A meta-analysis and systematic review of cardiac adverse events in patients with HER2-altered non–small cell lung cancer (NSCLC) enrolled in clinical trials.

Abstract

e24026 Background: HER2 inhibitors showed meaningful clinical activity in HER2-altered NSCLC. HER2 inhibitors-related cardiotoxicity remains inconsistently reported in NSCLC trials unlike in breast cancer trials. Cardiotoxicity can have a significant effect on patients’ morbidity and mortality. We conducted this meta-analysis to determine the incidence of cardiotoxicity with HER2-inhibitors in NSCLC. Methods: We performed a systematic search on MEDLINE, Embase, CENTRAL, Scopus, and Web of Science from inception to 10/10/25. Phase I-III clinical trials enrolling adults with HER2-altered NSCLC treated with HER2 inhibitors were included. The primary outcome was the pooled incidence of cardiotoxicity. Prespecified assumptions were applied for unreported cardiotoxicity outcomes, including conservative handling of missing data, zero-event imputation when appropriate, and classification of absent grade 5 events as no treatment-related cardiac deaths. Pooled incidence estimates were generated using random-effects meta-analyses of proportions with logit transformation and restricted maximum likelihood estimation in the R metafor package. In exploratory analyses, oncologic outcomes were compared between arms with and without reported grade ≥3 cardiotoxicity using nonparametric tests. Results: 48 arms from 39 trials were eligible, including 2035 patients. Most arms (79.2%) and patients (71.4%) were from phase II clinical trials. 19 (48.7%) trials reported cardiotoxicity. Patients were treated with tyrosine kinase inhibitors (TKIs) on 20 (42%) arms, monoclonal antibodies (mABs) on 15 (31%) arms, and antibody-drug conjugates (ADCs) on 15 (31%) arms. The incidence of severe grade cardiotoxicity was 5.1% (95CI, 2.4–10.3; I², 0%). No grade 5 cardiotoxicity was reported. The incidence of key cardiac events did not significantly differ by drug class. Pooled ORR was 34.6% (95CI, 29.3–40.3). ORR differed by drug class (p = 0.03), with the highest observed with ADC (44.5%, 95CI 35.0–54.4), followed by TKI (34.2%, 95CI 26.2–43.3), and mAB (30.3%, 95CI 24.3–37.0). Median OS was 13.8 (IQR 10.1–15.0) months. There were no significant differences in OS, PFS, and ORR between arms with versus without reported grade ≥3 cardiotoxicity. Conclusions: While not reported in all trials, the incidence of severe cardiotoxicity was rare in NSCLC patients treated with HER2-directed therapy. It was not associated with poor outcomes. Incidence of key cardiac adverse events. Cardiac adverse event arms, n Pooled incidence, % (95CI) I², % Cardiomyopathy 41 9.6 (3.8–14.1) 0 LVEF decline 42 9.2 (4.4–18.4) 39.1 Atrial fibrillation 41 5.5 (2.3–12.7) 0 Cardiorespiratory arrest 41 5.3 (0.7–29.7) 0 QTc prolongation 41 3.1 (5.7–10.4) 41.6 Acute heart failure 41 1.3 (4.1–12.0) 0 Abbreviations: 95CI, 95% confidence interval; LVEF, left ventricular ejection fraction.