Advertisement
Original Research|Articles in Press

High-throughput method to analyze the cytotoxicity of CAR-T Cells in a 3D tumor spheroid model using image cytometry

Open AccessPublished:February 07, 2023DOI:https://doi.org/10.1016/j.slasd.2023.01.008

      Abstract

      Solid tumors account for approximately 90% of all adult human cancers. As such, the development of novel cellular therapies has become of increasing importance to target solid tumor malignancies, such as prostate, lung, breast, bladder, colon, and liver cancers. One such cellular therapy relies on the use of chimeric antigen receptor T cells (CAR-T cells). CAR-T cells are engineered to target specific antigens on tumor cells. To date, there are six FDA-approved CAR-T cell therapies that have been utilized for hematologic B cell malignancies. Immune cell trafficking and immunosuppressive factors within the tumor microenvironment increase the relative difficulty in developing a robust CAR-T cell therapy against solid tumors. Therefore, it is critical to develop novel methodologies for high-throughput phenotypic and functional assays using 3D tumor spheroid models to assess CAR-T cell products against solid tumors. In this manuscript, we discuss the use of CAR-T cells targeted towards PSMA, an antigen that is found on prostate cancer tumor cells, the second most common cause of cancer deaths among men worldwide. We demonstrate the use of high-throughput, plate-based image cytometry to characterize CAR-T cell-mediated cytotoxic potency against 3D prostate tumor spheroids. We were able to kinetically evaluate the efficacy and therapeutic value of PSMA CAR-T cells by analyzing the cytotoxicity against prostate tumor spheroids. In addition, the CAR-T cells were fluorescently labeled to visually identify the location of the T cells as cytotoxicity occurs, which may provide more meaningful information for assessing the functionality of the CAR-T cells. The proposed image cytometry method can overcome limitations placed on traditional methodologies to effectively assess cell-mediated 3D tumor spheroid cytotoxicity and efficiently generate time- and dose-dependent results.

