1856P - Prediction of clinically significant bleeding in anticoagulated patients for cancer-associated venous thromboembolism: Validation of B-CAT score in a cohort from the TESEO study

Background

There is a lack of a validated scoring system for predicting clinically significant bleeding in patients anticoagulated for cancer-associated venous thromboembolism (Ca-VTE). The aim was to validate the B-CAT score, a new tool designed to assess bleeding risk in these patients.

Methods

Data came from the TESEO study, a national, multicenter and prospective registry that documents patients with Ca-VTE. We included patients anticoagulated for Ca-VTE and monitored them over 180 days for major or clinically relevant bleeding events. The variables of B-CAT score were selected: tumor location, metastasis, history of major or clinically relevant bleeding, anaemia, coagulopathies, and cerebrovascular and gastrointestinal disease. Data on minor trauma and minor surgery and clinically relevant bleeding without hospitalization after Ca-VTE could not be included as these were not available. Patients were stratified into three categories of bleeding risk, and a multivariate logistic regression model was developed using these variables to estimate bleeding risk.

Results

The study comprised 2,301 anticoagulated Ca-VTE patients. Over an equivalent period of 848 person-years, we identified 157 significant bleeding events (6.8%; 18.5 per 100 person-years), including 63 major (40.1%; 7.4 per 100 person-years) and 94 clinically relevant bleeding events (59.9%, 11.1 per 100 person-years). Patients classified as low (47.8%), medium (50.5%), and high (1.7%) risk according to the B-CAT score demonstrated different 6-month significant bleeding rates: 11.4, 24.4, and 100 per 100 person-years, respectively (p<0.001). The predictive model exhibited satisfactory calibration (Hosmer-Lemeshow test: p=0.886) and discrimination, as evidenced by C-statistic indices for significant bleeding, major bleeding, and clinically relevant bleeding of 0.63 (95% confidence interval: 0.58-0.67), 0.61 (0.53-0.69), and 0.63 (0.57-0.69), respectively.

Conclusions

We have validated the bleeding risk score B-CAT in patients with Ca-VTE receiving anticoagulation. This model has the potential to standardize decision-making in situations where there is limited robust evidence.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

M. Sanchez Canovas: Financial Interests, Personal, Speaker, Consultant, Advisor: Leo Pharma, Sanofi, Lundbeck, Angelini. I. Fernandez Perez: Financial Interests, Personal, Invited Speaker, training session: Merck; Financial Interests, Personal, Invited Speaker, training session: GSK. E.M. Brozos Vazquez: Financial Interests, Personal, Advisory Board: Pfizer, Servier, Bayer; Financial Interests, Personal, Invited Speaker: Servier, Amgen, LEO Pharma, Rovi, Merck, Roche, AstraZeneca, Kiowa Kirin, BMS, MSD. T. Quintanar Verduguez: Financial Interests, Personal, Advisory Board: Pfizer; Financial Interests, Personal, Invited Speaker: MSD oncology, Daiichi Sankyo, Lilly, Novartis; Financial Interests, Institutional, Local PI, pi clinical trial: AstraZeneca. S. Garcia-Adrian: Financial Interests, Personal, Speaker’s Bureau: AstraZeneca, Novartis, GSK, Pierre Fabre, Sanofi, Pfizer, Eisai. A.J. Munoz Martin: Financial Interests, Personal, Advisory Board: GSK, AstraZeneca, Sanofi, Celgene, Servier, MSD, Pfizer, Leo Pharma, Roche; Financial Interests, Personal, Invited Speaker: Lilly, Rovi, STADA, Menarini, BMS; Financial Interests, Institutional, Advisory Board, VTE risk assessment model: Genincode; Financial Interests, Institutional, Local PI: Celgene. All other authors have declared no conflicts of interest.