BackgroundTrastuzumab deruxtecan (T-DXd) is an effective HER2-directed antibody-drug conjugate used across multiple solid tumors. Interstitial lung disease (ILD) and pneumonitis are well-recognized as clinically significant toxicities associated with T-DXd. However, most real-world studies have focused on signal detection rather than the clinical context in which ILD is reported. The patterns by which ILD co-occurs with other respiratory adverse events in routine clinical practice remain poorly characterized.MethodsWe analyzed adverse event reports submitted to the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) between 2015 and 2025 in which T-DXd was listed as a suspect drug. ILD, pneumonitis, and respiratory adverse events were identified using MedDRA Preferred Terms. A binary case-event matrix was constructed, and hierarchical clustering was used to explore patterns of respiratory adverse event co-reporting. Subgroup analyses were conducted for breast, gastric/gastroesophageal junction (GEJ), and lung cancer. Cluster stability was assessed using bootstrap resampling.ResultsA total of 759 ILD cases and 289 pneumonitis cases associated with T-DXd were identified. In breast cancer reports, ILD consistently clustered with dyspnea and cough, forming a concentrated respiratory co-reporting phenotype. In gastric/GEJ cancer reports, ILD showed strong co-reporting with pneumonia, suggesting overlap between drug-related lung injury and infectious respiratory presentations. Lung cancer reports demonstrated more heterogeneous respiratory co-reporting patterns, with less consistent clustering of ILD with specific symptoms. Network analysis identified ILD as a central node within the respiratory adverse event network. Across all tumor types, respiratory clusters were reproducible in bootstrap analyses, indicating stable and non-random co-reporting structures.ConclusionsILD associated with T-DXd is reported alongside distinct respiratory adverse event patterns that differ across tumor types. Examining these co-reported respiratory events may help clinicians better interpret ILD reports in real-world practice, particularly in the setting of overlapping or non-specific pulmonary symptoms.
Evinç et al. (Thu,) studied this question.