How to Test Human CAR T cells in Solid Tumors, The Next Frontier of CAR T Cell Therapy

Abstract

Chimeric Antigen Receptor (CAR) T cell therapy has proven to be a successful treatment option for leukemias and lymphomas. These encouraging outcomes underscore the potential of adoptive cell therapy for other oncology applications, namely solid tumors. However, CAR T cells are yet to succeed in treating solid tumors. Unlike liquid tumors, solid tumors create a hostile tumor microenvironment (TME). CAR T cells must traffic to the TME, survive, and retain their function to eradicate the tumor. Nevertheless, there is no universal preclinical model to systematically test candidate CARs and CAR targets for their capacity to infiltrate and eliminate human solid tumors in vivo. Here, we provide a detailed protocol to evaluate human CAR CD4⁺ helper T cells and CD8⁺ cytotoxic T cells in immunodeficient (NSG) mice bearing antigen-expressing human solid tumors.

1. Introduction

Chimeric Antigen Receptors (CARs) are innovative synthetic immune receptors that have revolutionized the adoptive cell therapy field. CARs comprise three components: antigen binding extracellular domain, intracellular signaling domain, and transmembrane domain. Each domain is customizable for a given cell type, target, and function [1]. Extensive research has been conducted using CARs to redirect T cells to antigens on the surface of cancer cells [2–4]. Typically, T cells are restricted to seeing antigens presented by the target cell’s Major Histocompatibility Complex (MHC) [5]. Researchers have leveraged T cells by providing them with a new specificity determined by a CAR to eliminate elusive cancer cells that typically hide from the immune system. Additionally, the customization of signaling domains of the CAR constructs has allowed researchers to alter or improve T cell function.

Recently, CAR T cell therapy was approved for treating leukemias and lymphomas with revolutionizing results [6]. However, the translation of CAR technology for treating solid tumors has not reached the same level of success [7]. Unlike in the treatment of blood cancers, CAR T cells must traffic successfully to the solid tumor site, the so-called tumor microenvironment (TME). They must also penetrate the stromal components of the solid tumor and evoke a specific cytotoxic response against tumor antigens. However, solid tumors may selectively lose expression of chemokines, vascular-related factors, and other molecules that would otherwise allow for the recruitment and retention of circulating T cells. Moreover, solid tumor-associated antigens are either expressed at lower levels, less predominant, and/or not as unique as those found in liquid tumors. Even after success in all these aspects, CAR T cells must now survive the harsh environment of the TME, characterized by nutrient scarcity, hypoxia, and low pH [8, 9]. Luckily, the customizable nature of the CAR platform provides researchers the opportunity to develop a construct that may fulfill all these requirements for successful treatment.

Once a researcher has developed a CAR construct, it is of utmost importance to test the resulting CAR T cells’ ability to traffic to and kill solid tumors in vivo. The CAR candidate may have worked in vitro, but this does not necessarily mean it will work in vivo [10, 11]. It may not possess the required affinity, expression levels, and/or signaling strength required for CAR T cells to traffic to the site of tumor growth, survive in a complex environment where conditions may be more representative of a human tumor, and eliminate a three-dimensional tumor. Therefore, the ability of CAR T cells to traffic to a solid tumor ought to be tackled before addressing any further issues pertaining to the TME’s immunosuppressive factors. However, there is no universal model to evaluate human CAR T cell trafficking to a solid human tumor in vivo, with most studies solely measuring the effect of infused CAR T cells on tumor growth without investigating the infiltration of CAR T cells in the tumors.

Here, we provide a detailed protocol to test CAR constructs in human CD4⁺ helper T cells and CD8⁺ cytotoxic T cells in immunodeficient (NSG) mice bearing solid human tumors. This model provides information not only on the trafficking capacity of CAR T cells but also on their ability to survive in the metabolically harsh conditions of a TME and to reduce tumor burden.

2. Materials

2.1. T cell isolation

• 1/10 Leukopak (STEMCELL Technologies #200–0092)

• Ca²⁺ and Mg²⁺ free Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco #14190144)

• RPMI 1640 medium, no glutamine (Gibco #11875093)

• Fetal Bovine Serum (FBS) (Gibco #26140079)

• Penicillin-Streptomycin solution (Gibco #15140122)

• GlutaMAX (Gibco #35050061)

• Sodium pyruvate (Gibco #11360070)

• Non-essential amino acids (NEAA) solution (Gibco #11140050)

• 1M HEPES (Gibco #15630080)

• Ammonium chloride solution (STEMCELL Technologies #07850)

• Trypan Blue solution (Sigma #T8154–100ML)

• Cell counter (TC20 Automated Cell Counter, Bio-Rad #1450102)

• Cell Counting Slides (Bio-Rad #1450016)

• 0.5M EDTA, pH 8.0 (Gibco #15575020)

• EasySep magnet (STEMCELL Technologies #18000)

