Article Text
PDF
Abstract
Introduction/Background Despite accumulating knowledge of the underlying genetic fingerprints, ovarian cancer (OC) remains a major threat and cause of mortality in women worldwide. Concerted endeavors have been undertaken to pioneer innovative treatment modalities in various preclinical models to ultimately realize individualized precision medicine . Patient-derived xenograft (PDX) models are generally acknowledged to outperform conventional OC cell lines in multiple aspects. By transplanting fresh tumor tissue into immunodeficient mice, PDX maintains the fine pathological structure and genetic profiles of the corresponding tumor. Crucially, these models consistently mirror treatment responses observed in patients and therefore could be harnessed as ‘avatar models’ to access personalized therapy and advance understanding in drug resistance.
Methodology We established a living biobank of 66 OC PDX lineages derived from fresh tumor samples, encompassing primary and multi-site metastatic tumors collected from 53 patients during surgical procedures between April 2019 and August 2021. PDX models and paired patient samples were submitted to histological and genomic characterization. We evaluated the clonal evolution during model derivation and propagation. PARP and WEE1 inhibitor resistant models were established through iterative cycles of exposure to corresponding agents to facilitate the study of acquired drug resistance.
Results PDX models were successfully generated across the major pathological types of OC, faithfully recapitulating the characteristic histological traits of their corresponding primary tumors. Consistency was observed in the retention of genomic alterations within the PDX models. Despite an overall similarity, we identified certain genomic discrepancies between patients and PDXs in our panel which could be attributed to clonal evolution and/or intra-patient spatial tumor heterogeneity. Interrogation of drug resistant PDX revealed that activation of the Wnt/β-catenin signaling pathway contributes to acquired resistance of PARP and WEE1 inhibitors.
Conclusion Our findings demonstrated the fidelity of OC PDX models and the feasibility of recapitulating acquired resistance within personalized PDX models.
Disclosures All authors have no potential conflict of interest to report.
