Tumor-responsive, multifunctional CAR-NK cells cooperate with impaired autophagy to infiltrate and target glioblastoma

Jiao Wang, Sandra Toregrosa-Allen, Bennett D. Elzey, Sagar M. Utturkar, Nadia A. Lanman, Victor Bernal-Crespo, Matthew M. Behymer, Gregory T. Knipp, Yeonhee Yun, Michael C. Veronesi, Anthony L. Sinn, Karen E. Pollok, Randy R. Brutkiewicz, Kathryn Nevel, Sandro MatosevicbioRxiv 2020 · DOI

TL;DR

This paper introduces a novel, multi-functional CAR-NK cell therapy that simultaneously targets key resistance mechanisms in glioblastoma (GBM), including antigen escape, poor immune cell infiltration, and metabolic suppression. By combining engineered NK cells with a tumor-responsive drug release system and targeting autophagy, the therapy shows potent anti-tumor activity in preclinical models, paving the way for new GBM immunotherapies.

What Problem Does This Paper Address?

Glioblastoma (GBM) is a notoriously difficult-to-treat brain cancer due to several factors:

  • Tumor Heterogeneity: GBM tumors are not uniform; cells within them have diverse genetic and epigenetic profiles, making it hard for single-target therapies to be effective. This leads to "antigen escape," where cancer cells lacking the targeted antigen survive and regrow.
  • Immunosuppressive Tumor Microenvironment (TME): The environment around the tumor actively suppresses immune responses, preventing immune cells like Natural Killer (NK) cells from effectively attacking the cancer.
  • Poor Immune Cell Homing: Even if NK cells could overcome the immunosuppression, they struggle to reach the tumor site in sufficient numbers.
  • Immunometabolic Reprogramming: Cancer cells can manipulate the metabolic landscape of the TME to their advantage, further hindering immune cell function.

Existing immunotherapies, including CAR-T cells targeting single antigens, have shown limited success in GBM, failing to improve patient survival. This paper aims to overcome these challenges by developing a more sophisticated, multi-pronged approach.

Methods

The researchers developed a novel, "multifunctional" CAR-NK cell therapy. Here's a breakdown of the key components:

  • Engineered CAR-NK Cells: They genetically engineered human NK cells to express multiple targeting functions.
  • Dual Chimeric Antigen Receptors (CARs): These CARs allow the NK cells to recognize and bind to two different GBM-associated antigens: GD2 and NKG2D ligands. This dual targeting helps overcome antigen escape.
  • Tumor-Responsive Antibody Fragment Release: A key innovation is the inclusion of an antibody fragment targeting CD73. This fragment is linked to the CAR via a "cleavable, tumor-sensitive linker." This linker is designed to be broken down by proteases that are upregulated in the GBM TME. This means the CD73-blocking antibody is released specifically at the tumor site, minimizing systemic side effects.
  • Targeting Autophagy: The researchers investigated the role of autophagy (a cellular self-cleaning process) in GBM resistance. They used both genetic (shRNA to knock down the BECN1 gene) and pharmacological (chloroquine, CQ) methods to inhibit autophagy in GBM cells.
  • In Vitro Assays: Standard assays were used to evaluate NK cell function, including:
  • Killing Assays: To measure the ability of engineered NK cells to kill GBM cells.
  • Degranulation Assays: To assess NK cell activation.
  • Cytokine Production Assays: To measure the release of immune-signaling molecules like IFN-γ.
  • CD73 Enzymatic Activity Assays: To confirm the functional blockade of CD73 and its effect on adenosine production.
  • In Vivo Xenograft Models:
    • Subcutaneous Xenografts: Patient-derived GBM cells were implanted under the skin of immunocompromised mice (NSG mice) to test the anti-tumor efficacy of the engineered NK cells and CQ treatment.
  • Orthotopic Xenografts: GBM cells engineered to express luciferase were implanted directly into the brains of immunocompromised mice (NSG mice) to better mimic the natural tumor location and allow for bioluminescence imaging to track tumor growth.
  • Immunohistochemistry (IHC) and Immunofluorescence (IF): These techniques were used to visualize and quantify NK cell infiltration, as well as the expression of target antigens and chemokines within tumor tissues.
  • Bioinformatics Analysis: RNA-seq data from GBM patients (TCGA) was analyzed to correlate gene expression patterns with NK cell presence and identify potential therapeutic targets.

Datasets

  • TCGA (The Cancer Genome Atlas) Glioblastoma (GBM) RNA-seq data: This dataset, comprising 156 patients, was used for bioinformatics analysis to understand gene expression patterns related to NK cell function and potential therapeutic targets in GBM.
  • Patient-derived GBM xenografts: These were used to establish subcutaneous and orthotopic tumor models in mice for in vivo efficacy studies.

