Biomarkers of Thrombosis as Predictors of Venous Thromboembolism Risk in Cancer Patients

Overview

The main venous thromboembolism (VTE) risk prediction model for ambulatory cancer patients is Khorana. Cancer thrombosis is associated with elevated thrombin generation. Its quantification is a promising method for evaluating patient's thrombotic profile. This study aims to develop a predictive model of VTE risk in ambulatory cancer patients, combining thrombosis biomarkers (D-dimers and thrombin generation potential) with the Khorana score. This is a prospective observational study that includes newly diagnosed cancer patients proposed for anti-tumor treatment (chemotherapy, immunotherapy or targeted therapies). Patients with disease progression are allowed if chemotherapy-free for 3 months. A 6-month mean incidence of VTE 6-10% is expected, requiring a sample size of 600 patients. Blood sample is collected at inclusion to analyze thrombosis biomarkers and blood count. The primary endpoint is the occurrence of symptomatic or incidental VTE within 6 months of inclusion. Models will follow a logistic approach with K-fold cross-validation (k=10). Model quality will be assessed with Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). Decision for entering predictors in multivariate models will be based on p \<.10 in the univariate analysis.

The main aims will be the following: * Evaluate the utility of the combination of thrombosis biomarkers (D-dimers and thrombin generation potential) with the Khorana score in order to stratify VTE risk in ambulatory cancer patients; * Determine the potential of this new score in the stratification of cancer patients into high- and low-risk VTE groups, in order to identify patients who would benefit from primary thromboprophylaxis; * Determine the applicability of the thrombin generation test as an independent factor in the stratification of VTE risk in the cancer population under study; * Determine the predictive value of D-dimers in the cancer population under study (high versus low risk discrimination).

Conditions