Abstract A58: An ovarian patient-derived xenograft model to identify the chemoresistant population

Abstract

Abstract Introduction: A cornerstone of preclinical ovarian cancer research over the last 30 years has been the use of cell lines both in vitro and in vivo. This resource has underperformed in its ability to identify effective novel therapeutics and evaluate the heterogeneity of an ovarian neoplasm. Developing a patient-derived xenograft (PDX) allows for a comprehensive study of the heterogeneous tumor and potentially the identification of cellular populations responsible for chemoresistance and recurrence, but has not been fully characterized for ovarian cancer. We have developed an ovarian PDX model to demonstrate many important similarities and differences between these and primary patient tumors, and begun to explore its utility in identifying mediators of chemoresistance and personalizing therapy. Methods: Tumors removed from patients undergoing primary tumor reductive surgery were implanted into the subcutaneous flanks of SCID mice. Once a tumor developed, it was expanded into additional mice to propagate a PDX line. Tumors were analyzed for heterogeneity by examination of the tumor initiating cell populations, its stromal content with HLA immunohistochemistry, and proliferation with Ki-67. Cohorts of PDX lines were treated with maximum-tolerated dose of combined carboplatin and paclitaxel or vehicle and response was compared between the PDX and the patient. Chemoresistant PDX lines were generated by treatment with MTD carboplatin and paclitaxel until no evidence of disease (NED), observation until recurrence, and retreatment until fully resistant. RNAseq was conducted comparing treated PDX lines to the untreated PDX lines (n=6 pair) to identify important mediators of chemotherapy resistance. Finally, PDX lines were characterized for defects in homologous recombination (HR) repair with an ex vivo assay, correlating HR status with response to PARP inhibition in vivo. Results: Refinement of xenografting procedures yields an 85% success rate in establishing a PDX. PDX tumors maintain the patient's histology, but the stromal component is replaced by murine cells in the first generation. The xenografts retain primary tumor heterogeneity, at least in expression of putative tumor initiating cells (TIC) (ALDH: 17% vs 19%, CD44: 2.4% vs 5%, CD133 10% vs 3%). However, treatment with chemotherapy significantly increases these populations (ALDH1 increased to 36.1% from 16.2% (p=0.002); CD133 increased to 33.8% from 16.2% (p=0.026), suggesting a role in chemotherapy resistance. The treated PDX tumors showed a decrease in Ki67 staining (from 64% to 34%, p=0.001), suggesting dormancy is induced in the surviving population. The PDX tumors show similar response to chemotherapy as patient tumors, in that PDX tumors from patients with a partial response respond more slowly (if at all) than those from patients achieving a complete response (mean change in tumor volume -5.11% vs -63.73%, p=0.029). RNAseq comparing treated and untreated PDX lines indicate only 54 common genes with a significant change in mRNA (p<0.05), suggesting great heterogeneity among patients not only in primary tumor content, but also mediators of chemoresistance. Finally, patient tumors predicted to respond to PARP inhibition by an ex vivo assay show regression of tumor or stable disease with single-agent PARP inhibitor therapy. Conclusions: The ovarian PDX model recapitulates patient tumor heterogeneity and mirrors response to chemotherapy. This model may prove superior in predicting response to therapy in patients, and allows a better model to study complex tumor biology that is impacted by tumor heterogeneity, such as survival of small populations with chemotherapy. Citation Format: Zachary C. Dobbin, Ashwini K. Katre, Monjri M. Shah, Britt K. Erickson, Hauping Chen, Ronald D. Alvarez, Michael G. Conner, Dongquan Chen, Charles N. Landen. An ovarian patient-derived xenograft model to identify the chemoresistant population. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research: From Concept to Clinic; Sep 18-21, 2013; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2013;19(19 Suppl):Abstract nr A58.