Background: Antibody-drug conjugates (ADCs) are targeted cancer therapies that combine the specificity of monoclonal antibodies with the cytotoxic effects of chemotherapy. Trastuzumab, one of the earliest approved ADCs for breast cancer, is known to cause cardiotoxicity, however, significant cardiac adverse events (CAEs) have not been reported with the recently approved ADCs. We aim to explore the post-marketing reports of CAEs with these medications.

Methods:

  • Drugs of interest: We analyzed gemtuzumab, inotuzumab, polatuzumab, belantamab, and loncastuximab.

  • Data Source and Extraction: Data was obtained from the publicly available FDA Adverse Event Reporting System (FAERS). Quarterly data files containing adverse events (AEs), patient demographic information, and implicated drugs were downloaded. Duplicate reports were removed using key identifiers to ensure unique data values. Cases attributable to each drug were identified by conducting keyword searches for the drug's generic and brand name. For our analysis, only cases reported by healthcare providers and involving patients aged 18 and older were included. CAEs for each drug were queried from the date of drug approval through the last quarter of 2023. Cases associated with two or more drugs of interest were excluded from the analysis to eliminate possible confounding.

  • CAEs of interest: CAEs were flagged if the report fell under the system organ class (SOC) of cardiac disorders (SOC: 10007541) as defined by the Medical Dictionary for Regulatory Activities (MedDRA). Each CAE associated with ADCs of interest required at least three associated events.

  • Data Analysis: We used four metrics in our study: reporting odds ratio (ROR), proportional reporting ratio (PRR), information component (IC), and Empirical Bayes Geometric Mean (EGBM), all derived from a 2x2 contingency table. An adverse effect was considered statistically significant if all four metrics were positive. We used Fisher's Exact Test to assess statistical associations at two-tailed threshold of p <0.05. All data analysis was conducted using R version 4.4.1.

Results:

Descriptive Analysis:

  • CAEs constituted 3.2% (N = 46) of total AEs for gemtuzumab, 2.0% (N = 122) for polatuzumab, 0.96% (N = 12) for inotuzumab, and 1.3% (N = 32) for belantamab, all with sufficient sample sizes for further analysis. For loncastuximab, CAEs made up 3.0% (N = 5) of total AEs, but the sample size was too small for disproportionality analysis, resulting in its exclusion.

  • The most common indication for gemtuzumab in patients with CAEs was acute myeloid leukemia (N = 26, 76%). For inotuzumab, it was acute lymphocytic leukemia (N = 9, 90%). Polatuzumab (N = 73, 61%) and loncastuximab (N = 4, 80%) were primarily used for diffuse large B-cell lymphoma (DLBCL) in patients with CAEs. The main indication for belantamab in patients with CAEs was plasma cell myeloma (N = 24, 77%).

Disproportionality Analysis:

  • Cumulative CAEs associated with any of the five drugs were not disproportionately increased compared to non-cardiac AEs. However, individual CAEs showed positive signals with some of the ADCs.

  • For gemtuzumab, signals emerged in cardiotoxicity (ROR: 8.91, 3.99-19.89), cardiomyopathy (ROR: 7.73, 2.89-20.65), ventricular dysfunction (ROR: 78.42, 24.95-246.43), myocardial ischaemia (ROR: 19.39, 6.23-60.38), and tachycardia (ROR: 4.54, 1.46- 14.10).

  • Polatuzumab indicated signals in atrial tachycardia (ROR: 30.96, 16.40-58.48), ventricular tachycardia (ROR: 4.53, 1.88-10.92), and left ventricular failure (ROR: 10.42, 3.88-28.00). Cardiomyopathy did not meet our criteria for statistical significance as it only showed positive ROR (2.41, 1.00-5.80) and PRR values.

  • Inotuzumab was significantly associated with cardiotoxicity (ROR: 5.43, 2.43-12.09) across all four metrics. Only atrial fibrillation and acute myocardial infarction met sample size requirements for belantamab, but neither of these were statistically significant by our disproportionality analysis metrics.

Conclusion: Our analysis identifies some of the CAEs associated with newer ADCs based on the post-marketing reports from the FAERS database. Given the limited exploration in the trial data, there is a broader question of establishing causation between ADCs and CAEs. Findings of our analysis support the need for further post-marketing surveillance and potential inclusion of significant CAEs to the drug labels.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution