Risk Stratification for Trastuzumab-Induced Cardiac Dysfunction and Potential Implications for Surveillance

Abstract

Background: Although patient factors and sequential anthracycline use contribute to risk for cancer therapy-related cardiac dysfunction (CTRCD) with HER2-directed cancer therapy, frequent (every 3 months) left ventricular ejection fraction (LVEF) surveillance is recommended irrespective of baseline risk.

Objectives: The aim of this study was to examine the incidence of trastuzumab-associated CTRCD in a contemporary cohort with HER2-positive breast cancer and assess the performance of a risk assessment tool to identify patients at low risk for CTRCD to guide risk-based surveillance strategies.

Methods: A retrospective cohort of patients with HER2-positive breast cancer treated with trastuzumab at a tertiary cancer center was examined. Patients were categorized as low, medium, and high or very high risk for CTRCD by Heart Failure Association/International Cardio-Oncology Society risk assessment.

Results: Of 496 patients treated with trastuzumab, 29.8% also received anthracyclines. Over a median follow-up period of 51 months, 8.7% developed CTRCD, but only 1.6% had associated heart failure (HF). CTRCD rates were 3.6%, 12.8%, and 32.1% in low-risk, medium-risk, and high or very high risk groups, respectively. HF incidence was 0.4% in the low-risk group and 2.1% in the medium-risk group, with no HF in patients at low- or medium-risk who received trastuzumab without anthracyclines. HF was observed in 11% of high-risk patients. The risk assessment had a negative predictive value for CTRCD in low vs moderate- or high-risk patients of 96.4% (95% CI: 93.5%-98.3%).

Conclusions: The findings support the exploration of a prospective personalized risk-based approach to cardiac LVEF surveillance during trastuzumab therapy. Less frequent LVEF monitoring in low-risk patients may optimize resource use and reduce patient burden without compromising safety.

Keywords: HER2-targeted therapy; anthracycline; breast cancer; cancer therapy–related cardiac dysfunction; cardiomyopathy; echocardiography; heart failure; imaging; risk prediction; screening; surveillance; trastuzumab.

Graphical abstract

Figure 1

Flowchart of the Study Cohort Overview of the study cohort selection within the time period of the first baseline echocardiographic examination on January 18, 2013, to the last follow-up examination on December 29, 2022. MDA = MD Anderson Cancer Center.

Figure 2

Cumulative Incidence of CTRCD After Trastuzumab Initiation in Patients With and Those Without Anthracylines The incidence curves exclude patients who reinitiated trastuzumab at a later date. CTRCD = cancer therapy–related cardiac dysfunction; N = no; Y = yes.

Figure 3

Incidence of CTRCD and Symptomatic HF by Baseline Risk Groups Patients were stratified into risk groups using the European Society of Cardiology Heart Failure Association/International Cardio-Oncology Society risk tool and further subdivided within each risk category by those who received anthracyclines and sequential trastuzumab vs trastuzumab only. CTRCD = cancer therapy–related cardiac dysfunction; HF = heart failure.

Figure 4

Discrimination and Calibration of the Heart Failure Association/International Cardio-Oncology Society Risk Tool for Development of Trastuzumab-Related Cancer Therapy–Related Cardiac Dysfunction Receiver-operating characteristic curves with their areas under the curve (AUCs) (95% CIs) and the calibration plots are shown for the overall cohort (A, C) and for patients who received trastuzumab without anthracyclines (B, D). In C and D, the gray line indicates perfect calibration. The blue dots represent the observed risk for cancer therapy–related cardiac dysfunction for each risk group, while the blue bars represent the 95% CIs of the observed risks. The calibration plots indicate that the low-risk group had the lowest observed risk, while the high- and very high-risk group had the highest observed risk.

Figure 5

Changes in LVEF in Patients Treated With Trastuzumab With Incident CTRCD by Use of Anthracyclines Left ventricular ejection fraction (LVEF) is shown at baseline, time of cancer therapy–related cardiac dysfunction (CTRCD) diagnosis, LVEF nadir, and last follow-up visit. The median nadir LVEF for trastuzumab only was 45% (Q1-Q3: 38%-50%) and for anthracycline plus trastuzumab was 47% (Q1-Q3: 29%-51%).

Figure 6

Changes in LVEF in Patients Treated With Trastuzumab With Incident CTRCD by Baseline Risk LVEF is shown at baseline, time of CTRCD diagnosis, LVEF nadir, and last follow-up visit. Patients in the baseline low-risk group who developed CTRCD had lesser reductions in LVEF (almost all ≥40%) compared with those in the medium- and high-risk groups. Abbreviations as in Figure 5.

Central Illustration

Risk Stratification for Patients Receiving HER2-Targeted Therapy to Guide Personalized Risk-Based LVEF Surveillance In a contemporary cohort of patients with HER2-positive breast cancer treated with trastuzumab, the overall incidence of cancer therapy–related cardiac dysfunction (CTRCD) was 8.7%, with a low rate of symptomatic heart failure (HF) of 1.6%. Among patients categorized as low risk for CTRCD at baseline according to the European Society of Cardiology (ESC) Heart Failure Association (HFA)/International Cardio-Oncology Society (IC-OS) risk score, the incidence rates of CTRCD and symptomatic HF were very low at 3.6% and 0.4%, respectively, while they were much higher in those classified as high or very high risk (32% and 11%, respectively). This provides support to consider personalized risk-based left ventricular ejection fraction (LVEF) surveillance during treatment decreasing the frequency of LVEF monitoring in low-risk patients. sHR = subdistribution HR.