Development and Validation of an Individualized Model for Predicting COVID 19 Deterioration in Patients with Cancer: A Multicenter Retrospective Study

Background: The 2019 novel coronavirus has spread rapidly around the world. Cancer patients seem to be more susceptible to infection and disease deterioration, but the factors affecting the deterioration remain unclear. We aimed to develop and validate an individualized nomogram model for prediction of COVID-19 deterioration in cancer patients. Patients and Methods: The clinical data of 276 cancer patients who were primarily diagnosed with COVID-19 in 33 designated hospitals of Hubei, China from December 21, 2019 to March 18, 2020, were collected and randomly divided into a training and a validation cohort by a ratio of 2:1. The prediction model was developed in the training cohort. The LASSO regression analysis was used to select the characteristic variables from CT image features and clinical symptoms. Cox stepwise regression analysis based on AIC-minimum principle were conducted to select prognostic factors. The predictive accuracy of the model was quantified by C-index and time-dependent AUC. The calibration plot was performed to compare the actual and predicted probability of 2-week, 4-week and 8-week deterioration-free survival of COVID-19 (C-DFS) rate. Decision curve analysis (DCA) were used to evaluate the clinical usefulness of the model. Internal validation was assessed by the validation cohort. Risk stratification based on the model was performed. Findings: Cancer type, symptoms (dyspnea and fatigue), CT baseline image features (ground glass opacity and consolidation) were significantly associated with symptomatic deterioration in COVID-19 patients with cancer ( P 8·31) group. The Kaplan-Meier C-DFS curves presented the significant discrimination between the two risk groups in both training cohort ( P < 0·001) and validation cohort ( P = 0·034). Interpretation: This study presents an individualized nomogram model that incorporates cancer type, symptoms, CT baseline image features, and comorbidity, which can be used to individually predict the possibility of symptomatic deterioration of COVID-19 in patients with cancer. Funding Statement: National Natural Science Foundation of China (Grant No. 81670123) Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: The Ethics Committee of the National Cancer Center approved this study, and waived the informed consent due to the severity and rapid spreading of COVID-19.