Predictability of cardiotoxicity: Experience of a Belgian cardio-oncology clinic.

Background: New oncological treatments improved survival but also increased awareness of cardiovascular side-effects during and after cancer therapy.

Methods: We report the experience of the cardio-oncology clinic at a large Belgian tertiary care center and investigated the predictability of cardiotoxicity based on referring department, cardiovascular risk factors, cancer treatment and existing risk scores of the American Society of Clinical Oncologists (ASCO) and Mayo Clinic. Cardiotoxicity was defined as a 10% reduction in Left Ventricular Ejection Fraction (LVEF) compared to the baseline transthoracic echocardiography (TTE) in asymptomatic patients or 5% in symptomatic patients.

Results: Of the 324 patients included, 14.5% died during follow-up. Most deaths were oncological, yet 19% of deaths were attributable to cardiovascular diseases. Models based on cardiovascular risk factors alone and cardiovascular risk factors combined with cardiotoxic medication poorly predicted cardiotoxicity. Existing risk scores from ASCO and Mayo Clinic also poorly predicted cardiotoxicity. A weighed model based on the Mayo Clinic cardiotoxicity risk score was the best risk assessment tool with still a limited predictive value with an Area Under the Receiver Operating Characteristic curve of 0.654 (CI 95%: 0.601-0.715).

Conclusion: Cardiovascular morbidity and mortality are common in cancer patients and survivors and stress the unmet need of adequate risk prediction tools for systematic screening and rigorous cardiovascular follow-up. In our outpatient cohort, cardiotoxicity risk could not be adequately predicted by cancer type, using classic cardiovascular risk factors, nor by the combination of cardiovascular risk factors and the proposed cancer treatment. Furthermore, we showed that existing cardiotoxicity risk scores are suboptimal and should thus be interpreted with caution.