Abstract B061: In vivo benchmarking of trastuzumab deruxtecan (T-DXd) activity in panel of 400 XPDX models

Abstract Background: Trastuzumab deruxtecan (T-DXd) is an ADC consisting of an anti-HER2 antibody linked to a novel topoisomerase I inhibitor payload approved for several indications including HER2+/- pretreated breast cancers, HER2-mutated NSCLC, and HER2+ gastric, gastroesophageal, and other HER2+ solid tumors. Due to the broad activity of this targeted therapy in HER2+ and some HER2- cancers, whether the molecular profile of the tumor is also important in determining sensitivity is intriguing. To answer this question, we benchmarked single agent activity of T-DXd in a panel of well characterized solid and hematologic XPDX models representing most indications and subtypes and correlated results to various model characteristics including donor patient profile and treatment, model receptor expression, and identified driver variants. Methods: Four hundred models representing twenty-one indications were evaluated including breast, gastric, esophageal, lung, colorectal, ovary, urothelial, and pancreas cancers. Each model was characterized by donor patient treatment and outcome prior to sample collection for model development, model HER2 receptor staining, and molecular profiling by WES and RNAseq. For in vivo studies, all models were benchmarked against single agent T-DXd; endpoints included tumor volume and time from treatment initiation with %T/C values and tumor regression reported at study completion; a T/C of ≤ 20% versus control was considered sensitive. Tumor regression (%T/C<0%) versus Day 0 tumor volume was also reported. Results: HER2 staining was reported in all evaluated indications and receptor density matched clinically HER2+ donor positive patients. Models retained receptor expression and similar histology comparable with archival clinical or model samples. In vivo, T-DXd was most efficacious towards breast cancer models with 72% reporting sensitivity. 38% of lung models were found responsive to therapy with some tumor regressions and included models harboring ERBB2 mutations and fusions. However, only 13% of evaluated colorectal XPDX models were found sensitive to T-DXd and none with ERBB2 variants. Interestingly, several sensitive lung and colorectal models harbor the KRASG12C mutation. In addition, T-DXd was active towards 55% of ovary, 50% of gastric, 37% of urothelial, and 25% of pancreas models; sensitive models in these indications reported low to high HER2 staining and some known driver variants. Conclusion: We evaluated single agent T-DXd in a panel of four hundred models representing twenty-one indications and correlated drug sensitivity to HER2 expression and driver mutations. Results from this study are valuable in identifying additional cancer indications and subtypes which may be sensitive to T-DXd therapy. Citation Format: Alyssa Simonson, Johnnie Flores, Jennifer Garcia, Peter Forofontov, Crystal Moreno, Esteban Wiggins, Sarah Lopez, Amarissa Rojas, Mia Lopez, Kirstin McCormick, Raymona Artis, Amy Fredrickson, Jim Lund, Christopher Nelson, Ian Sturgill, Ashley Jamison, Ronan Pleass, Garrett Mercer, Kyriakos Papadopoulos, Amita Patnaik, Victor Moreno, Emiliano Calvo, David Smith, Andrew Cunningham, Chris Takimoto, Michael Wick. In vivo benchmarking of trastuzumab deruxtecan (T-DXd) activity in panel of 400 XPDX models [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr B061.