Machine Learning-Based Mortality Prediction Model for Critically Ill Cancer Patients Admitted to the Intensive Care Unit (CanICU)

Simple Summary This study describes a new machine-learning-based 28-day mortality prediction model in adult cancer patients admitted to the intensive care unit (ICU). A total of 6900 patients in three patient cohorts were used for the development, internal validation, and external validation, respectively, leading to the generation of a reliable model with high sensitivity and specificity. The CanICU uses nine variables that can be easily obtained in a practical ICU, with the potential benefit of critical care and avoiding unnecessary suffering. Furthermore, this is the largest patient cohort for developing a cancer patient-specific model. CanICU offers improved performance for predicting short- and long-term mortality in critically ill cancer patients admitted to the ICU. CanICU can help physicians determine how to allocate ICU care for patients with cancer according to objective mortality risk. Background: Although cancer patients are increasingly admitted to the intensive care unit (ICU) for cancer- or treatment-related complications, improved mortality prediction remains a big challenge. This study describes a new ML-based mortality prediction model for critically ill cancer patients admitted to ICU. Patients and Methods: We developed CanICU, a machine learning-based 28-day mortality prediction model for adult cancer patients admitted to ICU from Medical Information Mart for Intensive Care (MIMIC) database in the USA (n = 766), Yonsei Cancer Center (YCC, n = 3571), and Samsung Medical Center in Korea (SMC, n = 2563) from 2 January 2008 to 31 December 2017. The accuracy of CanICU was measured using sensitivity, specificity, and area under the receiver operating curve (AUROC). Results: A total of 6900 patients were included, with a 28-day mortality of 10.2%/12.7%/36.6% and a 1-year mortality of 30.0%/36.6%/58.5% in the YCC, SMC, and MIMIC-III cohort. Nine clinical and laboratory factors were used to construct the classifier using a random forest machine-learning algorithm. CanICU had 96% sensitivity/73% specificity with the area under the receiver operating characteristic (AUROC) of 0.94 for 28-day, showing better performance than current prognostic models, including the Acute Physiology and Chronic Health Evaluation (APACHE) or Sequential Organ Failure Assessment (SOFA) score. Application of CanICU in two external data sets across the countries yielded 79-89% sensitivity, 58-59% specificity, and 0.75-0.78 AUROC for 28-day mortality. The CanICU score was also correlated with one-year mortality with 88-93% specificity. Conclusion: CanICU offers improved performance for predicting mortality in critically ill cancer patients admitted to ICU. A user-friendly online implementation is available and should be valuable for better mortality risk stratification to allocate ICU care for cancer patients.