Successful drug development in oncology is grossly suboptimal, manifested by the very low percentage of new agents being developed that ultimately succeed in clinical approval. This poor success is in part due to the inability of standard cell-line xenograft models to accurately predict clinical success and to tailor chemotherapy specifically to a group of patients more likely to benefit from the therapy. Patient-derived xenografts (PDXs) maintain the histopathological architecture and molecular features of human tumors, and offer potential solution to maximize drug development success and ultimately generate better outcomes for patients. Over the past decade PDXs have been successfully used for pre-clinical evaluation of cancer therapeutics and to derive predictive biomarkers in various solid tumors. To assess the utility of tumor grafts in guiding therapy for individual patients, we analyzed a large set of correlations between PDX model drug responses and patient clinical outcomes.

We obtained 113 clinical correlations across 78 patients with various solid tumors for whom a PDX model was successfully developed and drug testing completed. Some patients were included in more than one response group, having undergone multiple treatments as cycles of disease progression and therapy occurred. Treatment outcomes and PDX responses were evaluated using Response Evaluation Criteria In Solid Tumors (RECIST) criteria and diagnostic parameters calculated. A significant association was observed between drug responses in the patient and the corresponding PDX model in 89% (101/113) of the therapeutic analyses. We further stratified our population by whether or not the patient's treatment was the first or further line received after resection of the tumor to develop the PDX model. We found positive responses in the tumorgrafts reflected positive outcomes in the patient in virtually every instance, suggesting that PDX model drug responses correlate with patient clinical outcomes, even as patients undergo numerous treatments overtime.

Although our data suggest that PDX models are a predictable platform for preclinical evaluation of a variety of treatments, the immunocompromised mice hosts reduce the similarity of PDX models to the original tumor environment and preclude the evaluation of cancer immunotherapeutic approaches. Hence, development of robust preclinical tools to test such drugs against human tumors in the context of an allogeneic immune system remains a challenge. Since it was demonstrated that transplantation of human hematopoetic stem cells (HSCs) into NOG or NSG mice result in reconstitution of the hematopoesis and functional human immunity, we have combined the PDX platform with the humanized NOG mice to create immunografts—tumorgrafts implanted into immunodeficient mice reconstituted with a homologous human immune system. Immunografts were treated with immune checkpoint inhibitors targeting CTLA4 and PD1 and human immune activation as well as tumor growth inhibition was evaluated. Our results indicate that immunografts produce a human tumor immune cell interaction and exhibit significant tumor growth inhibition in response to immune-oncology agents in several types of patient-derived tumors. Although this approach is technically challenging and may not be suitable for patients with aggressive, rapidly progressing disease, further development and optimization of this system may contribute to better selection of immune oncology drugs alone and in combination for pharmaceutical development.

Taken together, our studies demonstrate that application of PDX models to guide treatment decisions in oncology may improve the identification of drugs most likely to benefit an individual patient and lead to better patient tumor responses. Moreover, given the clinical relevance of these models, they could also be deployed as real-time patient surrogates in the drug development process.

Citation Format: Evgeny Izumchenko, David Sidransky. Patient-derived xenografts as translational tools to predict patient responses to oncology therapy. [abstract]. In: Proceedings of the AACR Special Conference: Patient-Derived Cancer Models: Present and Future Applications from Basic Science to the Clinic; Feb 11-14, 2016; New Orleans, LA. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(16_Suppl):Abstract nr IA03.