the time of diagnosis. Demographic variables and risk factors were compared between cases and matched controls using a univariate conditional logistic regression model. Multivariate conditional logistic regression models were applied to estimate the odds ratio (OR) and 95% confidence interval (CI) for risk factors in univariate analysis. The nomogram model and the calibration curve were developed to predict lung cancer probability for the probability of lung cancer.

Results: There was a total of 14604 subjects, comprising 7124 lung cancer cases and 7480 healthy controls included in the study. Marital status of unmarried persons, people with a history of lung-related disease, corporate personnel and production /service personnel were protective factors for lung cancer. People younger than 50 years old, people who were smoking and quit smoking, people who had been drinking consistently, people with family history of cancer and PM2.5 exposure were proven to be a risk factor for lung cancer. The risk of lung cancer varied with sex, smoking status and air pollution. Consistent alcohol consumption, persistent smoking and smoking quit were risk factors for lung cancer in men. By smoking status, male was risk factor for lung cancer in never smokers. Consistent alcohol consumption added risk for lung cancer in never smokers. The combined effects of PM2.5 pollution exposure and ever smoking aggravated the incidence of lung cancer. According to air pollution, lung cancer risk factors are completely different in lightly and heavily polluted areas. In lightly polluted areas, a history of lung-related disease was a risk factor for lung cancer. In heavily polluted areas, male, consistent alcohol consumption, a family history of cancer, ever smokers and smoking quit were all risk factors for lung cancer. A nomogram was plotted and the results showed that PM2.5 was the main factor affecting the occurrence of lung cancer.

Conclusions: The large-scale accurate analysis of multiple risk factors in different air quality environments and various populations, provide clear directions and guidance for lung cancer prevention and precise treatment.

Keywords: Air pollution; Case–control; Lung cancer; Never-smokers; Nomogram model.

Fig. 1

Five-year average concentration distribution of PM2.5 in 11 prefectures and cities

Fig. 2

Multivariate risks of lung cancer by sex. A Multivariate risk factors model in males. B Multivariate risk factors model in females

Fig. 3

Multivariate risks of lung cancer by smoking. A Multivariate risk factors model in ever smokers. B Multivariate risk factors model in never smokers

Fig. 4

Multivariate risks of lung cancer by air pollution. A Multivariate risk factors model in lightly polluted areas. B Multivariate risk factors model in heavily polluted areas

Fig. 5

A A nomogram was used to predict the lung cancer probability B Validation of the nomogram—calibration plots for predicting lung cancer. In the marriage variable, 1 referred to unmarried and 2 referred to married. In smoking status, 0 represented never smoking, 1 represented being smoking, and 2 represented quitting smoking after smoking. In the drinking status, 0 represented not drinking, and 1 represented being drinking. In the family history of cancer, 0 represented having no family history of cancer, and 1 represented having a family history of cancer. In the lung related disease, 0 represented having no lung related disease, and 1 represented having lung related disease. In the occupation, 1 represented Corporate personnel, 2 represented Production/Service and 3 represented Agriculture/fishing