Introduction: T-DXd is an antibody-drug conjugate consisting of an anti-HER2 antibody, cleavable tetrapeptide-based linker, and a membrane-permeable topoisomerase I inhibitor payload. In patients (pts) with HER2-expressing or -mutated solid tumors, T-DXd has demonstrated strong antitumor activity with a manageable safety profile and is approved for treatment (tx) of pts with HER2+ metastatic breast cancer (BC) that has progressed on ≥2 anti-HER2 therapies (US) or after chemotherapy (Japan); ILD is an important identified risk. We further characterize ILD in a pooled analysis of 8 single arm phase 1 and 2 T-DXd monotherapy studies.

Methods: Pts who received ≥1 dose of T-DXd monotherapy 5.4 mg/kg or higher across 8 single-arm studies were included (between Aug 28, 2015 and the data cutoff June 08, 2020). Potential ILD cases had imaging and clinical data (baseline through the time of ILD case) retrospectively reviewed by an independent adjudication committee to assess whether the reported ILD event was study drug related; events adjudicated as study drug related are reported.

Results: 879 pts were included; 510 with BC, 148 with lung cancer, 107 with colorectal cancer, 78 with gastric cancer, and 34 with other cancers (2 had missing tumor type). Median tx duration was 6.9 mo (range, 0.7-50.1 mo); 139 pts (15.8%) experienced an adjudicated ILD event, most were grade (G) 1/2 (G1/2=108 [12.3%]; G3=9 [1.0%]; G4=1 [0.1%]; G5=21 [2.4%]). Among pts treated with T-DXd 5.4 mg/kg (n=315), 48 (15.2%) experienced an ILD event (G1/2=39 [12.4%]; G3=2 [0.6%]; G4=1 [0.3%]; G5=6 [1.9%]). Additional details on ILD events by dose and tumor type will be presented. Among adjudicated ILD events (any grade), the onset date as assessed by investigator was delayed compared with the adjudication committee-assessed date in 71 cases (median difference in onset date, 49 d [range, 1-288 d]). The majority of ILD events occurred in the initial 12 mo of tx. After 12 mo the risk of new onset of ILD decreased with a conditional probability of developing ILD of 7.0% in pts treated to month 12 without ILD and 1.4% in pts treated to month 18 without ILD, which was likely driven by pts with longer tx duration in the BC cohort; however, additional follow-up is required to confirm these observations. Data on the use of steroids to manage ILD will be presented.

Conclusion: Pooled analysis showed that most ILD events were low grade (78% of pts with ILD had G1/2 events) and occurred in the first 12 mo of tx, with lowered risk for pts on tx > 12 mo. The adjudication committee identified ILD earlier than investigators in 48% of cases suggesting opportunity for early detection and intervention. Close monitoring, prompt recognition, and proactive management of ILD using current management guidelines may help to improve ILD outcomes.

Citation Format: Charles A. Powell, Shanu Modi, Hiroji Iwata, Shunji Takahashi, Egbert F. Smit, Salvatore Siena, Dwan-Ying Chang, Charles S. Fuchs, Kun Nie, Amy Qin, Jasmeet Singh, Corina Taitt, Sunil Verma, D. Ross Camidge. Pooled analysis of drug-related interstitial lung disease (ILD) in 8 single-arm trastuzumab deruxtecan (T-DXd) studies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr CT167.