Establishment of Patient-Derived Tumor Xenograft Models of High-Risk Endometrial Cancer.

Abstract

High-risk endometrial cancers (ECs), including high-grade EC, serous carcinoma (SC), clear cell carcinoma, and carcinosarcoma, account for 50% of deaths due to ECs. Therapies for these cancers are limited, and patient-derived tumor xenograft (PDTX) models are useful tools for preclinical drug evaluation, biomarker identification, and personalized medicine strategies. Here, we used and compared 2 methods to establish PDTX models. Fresh tumor tissues collected from 18 primary high-risk EC patients (10 high-grade ECs, 6 SCs, 1 clear cell carcinoma, and 1 carcinosarcoma) were engrafted subcutaneously and in the subrenal capsule in NOD/SCID for establishment and Balb/c-nu/nu mice for expansion. Histology and cytokeratin, estrogen receptor, progesterone receptor, and P53 expression were evaluated to assess the similarity of primary tumors and different generations of PDTX tumors. Whole-exome sequencing (WES) and RNA sequencing were used in 2 high-grade EC models to verify whether the genetic mutation profiles and gene expression were similar between primary and PDTX tumors. The total tumor engraftment rate was 77.8% (14/18) regardless of the engraft method. The tumor engraftment rate was increased in subrenal capsule models compared with subcutaneous models (62.5% vs 50%, P = 0.464). The time to tumor formation varied significantly from 2 to 11 weeks. After subrenal capsular grafting, grafted tumors could be successfully transplanted to subcutaneous sites. We observed good similarity between primary tumors and corresponding different passages of xenografts. The combination of 2 engrafting methods increases the tumor engraftment rate. The high tumor engraftment rate ensures the establishment of a high-risk EC biobank, which is a powerful resource for performing preclinical drug-sensitivity tests and identifying biomarkers for response or resistance.