Integrating genetic and clinical data to predict lung cancer in patients with chronic obstructive pulmonary disease

Chronic obstructive pulmonary disease (COPD) is closely linked to lung cancer (LC) development. The aim of this study is to identify the genetic and clinical risk factors for LC risk in COPD, according to which the prediction model for LC in COPD was constructed.

This is a case-control study in which patientis with COPD + LC as the case group, patientis with only COPD as the control group, and patientis with only LC as the second control group. A panel of clinical variables including demographic, environmental and lifestyle factors were collected. A total of 20 single nucleotide polymorphisms (SNPs) were genotyped. The univariate analysis, candidate gene study and multivariate analysis were applied to identify the independent risk factors, as well as the prediction model was constructed. The ROC analysis was used to evaluate the predictive ability of the model.

A total of 503 patients were finally enrolled in this study, with 188 patients for COPD + LC group, 162 patients for COPD group and 153 patients for LC group. The univariate analysis of clincial data showed compared with the patients with COPD, the patients with COPD + LC tended to have significantly lower BMI, higher smoking pack-years, and higher prevalence of emphysema. The results of the candidate gene study showed the rs1489759 in HHIP and rs56113850 in CYP2A6 demonstrated significant differences between COPD and COPD + LC groups. By using multivariate logistic regression analysis, four variables including BMI, pack-years, emphysema and rs56113850 were identified as independent risk factors for LC in COPD and the prediction model integrating genetic and clinical data was constructed. The AUC of the prediction model for LC in COPD reached 0.712, and the AUC of the model for predicting LC in serious COPD reached up to 0.836.

The rs56113850 (risk allele C) in CYP2A6, decrease in BMI, increase in pack-years and emphysema presence were independent risk factors for LC in COPD. Integrating genetic and clinical data for predicting LC in COPD demonstrated favorable predictive performance.