      Keywords

      Introduction

      Chimeric antigen receptor T cell (CAR-T) therapy is an antigen-targeted therapy that has gained considerable traction in the cancer immunotherapy field. CAR-T cell therapy involves engineering T cells to attack tumor cells by specifically binding a tumor antigen and inducing T cell activation, resulting in intracellular signaling, cytokine release, and tumor cell death [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Sun Y.-J.
      • Chen Y.-C.
      • Hua W.-K.
      • et al.
      Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
      ]. Currently, there are six FDA-approved CAR-T cell therapies – Kymriah (tisagenlecleucel), Yescarta (axiacabtagene ciloleucel), Tecartus (brexucabtagene), Breyanzi (lisocabtagene maraleucel), Abecma (idecabtagene vicleucel), Carvykti (ciltacabtagene autoleucel) [
      • Zheng P.-P.
      • Kros J.M.
      • Li J.
      Approved CAR T cell therapies: ice bucket challenges on glaring safety risks and long-term impacts.
      ,
      • Albinger N.
      • Hartmann J.
      • Ullrich E.
      Current status and perspective of CAR-T and CAR-NK cell therapy trials in Germany.
      ,
      • Cockroft A.
      • Wilson A.
      Comparability: what we can learn from the review of advanced therapy medicinal products.
      ,
      • Han D.
      • Xu Z.
      • Zhuang Y.
      • et al.
      Current Progress in CAR-T Cell therapy for hematological malignancies.
      ,
      • Huang W.
      • Li J.
      • Liao M.Z.
      • et al.
      Clinical Pharmacology Perspectives for Adoptive Cell Therapies in Oncology.
      ,
      • Borogovac A.
      • Keruakous A.
      • Bycko M.
      • et al.
      Safety and feasibility of outpatient chimeric antigen receptor (CAR) T-cell therapy: experience from a tertiary care center.
      ]. The target antigen for the aforementioned CAR-T cell therapies is either CD19 or BCMA, all of which target the hematologic B cell malignancies [
      • Maude S.L.
      • Frey N.
      • Shaw P.A.
      • et al.
      Chimeric antigen receptor T cells for sustained remissions in leukemia.
      ,
      • Srivastava S.
      • Riddell S.R.
      Engineering CAR-T cells: Design concepts.
      ,
      • Zhu Y.
      • Tan Y.
      • Ou R.
      • et al.
      Anti-CD19 chimeric antigen receptor-modified T cells for B-cell malignancies: a systematic review of efficacy and safety in clinical trials.
      ].
      In the recent years, a strong focus has been placed on CAR-T cell therapy discovery for solid tumor indications despite existing barriers that diminish efficacy of the approach. Some challenges faced by CAR-T cell therapy against solid tumors include the relative difficulty of immune cell trafficking and migration into the tumor microenvironment (TME), as well as the immunosuppressive factors, cells, and checkpoint inhibitors [
      • Sterner R.C.
      • Sterner R.M.
      CAR-T cell therapy: current limitations and potential strategies.
      ]. These challenges promoted the need for high-throughput phenotypic and functional assays to identify and characterize new targets for CAR-T cell therapy, the need for advanced co-culture assays to assess potency and specificity of CAR-T cell products, and the need for 3D assays that better recapitulate physiological conditions to investigate and overcome the immunosuppressive effects of the tumor microenvironment. The purpose is to reproduce the appropriate key aspects of the tumor microenvironment using various 3D model techniques [
      • Barbosa M.A.G.
      • Xavier C.P.R.
      • Pereira R.F.
      • et al.
      3D Cell Culture Models as Recapitulators of the Tumor Microenvironment for the Screening of Anti-Cancer Drugs.
      ].
      The conventional methods for characterizing the functionalities of CAR-T cells during the discovery process are plate readers, flow cytometry, or fluorescence microscopy [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Sun Y.-J.
      • Chen Y.-C.
      • Hua W.-K.
      • et al.
      Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
      ,
      • Li X.
      • Feng Y.
      • Shang F.
      • et al.
      Characterization of the therapeutic effects of novel chimeric antigen receptor T cells targeting CD38 on Multiple Myeloma.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ], however, there are some inherent limitations for each method. Plate readers are rapid and require relatively simple operations for multiple samples, but only provide an indirect bulk measurement of the cells, where it may not be sensitive for direct cell-based analysis [
