Abstract P3-06-31: Patient-derived xenografts accurately predict patient response in breast cancer patients

PURPOSE/OBJECTIVES: A growing body of evidence demonstrates that patient-derived xenografts (PDXs) represent living tumor models that accurately reflect the biology of the primary patient tumor. More importantly, we have previously shown that PDX models show responses to therapeutic agents that are concordant with patient clinical response and can be used to direct personalized cancer treatments (Stebbing, 2014). Here we report the ability of PDX models to predict for patient response to drug treatment in a cohort of breast cancer patients. MATERIALS/METHODS: Tumors were resected from patients with either primary or metastatic breast cancer and implanted into immunodeficient mice to establish PDX models. Successfully engrafted PDX models were expanded and randomized for drug sensitivity testing. Tumor growth inhibition and tumor regression were captured and results were correlated with a patient’s clinical response. In some cases, PDX results were used to personalize cancer treatment and some patients used PDX-directed treatments over multiple lines of therapy. RESULTS: A total of 42 tumors from 40 patients were implanted resulting in 21 successfully engrafted PDX models (50% engraftment rate). Notably, engraftment rates were much higher for patients with triple negative breast cancer (TNBC) and resulted in 7 successful PDX models from 8 TNBC patients (87.5% engraftment rate). Drug sensitivity testing was offered to patients with established PDX models. Drugs and drug combinations tested included standard and nonstandard chemotherapy as well as biologics. At that time of this publication, 4 patients (3 TNBC and 1 HER2+) with completed drug sensitivity tests also had clinical data available resulting in 7 clinical correlations; 4 retrospective and 3 prospective. In all 7 cases, the PDX model accurately predicted patient clinical response demonstrating an accuracy of 100%. Five of the drug tests predicted drug sensitivity and 2 tests predicted resistance, indicating the potential of the PDX platform to predict for both sensitivity and resistance to therapy. The 3 prospective correlations resulted in concordant clinical benefit in 2 patients for duration greater than 6 months each. CONCLUSIONS: These data support the use of the personalized PDX model as a platform for therapeutic decision making that can guide treatment for patients with breast cancer. A prospective clinical trial in TNBC is currently underway. Citation Format: Lisa Stow, Amanda Katz, Hanna Irie, Elisa Port, Justin Stebbing, Daniel Ciznadija, Angela Davies, Keren Paz. Patient-derived xenografts accurately predict patient response in breast cancer patients [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-06-31.