Abstract 1272: The development of PDX models from ADC-resistant breast cancer patient tissue for next-generation therapy evaluation

Abstract Introduction: Antibody-drug conjugates (ADCs) are monoclonal antibodies targeting tumor cells linked to cytotoxic drugs via chemical linkers. Trop-2 targeting ADCs like sacituzumab govitecan (SG), datopotamab deruxtecan (Dato-DXd), and SKB-264 have shown success in clinical settings. However, they face challenges such as complex pharmacokinetics, insufficient tumor targeting, suboptimal payload release, side effects and drug resistance. This study developed SG-resistant PDX models to replicate clinical resistance to support new therapy development and understanding of resistance mechanisms. Methods: Subcutaneous PDX models were established and characterized by histology and sequencing, with Trop-2 expression assessed by immunohistochemistry (IHC) and RNA sequencing (RNA-seq). For in vivo efficacy studies, models were treated 3-4 weeks after tumor inoculation with SG (3 or 5 mg/kg, i.v.) and Dato-DXd (5 mg/kg, i.v.) based on clinical SG treatment schedule. Tumor growth inhibition (TGI) was calculated as [1 - (mean treated tumor volume / mean control tumor volume)] * 100%. Homologous recombination deficiency (HRD) scores were calculated (using scarHRD and PureCN from whole exome sequencing, with a cutoff value of 38). Results A female TNBC patient underwent 8 rounds of chemotherapy and targeted treatment, including 2 cycles of SG in the sixth round. Tissues collected at different treatment stages following SG treatment were used to establish BR9690, BR9806 and BR9801 PDX models. High Trop-2 expression was detected at both gene and protein levels in all three models. To evaluate Trop-2 ADCs efficacy, SG and Dato-DXd were tested in the PDX models. Tumor progression was observed in BR9690 for both SG (3mg/kg TGI=9.40% and 5mg/kg TGI=16.77%) and Dato-DXd (TGI=32.15%). In BR9806, tumor progression was observed in SG with limited TGI (3mg/kg TGI=27.52% and 5mg/kg TGI=25.75%), while Dato-DXd showed a partial response (TGI=59.47%). For BR9801, both SG and Dato-DXd showed limited TGI (SG: 3mg/kg TGI=-2.60% and 5mg/kg TGI=0.21%, Dato-DXd: TGI=24.78%). These results indicate clinical consistency with SG resistance and resistance to other Trop2-ADCs as the disease progresses. Besides, HRD scores for all three PDX models were above the cutoff (BR9690: 42, BR9806: 39, BR9801: 50) indicating genomic instability. In this study, the limited response of the models despite HRD may suggest other factors affecting the drug’s efficacy targeting DNA repair mechanism, such as acquired resistance mechanisms or drug efflux activity, overriding the vulnerability provided by HRD. Conclusion The study successfully established longitudinal breast cancer PDX models from a patient with demonstrated resistance to Trop2-ADCs. These models will aid in evaluating new treatments or combination therapies and understanding drug resistance mechanisms. Citation Format: Qingzhi Liu, Jinxi Wang, Xu Zhang, Wubin Qian, Likun Zhang, Ludovic Bourre, Jingjing Wang. The development of PDX models from ADC-resistant breast cancer patient tissue for next-generation therapy evaluation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1272.