Abstract
PURPOSE: This study was designed to develop a nomogram for predicting neutropenia caused by chemotherapy in non-small cell lung cancer (NSCLC) patients. PATIENTS AND METHODS: Information was collected from 376 patients between November 2017 and November 2020. The endpoint was chemotherapy-induced neutropenia (absolute neutrophil count <2×10(9)/L). Logistic regression was performed to appraise the prognostic value of each potential predictor. Risk predictors from the final predictive model were used to generate a nomogram. C-index and calibration curve as well as decision curve analysis (DCA) was applied to evaluate model performance. RESULTS: The multivariate regression model ultimately included three predictors: previous radiotherapy, the current cycle of chemotherapy and neutrophil counts before current chemotherapy. A nomogram was developed and displayed better discrimination (with C-index of 0.875 in the development group and 0.907 in the validation group). Favorable consistency was shown between predicted probability and observed probability in the calibration curves. DCA illustrated that when the threshold probability was 8%-90%, the predictive model provided a net benefit relative to the intervention-all or the intervention-none strategy, indicating that the nomogram had favorable potential clinical utility. CONCLUSION: This nomogram will be an available tool to quantify the risk of neutropenia after chemotherapy in patients who suffer from NSCLC and deserves further external validation.