Risk factors analysis of lung cancer combined with venous thromboembolism and the construction of a clinic prediction model

Abstract Objective The purpose of this study is to explore the disparities among lung cancer patients who also experience venous thromboembolism (VTE) and to establish a predictive model utilizing patients' clinical data and standard laboratory indicators for accurate assessment of VTE risk. Method A retrospective analysis was conducted on the clinical records of lung cancer patients at the First Affiliated Hospital of Ningbo University from January 2018 to December 2023. The patients were randomly divided into a training set and a validation set in a 7:3 ratio. 27 clinical parameters were chosen. Independent risk factors were selected by lasso regression and multivariate logistic regression. A nomogram was constructed for all variables showing significance at p < 0.05 in the multiple variable logistic analysis, and it was internally validated. Result A total of 300 lung cancer cases were examined, with 64 cases of VTE and 236 cases without VTE. Following screening by Lasso regression and multiple logistic regression, 6 variables were determined to be significant for the final model, including metastasis, surgery, chemotherapy, targeted therapy, hemoglobin (HB), and platelet count (PLT). The modeling cohort (AUC 0.804) and the validation cohort (AUC 0.799) indicated strong discrimination. Calibration curve and decision curve analysis (DCA) demonstrated favorable consistency and clinical usefulness of the model. Conclusion The developed nomogram in this study has the capacity to predict the likelihood of VTE events in lung cancer patients to a certain degree, thus assisting healthcare providers in formulating appropriate prevention and treatment strategies.