Artificial Antigen-Presenting Cells for EnhancedCAR T Cell Persistence and Anti-Tumor Efficacy

1. Abstract

CAR T cell therapy is hampered by poor in vivo persistence, cytokine deprivation, and tumor microenvironment-driven exhaustion. Conventional bead-based expansion platforms provide only transient activation with no capacity for therapeutic co-delivery — leaving infused CAR T cells poorly equipped for durable anti-tumor function. We propose to engineer liposome-based artificial antigen-presenting cells (aAPC-Lipo) co-displaying anti-CD3, anti-CD28, 4-1BBL, and membrane-tethered IL-15 on their surface while encapsulating dominant-negative TGF-β receptor II mRNA in their core — simultaneously activating and therapeutically reprogramming CD19-targeted CAR T cells during manufacturing and after adoptive transfer. Efficacy will be assessed through ex vivo expansion, multiparameter phenotyping, serial cytotoxicity assays, and NSG xenograft tumor models with longitudinal persistence tracking.

This project will be conducted by a multidisciplinary team with expertise in lipid nanoparticle engineering, CAR T cell manufacturing, and tumor immunology, with access to institutional core facilities for nanoparticle characterization, flow cytometry, and small animal imaging. All animal studies will be conducted under IACUC-approved protocols. The estimated project timeline is 12 months.

All raw data, analysis scripts, and protocols will be deposited on OSF and linked to the ResearchHub preregistration record upon study completion.

2. Introduction

Chimeric antigen receptor (CAR) T cell therapy has emerged as a transformative approach in cancer immunotherapy, achieving remissions in hematologic malignancies such as B-cell acute lymphoblastic leukemia and diffuse large B-cell lymphoma. Despite these landmark successes, critical limitations continue to impede broader clinical translation. Chief among these are: (1) poor in vivo persistence of infused CAR T cells due to exhaustion and cytokine deprivation; (2) lack of sustained costimulatory signals following the initial activation phase; and (3) an immunosuppressive tumor microenvironment that attenuates effector function. Addressing these barriers is essential to realizing the full promise of CAR T cell therapy in solid tumors and late-relapse hematologic cancers.

Conventional ex vivo T cell expansion relies on anti-CD3/CD28-coated beads or irradiated feeder cells to provide activation and costimulatory signals. These approaches are cumbersome, introduce manufacturing variability, and do not readily enable controlled cytokine co-delivery. Moreover, once infused in vivo, CAR T cells rapidly lose access to these critical support signals, contributing to contraction and exhaustion. There is therefore a pressing need for next-generation artificial antigen-presenting platforms capable of delivering coordinated stimulatory and trophic signals both during manufacturing and within the in vivo therapeutic context.

Liposomal nanoparticles offer a compelling chassis for constructing artificial antigen-presenting cells (aAPCs). They are biocompatible, highly tunable in size and surface composition, and can be functionalized with membrane-bound proteins or antibodies through well-established lipid-protein conjugation chemistries. Critically, liposomes can simultaneously display Signal 1 (TCR/CD3 engagement via anti-CD3 antibodies), Signal 2 (costimulation via anti-CD28 or 4-1BBL), and Signal 3 (cytokine support via membrane-tethered IL-2, IL-15, or IL-21). This tripartite signaling architecture mimics the natural immunological synapse and can be tailored to selectively expand desired T cell subsets.

Recent work has demonstrated that liposomal aAPCs can expand primary human T cells with efficacy comparable to bead-based systems while offering superior control over stimulatory input ratios (Turtle et al., 2012; Steenblock & Fahmy, 2008). However, the potential of liposomal aAPCs specifically designed for the sustained support of CAR T cells — including in vivo re-stimulation and therapeutic payload co-delivery — remains largely unexplored. This proposal addresses this gap by developing and characterizing a library of aAPC liposomes engineered for maximal CAR T cell persistence and functional potency..

3. Hypotheses

We hypothesize that liposome-based artificial antigen-presenting cells (aAPC-Lipo) co-displaying activating antibodies (anti-CD3), co-stimulatory ligands (anti-CD28), and surface-conjugated homeostatic cytokines (IL-2, IL-15, IL-21) will significantly enhance the ex vivo expansion, in vivo persistence, and anti-tumor cytotoxicity of CAR T cells compared to conventional bead-based activation systems. We further hypothesize that genetic payload delivery (mRNA or plasmid DNA encapsulated within the liposomal core) will enable iterative CAR T cell re-programming in situ, overcoming exhaustion and enabling durable tumor control.