• EasySep Human CD8⁺ T cell Enrichment Kit (STEMCELL Technologies #19053)

• EasySep Human CD4⁺ T cell Enrichment Kit (STEMCELL Technologies #19052)

• Easy 50 EasySep magnet (STEMCELL Technologies #18002)

• Tissue culture 24 well plates (VWR #10861–558)

• Human CD3/28 T Cell Expansion and Activation Dynabeads (Gibco #11131D)

• Recombinant human interleukin-2 (rhIL-2) (Peprotech #200–02)

2.2. Transduction

• T cells isolated and activated in section 2.1

• Titrated CAR-encoding lentivirus

• Trypan Blue solution (Sigma #T8154–100ML)

• Cell counter (TC20 Automated Cell Counter, Bio-Rad #1450102)

• Cell Counting Slides (Bio-Rad #1450016)

• RPMI10 complete medium (see section 2.1)

• Recombinant human interleukin-2 (rhIL-2) (Peprotech #200–02)

• Flow cytometer (e.g., Beckman CytoFLEX)

2.3. K562 Subcutaneous Injection

• 8–12-week NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice (The Jackson Laboratory Strain #005557)

• 70% ethanol (VWR #97064–768)

• Hair Clipper (Wahl #79608)

• Alcohol Swabs (BD #326895)

• Scale (Mettler-Toledo #PB602-S)

• Trypan Blue solution (Sigma #T8154–100ML)

• Cell counter (TC20 Automated Cell Counter, Bio-Rad #1450102)

• Cell Counting Slides (Bio-Rad #1450016)

• Digital Caliper (Fisher Scientific #15-077-957)

• Ca²⁺ and Mg²⁺ free Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco #14190144)

• Insulin Syringes 3/10mL 8mm 31G (BD #328438)

2.4. CAR T Cell Intravenous Injection

• Digital Caliper (Fisher Scientific #15-077-957)

• Mouse Ear Tags (World Precision Instruments #501893)

• Easy 50 EasySep magnet (STEMCELL Technologies #18002)

• Trypan Blue solution (Sigma #T8154–100ML)

• Cell counter (TC20 Automated Cell Counter, Bio-Rad #1450102)

• Cell Counting Slides (Bio-Rad #1450016)

• Ca²⁺ and Mg²⁺ free Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco #14190144)

• Insulin Syringes 3/10mL 8mm 31G (BD #328438)

• Isoflurane 99.9% (McKesson #803250)

• Low-Flow Anesthesia System for Mice and Rats (Kent Scientific SomnoSuite, Fisher Scientific #13-005-111)

• Heating pad (Kent Scientific DCT-25)

2.5. Solid Tumor Measurement

• 70% Ethanol (VWR #97064–768)

• Digital Caliper (Fisher Scientific #15-077-957)

2.6. Tissue Dissection

• Digital Caliper (Fisher Scientific #15-077-957)

• 70% Ethanol (VWR #97064–768)

• Scale (Mettler-Toledo #PB602-S)

• Hair Clipper (Wahl #79608)

• Forceps (VWR #82027–406)

• Surgical Scissors (VWR #470315–214)

• Scalpel (VWR #82029–850)

• 100×15mm Petri Dish (VWR #25384–070)

• 60×15mm Petri Dish (VWR #25384–164)

• RPMI 1640 medium, no glutamine (Gibco #11875093)

2.7. Tissue Processing

• 100×15mm Petri Dish (VWR #25384–070)

• 60×15mm Petri Dish (VWR #25384–164)

• Heated Shaker (VWR # 76407–112)

• Scale (Mettler-Toledo #PB602-S)

• Razor blades (VWR #55411–050)

• Ca²⁺ and Mg²⁺ free Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco #14190144)

• 70μm cell filters (VWR #76327–100)

• 100μm cell Filters (VWR #76327–102)

• RPMI 1640 medium, no glutamine (Gibco #11875093)

• RPMI10 complete medium (see section 2.1)

• Collagenase P from Clostridium histolyticum (Roche #11215809103)

• DNase I, RNase Free (NEB M0303L)

• EDTA (Invitrogen #15575–038)

• 5ml Disposable Luer Lock Syringes (Thermo #S7510–5)

• ACK Lysing Buffer (Gibco #A1049201)

• Trypan Blue solution (Sigma #T8154–100ML)

• Cell counter (TC20 Automated Cell Counter, Bio-Rad #1450102)

• Cell Counting Slides (Bio-Rad #1450016)

2.8. Flow Cytometry

• 5ml FACS tubes (VWR #76449–666)

• Ca²⁺ and Mg²⁺ free Dulbecco’s Phosphate Buffered Saline (DPBS) (Gibco #14190144)

• Anti-human CD4 Alexa Fluor 700 (clone SK3, Biolegend #344621)

• Anti-human CD8 PerCP (clone SK1, Biolegend #344707)