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ,
      • Milone M.C.
      • Bhoj V.G.
      The Pharmacology of T cell therapies.
      ]. Flow cytometry has often been used not only to determine the relevant cell populations in a sample, but also to measure certain cell phenotypes such as activation, exhaustion, and the presence of biomarkers to determine the cytotoxicity of CAR-T cells in the presence of antigen [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ]. These methods are straightforward, but may require specially trained operators, analysis of a cell population in bulk as a percentage, and do not necessarily evaluate the mechanism of action [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ]. Finally, standard fluorescence microscopy provides effective live cell assay and in situ analysis, but lacks automated imaging and quantification [
      • Sun Y.-J.
      • Chen Y.-C.
      • Hua W.-K.
      • et al.
      Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ].
      Recently, image-based cytometry has emerged as a method to investigate and characterize CAR-T cell functions in a high-throughput manner. Plate-based image cytometry has demonstrated high efficiency in analyzing transduction efficiency [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ], cell proliferation [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ], and cytotoxicity for CAR-T cell therapy [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Sun Y.-J.
      • Chen Y.-C.
      • Hua W.-K.
      • et al.
      Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ]. With the development of three-dimensional (3D) spheroid models, image cytometry may provide the necessary tools and applications for CAR-T cell therapy discovery geared towards solid tumors [
      • Kessel S.
      • Cribbes S.
      • Déry O.
      • et al.
      High-Throughput 3D tumor spheroid screening method for cancer drug discovery using celigo image cytometry.
      ,
      • Cribbes S.
      • Kessel S.
      • McMenemy S.
      • et al.
      A novel multiparametric drug-scoring method for high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time apoptosis and viability high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time viability and apoptosis kinetic detection method of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Mukundan S.
      • Bell J.
      • Teryek M.
      • et al.
      Automated assessment of cancer drug efficacy on breast tumor spheroids in Aggrewell™400 Plates Using Image Cytometry.
      ,
      • Gonzalez-Exposito R.
      • Semiannikova M.
      • Griffiths B.
      • et al.
      CEA expression heterogeneity and plasticityconfer resistance to the CEA-targetingbispecific immunotherapy antibodycibisatamab (CEA-TCB) in patient-derivedcolorectal cancer organoids.
      ,
      • Grote S.
      • Chan K.C.-H.
      • Baden C.
      • et al.
      CD276 as a novel CAR NK-92 therapeutic target for neuroblastoma.
      ].
      Herein, we demonstrate the ability of image cytometry to perform cytotoxicity assays for prostate-specific membrane antigen (PSMA) CAR-T cells while simultaneously monitoring T cells. PSMA is a transmembrane glycoprotein with well-characterized expression in healthy prostate tissue and is known to be heavily upregulated in cancer tissue [
      • Paschalis A.
      • Sheehan B.
      • Riisnaes R.
      • et al.
      Prostate-specific Membrane Antigen Heterogeneity and DNA Repair Defects in Prostate Cancer.
      ]. First, cytotoxicity of PSMA CAR-T cells was measured using PC3-PSMA+GFP+ tumor spheroids. The specificity of the PSMA CAR-T cells was investigated by using an off-target cell line (MCF7-GFP+) as well as un-transduced (UTD) T cells. Next, PSMA CAR-T cells and UTD T cells were stained with CellTrace Far-Red to track the T cells during treatment. The results from each experiment show that PSMA CAR-T cells are specifically cytotoxic to PC3-PSMA+GFP+ tumor spheroids, and that CAR-T cells can be tracked using image cytometry during these experiments without affecting the cytotoxic functions. Utilizing the method outlined, it is possible to look at cytotoxicity of CAR-T cell therapy while monitoring the location of CAR-T cells for in vitro experimentations. The proposed image cytometry method can overcome limitations placed on traditional methodologies to effectively assess cell-mediated 3D tumor spheroid cytotoxicity and efficiently generate time- and dose-dependent results.

