Tumor organoid – T cell co-culture systems

Abstract

T cells are a key player in cancer immunotherapy, but strategies to expand tumor-reactive cells and study their interaction with tumor cells at the level of an individual patient are limited. This protocol describes the generation and functional assessment of tumor-reactive T cells based on co-cultures of tumor organoids and autologous peripheral blood lymphocytes. The procedure consists of an initial co-culture of two weeks in which tumor-reactive T cells are first expanded in the presence of (IFNγ-stimulated) autologous tumor cells. Subsequently, T cells may be evaluated for their capacity to carry out effector functions (IFNγ secretion, degranulation) upon recognition of tumor cells, and their capacity to kill tumor organoids. This strategy is unique in its use of peripheral blood as a source of tumor-reactive T cells in an antigen-agnostic manner. In two weeks, tumor-reactive CD8⁺ T cell populations can be obtained for ~33-50% of non-small cell lung cancer (NSCLC) and microsatellite instable (MSI) colorectal cancer (CRC) patient samples. It thereby allows the establishment of ex vivo test systems for T cell-based immunotherapy at the level of the individual patient.

Introduction

T cell-directed therapies, such as immune checkpoint inhibition and adoptive T cell transfer, have transformed the treatment landscape for solid tumors. Nonetheless, clinical efficacy is variable both between and within tumor types and the mechanisms that tumors use to escape elimination by the immune system are diverse^1,2,3,4. Pre-clinical models that incorporate both endogenous T cells and tumor cells are scarce, especially for epithelial cancers. Moreover, the expansion of tumor infiltrating lymphocytes (TIL) has been more challenging for epithelial tumors compared to melanoma⁵.

We have recently developed a co-culture system of tumor organoids and peripheral blood lymphocytes ⁶. Tumor-reactive T cells can be efficiently expanded from peripheral blood and evaluated for tumor reactivity and tumoricidal capacity. Here, we extend and detail the procedure for establishment of this co-culture system.

Applications of the protocol

We anticipate that this protocol will be primarily applied as an investigational tool in three different settings: (i) as a tool to evaluate the efficacy of investigational drugs in the immuno-oncology space; (ii) for translational studies for specific patients of interest, e.g. an exceptional responder within a clinical trial; or (iii) to identify or validate mechanisms of resistance in a physiologically relevant setting. As an example, samples could be obtained from patients who initially respond to immune checkpoint inhibition but then progress⁷. Hypotheses for potential resistance mechanisms of those patients that are generated based on more descriptive data sets (e.g. tumor DNA sequencing) may be validated using tumor organoids derived from pre- and post-therapy samples.

In addition to these research applications, this method offers a potential strategy for the generation of T cell products for adoptive T cell transfer. The clinical success of TIL therapy in melanoma has not been matched in other tumor types and there is a need for strategies that bypass the challenges of TIL therapy for epithelial cancers.

Comparison with other approaches

Several other strategies to generate tumor-reactive T cell products have been developed. First, expansion of tumor-infiltrating lymphocytes (TIL) has been used for many years to generate T cell products for adoptive T cell transfer⁸. This strategy has been successfully applied for the treatment of metastatic melanoma, but clinical experience in other tumor types has remained limited (reviewed in ref. 9). An advantage of using TIL as a source of tumor-reactive T cells is the straightforward protocol and relatively short time required to generate a cell product. Moreover, while it has recently become clear that a substantial fraction of TIL do not show tumor specificity^10,11, T cells present at the tumor site will nevertheless be enriched for tumor-specific cells relative to peripheral blood. In addition, strategies for the more selective expansion of tumor-reactive T cells from TIL could be conceived, either by pre-selecting T cell populations enriched for tumor reactivity^12,13,14, or by stimulation with tumor cells (e.g. organoids). However, there are also limitations to this technology: the absolute number of TIL may be too low in sparsely infiltrated tumors, which is particularly problematic when a tumor biopsy (rather than a surgical resection) is the only available material. Finally, how circulating tumor-reactive T cells and TIL compare with respect to fitness and TCR repertoire, and whether most tumor-expressed (neo-) antigens have already induced a tumor-resident T cell population, is presently unclear.

As a second strategy, neo-antigen-specific T cells have been generated by activating circulating T cells with antigen presenting cells (APCs) that present (a) neo-antigen(s) of interest. Such strategies can be successful, especially when a pre-selected population of T cells (e.g. PD1⁺) is used¹⁵. Neo-antigens that are loaded onto APCs can be predicted based on exome and RNA sequencing data of the tumor¹⁶. Although prediction pipelines continue to improve, they are presently imperfect (especially for less common HLA haplotypes) and there is a risk of missing relevant antigens¹⁷. Alternatively, mini-genes encoding potential neo-antigens can be introduced into APCs and used to screen T cells, including healthy donor T cells, for reactivity^15,18. This bypasses the need for antigen prediction but is a laborious approach, certainly in tumor types with a high mutational load, such as mismatch repair-deficient tumors. Instead, the use of tumor cells as a substrate for selection of tumor-reactive T cells is antigen-agnostic and can result in the establishment of a polyclonal tumor-reactive T cell population¹⁹. While the turnaround time of sequencing and antigen prediction may be similar to the time required to establish organoids, the cost of sequencing and the need for bio-informatic analysis can be a disadvantage. Importantly, the use of tumor organoids does not only allow expansion of tumor-reactive T cells, but also the analysis of the activity of the resulting T cells against matched tumor cells.

Other approaches that co-culture organoids or organotypic cultures with allogeneic or autologous lymphocytes have been described^20,21,22,23. These approaches typically make use of immune cells isolated from tumor fragments and are a very valuable tool to study immune cells from their native environment, in particular in their interaction with cells that are difficult to culture, such as macrophages. However, expansion of T cells and/or tumor cells is usually limited, and cryopreservation not always possible, which limits their use for approaches that require high numbers of T cells or repeated experiments.

Limitations

Tumor and healthy organoids can be established from both needle biopsies and surgical resections. The success rate depends on tumor type, amount of starting material (e.g. resection versus biopsy) and treatment history. Typically, organoid culture from resections of treatment-naïve primary tumors is more successful than from biopsies from heavily pre-treated metastases. For example, the success rate of establishing organoids from resection specimens of untreated primary colorectal cancer (CRC) is ~90%; from a biopsy of a metastasis, however the success rate ranges from ~70% for CRC to ~35% for non-small cell lung cancer^24,25,26,27.The time required to establish a frozen biobank of at least 1*10⁶ organoids takes on average 30 to 80 days for CRC or non-small cell lung cancer (NSCLC) samples, respectively (Fig. 1b-c). Because of this variability, the feasibility and cost-effectiveness of an organoid-based approach for both research purposes and for adoptive T cell transfer is expected to differ between tumor types and patient populations and should be carefully evaluated.

Figure 1. Overview of the procedure.

(a) Tumor organoids are established from tumor tissue and PBMC are isolated from peripheral blood prior to start of co-culture. Organoids are isolated from Geltrex 2 days prior to co-culture and stimulated with IFNγ 1 day prior to co-culture. On the day of co-culture, organoids are dissociated to single cells and plated together with PBMC on an anti-CD28-coated plate, in the presence of IL-2 and anti-PD-1. After 1 week of co-culture, PBMC are re-stimulated with tumor cells. After 2 weeks of co-culture with autologous tumor organoids (pink), T cells can be cryopreserved (yellow), T cell reactivity against tumor cells is evaluated (green), or a tumor organoid killing assay is performed (blue). (b) Time from start of organoid culture to establishment of a frozen biobank. (c) Passage number of organoids at time of freezing. (d) Percentage CD3⁺ T cells in PBMC after two weeks of co-culture. Horizontal bars in (b) and (c) indicate average, error bar s.e.m. Each symbol is an organoid culture from a separate tumor. NSCLC biopsy: n = 11; NSCLC resection: n = 13; CRC biopsy: n = 15; CRC resection: n = 25.

The evaluation of CD4⁺ T cell reactivity is complicated by the potential for recognition of non-tumor epitopes present in the culture system (e.g. Geltrex-derived epitopes)⁶. The current protocol describes an adaptation to our earlier method that serves to prevent the development of such undesired T cell responses. Further evaluation is required to determine the frequency with which true tumor-specific CD4⁺ T cell responses can be induced using this optimized strategy.

Experimental design

Overview

The following step-by-step procedure describes how to generate CD8⁺ T cells that recognize and kill autologous tumor organoids (Fig. 1a). In the first part, we describe how to perform co-cultures of peripheral blood lymphocytes and tumor cells, as well as methods to assess the reactivity of the resulting T cell populations against tumor cells at the end of the procedure. In the second part we focus on methods to test whether the obtained T cell populations are able to kill the tumor organoids.

Co-culture and reactivity assay

Before starting T cell – organoid co-cultures, human peripheral blood mononuclear cells (PBMC) are isolated from blood, and tumor organoids are generated and cultured as described previously^24,27,28 and in steps 1-31 of this protocol. Tumor organoids are isolated from Geltrex (basement membrane matrix) to avoid the presence of non-human protein material in the cultures. PBMC are then stimulated weekly with dissociated autologous tumor organoids for a total time of 2 weeks. Tumor organoids are pre-stimulated with IFNγ to maximize antigen presentation. Co-cultures are performed in the presence of anti-PD1 to counteract the induction of PD-L1 by IFNγ, IL-2 to support T cell proliferation, and plate-bound anti-CD28 to provide co-stimulation. T cell reactivity is assessed by evaluating CD107a and IFNγ expression or CD137 expression in the presence or absence of tumor organoids.

Killing assay

The procedure describes how to test whether the T cell populations obtained after two weeks of co-culture are able to kill autologous tumor organoids. Tumor organoids are co-cultured for 72 hours in the presence or absence of T cells, with the addition of anti-PD1, plate-bound anti-CD28, Y-27632 hydrochloride (to prevent anoikis), and a caspase dye (NucView488). Killing can be assessed by live imaging or a flow cytometry-based quantification that allows analysis of the number of live tumor cells at the end of the experiment.