• Anti-human CD3 PE Cy7 (clone SK7, Biolegend #344815)

• Ghost Dye Brilliant Violet 510 (TONBO #13–0870-T100)

• 5 ml FACS tubes with 35 um filter caps (VWR #76449–658)

• Flow cytometer (e.g., Beckman CytoFLEX)

3. Methods

The present protocol utilizes primary human T cells isolated from peripheral blood to generate CAR T cells, CAR antigen-expressing K562 cells, an HLA null human myelogenous leukemia cell line, as target cells for the CAR T cells, and NSG mice to test human CAR T cells in a CAR antigen-expressing K562 solid tumor in vivo (Figure 1A). This requires careful coordination to time the 1) generation, purification, and testing of antigen-expressing K562 cells, 2) injection of antigen-expressing K562 cells into immunodeficient NSG mice and await solid tumor formation, and 3) T cell purification from human peripheral blood, e.g., in the form of a concentrated leukocyte product – leukopak, activation, and transduction with CAR genes (Figure 1B, Figure 2).

Figure 1: Experimental timeline.

A. Steps needed to evaluate candidate CAR construct efficiency in human solid tumors. These can be divided into three components: Generation of CAR target-expressing tumor cells, production of human CAR T cells, and NSG mouse experimentation. B. Experimental timeline separating the three components of the experiment. Denoted days are based on the isolation of primary human T cells, the most time-sensitive component. Day 0 indicates leukopak arrival and human T cell isolation. The experimental timeline assumes the antigen-encoding tumor cell line has yet to be produced. Hence, a pure population of antigen-encoding tumor cells should be obtained 3 weeks before leukopak arrival to allow enough time to obtain the required number of tumor cells to inject on day 2.

Figure 2: Calendar view of experimental timeline.

A. Experimental timeline in a calendar format to provide a representative experiment weekly schedule. Antigen⁺ K562 cells are assumed to be already in place. Red text indicates T cell processing, while black text indicates mouse work.

3.1. T cell isolation

3.1.1. Leukopak Processing

• 1.Dilute leukopak in an equivalent volume of DPBS + 2% FBS. Mix by pipetting slowly.
• 3.Centrifuge at 300 g for 10 minutes at room temperature (RT) with the brake on.
• 4.Aspirate supernatant without disturbing the cell pellet.
• 5.Resuspend the cell pellet, add 2 mL DPBS + 2% FBS, and pipet up and down.
• 7.Add 8 mL ammonium chloride solution to cell suspension (4:1 ratio). Mix by pipetting slowly.
• 8.Incubate on ice for 15 minutes.
• 8.Centrifuge at 500 g for 10 minutes at RT with the brake on.
• 9.Aspirate supernatant.
• 10.Wash by adding 30 mL DPBS + 2% FBS.
• 11.Centrifuge at 150 g for 10 minutes at RT with the brake off.
• 12.Aspirate supernatant and resuspend the cell pellet in 30 mL DPBS + 2% FBS.
• 13.Count cells by diluting them first 1:100 with DPBS (10 μl cell suspension + 90 μl DPBS, and then 10 μl cells diluted 1:10 + 90 μl DPBS) and then 1:1 with Trypan Blue solution (10 μl cells diluted 1:100 + 10 μl Trypan Blue). See Note 4.1.

3.1.2. CD4⁺ T cell enrichment (Negative selection)

1. Collect 15×10⁶ PBMCs (peripheral blood mononuclear cells) and centrifuge at 500 g for 5 minutes at RT. The starting cell number will depend on your specific needs.

2. Resuspend cells in Cell Separation Buffer (DPBS + 10 mM EDTA + 2% FBS) at 50×10⁶ cells/ml.

3. Follow EasySep Human CD4⁺ T cell Enrichment Kit protocol.

4. Count cells in a 1:1 ratio with Trypan Blue.

3.1.3. CD8⁺ T cell enrichment (Negative selection)

1. Collect 50×10⁶ PBMCs and centrifuge at 500 g for 5 minutes at RT. The starting cell number will depend on your specific needs.

2. Resuspend cells in Cell Separation Buffer (DPBS + 10 mM EDTA + 2% FBS) at 50×10⁶ cells/ml.

3. Follow EasySep Human CD8⁺ T cell Enrichment Kit protocol.

4. Count cells in a 1:1 ratio with Trypan Blue.

3.1.4. T cell Activation and Culture

• 1.After T cell subset purification and counting, activate CD4⁺ and CD8⁺ T cells separately in 24-well plates with Human T-Activator CD3/28 for T cell Expansion and Activation Dynabeads and recombinant human IL-2 (rhIL-2) in RPMI10 complete medium by conducting the following steps:
• 2.Centrifuge T cells at 500 g for 5 minutes at RT
• 3.Decant supernatant
• 4.Resuspend at 5×10⁵ T cells per mL of RPM10 complete medium in a conical tube
• 5.Add 25 μL anti-CD3/CD28 Dynabeads for every 10⁶ T cells for a final 1:1 ratio of Dynabeads to T cells
• 3.
  Add recombinant human IL-2:

  • a.
    100 IU/ml of rhIL-2 to CD4⁺ T cells
  • b.
    300 IU/ml of rhIL-2 to CD8⁺ T cells
• 4.Pipette up and down 2–3 times to evenly resuspend T cells, Dynabeads, and rhIL2.
• 5.Plate 1ml per well of a 24-well plate so that a total of 5×10⁶ T cells are plated per well.
• 6.Incubate for 48 hours in a 37°C 5% CO₂ tissue culture incubator. See Note 4.2.