      Materials and methods

      Cell culture preparation

      PC3 (human prostate adenocarcinoma) cells stably expressing prostate specific membrane antigen (PSMA) and green fluorescent protein (GFP) were provided to the Tmunity company (Philadephia, PA) by the University of Pennsylvania. PC3 cells were cultured in RPMI 1640 media (Gibco, Grand Island, NY) with 10% FBS (Gibco) and 1% Penicillin-Streptomycin (Gibco) in TC-treated flasks at 37°C and 5% CO2. MCF7 (human breast adenocarcinoma) expressing GFP with puromycin antibiotic selection were acquired from GenTarget Inc. (San Diego, CA) and cultured in DMEM media (Gibco) with 10% FBS and 1% Penicillin-Streptomycin. DMEM was supplemented with puromycin (Gibco) at a concentration of 0.625 µg/mL. PC3 and MCF7 cultures were passaged every 3 – 4 days with a seeding density of 2 × 105 cells/mL to maintain cells in a logarithmic growth phase. The MCF7 cell line was measured and did not display PSMA antigen (Supplementary Figure 1).

      Preparation of PSMA CAR-T cells

      PSMA CAR-T cells were generated using cryopreserved leukapheresis from a healthy donor. Following leukapheresis thaw, CD4+ and CD8+ T cells were selected, activated, and the culture was initiated. The T cells were transduced with a lentiviral vector containing the PSMA CAR sequence and cultured for approximately 9 days. The PSMA CAR-T cells were harvested and cryopreserved using a controlled rate freezer for storage in the vapor phase of an LN2 freezer at a temperature ≤ -150°C. The general preparation of CAR-T cells have been described previously [
      • Schwab R.D.
      • Boesteanu A.C.;
      • et al.
      Engineered CAR T Cells Targeting the Cancer-Associated Tn-Glycoform of the Membrane Mucin MUC1 Control Adenocarcinoma.
      ].
      Cryopreserved PSMA CAR-T and UTD T cells were thawed approximately 24 hours prior to use in the assay, and subsequently centrifuged for 5 min at 300 x g. Following centrifugation, the cells were resuspended in RPMI 1640 media, and a small aliquot of each cell type was sampled to determine their respective cell concentration and viability. PSMA CAR-T and UTD T cell concentrations were adjusted to a total of 1 × 106 cells/mL and allowed to incubate overnight. After overnight incubation, the PSMA CAR-T and UTD T cells were again centrifuged for 5 min at 300 x g, where the cell concentrations were adjusted to 1 × 106 cells/mL prior to fluorescent labeling.

      CAR-T cell-mediated killing of 3D tumor spheroids

      To demonstrate the capability of image cytometry to evaluate CAR-T cell-mediated killing of 3D tumor spheroids, the cytotoxic effects of PSMA CAR-T and UTD T cells at various E:T ratios were compared. The 3D tumor spheroids were formed by seeding PC3-PSMA+GFP+ (antigen-expressing) or MCF7-GFP+ (negative control) cells into each well of Nexcelom3D 96-well ultra-low attachment treated round-bottom plates (ULA plates, Nexcelom) at 1 × 104 cells/well [
      • Kessel S.
      • Cribbes S.
      • Déry O.
      • et al.
      High-Throughput 3D tumor spheroid screening method for cancer drug discovery using celigo image cytometry.
      ,
      • Cribbes S.
      • Kessel S.
      • McMenemy S.
      • et al.
      A novel multiparametric drug-scoring method for high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time apoptosis and viability high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time viability and apoptosis kinetic detection method of 3D multicellular tumor spheroids using the celigo image cytometer.
      ]. Subsequently, the plates were centrifuged for 5 min at 150 x g. The PC3-PSMA+GFP+ and MCF7-GFP+ cells were then incubated at 37°C for 2 days to allow for tumor spheroid formation at one spheroid per well. After the 2-day spheroid formation, normalized number of effector cells were added to the plates at 10:1, 5:1, and 1:1 Effector (E):Target (T) ratios, including controls with target cells only. All conditions were tested in triplicate wells on three independently formed spheroids. The tumor spheroids were imaged using the image cytometer at 0, 24, 48, and 72 h post-addition of effector cells using bright field and green fluorescent channels (EX/EM: 483/536 nm) at an exposure time of ∼8,500 μs.
      The acquired images were analyzed using the “Zonal Tumorsphere 1 + 2 + Mask” application on the Celigo software. The green fluorescent intensity was utilized as the parameter to accurately assess the cell-mediated cytotoxicity effects of the effector cells. Under the application, the analysis parameters were set up as the following for the mask: Resolution (8 μm/pixel), Well Mask (100%), Colony Diameter (2000 μm), Precision (Low), Border Dilation (0), Minimum Thickness (2000 μm), Inner Radius (65%), Middle Radius (85%), Outer Radius (100%). The GFP mean fluorescence intensities of the inner radius were graphed over time for each tested condition using GraphPad Prism 9 (GraphPad Software, San Diego, CA).

