Trastuzumab-induced cardiotoxicity: predictors and clinical outcomes

Introduction

Trastuzumab significantly improves survival in HER2-positive breast cancer, but it carries a risk of cardiotoxicity that remains a major clinical challenge. Identifying patients at high risk is essential to optimize monitoring and improve outcomes.

Aim

To identify predictors of cardiotoxicity and assess clinical outcomes in patients treated with trastuzumab.

Methods

This retrospective, single-center study included patients who initiated trastuzumab between 2022 and 2023. The HFA-ICOS risk score was calculated, and predictors and outcomes were analyzed using multivariable logistic regression, Cox models, and Kaplan–Meier survival estimates.

Results

A total of 137 patients (92% female) were included, with a median age of 38 years (IQR 36–54). 83% received 120 mg trastuzumab and 17% received 150 mg. The majority had breast cancer (119; 87%), and 40% had stage IV disease. Baseline median LVEF was 57% (IQR 53–64), and median GLS was −17% (IQR −16 to −18); one patient had LVEF ≤50%. At baseline, 44% had intermediate and 24% high/very high risk of cardiotoxicity according to the HFA-ICOS score.

During a median follow-up of 31 months (IQR 19–37), 9 patients (7%) developed cardiotoxicity: 8 (6%) with LVEF reduction and 1 (1%) with new-onset atrial fibrillation. Six patients discontinued trastuzumab, and 2 did not resume therapy. The mean time to LVEF decline was 6 ± 1.5 months, and all patients recovered during follow-up.

In multivariable analysis, GLS ≤ −18% was an independent predictor of cardiotoxicity after adjustment for age and trastuzumab dose (OR 14, 95% CI 2–147, p=0.02). Patients with high or very high HFA-ICOS scores showed a trend toward increased cardiotoxicity that did not reach statistical significance (p=0.056).

No cardiovascular hospitalizations occurred. There were 29 deaths (21%), all non-cardiovascular. There was no statistically significant association between cardiotoxicity and all-cause mortality. Advanced cancer stage independently predicted all-cause mortality (OR 4, 95% CI 1–12, p=0.007), and high/very high HFA-ICOS scores were associated with an increased risk of all-cause mortality (HR 3, 95% CI 1.2–5.5, p=0.012).

Conclusion

Trastuzumab-related cardiotoxicity occurred in 7% of patients, mainly due to LVEF reduction, and all cases were reversible. Baseline GLS was the strongest predictor of cardiotoxicity, underscoring its value as a pre-treatment assessment tool. All deaths were non-cardiovascular, primarily due to disease progression, highlighting the importance of early risk stratification to optimize monitoring and avoid unnecessary interruptions of cancer therapy.For image description, please refer to the figure legend and surrounding text.