3.2. Transductions

3.2.1. T cell Transduction

Once T cells have been activated for 48 hours, they are at the peak of activation, ready to be transduced with the CAR construct of interest (See Note 4.3). For increased transduction efficiency it is important to transduce the T cells during this time window. There is no need to remove the Dynabeads for this process.

• 1.Count T cells with Trypan Blue.
• 2.Centrifuge at 500 g for 5 minutes at room temperature (RT).
• 3.Resuspend in RPMI10 complete medium at 1.25×10⁶ cells/ml.
• 4.Thaw lentivirus aliquots on ice.
• 5.Add lentivirus at a multiplicity of infection (MOI) of 2 to 2.5×10⁵ T cells in 200 μl in a 1.5 ml microcentrifuge tube. See Note 4.4.
• 6.Add rhIL-2 for a final concentration of 100 IU/ml for CD4⁺ T cells and 300 IU/ml for CD8⁺ T cells.
• 7.Centrifuge at 1000 g for 1 hour at 32°C.
• 8.Transfer transduced T cells from 1.5 ml microcentrifuge tubes to a 24-well plate, each 1.5 microcentrifuge tube to one well.
• 9.Place the plate with transduced T cells in tissue culture incubator overnight.
• 10.The following day, bring volume in each well up to 2 ml with RPMI10 complete medium with rhIL-2.
• 7.Expand and maintain T cells at 5×10⁵ − 1×10⁶ cells per ml for 1 week. Additionally, provide rhIL-2 every 48 hours for a final concentration of 100 IU/ml for CD4⁺ T cells and 300 IU/ml for CD8⁺ T cells.
• 8.On day 7 of T cell culture (Figure 1B), if a pure CAR⁺ population is required, sort CAR⁺ cells using fluorescence assisted cell sorting (FACS) to purify cells expressing a CAR reporter gene, e.g., GFP. If a pure CAR T cell population is not possible or needed, calculate the CAR⁺ T cell percentage in cultures using flow cytometry.

3.2.2. K562 Culture & Transduction

Due to the large amount of target-expressing K562 required for this experiment (5×10⁶ cells per mouse), it is essential to produce and expand the tumor cells well before leukopak arrival. It is crucial to have enough target-expressing K562 on day 2 of T cell culture (Figure 1B) to minimize the consequences of using older primary T cells. Thus, it is recommended to start expanding target-expressing K562 cells 14 days prior to leukopak arrival. Of note, the selected level of antigen expression on the K562 stable cell line can have an impact: too high expression may not be representative of a clinical phenotype, while too low expression might fail to trigger CAR T cell activation.

1. Three weeks (21 days) before leukopak arrival, count parental K562 cells with Trypan Blue and acquire 2×10⁵ cells.

2. Centrifuge at 500 g for 5 minutes at RT.

3. Aspirate supernatant.

4. Resuspend cell pellet with lentivirus aliquot in a 1.5 mL Eppendorf tube.

5. Centrifuge at 1000 g for 1 hour at 32°C.

6. Transfer cells to a well of a 24-well plate and put in tissue culture incubator.

7. Expand transduced K562 cells in a T75 flask in preparation for FACS.

3.2.3. Antigen⁺ K562 Sorting

1. One week (7 days) post transduction, count transduced K562 cells with Trypan Blue.

2. Centrifuge at 500 g for 5 minutes at RT in conical tube.

3. Aspirate supernatant.

4. Surface stain for the antigen of interest using 1 μL antibody / 10⁶ cells in 100 μl DPBS

5. Incubate at 4°C for 30 mins protected from light.

6. Wash by adding 500 μL PBS.

7. Centrifuge at 500 g for 5 minutes at RT.

8. Aspirate supernatant.

9. Resuspend pellet in DPBS at 15×10⁶ cells/mL for sorting.

10. Filter cell suspension through a 40 μm filter cap into a FACS tube and put on ice.

11. Prepare collection tubes for sorting by adding 3 mL RPMI10 complete medium per 15 mL conical tube and put on ice.

12. Sort antigen⁺ K562 cells using FACS.

13. Count sorted cells post-sort with Trypan Blue.

14. Plate sorted antigen⁺ K562 cells at 5×10⁵ cells per well of a 24-well plate in 1 mL RPMI10 complete medium and put in the tissue culture incubator.