      Monitoring of CAR-T cells during treatment

      To evaluate the ability of image cytometry to monitor the T cells during treatment, the PSMA-CAR and UTD T cells were labeled with CellTrace Far-Red (#C34564, Thermofisher, Waltham, MA) following manufacturer instructions. After labeling, the PSMA-CAR and UTD T cells were added to the plates at 10:1, 5:1, and 1:1 E:T ratios, including a negative control with target cells only. All conditions were tested in triplicate wells. The tumor spheroids were formed as described above. After spheroid formation, the tumor spheroids and T cells were imaged using the image cytometer at 0, 24, 48, and 72 h post-addition of fluorescently labeled effector cells using bright field, green (EX/EM 483/536 nm) and far red fluorescent channels (EX/EM: 628/688 nm) at an exposure times of ∼8,500 μs and ∼30,000 μs, respectively.
      The acquired images were analyzed using the “Zonal Tumorsphere 1 + 2 + Mask” application on the Celigo software. The green and far red fluorescent intensities were utilized as a parameter to measure cell-mediated cytotoxicity and to monitor the effector cells at 0, 24, 48, and 72 h, respectively. Under the application, the analysis parameters were set up as the following for the mask: Resolution (8 μm/pixel), Well Mask (100%), Colony Diameter (2000 μm), Precision (Low), Border Dilation (0), Minimum Thickness (2000 μm), Inner Radius (65%), Middle Radius (85%), Outer Radius (100%). The CellTrace Far-Red mean fluorescence intensities of the inner radius were graphed over time using GraphPad Prism 9.

      Celigo image cytometer

      The Celigo Image Cytometer (Nexcelom, Lawrence, MA) is a plate-based, high-throughput instrument that allows for rapid acquisition of whole-well images of tumor spheroids [
      • Kessel S.
      • Cribbes S.
      • Déry O.
      • et al.
      High-Throughput 3D tumor spheroid screening method for cancer drug discovery using celigo image cytometry.
      ,
      • Cribbes S.
      • Kessel S.
      • McMenemy S.
      • et al.
      A novel multiparametric drug-scoring method for high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time apoptosis and viability high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Kessel S.
      • Cribbes S.
      • Bonasu S.
      • et al.
      Real-time viability and apoptosis kinetic detection method of 3D multicellular tumor spheroids using the celigo image cytometer.
      ,
      • Mukundan S.
      • Bell J.
      • Teryek M.
      • et al.
      Automated assessment of cancer drug efficacy on breast tumor spheroids in Aggrewell™400 Plates Using Image Cytometry.
      ]. The system utilizes one brightfield and four fluorescent channels to capture images for analysis: blue (EX: 377/50 nm, EM: 470/22 nm), green (EX: 483/32 nm, EM: 536/40 nm), red (EX: 531/40 nm, EM: 629/53 nm), and far red (EX: 628/40 nm, EM: 688/31 nm). The image cytometer can rapidly acquire and analyze simultaneously in a 96-well ULA plate less than 3 min per plate. In this work, the “Zonal Tumorsphere 1 + 2 + Mask” application in Celigo software (Nexcelom) was used to acquire and analyze the GFP (green channel) and CellTrace Far-Red (far red channel) mean fluorescence intensities of tumor spheroids and T cells, respectively. This application allows for the acquisition of images in the bright field channel as well as two different fluorescent channels to perform high-throughput analysis of 3D tumor spheroids simultaneously. Most importantly, this application allows for the analysis in specific areas surrounding or within a tumor spheroid – inner ring, middle ring, and outer ring radius. The GFP and CellTrace Far-Red fluorescent signals in the inner and outer rings of the tumor spheroids were exported to Excel for downstream analysis.

