The Md Anderson Patient-Derived Xenograft Series For Modeling Precision Oncology In Biliary Tract Cancers.

Abstract Purpose: Biliary tract cancers (BTCs), including intrahepatic cholangiocarcinoma (ICC), extrahepatic cholangiocarcinoma (ECC), and gallbladder cancer (GBC), are rare malignancies which frequently present at advanced stage and have a poor survival. Incorporation of next-generation sequencing into clinical care has provided an avenue for the development of targeted therapies in BTCs which are rich in actionable mutations. However, clinically relevant models for BTC are limited, representing a major challenge in the field. We sought to develop a catalog of BTC patient-derived xenograft (PDX) models and representing diverse molecular profiles to create a resource for the modeling of precision oncology. Methods: Tumor tissue from surgical specimens or image-guided biopsies was collected from consenting patients at MD Anderson Cancer Center with histologically-confirmed BTCs for PDX development. Tumor fragments were implanted in the flank of NSG mice and passaged into nude mice. Whole-exome sequencing (WES) was performed on tumors from early-passage (P1-P2) PDXs and matching patient blood to determine somatic mutations/indels and copy number variations. Functional alterations in cancer-related genes that are targetable directly or indirectly with approved or investigational agents were considered “actionable”. Results: Samples were obtained from 97 patients undergoing image-guided biopsy (86/97, 89%) or surgical resection (11/97, 11%). Out of 97 tumors implanted, 33 (34%) PDXs from 30 patients were successfully established: 74% (24/33) ICC, 15% (5/33) ECC, and 12% (4/33) GBC. Relative take rates for PDX development were 34% (29/86) for those developed from biopsy samples and 36% (4/11) for those developed from surgical samples. In two patients, multiple PDXs were developed from biopsies at different time points during their care. WES demonstrated that the PDXs captured several key actionable alterations including alterations in ERRB2, FGFR2, IDH1, ATM, PIK3CA, PTEN and KRAS. Conclusion: Here, we describe a unique collection of clinically and molecularly annotated BTC PDX models which reflect the genomic heterogeneity present in ICC, ECC, and GBC. WES of early-passage PDXs provides evidence that these models capture the variety of actionable alterations harbored in the parental tumor and may enable the development of novel, biomarker-driven treatment strategies and rational combination therapies. To date, this is one of the largest collections of BTC PDX models available and will serve as an invaluable resource to guide the development of personalized treatments for patients with these aggressive malignancies. Citation Format: Timothy Philip DiPeri, Kurt W. Evans, Ching-Wei D. Tzeng, Lawrence Kwong, Michael P. Kahle, Xiaofeng Zheng, Dali Li, Hop S. Tran Cao, Thuy Vu, Sunhee Kim, Fei Su, Bryce Kirby, Chetna Wathoo, Gabriela Raso, Yasmin Rizvi, Huamin Wang, Filip Janku, Kenna Shaw, Timothy Yap, Milind Javle, Jordi Rodon, Funda Meric-Bernstam. The MD Anderson patient-derived xenograft series for modeling precision oncology in biliary tract cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2710.