15. Expand antigen⁺ K562 cells in a T75 cell culture flask for injection, liquid nitrogen stock freezing, and irradiation. See Note 4.5.

3.3. K562 Subcutaneous Injection

Two days after T cell isolation, target positive-K562 cells are subcutaneously injected into the right flank of NSG mice (Figure 1, Figure 2). Tumors are allowed to grow to a volume of 50mm³ to 150mm³ until CAR T cell injection. This will take approximately 7 days.

1. Six days before CAR T cell injections, measure tumor size with a caliper. This is done by measuring tumor length (along the animal, longitudinally) and tumor width (sideways, transversally).

2. Calculate tumor sizes using the formula: tumor volume = 0.5 × (length × width²)

3. Randomize NSG mice to be used for each condition based on tumor size.

4. Hold each mouse with one hand sanitize the shave area with 70% ethanol prep.

5. Shave the right flank of each mouse with hair clipper.

6. Ear tag each mouse.

7. Weigh each mouse for an initial weight measurement.

8. Collect 5×10⁶ antigen⁺ K562 cells per mouse being injected. See Notes 4.6. and 4.7.

9. Wash cells by adding 2 times the volume of DPBS.

10. Centrifuge cells at 500 g for 5 minutes at RT.

11. Decant supernatant.

12. Resuspend cells in ice-cold DPBS at 5×10⁷ cells/ml.

13. Aliquot 100 μl of cells into 1.5 ml microcentrifuge tubes so that each tube contains 5×10⁶ cells/ml. This will help reduce the variability of tumor growth across mice.

14. Store cells on ice.

15. Relocate to the clean mouse facility room where NSG mice are held, keeping cells on ice.

16. Briefly vortex each tube with K562 cells immediately before injection.

17. Fill a fresh 0.3ml syringe with attached 31G × 5/16” mm needle with cells from one aliquot. Do not allow syringe loaded with cell suspension to sit longer than a few minutes. Before injection, gently flick syringe to ensure cells are in suspension and there is no air.

18. Using thumb and index finger, scruff mouse such that flanked skin is taut.

19. With the other hand, fully slide needle at a 10-degree angle with bevel up underneath the skin near the right thigh of the mouse.

20. Gently pull up on the needle to form a ‘skin tent’ to ensure injection is in the subcutaneous space.

21. In one continuous motion, slowly inject 100 μl cell suspension. A ‘bulge’ should be visible where the cell suspension is located. It is important for ‘bulge’ to be far from the injection site as mice tend to rub the area after injection and cause leakage.

22. Slowly withdraw the needle, not allowing leakage of any cells.

23. Place mouse back into cage.

24. Repeat steps 16–23 for each mouse.

25. Follow up 7 days later, on the day of CAR T cell injection, to measure tumor volumes and identify those between 50mm³ and 150mm³.

3.4. CAR T Cell Intravenous Injection

One week (7 days) after tumor injections, either saline (DPBS) or 1×10⁶ CAR⁺CD4⁺ and 1×10⁶ CAR⁺CD8⁺ T cells are retro-orbitally intravenously injected into randomized mice bearing 50mm³-150mm³ tumors. Tumor volumes are then measured with a caliper every other day. Two weeks (14 days) after CAR T cell injection, i.e., three weeks (21 days) after tumor injection, mice are euthanized and tumors surgically removed, weighed, digested, and analyzed via flow cytometry.

1. Collect T cells. Ensure cells are in the logarithmic growth phase by harvesting cells from flasks between 50–80% confluent and at least 70% viable.

2. Remove Dynabeads with EasySep magnet for 5 minutes.

3. Count cells with 1:1 Trypan Blue.

4. Wash by adding 2 times volume of DPBS.

5. Centrifuge at 500 g for 5 minutes at 32°C.

6. Resuspend cells at 20×10⁶ CAR⁺ cells (either CD4⁺ T cells or CD8⁺ T cells) per ml of DPBS. If you did not FACS sort CAR⁺ cells, make sure to calculate CAR⁺ cell number by using transduction efficiency values and total cell numbers.

7. For each mouse being injected with CAR T cells, make a separate 1.5ml tube with 50 μl with 1×10⁶ CAR⁺ CD4⁺ T cells and 50 μl with 1×10⁶ CAR⁺ CD8⁺ T cells.

8. Store on ice.

9. Relocate to mouse facility clean room where NSG mice are held while keeping cells on ice throughout process.

10. Anesthetize mice by gas anesthesia (3% isoflurane) in a Plexiglas apparatus.

11. Briefly vortex CAR⁺ T cells.

12. Load a fresh 0.3ml syringe with attached 31G × 5/16’ mm needle with an aliquot of CAR T cells or saline (DPBS). Injected volumes should not exceed 150 μl or contain air bubbles. Do not allow syringes with cells to sit longer than a few minutes. Before injection, gently flick syringe to ensure cells are in suspension.

