Abstract 5016: High-throughput patient-derived 3-dimensional organoid cultures as personalized models to assess drug response and post-treatment residual disease

Abstract

Abstract Cell lines grown in 2-dimensional (2D) plastic surfaces have historically been the main models to identify cancer cell vulnerabilities. In contrast, malignant tumors in vivo grow as 3-dimensional (3D) cellular masses and manifest remarkably different drug-response patterns compared to cell lines. We have shown that breast cancer (BrCa) cells in 3D extracellular matrices acquire distinct phenotypic properties, compared to the 2D counterparts. Notably, serum-free culture medium supplemented with defined cocktails of developmental morphogens was sufficient to support the growth of BrCa cell lines in 3D, but not in 2D conditions; indicating a key role of 3D architecture in in vivo growth regulation. CRISPR/Cas9-based gene editing screens in BrCa cells indicated that distinct gene subset govern the growth in 3D vs. 2D conditions. Using similar culture conditions, we expanded several molecularly annotated BrCa patient-derived 3D Organoids (3D PDOs), adapted them into a high-throughput miniaturized drug screening platform, and interrogated them against a panel of ~500 target-annotated kinase inhibitors and FDA-approved anti-cancer agents. This pharmacological screen revealed high sensitivity to distinct classes of agents (eg. PI3K inhibitors), but also variable responses to others (eg. EGFR inhibitors); and provided insight into potentially driver vs. passenger mutations in the respective patients. Co-culture of PDOs with non-malignant stromal cells from bone metastatic niche reduced sensitivity to some classes of drugs (eg. PLK inhibitors). Importantly, long-term (up to 3-4 weeks) treatment of 3D organoids did not eradicate the cells in culture (in contrast to conventional 2D assays), but generated a cell subpopulation which remained viable for the duration of the experiment, reminiscent of the residual disease after initial clinical partial responses to a drug. This residual disease phenotype was also confirmed in vivo, when the respective PD-Xenografts were treated with the same agents. The majority of transcripts with differential expression in residual PDX tumor cells were also concordantly up- or down-regulated in the respective 3D PDO model. These transcripts included canonical regulators of cell cycle, chromatin signaling and epigenetic mechanisms, cytoskeletal function and metabolic pathways; as well as other transcripts not previously linked to drug resistance in conventional 2D models. To our knowledge this is the first description of an in vitro personalized cancer model that simulates both molecularly and phenotypically the post-treatment residual disease in vivo. Our high-throughput functional platform can potentially complement genomic approaches in personalized medicine and help discover novel drugs that overcome tumor drug resistance. Citation Format: Eugen Dhimolea, Ricardo De Matos Simoes, Yoko DeRose, Pallavi Awate, Xiang Weng, Huihui Tang, Aedin Culhane, Alana Welm, Constantine Mitsiades. High-throughput patient-derived 3-dimensional organoid cultures as personalized models to assess drug response and post-treatment residual disease [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5016. doi:10.1158/1538-7445.AM2017-5016