Abstract 3433: Genomic analysis of patient-derived xenograft (PDX) model from metastatic breast cacner

Background: Patient-derived xenograft (PDX) models have been shown to be predictive of clinical outcomes and are being used for preclinical drug evaluation, biomarker identification, biologic studies, and personalized medicine strategies. Molecular subtyping of metastatic breast cancer is specifically important for clinical decision for targeted therapy, endocrine therapy, and chemotherapy. The PDX model, reflecting the same molecular characteristics of a patient9s tumor, can provide valuable information prior to treatment. We explored the possibility of generating PDX from metasttic breast cancer and examined whether the primary or metastatic cancer of the patient and the tumor implanted in the mouse have the same genetic characteristics. Method: We established PDX models using small biopsy samples of tumors from primary breast or metastatic cancer using NSG mouse. Clinicopathologic characteristics and outcome were collected from the electronic medical records. Whole-exome sequencing was performed using Illumina platform. Single-nucleotide polymorphisms (SNPs) and small insertions and deletions (Indels) were called using muTect(1.1.7) and IndelGenotyper provided by GATK (3.6.0). We also investigated whether copy number variations (CNVs) are maintained in the PDX models. Conifer was used for counting the read framents. The read counts of normal and tumors were normalized by a log2 scale. Results: We established 24 PDX models from 34 biopsy samples (11 primary breast cancer and 23 metastatic cancer). 11 samples were hormone receptor-positive luminal type, 9 samples were HER2-positive type and 13 samples were triple-negative subtype. Our results indicate that the most frequent mutated genes are TP53, ARAF, GNAQ, ATRX, and PIK3CA in total samples. TP53 was the most frequently altered gene in all subtypes. Other mutated genes were slightly different for each subtype, followed by GNAQ and GATA3 in hormone receptor-positive subtype. ARAF and ATRX were in TNBC subtype. In HER2+ subtype, GNAQ, BRCA2, PIK3CA and ARID1B were same number of mutations. EIF3E and IKZF3 were the most frequently amplified genes and USP6 and SSX4 were the most frequently deleted genes. In addition, we compared the results in 5 cases that had all of the patient9s tumor sample and the xenografted tumor sample; the same mutation was found and reflected the molecular features of the patient. Conclusion: In this study, it was possible to establish PDX from biopsy tissues and PDX models maintain the genomic characteristics of the patient tumor. We believe that by using PDX models and bioinformatics analysis, it is possible to find new druggable targets and eventually the proper drug for personalized medicine. Citation Format: Dongjin Shin, Seock-Ah Im, Seongyeong Kim, Minjung Kim, Yu Jin Kim, Jinjoo Kang, Ahrum Min, Jieun Lee, Giyong Jang, Deukchae Na, Kyung-Hun Lee, Jongil Kim. Genomic analysis of patient-derived xenograft (PDX) model from metastatic breast cacner [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3433.