Controls

Evaluation of tumor reactivity is based on the quantification of different markers in the presence or absence of tumor organoids. As further controls to understand whether any observed reactivity is specific for the matched tumor cells the following additional controls should be considered when material is available: i) autologous healthy tissue organoids, to assess reactivity driven by any artifacts of organoid culture (prepared as described in Box 1), ii) co-culture with single-cell tumor digest, to test whether reactivity is also seen against tumor cells that have not been cultured in vitro, iii) co-culture with single cell digest of healthy tissue, to test for specificity for tumor as compared to healthy tissue.

Box 1. Establishment of healthy colon or airway organoids^25,26. * TIMING 2-3 h.

Procedure

1. Store tissue in Ad-DF+++ with 1X Primocin at 4 °C for no longer than 24 hours.

2. Remove the muscle layer and fat with forceps and razor blades.

3. Cut the tissue in small pieces.

4. Wash three times with complete chelation solution (CCS). Let the pieces sink to the bottom under normal gravity.

5. Add EDTA to a final concentration of 10 mM (200 uL of 0.5 M solution to 10 mL CCS).

6. Incubate 45 minutes at 4 °C in a rotating wheel and shake vigorously to liberate crypts. The solution turns cloudy when crypts are liberated. If crypts are not liberated, replace EDTA/CCS with fresh solution and repeat step 6.

7. Allow tissue fragments to settle. Transfer the supernatant containing the crypts to a new tube and pipette up and down 10 times.

8. Spin down the crypts (100g, 5’).

9. Continue with step 7 of the main protocol and culture similarly to tumor organoids. Use colon organoid medium for healthy colon organoids and NSCLC organoid medium for healthy airway organoids.

The killing assay is based on the quantification of the number of live tumor cells when organoids are cultured alone or in the presence of reactive T cells (for example at a 5:1 T cell:tumor cell ratio). The addition of T cells may lead to non-specific tumor cell death, e.g. due to competition for nutrients. For this reason, we consider the addition of a condition where tumor cells and T cells are co-cultured in the presence of MHC blocking antibodies essential.

Protocol

Materials

Reagents

• B27 supplement without vitamin A (Gibco, cat. no. C12587-010)

• B27 supplement (Gibco, cat. no. 17504-044)

• N-Acetylcysteine (Sigma-Aldrich, cat. no. A9165-5G)

• Nicotinamide (Sigma-Aldrich, cat. no. N0636)

• Human recombinant EGF (Peprotech, cat. no. AF-100-15)

• A83-01 (Tocris, cat. no. 2939)

• SB202190 (Cayman Chemicals, cat. no. 10010399)

• Prostaglandin E2 (Cayman Cehmicals, cat. no. 14010-1)

• Y-27632 dihydrochloride (Sigma-Aldrich, cat. no. Y-0503)

• Human recombinant FGF-7 (Peprotech, cat. no. 100-19)

• Human recombinant FGF-10 (Peprotech, cat. no. 100-26)

• R-spondin 1-conditioned medium produced by R-spondin 1 producing cell line (available from the C. Kuo laboratory; instructions for preparation in Box 1 of refs. 28^{, 29, 30})

• Noggin-conditioned medium produced by Noggin producing cell line (available from the H. Clevers laboratory; instructions for preparation in Box 3)

• Wnt-3a-conditioned medium produced by L-Wnt-3a producing cell line (available from the H. Clevers laboratory; instructions for prerpatio nin Box 1 of refs. 29^{, 30})

• Geltrex LDEV-free reduced growth factor basement membrane (Gibco, cat. no. A1413202)

• PBS tablets (Gibco, cat. no. 18912-014)

• EDTA buffer, 0.5 M, disodium salt (Lonza, cat. no. 51234)

• FBS (Sigma-Aldrich, cat. no. F7524)

• Dimethyl sulfoxide (DMSO) (Sigma-Aldrich, cat. no. 34943)

• Bovine serum albumin fraction V (Sigma-Aldrich, cat. no. 10735078001)

• Dispase type II (Sigma-Aldrich, cat. no. D4693)

• Advanced DMEM-F12 (Gibco, cat. no. 12634-028)

• Penicillin/streptomycin (Gibco, cat. no. 15070063)

• Ultraglutamine type I (Lonza, cat. no. BE17-605E)

• HEPES (Gibco, cat. no. 15630-056)

• TrypLE Express (Gibco, cat. no. 12604-013)

• Trypsin-EDTA 0.05% (Gibco, cat. No. 25300054)

• Collagenase II (Sigma, cat. No. C6885)

• Hyaluronidase type IV-S (Sigma, cat. no. H4272)

• RPMI 1640 (Gibco, cat. no. 11875093)

• Human serum, from human male AB plasma (Sigma-Aldrich, cat. no. H3667)

• Benzonase (Merck, cat. no. 70746-3)

• Human recombinant interferon gamma (Peprotech, cat. no. 300-02)

• GolgiSTOP (Monensin) (BD, cat. no. 554724)

• GolgiPLUG (Brefeldin A) (BD, cat. no. 555029)

• Phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich, cat. no. 19-144)

• Ionomycin (Sigma-Aldrich, cat. no. I9657)

• Mouse anti-human CD28 (eBioscience,CD28.2, cat. no. 16-0289-81, RRID:AB_468926)

• Interleukin-2 (Proleukin, Novartis)

• Anti-human PD1 / nivolumab (Merus / Selleckchem, cat. no. A2002, RRID:AB_2810223)

• Mouse anti-human CD107a (PE-conjugated) (BD, clone H4A3, cat. no. 555801, RRID:AB_396135)

• Mouse anti-human CD3 (PerCP-Cy5.5-conjugated) (eBioscience, clone SK7, cat. no. 332771, RRID:AB_1907379)

• Mouse anti-human CD4 (FITC-conjugated) (BD, clone RPA-T4, cat. no. 555346, RRID:AB_395751)

• Mouse anti-human CD8 (BV421-conjugated) (BD, clone RPA-T8, cat. no. 562429, RRID:AB_11153676)

• Mouse anti-human IFNγ (APC-conjugated) (BD, clone B27, cat. no. 554702, RRID:AB_398580)

• Mouse anti-human CD137 (APC-conjugated) (BD, clone 4B4-1, cat. no. 550890, RRID:AB_398477)

• Mouse anti-human HLA-ABC (MHC-I blocking antibody) (ThermoFisher, clone W6/32, cat. no. MA1-19027, RRID:AB_1076699)

• Mouse anti-human HLA-DR/DP/DQ (sodium azide free) (MHC-II blocking antibody) (BD, clone Tü39, cat. no. 555556, RRID:AB_395938)

• Mouse anti-human CD3 (Alexa Fluor 700-conjugated) (Invitrogen, cat. No. CD0329, RRID:AB_1470210)

• Mouse anti-human CD326 (EpCAM) (PE/Cy7-conjugated) (Biolegend, cat. No. 324222, RRID:AB_2561506)

• Live/dead fixable near-IR dead cell stain kit (Invitrogen, cat. no. L10119)

• Fixation/Permeabilization solution kit (BD, cat. no. 554714)

• Anti-mouse Ig, κ/negative control compensation particles set (BD, cat. no. 552843)

• Ammonium chloride (Sigma-Aldrich, cat. no. A9434)

• Sodium bicarbonate (Sigma-Aldrich, cat. no. S5761)

• Primocin (Invivogen, cat. no. PML-40-60)

• Lympholyte H (Tebu-bio, cat. no. CL5020)

• CellTrace Yellow Cell Proliferation Kit (Invitrogen, cat. No. C34567)

• CellTrace Far Red Cell Proliferation Kit (Invitrofen, cat. No. C34564)

• NucView488 Caspase-3 assay kit (Biotium, cat. No. 30029)

• AccuCount blank particles (7.0-7.9 μm) (Spherotech, cat. No. ACBP-70-10)

• Trypan Blue Solution (ThermoFisher, cat. no. 15250061)

• DAPI

• NaH₂PO₄ (Sigma-Aldrich, cat. no. S5011)

• KH₂PO₄ (Sigma-Aldrich, cat. no. P5655)

• NaCl (Sigma-Aldrich, cat. no. S7653)

• KCl (Sigma-Aldrich, cat. no. P9333)

• D-sorbitol (Sigma-Aldrich, cat. no. S1876)

• Sucrose (Sigma-Aldrich, cat. no. S7903)

• DL-dithiotreitol (Sigma-Aldrich, cat. no. D0632)

Box 3. Production of Noggin-conditioned medium * TIMING 10 d.

Additional materials required

DMEM (Gibco, cat. no. 41966). Store at 4 °C for up to 1 year.

Geneticin (G418 sulfate, 50 mg/mL) (Gibco, cat. no. 10131035). Store at -20 °C in 500 μL aliquots for up to 1 year.

HEK293-mNoggin-Fc cells (can be obtained from Hans Clevers Laboratory, Utrecht). Store at -80 °C for up to 1 year or in liquid nitrogen, indefinitely.

HEK293 culture medium Supplement DMEM with 1% (vol/vol) penicillin/streptomycin, 1% (vol/vol) Ultraglutamine I, and 10% (vol/vol) FBS. Store at 4 °C for up to 1 month.

Selection medium Supplement 50 mL HEK293 culture medium with 500 μL of 50 mg/mL geneticin (final concentration: 500 μg/mL). Store at 4 °C for up to 1 month.

