3D imaging analysis on an organoid-based platform guides personalized treatment in pancreatic ductal adenocarcinoma
Ya’an Kang, … , Jason B. Fleming, Michael P. Kim
Ya’an Kang, … , Jason B. Fleming, Michael P. Kim
Published October 25, 2022
Citation Information: J Clin Invest. 2022;132(24):e151604. https://doi.org/10.1172/JCI151604.
View: Text | PDF
Clinical Research and Public Health Development

3D imaging analysis on an organoid-based platform guides personalized treatment in pancreatic ductal adenocarcinoma

Abstract

BACKGROUND Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies, with unpredictable responses to chemotherapy. Approaches to assay patient tumors before treatment and identify effective treatment regimens based on tumor sensitivities are lacking. We developed an organoid-based platform (OBP) to visually quantify patient-derived organoid (PDO) responses to drug treatments and associated tumor-stroma modulation for personalized PDAC therapy.METHODS We retrospectively quantified apoptotic responses and tumor-stroma cell proportions in PDOs via 3D immunofluorescence imaging through annexin A5, α-smooth muscle actin (α-SMA), and cytokeratin 19 (CK-19) levels. Simultaneously, an ex vivo organoid drug sensitivity assay (ODSA) was used to measure responses to standard-of-care regimens. Differences between ODSA results and patient tumor responses were assessed by exact McNemar’s test.RESULTS Immunofluorescence signals, organoid growth curves, and Ki-67 levels were measured and authenticated through the OBP for up to 14 days. ODSA drug responses were not different from patient tumor responses, as reflected by CA19-9 reductions following neoadjuvant chemotherapy (P = 0.99). PDOs demonstrated unique apoptotic and tumor-stroma modulation profiles (P < 0.0001). α-SMA/CK-19 ratio levels of more than 1.0 were associated with improved outcomes (P = 0.0179) and longer parental patient survival by Kaplan-Meier analysis (P = 0.0046).CONCLUSION Heterogenous apoptotic drug responses and tumor-stroma modulation are present in PDOs after standard-of-care chemotherapy. Ratios of α-SMA and CK-19 levels in PDOs are associated with patient survival, and the OBP could aid in the selection of personalized therapies to improve the efficacy of systemic therapy in patients with PDAC.FUNDING NIH/National Cancer Institute grants (K08CA218690, P01 CA117969, R50 CA243707-01A1, U54CA224065), the Skip Viragh Foundation, the Bettie Willerson Driver Cancer Research Fund, and a Cancer Center Support Grant for the Flow Cytometry and Cellular Imaging Core Facility (P30CA16672).

Authors

Ya’an Kang, Jenying Deng, Jianhua Ling, Xinqun Li, Yi-Ju Chiang, Eugene J. Koay, Huamin Wang, Jared K. Burks, Paul J. Chiao, Mark W. Hurd, Manoop S. Bhutani, Jeffrey H. Lee, Brian R. Weston, Anirban Maitra, Naruhiko Ikoma, Ching-Wei D. Tzeng, Jeffrey E. Lee, Ronald A. DePinho, Robert A. Wolff, Shubham Pant, Florencia McAllister, Matthew H.G. Katz, Jason B. Fleming, Michael P. Kim

×

Figure 1

Construction and quantification of 3D cytoplasmic and nuclear algorithms in organoid models.

Options: View larger image (or click on image) Download as PowerPoint
Construction and quantification of 3D cytoplasmic and nuclear algorithms...
(A) Thick sections (40–120 μm) of organoids were reconstructed in 3D space and segmented based on the presence of DAPI+ nuclei (scale bar: 20 μm). 3D reconstructed images were generated from multiple organoids per sample and compiled to generate ACIs. Details related to number of examined organoids, number of examined sections, section thickness, etc., are listed in Supplemental Table 2. (B) Representative images of patient-derived xenograft organoids (PDXOs) and associated nuclear and cell segmentation using cytoplasmic algorithm buildups (scale bar: 20 μm). α-SMA (green), CK-19 (red), annexin A5 (gray). (C) PDXO cytoplasmic imaging analysis results, showing the compiled number of detected cells (2,349) and associated nuclei (2,260) in 8 different organoids (n = 8) derived from PATXO118. (D) Measured ACIs for α-SMA, CK-19, and annexin A5 measured in 8 different organoids (n = 8) from PATXO118. (E) Representative images of patient-derived organoids (PDOs) with PATO044 and associated nuclear and cell segmentation using cytoplasmic algorithm buildups (scale bar: 20 μm). (F and G) Representative PDO imaging of cytoplasmic algorithm buildups in 11 different organoids (n = 11) from PATO044 and associated ACIs for α-SMA, CK-19, and annexin A5. (D and G) Data are shown as the mean ± SEM by GraphPad Prism 9.0. All 3D images were captured and merged using an Andor Revolution XDi WD Spinning Disk Confocal microscope and analyzed using Imaris software.