13. Once mouse is properly anesthetized, as determined by breathing pattern and paw pinching, remove the mouse from plexiglass apparatus and place it on a nose cone providing 1.8% isoflurane.

14. To prevent hypothermia, place mouse on a warming pad at 37°C.

15. With thumb and index finger, protrude the anesthetized mouse’s eyeball by applying gentle downward pressure.

16. With the needle bevel facing down, not towards eyeball, insert the needle behind the eyeball at a 45-degree angle to the nose.

17. Without poking the eyeball, follow the edge of the eyeball until the needle tip reaches the back end of the eyeball.

18. In one continuous slow motion, inject the cells into the retro-bulbar sinus.

19. After a moment, very gently remove needle,

20. Close the eyelid and apply gentle pressure.

21. Place mouse in an individual lying on its back.

22. Monitor mouse until it recovers from anesthesia, should take ca. 60 seconds.

23. Repeat steps 10–22 for each mouse.

24. Follow up in 48 hours to begin measuring tumor volumes every other day.

3.5. Solid Tumor Measurements

After CAR T cells have been injected, it is important to monitor tumor volume growth over time. It is recommended to measure the tumor volumes every other day after CAR T injection. CAR T cells are not expected to influence tumor volume until 5–7 days after intravenous injection, yet still measure before then to record natural variability in tumor growth.

1. Using nondominant thumb and index finger, scruff mouse.

2. With the other hand, use a caliper to measure perpendicular tumor diameters in mm. See Note 4.8.

3. The initial longest measurement, longitudinal, should be noted as the length, while the smallest measurement, transversal, should be noted as the width. If tumor is hard to find, wet area with 70% ethanol to help reveal it.

4. Calculate tumor volume (mm³) = 0.5 × (length × width²).

5. Measure tumor every other day after CAR T cell injection.

3.6. Dissections

Three weeks (21 days) post tumor injection, tumors and spleens are ready to be removed for analysis. The removal of the tumor and spleen allow a final weight measurement, another metric to compare the tumor growth in addition to tumor volume. It will also allow the analysis of tumor-infiltrating T cells and thus confirmation of successful CAR T cell infusion and trafficking. Of note, it is possible that timeframes proposed here are shortened due to tumors reaching their humane endpoint earlier than anticipated.

1. Euthanize mouse by CO₂ asphyxiation, followed by cervical dislocation, or alternative approved method.

2. Measure final tumor volume with caliper.

3. Measure mouse final weight.

4. Spray mouse with 70% Ethanol.

5. Optionally, shave the area around the tumor to minimize hair from getting into sample in downstream tissue processing.

6. Remove tumor and spleen as described below.

3.6.1. Tumor dissection

1. Using forceps, pull up on the mouse’s right flank, creating a ‘skin tent’.

2. With scissors or scalpel, make an incision perpendicular to skin tent.

3. Insert closed scissors into the incision and open the scissors, spreading the incision opening with the dull ends of the scissors.

4. Keep spreading the incision site while gently pulling upward on the skin flap until some of the tumor is exposed.

5. Grab the mouse skin on top of tumor mass with forceps and place a scalpel in between skin and tumor.

6. Gently pull up on the skin as you move scalpel across the tumor edge until the subcutaneous tumor is completely exposed.

7. Move scalpel in between tumor and muscle wall while rotating the mouse to remove the tumor. The muscle wall should not be broken in the process. No hair, scabs, or skin should be left on the tumor and should be consistent across other tumors dissections.

8. Weigh tumor on a 100×15mm Petri dish.

9. Store tumor at 4°C until further processing.

10. Remove spleen as describe below (3.6.2).

11. Repeat for each mouse.

3.6.2. Spleen dissection

1. Using forceps, pull up on mouse’s left flank below ribcage, creating a ‘skin tent’.

2. With scissors or scalpel, make an incision perpendicular to skin tent.

3. Insert closed scissors into the incision site and open the scissors, spreading the incision opening with the dull ends of the scissors.

4. Keep spreading the incision site while gently pulling upward on the skin flap until left flank skin is removed

5. Make small incision on the muscle wall.

6. Remove the spleen with forceps and trim as much connective and adipose tissue as possible without rupturing the spleen.

7. Weigh spleen on a 60×15mm Petri dish.

8. Add cold RPMI medium.

9. Store at 4°C until further processing.

3.7. Tissue Processing

After the spleen and tumor are removed, they need to be processed into single cell suspensions for cell counting and antibody staining. ACK lysis is used to remove red blood cells from the spleen, while collagenase and DNase will be used to digest tumors into single cell suspensions.

3.7.1. Tumor Tissue

1. Obtain two representative sections of the tumor weighing 0.2 to 0.3 grams for a total of about 0.5 grams for the tumor sample. These sections will be digested separately, then combined once you count the cells. Take note of the total weight for representative section as this is used for the calculation of Tumor Cell Count.

