Developing a machine learning model for bleeding prediction in patients with cancer-associated thrombosis receiving anticoagulation therapy.

BACKGROUND: Only 1 conventional score is available for assessing bleeding risk in patients with cancer-associated thrombosis (CAT): the CAT-BLEED score. OBJECTIVES: Our aim was to develop a machine learning-based risk assessment model for predicting bleeding in CAT and to evaluate its predictive performance in comparison to that of the CAT-BLEED score. METHODS: We collected 488 attributes (clinical data, biochemistry, and International Classification of Diseases, 10th Revision, diagnosis) in 1080 unique patients with CAT. We compared CAT-BLEED score, Ridge and Lasso logistic regression, random forest, and Extreme Gradient Boosting (XGBoost) algorithms for predicting major bleeding or clinically relevant nonmajor bleeding occurring 1 to 90 days, 1 to 365 days, and 90 to 455 days after venous thromboembolism (VTE). RESULTS: The predictive performances of Lasso logistic regression, random forest, and XGBoost were higher than that of the CAT-BLEED score in the prediction of bleeding occurring 1 to 90 days and 1 to 365 days after VTE. For predicting major bleeding or clinically relevant nonmajor bleeding 1 to 90 days after VTE, the CAT-BLEED score achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.48 ± 0.13, while Lasso logistic regression and XGBoost both achieved AUROCs of 0.64 ± 0.12. For predicting bleeding 1 to 365 days after VTE, the CAT-BLEED score achieved a mean AUROC of 0.47 ± 0.08, while Lasso logistic regression and XGBoost achieved AUROCs of 0.64 ± 0.08 and 0.59 ± 0.08, respectively. CONCLUSION: This is the first machine learning-based risk model for bleeding prediction in patients with CAT receiving anticoagulation therapy. Its predictive performance was higher than that of the conventional CAT-BLEED score. With further development, this novel algorithm might enable clinicians to perform personalized anticoagulation strategies with improved clinical outcomes.

Measuring Quality of Life after Venous Thromboembolism: Who, When, and How?

There is a growing body of evidence revealing that many patients with a history of venous thromboembolism (VTE) suffer from long-lasting sequelae such as post-thrombotic syndrome and post-pulmonary embolism syndrome. These two syndromes are detrimental to patients as they affect their quality of life (QOL). From this perspective, monitoring QOL may play a crucial role to improve quality care in VTE patients. Many studies have explored possible temporal relations between VTE episodes and decreased functional status and/or QOL. However, studies exploring the implementation of QOL and functional status questionnaires in clinical practice are scarce. In this context, we discuss possible perspectives synthetized from available literature regarding in whom, when, and how QOL could be measured in clinical practice.

Health-related quality-of-life questionnaires for deep vein thrombosis and pulmonary embolism: A systematic review on questionnaire development and methodology.

To improve the quality and accuracy of the patient-reported outcome measures that assess health-related quality of life (HRQoL), guidelines have been developed to standardize the development and validation process. Considering the increasing importance of HRQoL questionnaires in research, we set out to review the literature and evaluate whether existing questionnaires developed for deep vein thrombosis (DVT) and pulmonary embolism (PE) fulfill state-of-the-art requirements. The literature search was conducted in March 2019 and updated in September 2020. Seven databases were searched. No time limit was set for the search to include all available questionnaires. The inclusion criteria were original publications describing the development of disease-specific HRQoL questionnaires specific to DVT or PE in adults and available in English. The questionnaires were assessed to determine whether they fulfill the requirements in the latest guidelines. A total of 3826 references were identified. After the exclusion process, 15 papers were reviewed in full, of which 7 were included. Four questionnaires were developed for chronic venous disease, two were specific to DVT, and one was specific to PE. Most questionnaires we found in this review, fulfilled some but none fulfilled all recommendations in existing guidelines. Because the development of current available HRQoL questionnaires specific to DVT or PE do not fulfil all recommendations of existing guidelines, there is room for improvements within this field. Such improvements could likely enhance the quality associated with the use of these end points in clinical trials and practice.