Specific predictions arising from this hypothesis include:

• aAPC-Lipo will yield a significantly greater expansion of central memory CAR T cells (TCM: CD45RO+CCR7+) compared to CD3/CD28 bead controls over a 10-day expansion protocol.

• Surface-tethered IL-15 on aAPC-Lipo will sustain CAR T cell viability and proliferative capacity in cytokine-depleted conditions by engaging the IL-15 receptor complex in trans.

• Co-encapsulation of mRNA encoding dominant-negative TGF-β receptor II (DNRII) within aAPC-Lipo will confer resistance to TGF-β-mediated suppression in the tumor microenvironment.

• In vivo administration of aAPC-Lipo will significantly extend CAR T cell persistence in NSG xenograft models compared with PBS control and improve tumor regression kinetics.

4. Methods

4.1 Liposome Formulation and Surface Engineering

aAPC-Lipo will be synthesized by thin-film hydration followed by extrusion through 100 nm polycarbonate membranes to achieve uniform size distributions. The core lipid composition will comprise POPC:DOPE-PEG2000-Maleimide:Cholesterol, providing membrane fluidity, steric stabilization, and functional handles for antibody/protein conjugation. Anti-CD3 and anti-CD28 antibodies will be covalently conjugated via thioether linkage to maleimide-terminated PEG anchors following partial reduction with TCEP. Recombinant IL-2, IL-15, and IL-21 will be conjugated in a similar manner using NHS-PEG4-maleimide crosslinkers. 4-1BBL will be incorporated as a lipid-anchored recombinant ectodomain. Three formulations will be generated with varying signal ratios to identify optimal stimulatory combinations.

Characterization will include dynamic light scattering (DLS) for hydrodynamic diameter and polydispersity index (PDI), nanoparticle tracking analysis (NTA) for particle concentration, cryo-TEM for morphological validation, and flow cytometry with fluorochrome-labeled secondary antibodies to confirm surface protein density and orientation.

4.2 CAR T Cell Generation

CD19-specific CAR T cells will be generated from healthy donor PBMCs using a second-generation CD19-CAR lentiviral vector. T cells will be activated using aAPC-Lipo or conventional CD3/CD28 beads (Dynabeads, positive control) and transduced with lentivirus. Transduction efficiency will be assessed by flow cytometry. Cultures will be maintained in medium with exogenous IL-2 for bead controls.

4.3 In Vitro Functional Assays

Expansion kinetics, phenotype (memory subsets, exhaustion markers), and functional capacity will be assessed. Key assays include:

• Proliferation and fold expansion (cell counting)

• Memory phenotyping (CD45RA, CCR7, CD62L, CD95) by flow cytometry

• Exhaustion profiling (PD-1, LAG-3, TIM-3, TOX expression)

• Cytotoxicity assays against CD19+ Nalm6 and OCI-Ly8 targets

• Cytokine secretion (IFN-γ, TNF-α, IL-2, granzyme B, perforin)

For mRNA delivery experiments, aAPC-Lipo will be loaded with DNRII mRNA or GFP mRNA control. Transfection efficiency and TGF-β resistance will be assessed by GFP expression, SMAD2/3 phosphorylation (western blot and flow), and maintained cytotoxic function in the presence of TGF-β1.

4.4 In Vivo Xenograft Studies

NSG mice will be inoculated intravenously with Nalm6-Luc or OCI-Ly8 cells (Day 0). On day 4, mice will receive CAR T cells expanded with either (a) CD3/CD28 beads, (b) optimal aAPC-Lipo formulation, or (c) vehicle control (PBS). A fourth group will receive CAR T cells followed by intravenous aAPC-Lipo booster injections post-transfer. Tumor burden will be monitored twice a week by bioluminescence imaging (IVIS). Peripheral blood CAR T cell persistence will be quantified by flow cytometry at days 7, 14, 21, and 35. Endpoint analysis (day 40) will include splenic CAR T cell counts, phenotypes and histopathological assessment. All animal studies will be conducted under IACUC-approved protocols.