Procedure

1. Pre-heat HEK293 growing medium to 37 °C.

2. Thaw a vial of HEK293-mNoggin-Fc cells in a 37 °C waterbath until a small clump of ice remains.

3. Transfer the cells to a 15 mL Falcon tube and add up to 10 mL pre-heated HEK293 growing medium.

4. Centrifuge (300g, 5’, room temperature) and discard supernatant.

5. Resuspend cell pellet in 50 mL selection medium and transfer to a T-175 culture flask.

6. Culture cells until confluency with the T-175 flask in a flat position.

7. Aspirate medium and wash flask with 20 mL PBS.

8. Collect cells by trypsinizing for up to 5 minutes with 5 mL TrypLE Express at 37 °C.

9. Transfer cells to 50 mL Falcon tube and add up 50 mL PBS.

10. Centrifuge (300g, 5’, room temperature) and discard supernatant.

11. Resuspend cells in HEK293 growing medium and split into 6x T-175 flasks. Culture 5 flasks in

    50 mL HEK293 growing medium and one flask in 50 mL selection medium.

12. When cells reach confluency, replace medium with 50 mL Ad-DF+++.

13. Harvest conditioned medium after one week and centrifuge (300g, 5’, room temperature) to pellet cells.

14. Collect supernatant and filter through 0.2 μm filter unit.

15. Store supernatant in aliquots of 50 mL at -20 °C. Conditioned medium can be stored for up to 6 months at -20 °C and for up to 2 weeks at 4 °C.

16. Use the cells cultured in the single flask with selection medium to repeat the procedure, or freeze the cells in small aliquots for later use.

    CRITICAL STEP Cells can be used for up to 12 passages to harvest conditioned medium.

Equipment

• Falcon tubes, 15 mL (Sarstedt, cat. no. 62.554.502)

• Falcon tubes, 50 mL (Sarstedt, cat. no. 62.547.254)

• Plates, 6 well (Greiner, cat. no. 657165)

• Plates, 96 well, U-bottom (Greiner, cat. no. 650180)

• Plates, 24 well (Greiner, cat. no. 662160)

• Plates, 96 well, Flat bottom, Non-Tissue Culture Treated (Falcon, cat. no. 351172)

• Parafilm (Bemis, cat. no. PM-992)

• Hemocytometer (Labor Optik, cat. no. 1910000)

• Nunc Cryovials (ThermoFisher, cat. no. 375418)

• No 21 sterile carbon steel surgical scalpel blade (Swann-Morton, cat. no. 0207)

• Microlance needles 20G x 1 1/2”, yellow (BD, cat. no. 301300)

• Culture dish, 100 x 20 mm (Greiner, cat. no. 664160)

• Flow cytometer (BD, Fortessa 1)

• -80 °C freezer (Panasonic, MOF-C8V1-PE)

• Mr Frosty freezing container (ThermoFisher, 5100-0036)

• Light microscope (Zeiss, Axiovert 25)

• 0.2 μm filter unit (Whatman, cat. no. 10462200)

• Syringe, 50 mL (Braun, cat. no. 8728844F-06)

• Water bath (37 °C)

• CO₂ incubator (5% CO₂, 37 °C)

• Centrifuge for 15- and 50-mL Falcon tubes and culture plates

• Biosafety cabinet

• Widefield microscope, recording PhaseContrast in combination with fluorescence and equipped with an AutoFocus system (Definite Focus – DF2), suitable for live cells

• ZeissZen Pro

• FIJI

• U-100 sterile insulin syringes (BD Micro-Fine, cat. no. 324891)

• Combitips advanced 5 mL (Eppendorf, cat. no. 0030089669)

• Combitips advanced 0.5 mL (Eppendorf, cat. no. 0030089634)

• Multipette E3x (Eppendorf, cat. no. 4987000029)

Reagent setup

Human material We have good experience with the use of cryopreserved human PBMC, stored at -80 °C. Tumor tissue can be obtained via surgical resection or a biopsy.

The biopsy tissue can be stored in Ad-DF+++ with 1X Primocin at 4 °C for no longer than 24 hours prior to starting the procedure.

! CAUTION Informed consent must be obtained for the use of human material. Ethical approval must be

obtained by and studies must conform to relevant institutional and national authorities.

! CAUTION It is strongly advised to authenticate human tissue (PBMC, organoid lines) using STR profiling or SNParrays and regularly test for contamination with Mycoplasma.

Ad-DF+++ Supplement Advanced-DMEM/F12 with 1% (vol/vol) penicillin/streptomycin, 1% (vol/vol) HEPES, and 1% (vol/vol) Ultraglutamine I. Store at 4 °C for up to 6 months.

Collagenase type II Dissolve 150 mg collagenase type II in 10 mL PBS + 0.2% (wt/vol) BSA for a 15 mg/mL stock. Store in 1 mL aliquots at -20 °C for up to 1 year.

Hyaluronidase Type IV-S Dissolve 30 mg hyaluronidase type IV-S in 3 mL distilled water to get a 10 mg/mL stock. Store in 1 mL aliquots at -20 °C for up to 1 year.

Wnt3a-conditioned medium Production of Wnt3a-conditioned medium has been described in Box 1 of refs. 29^,30. Wnt3a-conditioned medium can be stored at 4 °C for up to 6 months.

R-spondin 1-conditioned medium Production of R-spondin 1-conditioned medium has been described in Box 1 of refs. 28^,29,30. R-spondin 1-conditioned medium can be stored at -20 °C for up to 6 months or at 4 °C for up to 2 weeks.

Noggin-conditioned medium Production of Noggin-conditioned medium is described in Box 3. Noggin-conditioned medium can be stored in 50 mL aliquots at -20 °C for up to 6 months and for up to 2 weeks at 4 °C.

N-Acetylcysteine dissolve 40.79 g into 500 mL PBS to have a 500 mM stock solution. Store at -20 °C for up to 1 year.

Nicotinamide Dissolve 61.6 g into 500 mL PBS to have a 1 M stock solution. Store at -20 °C for up to 1 year.

Human recombinant EGF Dissolve 1 mg into 2 mL 0.1% (wt/vol) BSA/PBS to have a 500 μg/mL stock solution. Store at -20°C for up to 1 year. Do not refreeze; aliquots of 20 μL are recommended.

A83-01 Dissolve 1.25 μg into 50 mL DMSO to have a 500 μM stock solution. Store at -20 °C for up to 1 year.

SB202190 Dissolve 25 mg into 7.546 mL DMSO to have a 10 mM stock solution. Store at -20 °C for up to 1 year.

Prostaglandine E2 Dissolve 1 mg into 28.36 mL to have a 100 μM stock solution. Store at -20 °C for up to 1 year.

Y-27632 dihydrochloride Dissolve 32 mg into 10 mL DMSO to have a 10 mM stock solution. Store at -20 °C for up to 1 year.

Human recombinant FGF-7 Dissolve 0.1 mg into 2 mL 0.1% (wt/vol) BSA/PBS to have a 50 μg/mL stock solution. Store at -20 °C for up to 1 year. Do not refreeze; aliquots of 100 μL are recommended.

Human recombinant FGF-10 Dissolve 0.2 mg into 2 mL 0.1% (wt/vol) BSA/PBS to have a 100 μg/mL stock solution. Store at -20 °C for up to 1 year. Do not refreeze; aliquots of 100 μL are recommended.

Colorectal cancer organoid culture medium To prepare the medium, combine 5 mL Noggin-conditioned medium, 10 mL R-spondin 1-conditioned medium, 1 mL B27 minus vitamin A, 500 μL Nicotinamide (1 M in PBS), 125 μL N-Acetylcysteïne (500 mM in PBS), 5 μL EGF (500 μg/mL in PBS + 0.1% (wt/vol) BSA), 5 μL PGE2 (100 μM in PBS), 50 μL A83-01 (500 μM in DMSO), and 15 μL SB202190 (10 mM in DMSO). Add 5 μL Y-27632 (10 mM in DMSO) only for the first 3 days after thawing and passaging. Fill up to 50 mL with Ad-DF+++. Store at 4 °C for up to 2 weeks.

Colon organoid culture medium Same as colorectal cancer organoid medium, with an additional 25 mL of Wnt-3a-conditioned medium. Store at 4 °C for up to 2 weeks.

Non-small cell lung cancer and healthy airway epithelial medium To prepare the medium, combine 5 mL Noggin-conditioned medium, 5 mL R-spondin 1-conditioned medium, 1 mL B27, 500 μL Nicotinamide (1 M in PBS), 125 μL N-Acetylcysteïne (500 mM in PBS), 6.25 μL FGF-7 (50 μg/mL in PBS + 0.1% (wt/vol) BSA), 12.5 μL FGF-10 (100 μg/mL in PBS + 0.1% (wt/vol) BSA), 50 μL A83-01 (500 μM in DMSO), 5 μL SB202190 (10 mM in DMSO), and 25 μL Y-27632 (10 mM in DMSO). Fill up to 50 mL with Ad-DF+++. Store at 4 °C for up to 2 weeks.

T cell culture medium Supplement RPMI 1640 with 1% (vol/vol) penicillin/streptomycin, 1% (vol/vol) Ultraglutamine I, and 10% (vol/vol) human serum. Store at 4 °C for up to 1 month.

T cell thawing medium medium Supplement RPMI 1640 with 1% (vol/vol) penicillin/streptomycin, 1% (vol/vol) Ultraglutamine I, and 10% (vol/vol) FBS. Store at 4 °C for up to 1 month.

FACS buffer Supplement 500 mL PBS with 5 mL EDTA (0.5 M) and 5 mg BSA. Store at 4 °C for up to 6 months.

Freezing medium Combine 45 mL FBS and 5 mL DMSO. Store at 4 °C for up to 6 months.

Non-small cell lung cancer digestion buffer Combine 1 mL collagenase type II (15 mg/mL in PBS + 0.2% (wt/vol) BSA), 500 μL dispase type II (10 mg/mL in ddH₂O), 20 μL Primocin, 10 μL Y-27632 (10 mM in DMSO) and 8.5 mL PBS. Use directly after preparation.

