Incremental value of TAPSE added to HFA-ICOS for ESC-Defined Cancer Therapy–Related Cardiac Dysfunction

Background

The 2022 ESC cardio-oncology guidelines recommend the HFA-ICOS score for baseline cardiovascular risk stratification.1 However, data on its performance in contemporary practice and the potential contribution of right ventricular function remain limited.2,3 We aimed to evaluate the incremental value of TAPSE for discrimination of ESC-defined Cancer Therapy-Related Cardiac Dysfunction (CTRCD).

Methods

We conducted an observational study of consecutively evaluated patients at a cardio-oncology unit in 2024. Among 1,015 screened patients, 269 met inclusion criteria (Complete clinical and echocardiographic records).

Patients were categorized as HFA-ICOS risk at consultation. Echocardiographic parameters included left ventricular ejection fraction (LVEF), global longitudinal strain (GLS) and Tricuspid Annular Plane Systolic Excursion (TAPSE). CTRCD was defined per 2022 ESC guidelines.

Associations with CTRCD were modeled using Firth penalised logistic regression due to low event frequency. Three models were evaluated: TAPSE alone, HFA-ICOS alone, and HFA-ICOS combined with TAPSE. Incremental value was assessed using penalised likelihood penalised likelihood ratio testing. Discrimination was quantified using receiver operating characteristic (ROC) curves and area under the curve (AUC) with 95% confidence intervals derived from 2,000 stratified bootstrap samples. AUCs were compared using paired ROC tests.

Results

Among 269 patients, 60% were classified as low, 19% moderate and 19% high HFA-ICOS risk. Increasing risk category was associated with older age (p<0.001), higher prevalence hypertension (p<0.001) and diabetes (p = 0.028), and greater exposure to combination anthracycline-trastuzumab therapy (p<0.001).

LVEF progressively declined across risk strata (p<0.001), and GLS was impaired in high-risk patients (p<0.001). In contrast, TAPSE did not significantly differ across HFA-ICOS categories (p = 0.095). ESC-defined CTRCD was more frequent in the high risk group ( p<0.001) (Figure 1).

CTRCD was identified in 8 patients. TAPSE alone demonstrated moderate discrimination (AUC 0.79, 95% CI 0.57 - 0.96), while HFA-ICOS showed good discrimination (AUC 0.84, 95% CI 0.68 - 0.93). The combined model (HFA-ICOS + TAPSE) showed the highest discrimination (AUC 0.94, 95% CI 0.88 - 0.98). Pairwise AUC comparisons were not statistically significant. (Figure 2) However, addition of TAPSE significantly improved model fit beyond HFA-ICOS alone (penalised likelihood ratio χ²=6.05, p = 0.014).

Conclusion

Addition of TAPSE to HFA-ICOS improved discrimination for ESC-defined CTRCD, suggesting that right ventricular systolic assessment may refine cardiovascular risk stratification in cardio-oncology practiceFigure 1:Sociodemographic tableFigure 2.Discriminatory performance of TAPSE