Simple SummaryImmune checkpoint inhibitors (ICI) have been standard care for advanced bladder cancer patients who fail or are not eligible for chemotherapy. There is an urgent need for preclinical models to study immunotherapy responses because the majority of patients with advanced bladder cancer do not respond to ICI therapy. To develop an authentic bladder cancer preclinical model in immunocompetent mice, we generated a stable organoid and xenograft model by the ex vivo transduction of adenovirus-expressing Cre recombinase into normal mouse urothelial organoid cells isolated from mice engineered with LoxP sites flanking the Trp53, Pten, and Rb1 genes. The triple knockout (TKO) urothelial organoids developed into high-grade urothelial carcinomas of the basal subtype, both in vitro and in vivo. To study ICI anti-PD-1 treatment responses, the TKO tumors were treated with an anti-PD-1 antibody or a control IgG2a. A mixed pattern of treatment responses was observed. Single-cell analyses of immune cells revealed significantly different infiltration of immune cells between non-responders and responders. A higher percentage of immune cell infiltration, including macrophage and T cell tumor infiltration, was detected in responders compared to non-responders. Overall, these findings suggest that this preclinical TKO model will be a useful tool to study the factors influencing bladder cancer immunotherapy responses and suggest that modulating tumor-associated macrophages may help overcome ICI immunotherapy resistance.Approximately 80% of patients with advanced bladder cancer do not respond to immune checkpoint inhibitor (ICI) immunotherapy. Therefore, there is an urgent unmet need to develop clinically relevant preclinical models so that factors governing immunotherapy responses can be studied in immunocompetent mice. We developed a line of mouse triple knockout (TKO: Trp53, Pten, Rb1) urothelial carcinoma organoids transplanted into immunocompetent mice. These bladder tumors recapitulate the molecular phenotypes and heterogeneous immunotherapy responses observed in human bladder cancers. The TKO organoids were characterized in vivo and in vitro and compared to the widely used MB49 murine bladder cancer model. RNAseq analysis of the TKO tumors demonstrated a basal subtype. The TKO xenografts demonstrated the expression of urothelial markers (CK5, CK7, GATA3, and p63), whereas MB49 subcutaneous xenografts did not express urothelial markers. Anti-PD-1 immunotherapy resulted in a mixed pattern of treatment responses for individual tumors. Eight immune cell types were identified (basophils, B cells, dendritic cells, macrophages, monocytes, neutrophils, NK cells, and T cells) in ICI-treated xenografts. Responder xenografts displayed significantly increased immune cell infiltration (15.3%, 742 immune cells/4861 total cells) compared to the non-responder tumors (10.1%, 452 immune cells/4459 total cells, Fisher Exact Test p < 0.0001). Specifically, there were more T cells (1.0% vs. 0.4%, p = 0.002) and macrophages (8.6% vs. 6.4%, p = 0.0002) in responder xenografts than in non-responder xenografts. In conclusion, we have developed a novel preclinical model that exhibits a mixed pattern of response to anti-PD-1 immunotherapy. The higher percentage of macrophage tumor infiltration in responders suggests a potential role for the innate immune microenvironment in regulating ICI treatment responses.
Read full abstract