MP09-07 USE OF 3D VIDEO MICROSCOPY OF HUMAN-DERIVED PROSTATE ORGANOIDS TO ASSESS FEATURES OF EARLY INVASIVE PROSTATE CANCER

Abstract

INTRODUCTION AND OBJECTIVE: Patient-derived organoids provide a method for creating human prostate cancer cell lines in culture in three dimensions. We established living organoids using normal and prostate cancer tissue from hormone-naive patients obtained at the time of surgery. We assessed with an image analysis system, dynamic, 3D morphologic transformations of the organoids. Our long-term goal is to assess prognostic features of early invasive disease. METHODS: Needle cores from fresh radical prostatectomy specimens were harvested at the time of surgery. Gross tumor and normal prostate as controls were sampled. The de-identified samples are processed as described by Drost et al. Briefly, cells from the digested tissue are suspended within an extracellular matrix (ECM) gel, which give rise to differentiated prostate “mini-glands” or “mini-tumors” over ∼1-2 wks. Organoids are also biobanked. To assay dynamic spreading of cells from the compact mini-gland structure, mature organoids are isolated from the gel and plated on dishes coated with ECM. Time-lapse images are acquired and processed into binary masks for analysis. Spreading of prostate cells is estimated by the following formula of circularity: C = 4*pi*A/P^2, where C is circularity, A is area and P is perimeter. For a perfect circle, C = 1. RESULTS: We have biobanked prostate-derived organoids using primary tumor and matched normal controls from 47 patients. The samples represent a range of pathologic grades stratified by Gleason score (GS): GS6 n=8 (17%), GS7 n=33 (70%), GS8 n=1 (2%), and GS9 n=5 (11%). We tested 3D dynamic movement onto two ECM proteins: Laminin-332, normally secreted by epithelial cells, and Laminin-511, which coats smooth muscle cells and nerve axons within the gland. For control organoids, average time to dynamic change threshold was 12hr on Laminin-332 and 20hr on Laminin-511 using a circularity of 0.6 as a cut-off. CONCLUSIONS: This study demonstrates feasibility in assaying cellular behaviors of living human organoid cultures for prostate tumors and matched normal tissue. It will be interesting to correlate invasive behavior to the clinical grading and staging of the primary neoplasm. We predict further studies will reveal that organoids derived from higher grade tumors will spread on Laminin-511, reflective of their ability to invade through the smooth muscle stroma along nerves to escape the prostate gland. Source of Funding: NCI P30 CA 23074, Arizona Biomedical Research Centre ADHS15-090301, and DoD PCRP W81XWH-19-1-0455. K.D.M. is supported by the ARCS Foundation.