ESTABLISHMENT OF A DOUBLE-HUMANIZED MOUSE MODEL USING PATIENT-DERIVED XENOGRAFT BLADDER CANCER TUMOR CELL LINE AND HUMAN γδ T-CELLS.

Despite recent breakthroughs of immune-checkpoint inhibitors in the management of bladder cancer (BC), only 20% of patients with non-muscle invasive BC (NMIBC) completely respond, leaving room for improvement. Due to challenges in developing a preclinical model which reflects the heterogeneity and biology of bladder tumors, patient-derived xenograft (PDX) models were created. PDX models involve implanting tumor extracted from patients into immunocompromised mice, which therefore keep the morphology and genomic fidelity of their parental patient cancers. The state of immunodeficiency, however, precludes study of immunotherapy and consequently humanized mice models engrafted with a human immune system were developed, though this led to spontaneous xenogenic graft vs host disease (GVHD). As a potential solution to GVHD in humanized mouse models, this paper describes successful creation of a double-humanized mouse (DHM) by simultaneously engrafting human immune cells and human tumors from the same donor. BC tissue was processed +/- digested into single-cell suspensions (SCS, F0 tissue) and injected orthotopically or subcutaneously into flanks of NSG™ mice. Subsequent tumor growth (PDX-F1) was measured, and PDX-F1 tumors reaching >200mm3 were processed into SCS and reimplanted subcutaneously into a new NSG™ (PDX-F2). Continuous passages (PDX-Fn) were repeated.; Extra SCS-F1 and Fn was analyzed for human epithelial cell adhesion molecule (EpCAM) expression. From an individual F1, a stable cell line (PDX257S) was cultured, and also analyzed for EpCAM expression by flow cytometry. For the DHM model, autologous γδ T-cells expanded in vitro from PDX257S donors were injected with PDX257S cells (effector: target = 2:1) into the NSG™ flanks and were observed for body condition/hematuria for 45 days. Total RNA-sequencing was performed on F0, PDX-Fn and PDX257S cell line. Tumor growth curves were analyzed with 2-way analysis of variance (ANOVA). F0 tissue was obtained from 60 patients. Tumor acceptance in orthotopic injection was not observed in the first 11 suitable tumors, and this method was ceased. Subcutaneous tumor acceptance rate improved with digestion and processing of F0 rather than processing alone (11% vs 5%). EpCAM expression was 98% vs 70% in PDX257S vs its F1 tissue, and < 30% in F1 tissue from other donors that did not yield further passages. In the DHM, while tumor growth was significantly reduced compared to control group after 45 days (mean vol 200 mm3 vs 580 mm3, p=0.02) post injection, it was not completely suppressed. RNA sequencing revealed increased expression of genes including BCL2L1, KRAS, MYC, E2F1, and E2F4 in both PDX257 subcutaneous tumor and cell line compared with donor tumor tissue, and very low/no expression of CDKN2A and FHIT genes in both PDX257 cell line and donor tumor tissue. Creation of a double-humanized mouse model is feasible, and tumor growth is not completely suppressed with the addition of autologous γδ T-cells. PDX257S tumor and cell line is associated with increased EpCAM expression and on RNA sequencing, increased expression of BCL2L1, KRAS, MYC, E2F1, and E2F4. [ABSTRACT FROM AUTHOR]