Prostate cancer (PCa) incidence and mortality is nearly twice that in Black men than any other race (Non-Hispanic White, Asian/Pacific Islander, American Indian/Alaska Native, Hispanic (any race)). Characterization of the underlying biological factors that contribute to this cancer health disparity (CHD) is required to close the gap and improve patient outcome for Black men. In vitro PCa research tools which reflect the racial/ethnic diversity of this patient population are urgently needed, as most PCa cell lines are of Non-Hispanic White-origin. PCa patient-derived xenografts (PDXs) retain many of the characteristics of patient tumors and can be isolated from specific racial/ethnic groups to address this concern. Rodent PDX models provide a valuable resource for studying human cancer, however recent trends in reducing animal usage have instituted a search for alternative methods for studying PDXs that also maintain their complexity. While 2D in vitro culture of PCa PDXs has been largely unsuccessful, recent evidence suggests 3D in vitro culture of PCa PDXs may enable better long-term culture of the tumor cells. Here, we present a racially/ethnically diverse PCa PDX library of specimens (Black, Non-Hispanic White, Hispanic) developed at MD Anderson Cancer Center (the MDA PCa PDX series) compatible with 3D in vitro culture. In order to confirm self-reported race/ethnicity, each PCa PDX was characterized by whole-exome sequencing and compared to reference populations for a genetic ancestry estimation. For 3D in vitro culture, we utilized a high throughput microfluidic culture platform with 96 chips, the MIMETAS OrganoPlate® 2-lane. This platform suits both 3D tissue/cell model development and throughput needs required for drug discovery. MDA-PCa PDX cell clusters were suspended in hyaluronic acid-based hydrogel solutions, seeded into the OrganoPlate, and cultured under continuous perfusion. Genetic ancestry estimation studies revealed that patient-reported race/ethnicity often aligned with the same racial/ethnic population genetic signatures. For example, two self-reported Black PDXs were primarily of West African origin (74.6-84.6%). When cultured in 3D, PCa PDX cultures were stable and viable for at least 7 days, as determined by high content fluorescence imagining coupled with cell viability dyes. By immunofluorescent staining, PCa PDX cultures exhibited appropriate expression of phenotypic prostate-specific antigen (PSA) and androgen receptor (AR) which was maintained over the life of the culture. Studies are ongoing to screen a panel of chemotherapy drugs and determine the predictivity of the platform. This well characterized racially/ethnically diverse PCa PDX library, together with the 3D in vitro platform and methods is a valuable resource for evaluating population-based tumor response. Citation Format: Peter Shepherd, Andrei Bonteanu, Stanley Hooker, Kristin Bircsak, Divya Iyer, Dwayne Dexter, Daniel A. Harrington, Rick Kittles, Nora M. Navone. 3D in vitro prostate cancer PDX resource for studying cancer health disparities [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2625.

Abstract Prostate cancer (PCa) is the most common occurring cancer in men and the second-most leading cause of cancer-related deaths in the United States. To improve patient outcome, research tools which mimic the prostate tumor microenvironment (TME) and accurately predict drug response are urgently needed. However, current in vitro PCa models do not recapitulate the complexities of the prostate TME including three-dimensional (3D) orientation, perfusion, extracellular matrix (ECM), and the presence of multiple cell types such as tumor cells, fibroblasts, and endothelium. Further, many complex in vitro model systems fail to maintain the throughput required for robust drug screening. To address this need, here, we highlight the stepwise development of a 3D in vitro PCa model by optimizing the individual culture conditions for each cell type (PCa cells, fibroblasts (FB), endothelial cells (EC)) within MIMETAS' high throughput perfusable 3D cell culture platform, the OrganoPlate® 2-lane 96. All monocultures (PCa, fibroblast, endothelial cell) were viable and expressed respective phenotypic markers (PCa: AR and PSA; FB: vimentin; EC: CD-31, VE-cadherin) as detected by immunofluorescent staining and high-content imaging. To demonstrate the statistical retention of heterogeneity within PCa populations, PCa cells (MDA-PCa-2b) were pre-labeled with 4 different tracking dyes and clustered into multicellular aggregates using a multiwell, ultra-low attachment plate. After 3D encapsulation within a migration permissive hyaluronic acid (MP-HA) hydrogel, PCa clusters retained uniform size clusters and even distribution of the pre-labeled populations within each cluster. Additionally, use of this pre-clustering reduced cell debris in the encapsulated 3D cultures. 3D FB cultures, mimicking normal or reactive stroma, were established with bone stroma cell lines and primary human mesenchymal stem cells. The cells' phenotypic stretched morphology and migration through the ECM were tuned by tailoring the hydrogel crosslinking assessed further by measuring the permeability of 2,000 kDa FITC-dextran. EC (primary human lung microvascular endothelium) were seeded against MP-HA to form a tubule structure within the perfusion channel. Culture conditions were optimized to reduce the permeability of smaller molecules (4.4, 150 KDa Dextran) through EC barriers and further imaging revealed the formation of a 3D blood vessel-like structure within the perfusion channel. Studies are ongoing to combine all three cell types into a single model. This stepwise approach, for building a complex, 3D prostate tumor model, will enable the full incorporation of all cell types and ECM into a single model which will have the potential to better recapitulate the in vivo prostate TME and improve the predictivity of PCa in vitro models. Citation Format: Divya Iyer, Andrei Bonteanu, Jedidiah Z. Zhu, Peter Shepherd, Rick Kittles, Nora M. Navone, Daniel A. Harrington, Dwayne Dexter, Kristin M. Bircsak. Development of a 3D in vitro model of the prostate tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2641.