2. With a fresh razor blade, mince each tissue section to a minimum of 1mm sections inside a 100×15mm Petri dish.

3. Optionally, add 1ml of cold PBS halfway through to help with mincing.

4. Transfer minced tumor pieces into a 50ml conical tube.

5. Rinse Petri dish with cold DPBS to collect all tumor pieces and cells.

6. Optionally, run through a 100um cell filter to help remove fat, connective tissue, and skin debris. Use a syringe plunger to push minced tumor through filter.

7. Bring up volume to at least 10 ml of cold DPBS.

8. Centrifuge cells at 500 g for 5 minutes at 4°C.

9. Decant supernatant. You may observe fat and debris in the supernatant.

10. Briefly vortex cell pellet.

11. Resuspend in 500 μl of serum-free RPMI medium.

12. Transfer resuspended cells into a 1.5ml microcentrifuge tube.

13. Add 500 μl of 2mg/ml of PBS Collagenase P to each tube.

14. Add 1 μl DNase I (RNase Free) at 2000U/ml to each tube.

15. Briefly vortex.

16. Shake samples at 37°C at 225 rpm for 1.5 hours. Longer times can result in the cleaving of surface T cell markers, such as CD4 and CD8α.

17. Transfer processed sample into a 50 ml conical tube and add 10ml of PBS with 2 mM EDTA to stop collagenase digestion.

18. Combine digested tumor sections from same tumor.

19. Filter solution through a 70μm filter to a new 50 ml conical tube.

20. Rinse 70μm filter with an additional 10ml of DPBS with 2 mM EDTA.

21. Keep on ice.

22. Count cells using 1:1 Trypan Blue. See Note 4.9.

3.7.2. Spleen

1. Place a 70μm filter on a 50ml conical tube.

2. Pour spleen and RPMI from 60×15mm Petri dish onto 70μm filter.

3. Grind, and push spleen against the filter using a syringe plunger.

4. Rinse filter with RPMI to obtain a final volume of ca. 10ml RPMI.

5. Centrifuge at 500 g for 5 minutes at RT.

6. Decant supernatant.

7. Briefly vortex cell pellet.

8. Gently resuspend in 1ml of ACK lysing buffer.

9. Incubate for 5 minutes at RT.

10. Stop the reaction by adding 10 ml RPMI.

11. Centrifuge at 500 g for 5 minutes at 4°C.

12. Resuspend cell pellet in 3 ml cold DPBS.

13. Keep on ice.

14. Count cells with 1:1 Trypan blue. See Note 4.9.

3.8. Flow Cytometry

After cells from tumor and spleen have been processed and counted, it is time to phenotype them using flow cytometry. Depending on the antibodies on hand, what reporter gene is incorporated downstream of the CAR gene, and flow cytometer used, the panel used may be different from the one described below. In the current protocol, we used GFP as the CAR reporter gene and stained cells for human CD3, CD4, and CD8, as well as with a viability dye.

1. Obtain a maximum of 1×10⁷ total viable tumor cells and splenocytes for flow cytometry analysis into a 5ml FACS tube. Take note of how many cells were used to later calculate percentage of tumor-infiltrating leukocytes and engraftment, respectively.

2. Centrifuge at 500 g for 5 minutes at RT.

3. Prepare antibody master mix with DPBS, anti-human CD4 A700 1:100, anti-human CD8 PerCP 1:100, anti-human CD3 PE/Cy7 1:100, and Ghost Viability Dye BV510 1:500. Each sample will be stained with 100 μl antibody master mix. Hence, if staining a total of 12 samples as an example, add 12 μl CD4 A700, 12 μl CD8 PerCP, 12 μl CD3 PE/Cy7, and 2.4 μl Ghost BV510 to 1200 μl DPBS. Store on ice in the dark. See Note 4.10.

4. Decant supernatant.

5. Briefly vortex.

6. Add 100 μl antibody master mix to each sample.

7. Briefly vortex and incubate at 4°C in the dark for 30 minutes.

8. Wash by adding 500 μl DPBS.

9. Centrifuge at 500 g for 5 minutes at RT.

10. Decant supernatant.

11. Resuspend the cell pellet in 300 μl DPBS.

12. Keep tubes on ice in the dark.

13. Analyze by flow cytometry. Filter samples through 40μm filter cap into flow tubes immediately before reading to guarantee single cell suspensions. See Note 4.11.

14. Using total tumor cell number and percentage of CAR⁺ T cells in tumor sample, calculate the number of tumor-infiltrating CAR⁺ T cells per tumor.

4. Notes

4.1. If using an automated cell counter, e.g. Bio-rad TC20 Automated Cell Counter, adjust cell count to this 200-fold cell dilution by multiplying the result by 100, as most cell counters assume a 1:1 dilution.