      Results

      PSMA CAR-T cells induced specific cytotoxic effects in PC3 tumor spheroids

      Image cytometry was employed to assess the cytotoxicity effects of PSMA CAR-T and UTD T cells on PC3-PSMA+GFP+ and MCF7-GFP+ tumor spheroids 96-well ULA plates. The cytotoxic potency and specificity of the effector cells were tested at E:T ratios 10:1, 5:1, and 1:1 E:T ratios and monitored by measuring the average GFP mean fluorescent intensity of the PC3-PSMA+GFP+ and MCF7-GFP+ spheroids from 0 to 72 h in 24-hour intervals. It is important to note that the average mean fluorescent intensity represented the averages of the mean fluorescent intensity of the three replicate spheroids.
      The GFP fluorescent images of the PC3-PSMA+GFP+ tumor spheroids are shown in Fig. 1, which showed high cytotoxic potency of PSMA CAR-T cells against the PC3-PSMA+GFP+ spheroids (Fig. 1 Left), and no noticeable effects with UTD T cells (Fig. 1 Right). The control with target cells only showed a consistent spheroid diameter at ∼1189 μm over 72 h. In the PSMA CAR-T cell treatment, both E:T ratio- and time-dependent showed GFP fluorescence decrease demonstrating cytotoxicity effects, whereas the UTD T cell treatment showed consistent GFP fluorescence in all conditions. In Fig. 2, the GFP fluorescent intensities were quantified over time at each E:T ratio (Fig. 2 Top). The PSMA CAR-T cells demonstrated a high level of cytotoxicity against PC3-PSMA+GFP+ spheroids, with the 10:1 and 5:1 E:T ratios showing a 66% and 45% reduction in GFP fluorescence intensity at 24 hours after treatment, respectively. The lowest E:T ratio of 1:1 exhibits a delay in reduction of GFP fluorescence with a 40% reduction by 48 hours post-addition of PSMA CAR-T cells, while the GFP fluorescence of the 10:1 and 5:1 E:T ratios further decreased by an additional 22% and 41%, respectively. By the 72-h time point, all three E:T ratios for the PSMA CAR-T cell treatment resulted in significant reduction of GFP signals (10:1 ≅ 90% reduction; 5:1 ≅ 89% reduction; 1:1 ≅ 78% reduction). In contrast, the negative control treatments with UTD T cells did not show any significant change in the GFP fluorescence intensities, demonstrating that the engineering of PSMA CAR-T effectively allowed for targeting the PSMA antigen-expressing PC3-GFP+ tumor spheroids.
      Fig 1
      Fig. 1GFP fluorescent images of PSMA CAR-T and UTD T cells treated PC3-PSMA+GFP+ tumor spheroids. (Left) PSMA CAR-T cells specifically targeted and induced cytotoxicity in PC3-PSMA+GFP+ spheroids. (Right) UTD T cells did not induce cytotoxicity for all conditions and time points. The “No Treatment” images in the panel are the same sample.
      Fig 2
      Fig. 2E:T ratio- and time-dependent GFP fluorescent intensities for PSMA CAR-T and UTD T cells on (Top) PC3-PSMA+GFP+ and (Bottom) MCF7-GFP+ tumor spheroids.
      The specificity of PSMA CAR-T cell-mediated cytotoxicity was assessed by comparing the MCF7-GFP+ to PC3-PSMA+GFP+ tumor spheroids. Similarly, we examined the cytotoxic effects induced by the PSMA-CAR and UTD T cells at 10:1, 5:1, and 1:1 E:T ratios over the 72 h time-course. The GFP fluorescent images of the MCF7-GFP+ tumor spheroids are shown in Fig. 3, which showed no noticeable cytotoxic effects from both PSMA CAR-T and UTD T cells. All the conditions showed consistent spheroid diameter at ∼486 μm over 72 h. In Fig. 2, MCF7-GFP+ spheroids did not show a reduction in GFP fluorescence when treated with PSMA CAR-T and UTD T cells at all E:T ratios and time points (Fig. 2 Bottom).
      Fig 3
      Fig. 3GFP fluorescent images of PSMA CAR-T and UTD T cells treated MCF7-GFP+ tumor spheroids. Both (Left) PSMA CAR-T and (Right) UTD T cells did not induce cytotoxicity for all conditions and time points.

      Monitoring CAR-T cells during co-culture cytotoxicity assay

      Monitoring CAR-T cells during co-culture cytotoxicity assays can be challenging because fluorescent labeling of CAR-T cells may affect their functionalities [
      • Muirhead K.A.
      • Pan Dalin
      • M.; M.E.
      • et al.
      Tracking immune cell proliferation and cytotoxic potential using flow cytometry.
      ]. In order to determine whether the effector cells could be monitored without affecting their cytotoxic function, the cell-mediated cytotoxicity of PC3-PSMA+GFP+ spheroids were evaluated using the PSMA CAR-T and UTD T cells with or without staining with the CellTrace Far-Red.
      The bright field and far red fluorescent overlay images of PC3-PSMA+GFP+ spheroids are shown in Fig. 4, which showed migration of PSMA CAR-T cells to the center of the spheroids (Fig. 4 Left). In contrast, the UTD T cells seemed to spread out across the spheroid (Fig. 4 Right). The cross-section of the far red fluorescence was measured with ImageJ and shown in Supplementary Fig. 2. The morphology of the spheroids displayed in bright field also showed noticeable differences between the PSMA CAR-T and UTD T cells, where the cells in the PSMA CAR-T cell treated spheroid became dark with high density, while the UTD T cell treated spheroid remained bright and loosely packed.
      Fig 4
      Fig. 4Bright field and CellTrace Far-Red fluorescent overlay images of (Left) PSMA CAR-T and (Right) UTD T cells treated PC3-PSMA+GFP+ tumor spheroids.
      The time- and E:T ratio-dependent GFP fluorescent intensity results are shown in Fig. 5, where the CellTrace Far-Red did not have noticeable effects on the cytotoxic function of PSMA CAR-T cells (Fig. 5 Top) or their ability to specifically target PC3-PSMA+GFP+ tumor spheroids when compared with UTD T cells (Fig. 5 Bottom). The GFP fluorescence showed expected reductions of 89%, 87%, and 74% for the 10:1, 5:1, and 1:1 E:T ratios at 72 h time point, respectively. In comparison to the unstained CAR-T cells, the differences were 1%, 2%, and 4% for the 10:1, 5:1, and 1:1 E:T ratios, respectively. Interestingly, larger differences in the GFP fluorescent intensities were observed at the 24 h time point for the 10:1 and 5:1 E:T ratios, where the stained PSMA CAR-T cells showed reductions in GFP fluorescence of 56% and 30%, respectively. The bright field and far red fluorescent overlay images for MCF7-GFP+ spheroids and intensity results are shown in Supplementary Fig. 3 and 4.
      Fig 5
      Fig. 5E:T ratio- and time-dependent GFP fluorescent intensities for CellTrace Far-Red stained or unstained (Top) PSMA CAR-T and (Bottom) UTD T cells on PC3-PSMA+GFP+ tumor spheroids.
      The measured far red fluorescence in the outer ring of the PC3-PSMA+GFP+ spheroids showed higher reduction for the PSMA CAR-T in comparison to the UTD T cells. In contrast, the MCF7-GFP+ spheroids showed no noticeable far red fluorescence reduction in neither PSMA CAR-T nor the UTD T cells treatments.

