High-sensitivity troponin T as a predictor of cardiovascular events in colorectal cancer patients undergoing chemotherapy: results of the COLOVASC prospective cohort study
Abstract
Background Fluoropyrimidine-based chemotherapy is associated with cardiovascular events in patients with colorectal cancer, yet validated tools for pre-treatment cardiovascular risk stratification are lacking. Cardiac biomarkers are not routinely incorporated into baseline assessment and prospective evidence supporting their use remains limited. Objectives Our aim was to determine whether baseline high-sensitivity troponin T (hs-TnT) provides prognostic information for major adverse cardiovascular events during chemotherapy in patients with colorectal cancer. Materials and methods Within a prospective single-center observational cohort study at the Department of Internal Medicine and Hematology, Semmelweis University, consecutive patients with histologically confirmed newly diagnosed or recurrent colorectal cancer scheduled to receive 5-fluorouracil (5-FU)-based chemotherapy were enrolled between September 1, 2020, and June 2, 2024. Baseline hs-TnT concentrations were measured before treatment initiation, and patients were followed for the occurrence of major adverse cardiovascular events (MACEs). Independent predictors of MACE were identified using penalized multivariable logistic regression. Receiver operating characteristic (ROC) curve analysis evaluated the discriminatory performance of baseline hs-TnT and determined the optimal cut-off value using the Youden index. Adequacy index analysis assessed the relative prognostic contribution of hs-TnT to the multivariable model. Results 180 patients were enrolled, with a median follow-up of 185 days, 44 patients (24.4%) experienced MACE, more than half of them occurred within the first two cycles of chemotherapy. Baseline hs-TnT demonstrated modest discrimination for predicting MACE (AUC 0.628, 95% CI 0.513–0.742). The optimal hs-TnT cut-off was 14 ng/L. In a multivariable analysis, elevated baseline hs-TnT (OR 2.35 per +10 ng/L, 95% CI 1.37–4.05) and Vascular Endothelial Growth Factor (VEGF)-inhibitor therapy (OR 2.38, 95% CI 1.23–4.60) were associated with MACEs. In a sensitivity analysis, excluding treatment-related hypertension from the composite endpoints, elevated baseline hs-TnT remained the only independent predictor. According to the adequacy index analysis, hs-TnT contributed 38.3% of the total prognostic information provided by the multivariable model. Conclusions Baseline hs-TnT was independently associated with MACEs in patients with colorectal cancer receiving fluoropyrimidine-based chemotherapy. Although its discriminatory performance as a standalone biomarker was modest, hs-TnT provided incremental prognostic information for baseline cardiovascular risk assessment and may help identify patients who warrant closer cardiovascular surveillance.