Prostate cancer (PC), especially castration resistant PC (CRPC), is known to exhibit a high degree of intra- and inter-tumoral (inter-individual) heterogeneity, which have been linked to treatment resistance and disease… Click to show full abstract
Prostate cancer (PC), especially castration resistant PC (CRPC), is known to exhibit a high degree of intra- and inter-tumoral (inter-individual) heterogeneity, which have been linked to treatment resistance and disease recurrence. Yet, models employed for PC drug development so far have been derived from cell lines or individual (single) patients and screened independently without consideration of heterogeneity. We hypothesized that the inability to address this critical aspect of tumor biology, has in part led to translational failure in broader patient cohorts. Therefore, we created a number of mixture cell models where multiple PC cell lines, were labeled and co-cultured to capture genomic heterogeneity observed clinically in CRPC patients. Specifically, expression profiles of PC cell lines were clustered with patient-derived single-cell (intra-) or bulk RNA-seq datasets (inter-) to identify cell lines representative of heterogeneous PC tumors, based on their transcriptomic signatures. Cell lines assigned to distinct patient clusters were systematically selected and combined for the mixture model, to reflect the complex in-vivo genomic landscape. To apply these models to drug combination testing, we employed a computational approach to nominate efficacious combinations with Docetaxel (Doc), current Standard of Care. Our mixed-cell model created to recapitulate inter-patient diversity, successfully validated the efficacy of a novel combination of Doc with COL3 (Incyclinide), which is currently in Phase I trial for Refractory Metastatic Cancer. The combination was confirmed to be more potent than individual monotherapies over a range of concentrations. We further tested Doc +Olaparib (clinically approved for patients harboring BRCA mutations), a combination predicted to be inefficacious in the heterogeneous patient cohort as negative control. Indeed, addition of Olaparib did not improve on the inhibitory effect of Doc. To address intra-tumoral heterogeneity, we created a mixture of cell lines which were representative of distinct clonal fractions in single-cell patient datasets. In both models, it was evident that cell lines with similar molecular and phenotypic attributes (such as AR dependency) co-clustered, lending further credibility to our pipeline. We further explored intra-tumoral heterogeneity in the context of acquired phenotypes and advanced disease states, by mixing parent and Doce-resistant clones of DU145 and 22RV1. The binary mixtures were challenged with combinations of Doc and Rigosertib/Selumitinib, drugs that were predicted to selectively target the resistant clone. The combinations produced a higher tumor kill than either single agent drug. Additionally, we observed the effects of altering the relative proportion of constituent clones on drug response. Development of these preclinical models holds enormous promise not only for advancing combinatorial therapies for cancer, but also for providing valuable insights into the mechanisms of resistance and recurrence, by accurately modeling genomic heterogeneity in cancer. Citation Format: Sampreeti Jena, Daniel Kim, Adam M. Lee, Weijie Zhang, R. Stephanie Huang. Development of novel in-vitro mixed-cell models to capture genomic heterogeneity in prostate tumors [abstract]. In: Proceedings of the AACR Special Conference: Advances in Prostate Cancer Research; 2023 Mar 15-18; Denver, Colorado. Philadelphia (PA): AACR; Cancer Res 2023;83(11 Suppl):Abstract nr A041.
               
Click one of the above tabs to view related content.