Rapid digital patient-derived organoid testing for guiding therapy in patients with metastatic breast cancer receiving antibody drug conjugates (ADCs).

Abstract

1033 Background: MBC remains incurable despite recent incorporation of ADCs. The optimal sequencing of FDA-approved ADCs such as trastuzumab deruxtecan (T-DXd) or sacituzumab govitecan (SG) are lacking and real-world data have shown cross resistance between the ADCs. The current study aims to develop a clinical diagnostic assay to guide optimal treatment for patients with MBC using digital patient organoid (DPO) in patients with MBC receiving or progressed on ADCs for a precision tool to guide therapy in post-ADC resistance setting. Methods: An institutional IRB was established for prospective collection of fresh tumor biopsies in patients with MBC undergoing ADC therapy. DPO drug sensitivity tests were conducted and analyzed for association with clinical outcome using progression free survival (PFS). DPO tests include ADCs (T-DXd, SG, Dato-DXd) or single agent chemo dosing experiments to determine drug sensitivities (AUC values calculated from CTG readouts) for downstream correlation analyses. Results: Between 09/2024–01/2026, 41 fresh tumor tissues were collected and processed. 22 samples had sufficient viable cells to pass QC for DPO establishment and were successfully assayed and 12 failed QC, giving a DPO establishment success rate of 64.71%. DPO success rates varies by breast cancer subtypes: 100% in HER2+, 73% in TNBC and 56% in HR+HER2- BC. and sites of metastasis. Sites of metastasis (met) also impacted DPO success rates: lymph node 0%, liver met 33%, skin met 60%, breast/ascites 100%. Organoids typically established and expanded within 7-14 days post-biopsy. The DPO drug assays were conducted for ADCs, payloads, and chemotherapy agents testing. Analysis revealed that prior clinical exposure and progression on an ADC correlated with reduced DPO sensitivity to the same agent, consistent with acquired resistance. The platform was capable of delineating between target- versus payload-based resistance. Notably, DPOs from heavily pretreated patients were often resistant to both ADCs but remained responsive to alternative agents, highlighting potential therapeutic opportunities. Conclusions: The DPO platform shows promise in predicting treatment responses and differentiating resistance mechanisms to ADCs in MBC. This approach provides a foundation for rational treatment selection in patients with MBC with prior ADC resistance.