      Discussion

      The current challenges presented in the field of cell and gene therapy have increased the need to develop new and novel detection methods for high-throughput phenotypic and functional assays, advanced co-culture assays, and 3D models for investigating tumor microenvironment. The technique of forming a single spheroid in the ultra-low attachment plates enabled the representation of the 3D spatialization of the TME, which can be utilized in a high-throughput screening format for large libraries of drug compounds or immunological factors. In this study, we have demonstrated the detection of cytotoxic potency and specificity of PSMA CAR-T cells on 3D tumor spheroid models, as well as simultaneously monitoring the CAR-T cells using a high-throughput image cytometry method. In comparison to the conventional methods such as the release assays (Cr51, calcein, LDH), MTT assay, luciferase, or flow cytometry, image cytometry may be utilized to investigate multiple parameters throughout CAR-T cell discovery for solid tumors [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Sun Y.-J.
      • Chen Y.-C.
      • Hua W.-K.
      • et al.
      Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ,
      • Singleton C.S.
      • Chan L.L.-Y.
      • McCulley K.J.
      • et al.
      ER+ Breast Cancer Mammosphere Formation and Analysis.
      ]. Parameters such as tumor spheroid counts, morphology, size, viability, and fluorescence intensities can be generated. Furthermore, image cytometry allows for kinetic assays, instead of the conventional end-point assays [
      • Wang Y.
      • Chan L.L.-Y.
      • Grimaud M.
      • et al.
      High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
      ,
      • Maldini C.R.
      • Love A.C.
      • Tosh K.W.
      • et al.
      Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
      ].
      In order to demonstrate the ability to detect cytotoxic potency and specificity of CAR-T cell-mediated killing for 3D tumor spheroids, we compared the cytotoxic effects induced by PSMA CAR-T and UTD T cells on PC3-PSMA+GFP+ and MCF7-GFP+ spheroids. The PSMA CAR-T cells showed clear time- and E:T ratio-dependent killing of the PC3-PSMA+GFP+ spheroids. In contrast, no cytotoxicity was observed when UTD T cells or non-antigen expressing MCF7-GFP+ spheroids were used. Taken together with the demonstrated cytotoxicity effects observed in this experiment, the finding suggests that PSMA CAR-T cell treatment can specifically target PC3-PSMA+GFP+ tumor spheroids (Fig. 2).
      In addition to assessing potency and specificity of the PSMA CAR-T cells on PC3-PSMA+GFP+ spheroid model, we hypothesized the use of a long-term tracer dye can be used to monitor the approximate location of the CAR-T cells during the cell-mediated cytotoxicity assay. However, it was important to ensure that the CellTrace Far-Red dye did not affect the cytotoxic function of the PSMA CAR-T cells. As such, we compared cytotoxicity effects between stained and unstained CAR-T cells on PC3-PSMA+GFP+ spheroids. All of the E:T ratios and time points showed no differences between the stained and unstained CAR-T cells, except the 24 h time point for 10:1 and 5:1 ratios. The unstained PSMA CAR-T cells did show approximately 10 – 15% improvement in cytotoxicity effects, which may not be concerning. The differences observed at the 24-h time point suggest that the tracer dye may initially affect cytotoxicity; however, the cytotoxic affect is restored by 48 hours following treatment with the difference in GFP fluorescent intensity being reduced to 1% and 3% for the 10:1 and 5:1 E:T ratios, respectively (Fig. 5).
      Interestingly, we were able to visually observe the location of the stained T cells during the cytotoxicity assay, where the PSMA CAR-T cells seemed to surround the tumor spheroids and then mostly migrated to the center, while the UTD T cells seemed to remain on the outside of the spheroids (Fig. 4). In addition, both PSMA CAR-T and UTD T cells may be migrating into the PC3-PSMA+GFP+ spheroids as evidenced by a decrease in far red fluorescence in the outer ring of the spheroids over the course of experiment. Other hypotheses could be the stimulation and proliferation of T cells or cell death, which can also reduce the cytoplasmic far red fluorescence. The PSMA CAR-T and UTD T cells exhibited a 50% and 36% reduction in far red signal, respectively, for the 10:1 ratio at the 48-h time point (Fig. 6). However, the same PSMA CAR-T and UTD T cells did not exhibit the same decrease over time for the MCF7 spheroids (Supplementary Fig. 3), which could be due to the complete lack of PSMA antigen expression eliminating T cell proliferation/migration and ultimately cell death (Supplementary Fig. 1). Since there is no interaction between the T cells and MCF7 spheroids, they can remain settling and resting in the round bottom well. This method may allow researchers to gain an initial understanding of tumor microenvironment when investigating various CAR constructs.
      Fig 6
      Fig. 6E:T ratio- and time-dependent CellTrace Far-Red fluorescent intensities for stained or unstained PSMA CAR-T and UTD T cells on (Top) PC3-PSMA+GFP+ and (Bottom) MCF7-GFP+ tumor spheroids.
      The acquired bright field images using image cytometry can provide qualitative morphological information for the 3D co-culture cytotoxicity assay. Although a relatively similar reduction in far red signals were measured for the stained PSMA CAR-T and UTD T cells, we observed an obvious spheroid morphological difference between the PSMA CAR-T and UTD T cells treatments. The PSMA CAR-T cell treated PC3-PSMA+GFP+ spheroids showed an increase in circularity over the treatment time for 5:1 and 10:1 ratios. The brightness also reduced over time potentially due to an increase in dead cell density, whereas the UTD T cells treated spheroids showed consistent morphology with loose density of cells for all conditions and time points. Therefore, we hypothesize that the UTD T cells may be infiltrating the spheroids but were not cytotoxic. Although the decrease in signal on the outer ring of the spheroids may indicate T cell infiltration, further investigation is needed to determine if this is indeed the case. To confirm T cell infiltration, we propose a future comparison study using flow cytometry to analyze digested spheroids to determine the amount of T cells within the spheroids.
      In the last decade, in vitro assays for CAR-T cell therapy discovery have mainly been performed using 2D tumor cell model, however, it is critical to study the potency and specificity of CAR constructs against 3D tumor spheroid models that can better recapitulate the 3D spatial features and organizations of the solid tumors in cancer patients. The proposed plate-based image cytometry method can provide a robust and high-throughput method to characterize CAR-T cells treatment using a 3D spheroid model, which may improve the efficiency CAR-T cell discovery for treatment of solid tumors, and more rapidly identify more suitable CAR construct candidates for downstream process.