Colorectal cancer digestion buffer Combine 1 mL collagenase type II (15 mg/mL in PBS + 0.2% (wt/vol) BSA), 10 μL hyaluronidase type IV (10 mg/mL in ddH₂O), 20 μL Primocin, 10 μL Y-27632 (10 mM in DMSO) and 8.5 mL PBS. Use directly after preparation.

Red blood cell lysis buffer Add 8.26 g ammonium chloride, 1 g sodium bicarbonate and 200 μL EDTA (0.5 M) to 1 L of ddH₂O. Stir for 30 minutes. Pass through 0.2 μm filter to sterilize. Store at 4 °C for up to 6 months.

NaH₂PO₄ Dissolve 28.39 g NaH₂PO₄ in 200 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

KH₂PO₄ Dissolve 27.22 g KH₂PO₄ in 200 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

NaCl Dissolve 58.44 g NaCl in 200 mL distilled water for a 5 M stock. Store at room temperature for up to 1 year.

KCl Dissolve 14.91 g KCl in 200 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

D-sorbitol Dissolve 36.44 g D-sorbitol in in 200 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

Sucrose Dissolve 68.46 g sucrose in 200 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

DL-dithiotreitol Dissolve 1.54 g DL-dithiotreitol in 10 mL distilled water for a 1 M stock. Store at room temperature for up to 1 year.

Complete chelation solution Prepare 200 mL complete chelation solution by combining 1.12 mL of 1 M Na₂HPO₄ (for final concentration 5.6 mM), 1.6 mL of 1 M KH₂PO₄ (for final concentration 8.0 mM), 23.86 mL of 5 M NaCl (for final concentration 96.2 mM), 320 μL of 1 M KCl (for final concentration 1.6 mM), 6.68 mL of 1 M sucrose (for final concentration 43.4 mM) and 10.98 mL of 1 M D-sorbitol (for final concentration 54.9 mM) and 153.44 mL cold distilled water. Adjust to pH 7.0-7.3. Add 100 μL of 1 M DL-dithiothreitol (for final concentration 0.5 mM). Filter through 0.22 μm filter to sterilize. Store at 4 °C for maximum 1 day.

Procedure

Establishment of tumor organoid culture from biopsy or resection material^24,25,26 * TIMING: 2-3 hours

• 1)Dissect tissue with razor blade or needles on Petri dish to cubes of 1-2 mm³.
• 2)Transfer tissue fragments to 15 mL Falcon tube and wash once in PBS. Centrifuge at 200g, for 5 minutes, at RT.
• 3)Discard supernatant and incubate tissue fragments for 15-60’ at 37 °C in 10 mL colorectal or nonsmall cell lung cancer digestion buffer. Resuspend 10-15 times every 10 minutes and check the status of the digestion by evaluating using a microscope whether the tissue has dissociated into single cells.

  △ CRITICAL STEP The total incubation time can vary depending on specimen size and tumor type: generally it can range from 15 to 30 minutes for biopsies, and from 45 to 60 minutes for resections.

• 4)Block the reaction by adding 2 mL of FBS when the tissue has dissociated so that >70% are single cells, or after 30 minutes for biopsies or 60 minutes for resections.
• 5)Pellet cells by centrifugation at 300g, 5’, RT.
• 6)If the pellet is red, lyse erythrocytes by incubating for 5’ in 5 mL red blood cell lysis buffer at room temperature. After lysis, fill up the 15 mL Falcon tube with PBS.

  △ CRITICAL STEP This is not necessary if the pellet is not red.

• 7)Wash once more in PBS, centrifuge at 300g, 5’, RT and discard the supernatant.
• 8)Count cells using a hemocytometer or automated cell counter.
• 9)Resuspend the pellet of cells in ice-cold complete organoid medium (roughly 1.5-3 * 10⁴ cells/10 μL medium), and add 2 times the same volume of Geltrex.

  △ CRITICAL STEP Geltrex needs to be kept ice cold at all times.

• 10)Plate digest/Geltrex drops in a 6-well culture plate (roughly 20 μL per drop).
• 11)Place the plate upside-down in a 37 °C incubator for 30 minutes, to let the Geltrex solidify.
• 12)Place plate in upright position and add complete organoid medium, supplemented with 1x Primocin.

  △ CRITICAL STEP Add Primocin to all medium until a frozen biobank is stored (step 20).

• 13)Replace medium with complete organoid medium twice per week.

  Passaging organoids * TIMING: 1 hours

  △ CRITICAL STEP Organoids can be passaged once every 1-2 weeks at a 1:2-1:6 split ratio. Split organoids when confluent or when their diameter exceeds 300 μm.

• 14)Remove complete organoid medium.
• 15)Resuspend organoid/Geltrex drops in pre-warmed TrypLE Express (1 mL/well).
• 16)Incubate at 37 °C for 5-15 minutes, resuspending with a p1000 pipette every 5 minutes.

  △ CRITICAL STEP Avoid overtrypsinization. Check under microscope when organoids are dissociated into single cells: a mix of 70% single cells and 30% clusters of 7-10 cells is ideal.

• 17)Transfer TrypLE mix into a 15 mL Falcon tube.
• 18)Inhibit TrypLE by diluting 5-10 fold with cold Ad-DF+++.
• 19)Spin 300g, 5’ and remove supernatant.
• 20)Repeat step 9-13.

  Freezing organoids * TIMING: 30 minutes

  △ CRITICAL STEP Organoids are typically frozen after expansion to >5*10⁵ organoids, which typically requires 5-11 weeks of culture with 2-4 passages (Fig. 1b-c).

  △ CRITICAL STEP Not all cells need to be frozen. If you so wish, retain some cells, these are ready for use in step 44.

• 21)Remove complete organoid medium and resuspend in ice cold Ad-DF+++ (1 mL Ad-DF+++ per well)

  △ CRITICAL STEP Freezing (and thawing) steps are most efficient if organoids are frozen 3-5 days after passaging.

• 22)Transfer the suspension into a 15 mL Falcon tube and fill up tube with Ad-DF+++.
• 23)Spin 300g, 5’, 4 °C.
• 24)Resuspend the pellet in ice cold 10% (vol/vol) DMSO/FCS. Use 1 mL of freezing medium per 5*10⁴ organoids.
• 25)Add 1 mL of the freezing mix per cryovial and quickly place the cryovials into a freezing container (e.g. Mr Frosty). Store at -80 °C overnight.

  △ CRITICAL STEP Keep freezing medium ice cold during freezing and place the freezing container into a -80 °C freezer without delay.

• 26)Transfer cryovials to liquid nitrogen for long-term storage.

  □ PAUSE POINT Cells can be stored in liquid nitrogen indefinitely.

  Thawing organoids * TIMING: 1 hour

• 27)Pre-heat 10 mL Ad-DF+++ at 37 °C.
• 28)Quickly thaw cryovials containing organoids in a 37 °C water bath.

  △ CRITICAL STEP Do not completely thaw the frozen cells: a small clump of ice should remain when the vial is removed from the water bath.

• 29)Quickly transfer the content of the vial into a 15 mL Falcon tube and add 10 mL pre-warmed Ad-DF+++ in a dropwise manner.
• 30)Spin 300g, 5’, room temperature. Remove supernatant.
• 31)Plate as described in steps 9-13.

  Isolation of human peripheral blood mononuclear cells from blood * TIMING: 2 h

• 32)Prepare 0.1% (wt/vol) BSA in PBS (PBSA). Pass through 0.2 μm filter to sterilize.
• 33)Dispense blood into 50 mL Falcon tubes (15 mL per tube).
• 34)Add equal volume of 0.1% PBSA to each tube.
• 35)Transfer to 50 mL tubes containing 15 mL Lympholyte H. Lock the two tubes in a horizontal position and very carefully tilt the tubes so that a slow flow of blood starts and forms a layer on top of the Ficoll.
• 36)Spin 15’, 1200g, no break, no acceleration, room temperature.
• 37)There will be a bottom layer of red blood cells, followed by a thin layer of PBMC and then a layer of plasma. Remove plasma to reach the PBMC.

  △ CRITICAL STEP Minimize movement of tubes to maintain separation of the layers.

• 38)Collect PBMC in a separate tube by carefully pipetting up the cells from the layer.
• 39)Wash 2x with 0.1% PBSA (between washes spin at 1,000 g, 7’, RT and discard the supernatant).
• 40)Wash once with PBSA followed by a spin at 200g, 7’, RT to remove platelets. Discard supernatant.
• 41)If necessary, remove erythrocytes by incubating for 5 min in red blood cell lysis buffer at 37 °C. Wash again with PBSA and centrifuge (200g, 7’, RT).
• 42)Count cells using hemocytometer or automated cell counter.
• 43)Cryopreserve cells for later use (as described in step 81A).

  Organoid isolation for co-culture of tumor cells and peripheral blood mononuclear cells (day -2) * TIMING: 30 min

• 44)Aspirate medium from 6-well plate with organoids (containing 1-5*10⁴ organoids per well, from step 12). Use one well for every 2 * 10⁶ PBMC to be used for co-culture in step 71.
• 45)Add 1 mL/well of pre-heated (37 °C) dispase (2 mg/mL in PBS).

  △ CRITICAL STEP Dispase treatment allows isolation of organoids from Geltrex. This step is essential to prevent recognition of Geltrex-derived epitopes by CD4⁺ T cells.

• 46)Resuspend organoids by gently pipetting up and down with a P1000 pipette.
• 47)Incubate for 15’ at 37 °C.
• 48)Transfer organoid suspension to a 15 mL Falcon tube.
• 49)Add 100 μL EDTA (0.5 M) for every 1 mL of dispase used and fill up tube up till 10 mL with PBS.
• 50)Pellet organoids (300g, 5’, RT) and aspirate supernatant.
• 51)Resuspend organoids in complete organoid culture medium and plate 2-4 mL/well of a tissue culture-treated 6-well plate. Use 2 mL for every well used in step 44.