4.5 Plan for Statistical Analysis

All experiments will be performed with a minimum of n = 3 independent biological replicates per condition (donors for in vitro; individual animals for in vivo), with technical triplicates for each assay where applicable. Sample sizes for in vivo studies are based on power calculations to detect ≥50% difference in tumor burden (α = 0.05, power = 0.80, estimated σ from pilot data), yielding n = 8 mice per group.

Statistical comparisons will be conducted as follows:

• Continuous parametric variables (fold expansion, cytotoxicity, cytokine levels): one-way ANOVA followed by Tukey's post-hoc test for multiple group comparisons; unpaired Student's t-test for two-group comparisons.

• Non-parametric data (skewed distributions, small n): Kruskal-Wallis test with Dunn's correction.

• Survival and tumor regression curves: Kaplan-Meier analysis with log-rank (Mantel-Cox) test.

• Longitudinal CAR T cell persistence: two-way ANOVA with repeated measures and Bonferroni correction.

• Correlation analyses (e.g., surface cytokine density vs. expansion fold): Pearson or Spearman correlation depending on normality (Shapiro-Wilk test).

All analyses will be performed using GraphPad Prism v10 and R (v4.3). Statistical significance will be defined as p < 0.05. Data will be presented as mean ± SEM unless stated otherwise. Outliers will be identified using Grubbs' test (α = 0.05) and reported transparently. All raw data, analysis scripts, and protocols will be deposited on OSF and linked to the ResearchHub preregistration upon completion.

4.6 Timeline

Development of liposomes: Months 1-2

In vitro CAR T generation and function validation: Months 2-5

In vivo studies: Months 6-12

4.7 Ethics and Data Management

All animal experiments will be conducted under an IACUC-approved protocol in compliance with institutional and federal guidelines for the humane care and use of laboratory animals. No human subjects research is involved. Primary human PBMCs will be sourced from the Blood Bank of UChicago Medicine using fully anonymized, consent-obtained donor material under an Institutional Review Board approved protocol.

All processed data and statistical results will be shared openly on the Open Science Framework (OSF) and GitHub. A preprint will be submitted to bioRxiv concurrently with journal submission. The study will be preregistered on ResearchHub prior to any analysis of outcome data.

5. Budget

The requested budget of $10,000 supports a 12-month research program. Costs are itemized below:

Category	Item	Cost (USD)	Justification
Materials & Supplies	Lipid components (POPC, DOPE-PEG, cholesterol)	$1,000	Formulation of aAPC liposomes across 3 designs
	Recombinant cytokines (IL-2, IL-15, IL-21)	$4,000	Surface conjugation and bioactivity assays
	Anti-CD3/CD28 antibodies for conjugation	$2,000	Activating signal display on liposome surface
	CAR T cell manufacturing reagents & vectors	$500	CD19-CAR lentiviral production and T cell activation
	In vitro cell line costs (Nalm6, OCI-Ly8, PBMC)	$500	Target and effector cell co-culture experiments
Equipment & Services	DLS nanoparticle characterization (core facility)	$300	Size, PDI, and zeta potential measurements
	Cryogenic TEM imaging (core facility)	$300	Morphological validation of liposome structures
	Flow cytometry (core facility)	$1400	Mandatory preprint + peer-reviewed journal
TOTAL		$10,000

6. References

1. Turtle CJ et al. (2012). Artificial antigen-presenting cells for use in adoptive immunotherapy. Cancer J. 16(4): 374-381.

2. Steenblock ER & Fahmy TM (2008). A comprehensive platform for ex vivo T-cell expansion based on biodegradable polymeric artificial antigen-presenting cells. Mol Ther. 16(4): 765-772.

3. June CH et al. (2018). CAR T cell immunotherapy for human cancer. Science. 359(6382): 1361-1365.

4. Lim WA & June CH (2017). The principles of engineering immune cells to treat cancer. Cell. 168(4): 724-740.

5. Kosmides AK et al. (2017). Biomimetic biodegradable artificial antigen presenting cells synergize with PD-1 blockade to treat melanoma. Biomaterials. 118: 16-26.

6. Schaft N et al. (2022). T cell costimulation and differentiation in the context of CAR T cell therapies. Cells. 11(3): 392.

Artificial Antigen-Presenting Cells for EnhancedCAR T Cell Persistence and Anti-Tumor Efficacy

Artificial Antigen-Presenting Cells for EnhancedCAR T Cell Persistence and Anti-Tumor Efficacy

Funding Progress

Funders

Nonprofit Recipient

Topics