Evaluation of novel microscale cell culture platform for translational drug development in prostate cancer

Prostate cancer is a disease with a remarkably high impact on Veteran health. Not only is it the most common cancer diagnosed in Veterans, with nearly 15,000 new cases diagnosed each year, but Veteran men get diagnosed with prostate cancer at nearly twice the rate of the general population 1. Unfortunately, even with treatment, many of these Veterans will ultimately die from

their disease. New, effective treatments are therefore desperately needed to improve outcomes for Veterans with prostate cancer. Although the development of new prostate cancer research is currently under way, the development of such therapies is heavily limited by current preclinical research models, which do a notoriously poor job of identifying therapies that will be effective at

the clinical level. We have therefore developed a novel open microfluidic cell culture platform that enables multi-culture tumor models in vitro using primary, patient-derived cells. The central hypothesis in this proposal is that primary cell derived multi-culture TME models in Stacks will more closely model patient tumor biology and can better predict clinical therapeutic efficacy in

prostate cancer than traditional preclinical models. The primary objective of this proposal will be to test this hypothesis through three Specific Aims: Aim 1: To determine whether the gene expression profiles of prostate tumor cells in multi-culture Stacks models more closely correlate with patient expression profiles than tumor cells in traditional in vitro models.

Tumor cells (cell line and patient-derived organoids) will be cultured in traditional in vitro platforms in mono-culture and in co-culture with primary macrophages/cancer-associated fibroblasts. The same mono- and co-culture models will also be established in Stacks along with the addition of a tri-culture model with all 3 cell populations. RNA-seq will then be performed on the tumor cells

from each model. Transcription profiles will be compared to patient datasets to determine which model most closely correlates with patient tumors. Aim 2: To establish whether multi-culture tumor models in Stacks can more accurately predict the efficacy of therapies in patients with prostate cancer than standard in vitro models. Using the same models in Aim 1, each

model will be treated with 3 therapies known to be effective in patients with prostate cancer and 3 therapies known to be ineffective in patients with prostate cancer. The cytotoxic effect of the therapies will be evaluated in each model and compared to clinical trial data to determine which system most accurately predicts therapeutic efficacy at the clinical level. Aim 3: To evaluate

whether patient-derived TME models in Stacks can predict therapeutic response to docetaxel in Veterans with prostate cancer. Co-culture models will be established in Stacks using tumor cell lines and primary monocyte-derived macrophages from Veterans with prostate cancer about to initiate docetaxel treatment. Stacks models will then be treated with docetaxel

and evaluated for the cytotoxic effect of the treatment. The data from each patient-derived Stacks model will then be compared to the docetaxel response in the corresponding patient to determine if the Stacks co-culture models can predict therapeutic response in patients. The data from each of the Aims will be analyzed to determine if multi-culture models are more biologically and clinically

relevant and if the Stacks platform is a more effective tool than traditional in vitro platforms for translational prostate cancer research.