  △ CRITICAL STEP Add 10 μM Y-27632 dihydrochloride to prevent cell death due to detachment from Geltrex.

• 52)Culture organoids for 24h at 37 °C.

  Co-culture preparation (day -1) * TIMING: 1 h

• 53)Add 200 ng/mL IFNγ to organoids to enhance antigen presentation.
• 54)Coat tissue culture-treated 96-well U-bottom plate with 5 μg/mL anti-CD28 in PBS (50 μL/well). Wrap plate in parafilm and incubate for 24 h at 4 °C. Anti-CD28 provides co-stimulatory signals during co-culture.
• 55)Quickly thaw cryopreserved PBMC at 37 °C.
• 56)Transfer PBMC to 15 mL Falcon tube and add up to 15 mL pre-heated (37 °C) T cell thawing medium in a drop-wise manner.

  △ CRITICAL STEP Cell loss during thawing can be up to 50%. Pre-heating of medium is essential for good recovery upon thawing.

• 57)Pellet cells (200g, 12’, RT, intermediate deceleration).
• 58)Remove supernatant and incubate for 15’ at 37 °C in 5 mL T cell thawing medium with 1:1,000 benzonase.

  △ CRITICAL STEP Benzonase prevents cell clumping due to DNA released by dying cells. This is particularly important when thawing PBMC.

• 59)Fill up tube with T cell thawing medium and pellet cells (200g, 12’, RT, intermediate deceleration). Discard supernatant.
• 60)Resuspend cells at 2 * 10⁶/mL in T cell culture medium and add 150 u/mL of IL-2 and incubate overnight. To prevent adhesion of myeloid cells to the culture dish, leave cells in a 15 mL Falcon tube with a slightly opened lid wrapped in parafilm to allow gas exchange. Do not add >4 mL/tube to ensure sufficient oxygenation.

  Co-culture (day 0) * TIMING: 14 d

• 61)Collect medium containing organoids in suspension that have been incubated with IFNγ overnight and spin down (300g, 5’, RT). Remove supernatant.
• 62)Resuspend pellet in 1 mL/well TrypLE and combine with remaining cells that have adhered to the culture plate.
• 63)Incubate for 5-15’ at 37 °C. Resuspend every few minutes and check using a microscope if organoids or big clusters of cells are still present. Stop when they have dissociated into single cells.

  △ CRITICAL STEP Take care to prevent overtrypsinization. Some small cell clusters are acceptable. ? TROUBLESHOOTING

• 64)Transfer dissociated organoids to 15 mL Falcon tube and fill up with PBS.
• 65)Pellet dissociated organoids (300g, 5’, RT).
• 66)Remove supernatant and resuspend pellet in 1 mL T cell culture medium.
• 67)Count cells using a hemocytometer.

  △ CRITICAL STEP Resuspend well before taking a count sample to prevent dissociated organoids from sinking to the bottom of the tube.

• 68)Resuspend dissociated organoids at 5 * 10⁴ cells/mL in T cell culture medium.
• 69)Resuspend PBMC from step 60 and count using hemocytometer or automated cell counter.
• 70)Wash PBMC in PBS (500g, 5’, RT).
• 71)Resuspend PBMC at 1 * 10⁶ cells/mL in T cell culture medium, supplemented with 300 U/mL IL-2 (2x concentrated) and 40 μg/ml anti-PD1 (2x concentrated).
• 72)Mix equal volume of dissociated organoids and PBMC for a PBMC:tumor cell ratio of 20:1.
• 73)Wash anti-CD28-coated plate 2x with PBS. Do not leave plate dry.
• 74)Plate 200 μL dissociated organoid / PBMC suspension per well and incubate at 37 °C.
• 75)Refresh or split co-culture every 2-3 days. If medium does not turn yellow after 2-3 days, follow option A. If medium turns yellow, follow option B.
  • (A) Refreshing co-culture
    • (i)Gently remove 100 μL of culture medium per well.
    • (ii)Add 100 μL T cell culture medium, supplemented with 300 U/mL IL-2 (2x concentrated) and 40 μg/ml anti-PD1 (2x concentrated).
    • (B) Splitting co-culture
    • (i)Split cells 1:1 by resuspending cells and transferring 50% of volume to a new well. Note that there is no need to separate tumor cells from T cells at this stage.
    • (ii)Add 100 μL T cell culture medium, supplemented with 300 U/mL IL-2 (2x concentrated) and 40 μg/ml anti-PD1 (2x concentrated).

      Re-stimulation with tumor cells (day 7)

• 76)After 5 days of co-culture, repeat steps 44-52.
• 77)After 6 days of co-culture, repeat steps 53-54.
• 78)After 7 days of co-culture, resuspend co-cultured PBMC and transfer to 15 mL Falcon tube. Note that there is no need to separate tumor cells from PBMC at this stage; resuspension is sufficient to enrich for PBMC over tumor cells, since most tumor cells either adhered to the plastic, or have died during co-culture.
• 79)Repeat steps 61-75.
• 80)After 14 days of co-culture, collect PBMC and proceed with downstream assays.

  Note that after 2 weeks of co-culture, T cells should represent 90% of the total PBMC population (Fig. 1d).

Downstream assays

• 81)
  After co-culture, cells can be cryopreserved (option A), evaluated for tumor reactivity using IFNγ and CD107a as read-outs (option B), evaluated for tumor reactivity using CD137 as a read-out (option C), or can be used in tumor killing assays (option D). Typically, reactivity is assessed before cryopreservation, initially based on IFNγ and CD107a and if desired, confirmed using CD137 as a read-out.

  • (A)
    Cryopreservation of co-cultured T cells * TIMING 30 min

    • (i)
      Count T cells using hemocytometer or automated cell counter.
    • (ii)
      Pellet T cells (500 g, 5’, 4 °C).
    • (iii)
      Resuspend at 5-10 * 10⁶/mL in ice cold freezing medium.
    • (iv)
      Quickly transfer 1 mL/vial to cryovials.
    • (v)
      Freeze immediately in Mr Frosty or comparable freezing container at -80 °C.
      □ PAUSE POINT Cells can be stored at -80 °C for 1 month. For long-term storage, transfer cells to liquid nitrogen.
  • (B)
    Tumor reactivity assay using IFNγ and CD107a as a read-out * TIMING 4 d, 24h required for reactivity assay

    • (i)
      Two days before reactivity assay: repeat steps 44-52. If required for experimental design, prepare healthy tissue organoids as a control as described in Box 1 and processed as described above for tumour organoids.
    • (ii)
      One day before reactivity assay: repeat steps 53-54. If required for experimental design, prepare healthy tissue organoids as a control as described in Box 1 and processed as described above for tumour organoids.
    • (iii)
      Proceed with reactivity assay.
      Repeat steps 61-67.
    • (iv)
      Resuspend dissociated organoids at 1 * 10⁶ cells/mL in T cell culture medium.
    • (v)
      Count T cells using hemocytometer or automated cell counter.
    • (vi)
      Pellet T cells (500 g, 5’, 4 °C).
    • (vii)
      Resuspend T cells at 2 * 10⁶ cells/mL in T cell culture medium.
    • (viii)
      Prepare 3X anti-CD107a-PE/anti-PD1 master mix by adding 1:50 anti-CD107a-PE and 60 μg/mL anti-PD1 to T cell culture medium.
      CD107a expression at the cell surface is transient due to recycling of cytotoxic granules and
      therefore the antibody is added throughout the co-culture.
      △ CRITICAL STEP Keep anti-CD107a-APC protected from light.
    • (ix)
      Prepare 3X PMA/Ionomycin solution by adding 150 ng/mL PMA and 3 μg/mL ionomycin to T cell culture medium.
    • (x)
      If using MHC-blocked organoids as controls, incubate dissociated organoids with 150 μg/mL of mouse anti-human HLA-ABC, clone W6/32, to block MHC-I and/or 30 μg/mL of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block MHC-II for 30’ at 37 °C.
    • (xi)
      Wash anti-CD28-coated (step 54) tissue culture-treated 96-well U-bottom plate 2x with PBS.
      △ CRITICAL STEP Do not leave plate dry for >5 minutes.
      We chose to use anti-CD28-coated plates also during the tumor reactivity assay to maintain co-stimulatory signals. If this is not desired, use a non-coated plate.
    • (xii)
      Plate 50 μL (= 1 * 10⁵) T cells per well, and add 50 μL of anti-CD107a-APC/anti-PD1 master mix per well.
    • (xiii)
      Add 50 μL (= 5 * 10⁴) tumor cells (test condition), 50 μL of T cell culture medium (negative control), or 50 μL of PMA/Ionomycin solution (positive control). Additional controls can include dissociated healthy organoids, tumor cells incubated with MHC-I/II blocking antibodies, from step (x), or single cell digests of tumor or healthy tissue (for preparation, follow steps 1-8).
    • (xiv)
      Mix well and spin (200g, 1’, RT) to loosely pellet cells.
    • (xv)
      Incubate for 1 h at 37 °C.
    • (xvi)
      Prepare 4x GolgiStop/GolgiPlug solution by adding 1:375 GolgiStop (BD) and 1:250 GolgiPlug (BD) to T cell culture medium.
    • (xvii)
      Add 50 μL GolgiStop/GolgiPlug to each well.
    • (xviii)
      Mix well and spin (200g, 1’, RT) to loosely pellet cells.
    • (xix)
      Incubate for 4 h at 37 °C.
    • (xx)
      Pellet cells (330g, 5’, 4 °C) and remove supernatant.
      △ CRITICAL STEP Work on ice from this step on and keep fluorochromes protected from light.
    • (xxi)
      Wash cells in 200 μL/well FACS buffer. Centrifuge at 330g, 5’, 4 °C and remove supernatant. (xxii) Prepare cell surface staining solution with 1:20 anti-CD3-PerCP-Cy5.5, 1:20 anti-CD4-FITC,
      1:200 anti-CD8-BV421 and 1:2,000 near-IR viability dye in FACS buffer.
    • (xxiii)
      Add 20 uL of the solution of step (xxii) to each well. For preparation of compensation controls, see Box 2.
    • (xxiv)
      Mix well without causing air bubbles.
    • (xxv)
      Incubate for 30’ at 4 °C, protected from light.
    • (xxvi)
      Wash 2x with 200 μL FACS buffer (after each wash spin down cells at 330g, 5’, 4 °C and then remove supernatant).
    • (xxvii)
      Fix cells in 80 μL Fix/Perm buffer (BD) on ice. Mix immediately.
    • (xxviii)
      Incubate for 20’ at 4 °C, protected from light.
      □PAUSE POINT After fixation, cells can be stored for 24h. Wash cells 1x with 200 μL FACS buffer to remove fixative and store in 20-100 μL FACS buffer. If desired, continue with the following steps the following day.
    • (xxix)
      Wash 2x with 200 μL Perm/Wash buffer (Fixation/Permeabilization kit, BD) (after each wash, spin down cells at 500g, 5’, 4 °C and remove supernatant).
    • (xxx)
      Prepare intracellular staining solution with 1:40 anti-IFNγ-APC in Perm/Wash buffer (Fixation/Permeabilization kit, BD).
    • (xxxi)
      Add 20 uL of the solution made in step (xxx) to each well.
    • (xxxii)
      Mix well without causing air bubbles.
    • (xxxiii)
      Incubate for 30’ at 4 °C, protected from light.
    • (xxxiv)
      Wash 2x with 200 μL Perm/Wash buffer (after each wash spin down at 500g, 5’, 4 °C and discard supernatant).
    • (xxxv)
      Resuspend cells in 50 -100 μL FACS buffer and record on a suitable flow cytometer.
      □PAUSE POINT It is possible to store the samples in 50-100 μL FACS buffer overnight at 4 °C in the dark and continue with the recording the morning after.
      Examples of flow cytometry plots are shown in Figure 2a.
      ? TROUBLESHOOTING
  • (C)
    Tumor reactivity assay using CD137 as a read-out * TIMING 4 d including 24h for reactivity assay

