PB1455 - Evaluating Predictive Tools for Thrombosis Risk Assessment in Breast Cancer- An Unmet Need

Background: Venous thromboembolism (VTE) is a common, and potentially fatal complication among cancer patients. Various VTE risk assessment tools have been developed to identify high-risk cancer patients who benefit from thromboprophylaxis, however, it has not been extensively studied in breast cancer (BC) populations.

Aims: To assess the validity of the Khorana (KS), PROTECHT (PT), COMPASS-CAT (CC), MDACC CAT (MD), and thromboelastographic (TEG) tools for predicting VTE in BC patients prior to the initiation of chemotherapy.

Methods: Female BC patients planned for chemotherapy were recruited between January 2022 and December 2023. Patients were followed for 12 months for VTE events. Coagulation profiles were analyzed via thromnolastography TEG 5000® parameters R, K, MA, Alpha and CI, defined as hypercoagulable based on Kaufmann’s classification. Each patient’s thrombosis risk was scored according to the KS, PT, CC, and MD scoring systems. Risk tool sensitivity and specificity were analyzed via ROC curves, and independent predictors were analyzed via binary logistical regression.

Results: 42 BC patients were enrolled prior to chemotherapy, of whom 5 (11.9%) developed VTE within the 12-months of the study. PT was indicated as the most important predictor of VTE in comparison to the other risk tools with excellent sensitivity (100%), specificity (66%), and significant odds ratio (1.33). Binary logistic regression of PT shows significance with (β=1.8 and p=0.02). While, KS, MD and CC showed comparatively worse performance with (p=0.04, 0.3 and 0.4) respectively. Moreover, PT obtained the greatest AUC of 0.857 while, KS, MD and CC showed AUC (0.692, 0.695 and 0.665) respectively (Fig 1). The predictive probability of the combination of TEG parameters in binary logistic regression analysis shows significant AUC= 0.839 (Fig. 2)

Conclusion(s): PT is an effective thrombosis risk assessment tool in breast cancer and thromboelastography has potential in predicting VTE. Larger studies are needed in order to begin moving this to clinical application.