4.2. If T cells are not elongated and clustering by 24 hours, then activation and/or isolation was not performed properly.

4.3. If lentivirus coding for the CAR or antigen of choice is to be produced in lab rather than purchased, please refer to Zimmerman et al. 2022, this issue, for methods on producing and titrating lentivirus.

4.4. CD8⁺ T cells have a lower transduction efficiency with lentivirus than CD4⁺ T cells at the same MOI. Therefore, transduce 2–3 times more CD8⁺ T cells than CD4⁺ T cells to have enough CAR⁺ cells for in vivo experiments.

4.5. Irradiated antigen-expressing K562 can be used to stimulate and expand CAR T cells in vitro, please see Zimmerman et al., 2022, this issue.

4.6. On the day of subcutaneous tumor injection, i.e., seven days before CAR T cell intravenous injection, ensure antigen-expressing K562 cells are in the logarithmic growth phase by harvesting cells from flasks 50–80% confluent and at least 90% viable.

4.7. Due to the variability of tumor growth, it is recommended to inject more mice than the intended number of mice to receive CAR T cells. This will ensure there are enough mice with tumors between 50mm³ and 150mm³ on the day of CAR T cell injection.

4.8. As it is crucial to be consistent with the tumor measurements each time, it is recommended that the same person make all measurements across all days and take note of how tight the caliber is placed against the tumor for maximum consistency.

4.9. If tumor and spleen processing need to be carried out over two days, then this is a good place to stop. If this is the case, then cells should be given in RPMI10 in step 17 instead of DPBS with 2 mM EDTA to provide nutrients to single cell suspension. Place single cell suspension at 4°C overnight until antibody staining for flow cytometry. Note that a 30–50% loss of total cells is expected if cells are kept overnight instead of analyzed on the same day.

4.10. Multi-color flow cytometry requires preparation of single-color controls for the cytometer to perform compensation, i.e., correct for spectral overlap between different channels. Hence, for the described panel, prepare the following single-color controls: unstained, A700, PerCP, PE/Cy7, BV510, GFP.

4.11. Take into consideration that CD4 and CD8α may be downregulated due to collagenase treatment [12]. These effects can be exacerbated depending on incubation time and type of collagenase used.

Figure 3: CAR T cell impact on human solid tumor growth in NSG mice.

A. NSG mice were injected with 5×10⁶ CD19⁺ K562 cells. After 7 days, mice with tumor volumes between 50mm3 and 150mm3 were injected with either saline (n=3) or 1×10⁶ human CD4⁺ T cells and 1×10⁶ human CD8⁺ T cells (n=3) expressing a CD19CAR-2A-GFP construct (Addgene #1359910). Tumor volumes were measured every other day for the following 14 days. Mice were euthanized 21 days after initial tumor subcutaneous (s.c.) injection. The graph displays the fold change in tumor volume relative to initial volume on day 7. B. Final tumor volumes, weights, and cell numbers from experiment described in A. C. Correlation between final tumor volumes (mm3) and final tumor weights (grams) (R²=0.8956) demonstrating that caliper measurements are an accurate representation of tumor growth. Correlation between final tumor weights (grams) and tumor cell number (R² = 0.7249) demonstrating consistency of the tumor digestion protocol.

Figure 4: CAR T cell engraftment in NSG murine spleen.

Three weeks (21 days) after tumor injection, single-cell suspensions were collected from saline (DPBS) or CAR T cell-treated mouse spleens and surface stained with anti-human CD4 Alexa700 1:100, anti-human CD8 PerCP 1:100, anti-human CD3 PE/Cy7 1:100, and Ghost Viability Dye BV510 1:500. Positive control was created by adding 1×10⁶ CAR⁺CD4⁺ and CAR⁺CD8⁺ T cells on top of a nontreated spleen sample.

Figure 5: Tumor-infiltrating T cells in human solid tumors in NSG mice.

Three weeks (21 days) after tumor injection, single-cell suspensions were collected from saline (DPBS), or CAR T cell-treated human solid tumors and surface stained with anti-human CD4 Alexa700 1:100, anti-human CD8 PerCP 1:100, anti-human CD3 PE/Cy7 1:100, and Ghost Viability Dye BV510 1:500. Positive control was created by adding 1×10⁶ CAR⁺CD4⁺ and CAR⁺CD8⁺ T cells on top of a nontreated tumor sample.

Figure 6. Detecting CAR T cells in the spleen and tumor in NSG mice.

A. CAR T cells (CAR⁺CD4⁺ and CAR⁺CD8⁺ T cells) were detected in the spleen 2 weeks after intravenous T cell injection by assessing the expression of the CD19CAR-2A-GFP construct (Addgene #1359910) reporter gene, GFP, within viable human CD3⁺CD4⁺ and viable human CD3⁺CD8⁺ cells. B. Analogously to A., CAR T cells were detected in the tumor 2 weeks after intravenous T cell injection by assessing GFP expression.