Abstract Number: OC 12.3
Meeting: ISTH 2021 Congress
Background: Bleeding risk is highly relevant when deciding on type and duration of anticoagulation for cancer-associated thrombosis (CAT). Several bleeding risk scores already exist, but have never been validated in the subpopulation of patients with CAT and are not recommended for practice.
Aims: To compare the following methods of estimating clinically relevant (major and clinically relevant non-major) bleeding risk in patients with CAT: 1) existing risk scores for bleeding in venous thromboembolism, 2) pragmatic classification based on cancer type, 3) newly derived prediction model.
Methods: In a post-hoc analysis of Hokusai VTE Cancer, a randomized trial comparing edoxaban with dalteparin for treatment of CAT, 7 bleeding risk scores were externally validated (ACCP-VTE, HAS-BLED, Hokusai, Kuijer, Martinez, RIETE, VTE-BLEED). Further, the predictive performance of these existing risk scores was compared with a pragmatic classification based on cancer type (i.e. gastrointestinal, genitourinary or other) and a newly derived, competing risk-adjusted model based on clinical predictors that predicts clinically relevant bleeding within 6 months after CAT diagnosis with non-bleeding related mortality as competing event (‘CAT-BLEED’).
Results: Data of 1046 patients (149 outcome events) were analyzed. Predictive performance of existing scores was poor to moderate, with a limited range of predicted risks, C-statistics of 0.50-0.57 and poor calibration. The pragmatic classification performed slightly better (C-statistic 0.61 (95%CI 0.56-0.66)). Internal validation of the new model ‘CAT-BLEED’ showed moderate discrimination and good calibration (Figure 1).
|Predictors||Subdistribution Hazard Ratios (95% Confidence Interval)|
|Genitourinary cancer||2.48 (1.14-5.38)|
|Gastrointestinal cancer * edoxaban treatment||2.20 (1.07-4.53)|
|Recent use of anticancer therapies associated with gastrointestinal toxicity (<4 weeks)||1.74 (1.03-2.92)|
|Regionally advanced or metastatic cancer||1.21 (0.82-1.80)|
|Age >75 years||1.02 (0.98-1.08)|
|Creatinine clearance (mL/min)||1.00 (0.99-1.00)|
Conclusions: Existing risk scores for bleeding during anticoagulation in patients with VTE perform insufficiently in patients with CAT. The optimal-fit ‘CAT-BLEED’ model performs better, but external validation in practice-based settings and with other DOACs is warranted to assess its clinical value. In the meantime, we suggest a pragmatic distinction based on type of cancer and other well-established risk factors (e.g. history of bleeding, severe thrombocytopenia or anemia).
To cite this abstract in AMA style:de Winter MA, Dorresteijn JAN, Ageno W, Ay C, Beyer Westendorf J, Coppens M, Klok FA, Moustafa F, Riva N, C Ruiz Artacho P, Vanassche T, Nijkeuter M. Estimating Bleeding Risk in Patients with Cancer-associated Thrombosis: External Validation of Existing Risk Scores and Development of a New Risk Score [abstract]. Res Pract Thromb Haemost. 2021; 5 (Suppl 1). https://abstracts.isth.org/abstract/estimating-bleeding-risk-in-patients-with-cancer-associated-thrombosis-external-validation-of-existing-risk-scores-and-development-of-a-new-risk-score/. Accessed September 25, 2021.
« Back to ISTH 2021 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/estimating-bleeding-risk-in-patients-with-cancer-associated-thrombosis-external-validation-of-existing-risk-scores-and-development-of-a-new-risk-score/