    • (i)
      Two days before reactivity assay: repeat steps 44-52. If required for experimental design, prepare healthy tissue organoids as a control as described in Box 1 and processed as described above for tumour organoids.
    • (ii)
      One day before reactivity assay: repeat steps 53-54. If required for experimental design, prepare healthy tissue organoids as a control as described in Box 1 and using the same steps as used for tumor organoids.
    • (iii)
      Proceed with reactivity assay by repeating steps 61-67.
    • (iv)
      Resuspend dissociated organoids at 5 * 10⁵ cells/mL in T cell culture medium.
    • (v)
      Count T cells using hemocytometer or automated cell counter.
    • (vi)
      Pellet T cells (500 g, 5’, 4 °C) and discard supernatant.
    • (vii)
      Resuspend T cells at 1 * 10⁶ cells/mL in T cell culture medium. Add 40 μg/mL anti-PD1 (2X final concentration).
    • (viii)
      Prepare 2X PMA/Ionomycin solution by adding 100 ng/mL PMA and 2 μg/mL ionomycin to T cell culture medium.
    • (ix)
      If using MHC-blocked organoids as controls, incubate dissociated organoids with 150 μg/mL of mouse anti-human HLA-ABC, clone W6/32, to block MHC-I and/or 30 μg/mL of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block MHC-II for 30’ at 37 °C.
    • (x)
      Wash anti-CD28-coated (step 54) tissue culture-treated 96-well U-bottom plate 2x with PBS.
      △ CRITICAL STEP Do not leave plate dry for >5 minutes.
      We chose to use anti-CD28-coated plates also during the tumor reactivity assay to maintain co-stimulatory signals. If this is not desired, use a non-coated plate.
    • (xi)
      Plate 100 μL (= 1 * 10⁵) T cells per well, and add 100 μL (=5 * 10⁴) tumor cells (test condition), 100 μL of T cell culture medium (negative control), or 100 μL of PMA/Ionomycin solution (positive control). Additional controls can include healthy organoids, organoids incubated with MHC-I/II blocking antibodies, or single cell digests of tumor or healthy tissue (for prepration, follow steps 1-8).
    • (xii)
      Mix well and spin (200g, 1’, RT) to loosely pellet cells.
    • (xiii)
      Incubate for 24 h at 37 °C.
    • (xiv)
      Pellet cells (330g, 5’, 4 °C) and remove supernatant.
      △ CRITICAL STEP Work on ice from this step on and keep fluorochromes protected from light. Wash cells in 200 μL/well FACS buffer (330g, 5’, 4 °C).
    • (xv)
      Prepare cell surface staining solution with 1:20 anti-CD3-PerCP-Cy5.5, 1:20 anti-CD4-FITC, 1:200 anti-CD8-BV421, 1:2,000 near-IR viability dye and 1:30 anti-CD137-APC in FACS buffer.
    • (xvi)
      Add 20 uL of the solution of step (xv) to each well. For preparation of compensation controls, see Box 2.
    • (xvii)
      Mix well without causing air bubbles.
    • (xviii)
      Incubate for 30’ at 4 °C, protected from light.
    • (xix)
      Wash 2x with 200 μL FACS buffer (after each wash, spin at 330g, 5’, 4 °C and discard supernatant).
    • (xx)
      Resuspend cells in 50 -100 μL FACS buffer and record on a suitable flow cytometer. Examples of flow cytometry plots are shown in Figure 2b.
      ? TROUBLESHOOTING
      □PAUSE POINT Cells can be rested for up to 4 hours before recording on a flow cytometer if kept at 4 °C, protected from light.
  • (D)
    Tumor organoid killing assay * TIMING 5 d

    • (i)
      Organoid isolation (day -2). Aspirate medium from 6-well plate with organoids.
      △ CRITICAL STEP Two wells of a 6-well plate (1-5*10⁴ organoids per well) is sufficient to perform the experiment.
    • (ii)
      Add 1 mL/well of pre-heated (37 °C) dispase (2 mg/mL in PBS).
    • (iii)
      Resuspend organoids by gently pipetting up and down with a P1000 pipette
    • (iv)
      Incubate for 5-15’ at 37 °C.
      △ CRITICAL STEP Take care to prevent overincubation with dispase II. Dispase can lead to dissociation of organoids into single cells for a small fraction of organoid lines. Monitor organoids closely, and when detachment of single cells is observed, stop the reaction and continue with the following step.
    • (v)
      Transfer organoid suspension to a 15 mL Falcon tube.
    • (vi)
      Add 100 μL EDTA (0.5 M) for every 1 mL of dispase used and fill up tube up till 10 mL with PBS.
    • (vii)
      Pellet organoids (300g, 5’, RT) and aspirate supernatant.
    • (viii)
      Resuspend organoids in complete organoid culture medium and plate 2-4 mL/well of a tissue culture-treated 6-well plate. Use 2 mL for every well used in step (i).
      △ CRITICAL STEP Add 10 μM Y-27632 dihydrochloride to prevent cell death due to detachment from Geltrex.
    • (ix)
      Culture organoids for 24h at 37 °C.
    • (x)
      Co-culture preparation (day -1). Add 200 ng/mL IFNγ to organoids to enhance antigen presentation.
    • (xi)
      Coat a non-tissue culture-treated 96-well flat-bottom plate with 5 μg/mL anti-CD28 in PBS (50 μL/well). Wrap plate in parafilm and incubate for 24 h at 4 °C. Anti-CD28 provides costimulatory signals during co-culture.
    • (xii)
      Killing assay (day 0). Collect organoids from the 6-well plate. Most organoids will either be in suspension or are loosely attached to the culture plate, and can be collected by resuspending the solution several times. Transfer the suspension of organoids and medium into a 15 mL tube. Split the total volume into two aliquots: 1/11 of the total volume will be used for dissociation into single cells and counting; 10/11 of the total volume will be used as fully formed organoids for the experiment.
      ? TROUBLESHOOTING
    • (xiii)
      Pellet the 1/11 aliquot (300g, 5’, RT). Resuspend the pellet in 500 uL TrypLE. Incubate 5-15’ minutes at 37 °C. Resuspend every few minutes and check if organoids have dissociated. Stop when they have dissociated into single cells.
      △ CRITICAL STEP Take care to prevent overtrypsinization. Some small cell clusters are acceptable.
    • (xiv)
      Count cells using a hemocytometer.
      △ CRITICAL STEP Resuspend well before taking a count sample to prevent dissociated organoids from sinking to the bottom of the tube.
    • (xv)
      The 10/11 aliquot will contain 10 times the number of cells that was counted in the 1/11 aliquot, but in the form of organoids (“single cell equivalents”)
    • (xvi)
      Wash organoids in PBS (300g, 5’, RT) to remove IFNγ (added in step 81D(x)) from the medium.
    • (xvii)
      If desired, label organoids with fluorescent dyes by resuspending organoids in 1 mL 1:1,000 CellTrace FarRed or 1:10,000 CellTrace Yellow (in PBS). If T cells are also labelled with a fluorescent dye, make sure to use different dyes for T cells and organoids. Mix immediately. Incubate 20’ at 37 °C, dark. Resuspend well every 5-10’. Block the labelling with 10 mL 10% (vol/vol) FBS/PBS (5’, RT). Wash twice with 10 mL PBS (300g, 5’, RT).
      △ CRITICAL STEP Organoids can be distinguished from T cells by labelling with fluorescent dyes as an alternative to staining with fluorescent antibodies. This may be preferred if separation by flow cytometry is difficult after staining with antibodies.
      △ CRITICAL STEP Before starting staining, take aside an aliquot of organoids to be used as unstained control during compensation (see Box 2).
      △ CRITICAL STEP Before continuing, take aside an aliquot of organoids to be used as single stained control during compensation (see Box 2).
    • (xviii)
      Resuspend organoids at 10⁵ single cell equivalents/mL in T cell culture medium. Add: anti-PD1 at 40 μg/mL concentration (2X final concentration); Y-27632 dihydrochloride at 20 μM concentration (2X final concentration); and 1:1,000 NucView488 caspase 3/7 dye (2X final concentration)
    • (xix)
      Block MHC molecules on tumor organoids by taking aside three aliquots and incubating for 30’ at 37 °C with the following: in the first, 50 μg/mL of mouse anti-human HLA-ABC, clone W6/32, to block MHC-I; in the second 10 μg/mL of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block MHC-II; and in the third 50 μg/mL of mouse anti-human HLA-ABC, clone
      W6/32, and 10 μg/mL of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block both MHC-I and MHC-II. Resuspend all three aliquots well every 5-10’.
    • (xx)
      Resuspend T cells and count using hemocytometer or automated cell counter.
    • (xxi)
      Pellet T cells (300g, 5’, RT)
    • (xxii)
      If required, label T cells with fluorescent dyes by resuspending T cells in 1 mL 1:10,000 CellTrace FarRed or 1:100,000 CellTrace Yellow (in PBS). If organoids are also labelled with a fluorescent dye, make sure to use different dyes for T cells and organoids.. Mix immediately. Incubate 20’at 37 °C, dark. Resuspend well every 5-10’. Block the labelling with 10 mL 10% (vol/vol) FBS/PBS (5’, RT). Wash twice with 10 mL PBS (300g, 5’, RT)
      △ CRITICAL STEP Before starting, take aside an aliquot of T cells to be used as unstained control during compensation (see Box 2).
      △ CRITICAL STEP Before continuing to the next step, take aside an aliquot of T cells to be used as single stained control during compensation (see Box 2).
    • (xxiii)
      Resuspend T cells at 5 * 10⁵ cells/mL in T cell culture medium.
      This results in a 5:1 T cell:tumor cell ratio. Other ratios can be used to measure dose dependent effects.
    • (xxiv)
      Block MHC molecules on T cells by taking aside three aliquots and incubating the first with 50 ug/mL of of mouse anti-human HLA-ABC, clone W6/32, to block MHC-I for 30’ at 37 °C; the second with 10 ug/mL of of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block MHC-II for 30’ at 37 °C; and the third with 50 ug/mL of mouse anti-human HLA-ABC, clone W6/32, and 10 μg/mL of mouse anti-human HLA-DR/DP/DQ, clone Tü39, to block both MHC-I and MHC-II for 30’ at 37 °C. Resuspend all three aliquots well every 5-10’.
    • (xxv)
      Wash the anti-CD28-coated plate twice with 200 μL PBS per well. Do not leave the wells dry. We chose to use anti-CD28-coated plates also during the tumor killing assay to maintain co-stimulatory signals. If this is not desired, use a non-coated plate.
    • (xxvi)
      Plate each of the following in triplicate: organoids alone: 100 μL (= 1 * 10⁴) tumor organoids + 100 μL T cell culture medium per well; organoids + T cells: 100 μL (= 1 * 10⁴) tumor organoids + 100 μL (= 5 * 10⁴) T cells per well; organoids + T cells + MHC-I block: 100 μL (= 1 * 10⁴) MHC-I blocked organoids + 100 μL (= 5 * 10⁴) MHC-I blocked T cells per well; organoids + T cells + MHC-II block: 100 μL (= 1 * 10⁴) MHC-II blocked organoids + 100 μL (= 5 * 10⁴) MHC-II blocked T cells per well; organoids + T cells + MHC-I/II block: 100 μL (= 1 * 10⁴) MHC-I/II blocked organoids + 100 (= 5 * 10⁴) μL MHC-I/II blocked T cells per well; unstained and single stained controls for compensation setup: 100 μL (= 1 * 10⁴) of unstained or single stained organoids + 100 μL T cell culture medium per well.
    • (xxvii)
      Take picture with regular microscope and incubate 72 h at 37 °C or proceed to next step with live cell imaging.
    • (xxviii)
      This and the following steps provide specific instructions for live cell imaging using ZeissPro imaging software. These can be taken as guidance for use with alternative software. General steps are indicated. First insert the plate in the appropriate location at the microscope. Remove lid, place the protection cover on top, and attach the CO₂ cover.
      △ CRITICAL STEP Keep the plate in the dark during the setup of the experiment
      △ CRITICAL STEP A widefield microscope, allowing recording of phase contrast in combination with fluorescense, suitable for live cells, and equipped with AutoFocus system is necessary for this step.
    • (xxix)
      For any software used, set magnification to 10X
    • (xxx)
      For any software used, calibrate plate and optimize exposure time. When using ZeissPro, load ZeissZenPro program, acquisition mode. Check all the following parameters:

      • -
        Camera: HDCam full 10X
      • -
        Carrier setup: 96 wells plate
      • -
        Calibration: 7-points method. Follow the instructions provided by the software.
      • -
        Focus strategy: absolute fixed. Check “use z-position from tiles set up”
      • -
        Tiles: create a 3x3 tile for each scene you want to record, fix the z position for each position
      • -
        Setup exposure time for bright-field and fluorochromes: ensure to maximize the signal without reaching saturation levels
      • -
        For any software used: set time series to desired frequency (30 minutes)
      • -
        Enable autosave to a specified folder
      • -
        Start experiment and record for 72 hours.
        △ CRITICAL STEP Ensure there will be enough storage space on the hard drive.
    • (xxxi)
      Killing quantification (day 3). Take endpoint pictures using regular microscope or finish live imaging.
    • (xxxii)
      Wash cells twice with 200 μL PBS per well to remove serum from the medium.
    • (xxxiii)
      Resuspend and transfer cell suspension into a 96-well U-bottom plate. Spin 500g, 5’, RT. Decant supernatant.
    • (xxxiv)
      During the spinning of the U-bottom plate, add 150 μL TrypLE per well to the empty 96-well flat-bottom plate, to be sure to detach any remaining cells from the bottom of the plate. Incubate 5’ at 37 °C.
    • (xxxv)
      Combine TrypLE from flat-bottom plate with U-bottom plate to dissociate both the organoids that were in suspension and those that attached to the plate.
    • (xxxvi)
      Incubate 5’-15’ at 37 °C. Resuspend well every 5’.
      △ CRITICAL STEP Carefully check very often under the microscope whether organoids have dissociated into single cells. When dissociated, immediately proceed to step (xxxvii).
      ? TROUBLESHOOTING
    • (xxxvii)
      After cells are dissociated, inhibit TrypLE with 50 μL FBS per well. In this case, FBS is used to inhibit TrypLE because the volume in the well is too low to sufficiently dilute and wash out TrypLE.
    • (xxxviii)
      Add 5 μL per well of counting beads.
      △ CRITICAL STEP Ensure the number of beads per well is the same. If available, use a repetitive automated pipette.
    • (xxxix)
      Pellet the cells (500g, 5’, RT). Discard supernatant.
    • (xl)
      Wash the cells twice with 200 μL PBS per well (after each wash spin at 500g, 5’, RT and then discard supernatant).
    • (xli)
      If cells have been stained with CellTrace dyes, skip this step and proceed directly to step 82D(xlii). Otherwise add 20 μL of staining mix containing 1:20 anti-CD3-AF700 and 1:200 anti-EpCAM-PE-Cy7 to each well. Incubate 30’, 4 °C, dark. Wash the plate twice with 200 μL FACS buffer per well.
    • (xlii)
      Resuspend each well in 50 μL FACS buffer.
    • (xliii)
      Add DAPI (1:50) prior to recording on flow cytometer.
    • (xliv)
      Prepare compensation controls as described in Box 2
    • (xlv)
      Record on flow cytometer.
      △ CRITICAL STEP It is very important to resuspend cells right before recording to prevent cells from settling on the bottom of the tube.
    • (xlvi)
      Calculate the amount of alive tumor cells per bead per well. An example of the gating strategy using antibodies or CellTrace dyes is shown in Fig. 4 and Suppl. Fig. 1, respectively.
      ? TROUBLESHOOTING
    • (xlvii)
      Optional: processing of live imaging data. Open the file with Zeiss ZenPro, processing mode.
      It will be a large file that has two dimensions: scenes (the well or condition) and time points (if one image was taken every 30 minutes for 72 hours, there will be 144 different time points).
      △ CRITICAL STEP Here, specific instructions are provided for ZeissPro imaging software. These can be taken as guidance for use with alternative software.
    • (xlviii)
      Use the function “split scenes” of the program to split scenes: instead of a large unique file, it will create several files (one per scene).
      △ CRITICAL STEP It is an important step to reduce the size of the file to be able to work with the recorded data.
    • (xlix)
      Create region of interest. If not all tiles are needed, it is possible to further reduce the size of each scene by creating a smaller area of interest. If using Zeiss ZenPro, use the function “create image subset”. Check the following parameters: All channels, to extract data on all the different channels used for the experiment; Extract all time points (unless you want to focus on a particular time frame; in that case, select time of interest); and rectangular region. Select and apply spatial rectangular region of interest.
      △ CRITICAL STEP It is possible to perform this step in batch, for all scenes at the same time. To do this select batch analysis on the program, and select all the different files of interest.
    • (l)
      Stitch tiles by using the function “Stitch” to fuse the 3x3 tiles in a unique file. Check “fuse tiles” and unselect the automatic function. In the “Channel to stitch” field select “bright field” because it is the channel that contains most of the data, making the stitch function more accurate. Choose to stitch multiple dimensions: c-dimension (channel) = reference, to stitch all the different channels in accordance with the bright field; and t-dimension (time) = reference, it is easier if a time point in the middle of experiment is selected (organoids may have moved in the field since the beginning of the experiment). Check and apply “create a new output”.
      △ CRITICAL STEP If you do not select a new output it will overwrite the original data.
      △ CRITICAL STEP It is possible to perform this step in batch, for all scenes at the same time. To do this, select batch analysis on the program, and select all the different files of interest.
    • (li)
      If you need to remove background, for example if for some of the fluorochromes the background signal is too intense, use background removal function for that channel (function available also in other programs, such as FiJi or Imaris). An example of the processing of live imaging data is shown in Fig. 5.
      ? TROUBLESHOOTING

Box 2. Preparation of compensation controls. * TIMING 30 min.

For compensation, cells (option A) or compensation beads (option B) can be used. If cells are used, ensure that the relevant antigen is expressed. If necessary, stimulate cells to increase antigen expression. Compensation beads can only be used for antibody-based stains and not for dyes. A combination of compensation beads (for antibody-based stains) and cells (for dyes) is acceptable.

Procedure

• (A)
  Preparation of compensation controls using cells

  1. Prepare single cell suspension of cells of interest and plate in a 96-well plate. The number of wells needed is equal to the number of different fluorochromes used in the staining panel plus one.
     △ CRITICAL STEP Work on ice from this step on and keep fluorochromes protected from light.
  2. Prepare single stain dilutions for every fluorochrome used in the staining panel, at the same dilution as used in the staining panel.
  3. Wash cells in FACS buffer. Spin down at 330g, 5’, 4 °C, discard supernatant.
  4. Add 20 μL of staining solution to separate wells for each stain. Leave one well unstained. For example, if the staining panel consists of anti-CD4-FITC and anti-CD8-BV421, stain one well with
     anti-CD4-FITC alone, one well with anti-CD8-BV421 alone, and leave one well unstained. Mix well, without causing air bubbles.
  5. Incubate for 30’, 4 °C, dark.
  6. Wash cells twice in FACS buffer. Spin down at 330g, 5’, 4 °C and discard supernatant.
  7. Resuspend cells in 20-100 μL FACS buffer and set up compensation on suitable flow cytometer. ? TROUBLESHOOTING
• (B)
  Preparation of compensation controls using compensation beads

  1. Vortex beads for 5 s.
  2. Mix 1-3 drops of negative and positive compensation beads.
  3. Pipette 20 μL of beads mixture into as many wells as antibody-fluorochrome conjugates are used in the staining panel
  4. Add 2 μL of antibody-fluorochrome conjugate per well. Use a different well for each different fluorochrome.
     △ CRITICAL STEP Work on ice from this step on and keep fluorochromes protected from light.
  5. Mix well by resuspending.
  6. Incubate 5’ – 60’ (4 °C, dark).
  7. Set up compensation on suitable flow cytometer.
     ? TROUBLESHOOTING

Figure 2.

Examples of flow cytometry plots showing tumor reactivity of CD8⁺ T cells. After two weeks of co-culture, T cells were restimulated with tumor organoids as described in step 81B or 81C of this protocol. Intracellular IFNγ and cell surface CD107a (a) or CD137 (b) are used as a read-out of reactivity. Gating strategy is shown.

Figure 4. Quantification of the tumor organoid killing assay.

(a) Examples of flow cytometry plots showing tumor cells after 72 hours incubation with or without previously obtained autologous T cell populations, in the presence or absence of MHC-I blocking antibodies. The number of live tumor cells, defined as EpCAM⁺, DAPI^- and NucView488^-, is evaluated. Gating strategy is shown. (b) The absolute count of events in the “Alive cells” gate is divided by the number of events in the “Counting beads” gate (red box in (a)). (c) Bar graph of the number of live cells, normalized to counting beads.

Figure 5.

Example of processing of live imaging data. After tiles have been fused with the stitching fuction of ZenPro (a), an image subset containing the region of interest is created (b). Background is then removed from each channel (c) and a merged composite picture is created (d). Red = mCherry-labelled tumor organoids. Green = NucView488 Caspase 3/7 dye. In this example, T cells are surrounding mCherry⁺ tumor organoids, associated with tumor organoid dissociation, release of debris and NucView488⁺ apoptotic cells. Scale bar = 100 μm.

Troubleshooting

For troubleshooting guidance see Table 1.

Table 1. Troubleshooting guidance.

Step	Problem	Possible reason	Solution
63, 81D(xxxvi)	Organoids do not dissociate into single cells	Human serum is inhibiting the reaction	Wash the plate two more times with 200 μL of PBS per well
		TrypLE is not effective in reducing organoids into single cells	Remove TrypLE and use Trypsin/EDTA 0.05% instead
		Trypsin/EDTA 0.05% is not effective in reducing organoids into single cells	Use mechanical methods: pass organoid suspension (in Trypsin/EDTA 0.05%) through a 29G needle, pipetting 5-7 times with gentle pressure
		Tumor organoids were too large at the end of the experiment	Start with smaller organoids
81B(xxxv),81C(xx)	Few cells recorded	Cell loss during washing steps	Plate more cells, prevent air bubbles during staining, or use V-bottom plates
81B(xxxv), 81C(xx)	Low-intensity signal	Laser settings on flow cytometer suboptimal	Optimize laser settings on flow cytometer
		Decrease in fluorescence	Protect from light and keep on ice during staining procedure
81D(xii)	Organoids have attached to culture plate	The degree of attachment differs between samples	Use dispase to collect attached organoids (step 45-50)
81D(xlvi)	Too few cells recorded	Organoids may suffer from absence of Geltrex during the experiment, and only a few of them may survive	Plate more cells (maintain T cell: tumor cell ratio)
	Poor separation between CD3 positive and EpCAM positive cells	Difficult dissociation into single cells: long exposure to TrypLE or Trypsin/EDTA 0.05% can affect expression of surface markers	Label organoids and/or T cells with dyes before starting experiment (see optional step 81D(xvii) and 81D(xxii))
		Low expression of surface markers
81D(li)	High background despite performing background removal	Background intensity is not even over time	Background needs to be subtracted at each time point individually.
Box 2, A(vii)	No positive signal for dead cell dye	No dead cells in sample	Kill 50% of cells prior to staining by incubating at 95 °C for 45 s
Box 2, A(vii)	No positive signal for antibody-fluorochrome conjugate	Antigen not expressed on cells used for compensation	Use compensation beads for compensation
Box 2, B(vi)	No positive signal	Compensation beads used for non-antibody-coupled fluorochromes	Use cells for compensation

Timing

Steps 1-26: generation of frozen biobank of tumor organoids: variable, average 5 (CRC) to 10 weeks (NSCLC), Fig. 1b-c.

Steps 32-43, isolation of human PBMC from peripheral blood: 2 hours.

Steps 44-80, co-culture of tumor cells and peripheral blood mononuclear cells: 16 days. Assays are best started on Mondays or Wednesdays for optimal timing of medium refreshing.

Step 81A, cryopreservation of cultured T cells: 30 minutes.

Step 81B, tumor reactivity assay using IFNγ and CD107a as a read-out: 4 days. This consists of 2 days for isolation of organoids from Geltrex, followed by 24 hours for reactivity assay, staining, and flow cytometry.

Step 81C, tumor reactivity assay using CD137 as a read-out: 4 days. This consists of 2 days for isolation of organoids from Geltrex, followed by 24 hours for reactivity assay, staining, and flow cytometry.

Step 81D, tumor organoid killing assay: 5 days. This consists of 2 days for isolation of organoids from Geltrex, followed by 3 days co-culture, and flow cytometry on the fifth day.

Anticipated results

This protocol describes an efficient method to generate tumor-reactive T cells from peripheral blood for a patient (population) of interest. After two weeks of co-culture, we observed a ~90% enrichment for CD3⁺ cells in the PBMC population (Fig. 1d). Furthermore, we observe induction of tumor-reactive CD8⁺ T cells in 33-50% of cases⁶. The proportion of CD8⁺ T cells that are tumor-reactive can vary substantially between samples, ranging from 1% to 40%⁶. Given the high variability the proportion of reactive cells, methods to enrich for tumor-reactive cells can be considered (e.g. FACS-based sorting of the reactive population).

Note that loss of MHC class I expression can be frequent in some tumor types (e.g. up to 50% in MSI-CRC) and that omission of MHC-I deficient samples may help in increasing the proportion of successful cases. The evaluation of CD4⁺ T cell reactivity is limited by the potential for recognition of foreign epitopes present during culture, in particular Geltrex-derived epitopes. The current protocol describes an adaptation to our earlier method that serves to prevent such reactivity. The frequency with which tumor-specific CD4⁺ T cell reactivity can be observed using the current strategy still needs to be evaluated more thoroughly.

Co-cultures can be initiated with as little as 10 cc of blood (containing 2-10*10⁶ T cells). To avoid sampling bias, it is recommended to start co-cultures with at least 1*10⁶ PBMC. T cell expansion during co-culture can range from 3- to 20-fold. After co-culture, T cells can be used for downstream assays as described in this protocol, or further expanded using protocols established for TIL expansion³¹.

In T cell - organoid killing assays, variable levels of cytotoxicity can be observed, but typically range between 20-80%. Changes in target : effector ratio can substantially alter the extent of tumor killing observed. Non-specific cell death of up to 20% can be seen (depending on the target : effector ratio) and, to control for this, MHC blocking antibodies are strongly recommended.

Figure 3.

Killing experiment overview. Experimental set up is described in (a). Flow cytometry-based quantification of the amount of alive tumor organoids is described in (b). Live imaging set up and processing is described in (c).

Data availability statement

The data that provide examples of the results that can be generated with this protocol are available from the corresponding author on reasonable request.