Key Insights and Ideas

  • Multifunctionality is Key: A single-cell therapy that can simultaneously address multiple GBM resistance mechanisms (antigen escape, TME suppression, poor homing) is more effective than single-target approaches.
  • Tumor-Responsive Drug Release: Localizing the release of therapeutic agents (like the anti-CD73 antibody) to the tumor site is crucial for maximizing efficacy and minimizing systemic toxicity. This is achieved through the clever use of tumor-specific proteases to cleave a linker.
  • Autophagy Inhibition as an Immuno-Modulator: Blocking autophagy in GBM cells doesn't just directly affect the cancer; it also reprograms the TME to be more favorable for NK cell infiltration and function. This is a novel way to enhance immunotherapy.
  • Synergistic Effects: Combining the engineered CAR-NK cells with autophagy inhibition (CQ) leads to significantly enhanced anti-tumor responses compared to either treatment alone.
  • Targeting CD73 is Crucial for NK Cell Function: CD73's role in producing adenosine creates an immunosuppressive environment that impairs NK cell activity. Blocking CD73 locally helps restore NK cell function and survival within the tumor.

Key Results

  • Engineered NK Cells Show Potent Cytotoxicity: Both NK-92 cell lines and primary NK (pNK) cells engineered with the multifunctional construct demonstrated significantly higher killing of patient-derived GBM cells compared to control cells.
  • CD73 Blockade Reduces Adenosine: The release of the anti-CD73 antibody fragment effectively reduced extracellular adenosine levels in the TME, mitigating immunosuppression.
  • Autophagy Inhibition Enhances NK Cell Homing: Inhibiting autophagy in GBM cells led to increased expression of chemokines (CCL5 and CXCL10), which promoted NK cell migration into the tumor bed.
  • Combined Therapy Significantly Inhibits Tumor Growth: In both subcutaneous and orthotopic GBM xenograft models, the combination of CQ and CD73.mCAR-pNK cells resulted in the most significant reduction in tumor growth and improved survival.
  • No Obvious Toxicity to Normal Cells: The engineered NK cells showed no preferential killing of normal brain cells, indicating a good safety profile.

Limitations and Future Work

  • Administration Regimen: The paper notes that optimizing the administration regimen, dosage, and frequency of adoptive transfer is an area for future investigation.
  • Immune Landscape of GBM Tumors: Further studies are needed to fully understand how targeting autophagy affects the broader immune landscape within GBM tumors and the infiltration of other immune cell types. This could be explored in immunocompetent GBM models.
  • Longer-Term Efficacy and Survival: While the current study shows promising results in tumor growth inhibition, longer-term efficacy studies focusing on survival will be pursued in future investigations.
  • Clinical Translation: The ultimate goal is to advance this immunotherapy toward human trials, which will require further validation and optimization.

Summary updated

Abstract

Abstract Tumor antigen heterogeneity, a severely immunosuppressive tumor microenvironment (TME) and lymphopenia resulting in inadequate immune intratumoral trafficking have rendered glioblastoma (GBM) highly resistant to therapy. As a result, GBM immunotherapies have failed to demonstrate sustained clinical improvements in patient overall survival (OS). To overcome these obstacles, here we describe a novel, sophisticated combinatorial platform for GBM: the first multifunctional immunotherapy based on genetically-engineered, human NK cells bearing multiple anti-tumor functions, including local tumor responsiveness, that addresses key drivers of GBM resistance to therapy: antigen escape, poor immune cell homing, and immunometabolic reprogramming of immune responses. We engineered dual-specific CAR-NK cells to bear a third functional moiety that is activated in the GBM TME and addresses immunometabolic suppression of NK cell function: a tumor-specific, locally-released antibody fragment which can inhibit the activity of CD73 independently of CAR signaling and decrease the local concentration of adenosine. The multifunctional human NK cells targeted patient-derived GBM xenografts, demonstrated local tumor site specific activity in the tissue and potently suppressed adenosine production. We also unveil a complex reorganization of the immunological profile of GBM induced by inhibiting autophagy. Pharmacologic impairment of the autophagic process not only sensitized GBM to antigenic targeting by NK cells, but promoted a chemotactic profile favorable to NK infiltration. Taken together, our study demonstrates a promising new NK cell-based combinatorial strategy that can target multiple clinically-recognized mechanisms of GBM progression simultaneously.
BibTeX
@misc{wang2020tumor,
  author = {Jiao Wang and Sandra Toregrosa-Allen and Bennett D. Elzey and Sagar M. Utturkar and Nadia A. Lanman and Victor Bernal-Crespo and Matthew M. Behymer and Gregory T. Knipp and Yeonhee Yun and Michael C. Veronesi and Anthony L. Sinn and Karen E. Pollok and Randy R. Brutkiewicz and Kathryn Nevel and Sandro Matosevic},
  title = {{Tumor-responsive, multifunctional CAR-NK cells cooperate with impaired autophagy to infiltrate and target glioblastoma}},
  year = {2020},
  doi = {10.1101/2020.10.07.330043},
  howpublished = {bioRxiv},
}
Figure