      Declaration of Competing Interest

      The authors SP, CH, ACL, BL, and LLC declare competing financial interests. The CAR-T cell-mediated cytotoxicity methodology using 3D tumor spheroid model in this manuscript was developed using the Celigo Image cytometer from Nexcelom from PerkinElmer.

      Acknowledgment

      We would like to thank Tmunity Therapeutics for the kind gift of the PSMA CAR for assay purposes.

      Appendix. Supplementary materials

      References

        • Wang Y.
        • Chan L.L.-Y.
        • Grimaud M.
        • et al.
        High-throughput image cytometry detection method for CAR-T transduction, cell proliferation, and cytotoxicity assays.
        Cytometry Part A. 2020; https://doi.org/10.1002/cyto.a.24267
        • Sun Y.-J.
        • Chen Y.-C.
        • Hua W.-K.
        • et al.
        Comparison of chimeric antigen receptor-T cell-mediated cytotoxicity assays with suspension tumor cells using plate-based image cytometry method.
        Cytometry Part A. 2022;
        • Zheng P.-P.
        • Kros J.M.
        • Li J.
        Approved CAR T cell therapies: ice bucket challenges on glaring safety risks and long-term impacts.
        Drug Discovery Today. 2018; 23: 1175-1182
        • Albinger N.
        • Hartmann J.
        • Ullrich E.
        Current status and perspective of CAR-T and CAR-NK cell therapy trials in Germany.
        Gene Ther. 2021; 28: 513-527
        • Cockroft A.
        • Wilson A.
        Comparability: what we can learn from the review of advanced therapy medicinal products.
        Regenerat Med. 2021; 16: 655-667
        • Han D.
        • Xu Z.
        • Zhuang Y.
        • et al.
        Current Progress in CAR-T Cell therapy for hematological malignancies.
        J Cancer. 2021; 12: 326-334
        • Huang W.
        • Li J.
        • Liao M.Z.
        • et al.
        Clinical Pharmacology Perspectives for Adoptive Cell Therapies in Oncology.
        Clin Pharmacol Ther. 2021; 0: 1-14
        • Borogovac A.
        • Keruakous A.
        • Bycko M.
        • et al.
        Safety and feasibility of outpatient chimeric antigen receptor (CAR) T-cell therapy: experience from a tertiary care center.
        Bone Marrow Transplant. 2022; 57: 1025-1027
        • Maude S.L.
        • Frey N.
        • Shaw P.A.
        • et al.
        Chimeric antigen receptor T cells for sustained remissions in leukemia.
        N Engl J Med. 2014; 371: 1507-1517
        • Srivastava S.
        • Riddell S.R.
        Engineering CAR-T cells: Design concepts.
        Trends Immunol. 2015; 36: 494-502
        • Zhu Y.
        • Tan Y.
        • Ou R.
        • et al.
        Anti-CD19 chimeric antigen receptor-modified T cells for B-cell malignancies: a systematic review of efficacy and safety in clinical trials.
        Eur J Haematol. 2016; 96: 389-396
        • Sterner R.C.
        • Sterner R.M.
        CAR-T cell therapy: current limitations and potential strategies.
        Blood Cancer J. 2021; 11: 1-11
        • Barbosa M.A.G.
        • Xavier C.P.R.
        • Pereira R.F.
        • et al.
        3D Cell Culture Models as Recapitulators of the Tumor Microenvironment for the Screening of Anti-Cancer Drugs.
        Cancers. 2022; 14: 1-30
        • Li X.
        • Feng Y.
        • Shang F.
        • et al.
        Characterization of the therapeutic effects of novel chimeric antigen receptor T cells targeting CD38 on Multiple Myeloma.
        Front Oncol. 2021; 11: 1-10
        • Maldini C.R.
        • Love A.C.
        • Tosh K.W.
        • et al.
        Characterization of CAR T cell expansion and cytotoxic potential during Ex Vivo manufacturing using image-based cytometry.
        J Immunol Methods. 2020;
        • Milone M.C.
        • Bhoj V.G.
        The Pharmacology of T cell therapies.
        Mol Therapy Methods Clin Develop. 2018; 8: 210-221
        • Kessel S.
        • Cribbes S.
        • Déry O.
        • et al.
        High-Throughput 3D tumor spheroid screening method for cancer drug discovery using celigo image cytometry.
        J Laboratory Automation. 2016; 22: 454-465
        • Cribbes S.
        • Kessel S.
        • McMenemy S.
        • et al.
        A novel multiparametric drug-scoring method for high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
        SLAS Discovery. 2017; 22: 547-557
        • Kessel S.
        • Cribbes S.
        • Bonasu S.
        • et al.
        Real-time apoptosis and viability high-throughput screening of 3D multicellular tumor spheroids using the celigo image cytometer.
        SLAS Discovery. 2017; : 1-9https://doi.org/10.1177/2472555217731076
        • Kessel S.
        • Cribbes S.
        • Bonasu S.
        • et al.
        Real-time viability and apoptosis kinetic detection method of 3D multicellular tumor spheroids using the celigo image cytometer.
        Cytometry Part A. 2017; https://doi.org/10.1002/cyto.a.23143
        • Mukundan S.
        • Bell J.
        • Teryek M.
        • et al.
        Automated assessment of cancer drug efficacy on breast tumor spheroids in Aggrewell™400 Plates Using Image Cytometry.
        J Fluoresc. 2022;
        • Gonzalez-Exposito R.
        • Semiannikova M.
        • Griffiths B.
        • et al.
        CEA expression heterogeneity and plasticityconfer resistance to the CEA-targetingbispecific immunotherapy antibodycibisatamab (CEA-TCB) in patient-derivedcolorectal cancer organoids.
        J Immunother Cancer. 2019; 7: 1-14
        • Grote S.
        • Chan K.C.-H.
        • Baden C.
        • et al.
        CD276 as a novel CAR NK-92 therapeutic target for neuroblastoma.
        Adv Cell Gene Therapy. 2021; : 4
        • Paschalis A.
        • Sheehan B.
        • Riisnaes R.
        • et al.
        Prostate-specific Membrane Antigen Heterogeneity and DNA Repair Defects in Prostate Cancer.
        Eur Urol. 2019; 76: 469-478
        • Schwab R.D.
        • Boesteanu A.C.;
        • et al.
        Engineered CAR T Cells Targeting the Cancer-Associated Tn-Glycoform of the Membrane Mucin MUC1 Control Adenocarcinoma.
        Immunity. 2016; 44 (Jr., A. D. P.): 1444-1454
      1. (Jr., J. D. T.)
        • Muirhead K.A.
        • Pan Dalin
        • M.; M.E.
        • et al.
        Tracking immune cell proliferation and cytotoxic potential using flow cytometry.
        Flow Cytometry Protocols; Methods in Molecular Biology. Humana Press, 2010: 119-164
        • Singleton C.S.
        • Chan L.L.-Y.
        • McCulley K.J.
        • et al.
        ER+ Breast Cancer Mammosphere Formation and Analysis.
        in: Valle L.D. Immunohistochemistry and Immunocytochemistry. 2022: 233-245 (Methods in Molecular Biology; Walker, J. M., Series Ed.Chapter 16)