OPS-γδ: allogeneic opsonin-secreting γδT cell immunotherapy for solid tumours mediates direct and bystander immunity

Daniel Fowler, Marta Barisa, Alba Southern, Callum Baird Nattress, Elizabeth Hawkins, Eleni Vassalou, Angeliki Kanouta, John R. Counsell, Enrique Miranda, Petra Vlckova, Benjamin Draper, Christopher J. Tape, Kerry Chester, John Anderson, Jonathan FisherbioRxiv 2022 · DOI

Abstract

Traditional T cell therapies for solid tumors face significant challenges, including manufacturing complexity, limited efficacy of non-transduced cells, and the immunosuppressive tumor microenvironment. We present OPS-γδ, an allogeneic platform utilizing Vγ9Vδ2 γδ T cells engineered to secrete synthetic opsonins and stabilized IL-15 (stIL15). This approach leverages the innate anti-tumor properties of γδ T cells while recruiting bystander immunity through secreted payloads. Our results demonstrate that OPS-γδ cells exhibit superior cytotoxicity, persistence, and the ability to "license" unmodified cells within the therapeutic product, offering a potent, off-the-shelf solution for solid tumor immunotherapy.

Introduction: Addressing the Hurdles of Solid Tumor Immunotherapy

Current T-cell therapies, particularly chimeric antigen receptor (CAR) T cells, primarily rely on membrane-bound receptors. While successful in hematological malignancies, they face three major barriers in solid tumors:

  1. Manufacturing Complexity: Most CAR-T products are autologous, requiring bespoke manufacturing for each patient, which is costly and slow.
  2. The "Inactive Fraction": Only cells successfully expressing the transgene provide therapeutic benefit, leaving a large portion of the manufactured product as "dead weight."
  3. Solid Tumor Resistance: The dense, immunosuppressive environment of solid tumors often leads to T-cell exhaustion and poor penetration.

To overcome these, we developed an allogeneic (off-the-shelf) platform using γδ T cells. These cells are ideal candidates as they do not cause Graft-versus-Host Disease (GvHD) and naturally possess Antibody-Dependent Cellular Cytotoxicity (ADCC) capabilities.

Methods and Platform Design

We engineered Vγ9Vδ2 γδ T cells to serve as "micro-factories" for two key secreted payloads:

  • Synthetic Opsonins (SFPs): Fusion proteins designed to bind tumor antigens (e.g., GD2) and provide an "eat me/kill me" signal via an Fc region.
  • Stabilized IL-15 (stIL15): A cytokine module that enhances T-cell survival and activation without the need for exogenous supplementation.
Engineering and functional validation of the OPS-γδ platform, including construct design (A), binding assays (B), and signaling heatmaps (E-L).

The platform was validated across various cell lines (SupT1, Kelly) and 3D patient-derived osteosarcoma (PDOS) models. We utilized CyTOF (Mass Cytometry) and the DREVI/DREMI computational framework to analyze signaling dynamics, specifically investigating how secreted modules affect the entire cell population regardless of individual transgene expression levels.

Key Insights: The Neighborhood Effect and Bystander Recruitment

A defining feature of the OPS-γδ platform is the "Neighborhood Effect." Unlike membrane-bound CARs, secreted opsonins and IL-15 benefit the entire therapeutic batch and the surrounding environment.

  • Licensing the Unmodified Fraction: Non-transduced cells (often up to 50% of a batch) are "licensed" to kill by the secreted payloads, maximizing the utility of the entire manufactured product.
  • Bystander Activation: Secreted opsonins effectively "paint" tumor cells, recruiting bystander immune cells like NK cells and macrophages to participate in the attack.
  • Decoupled Signaling: DREMI analysis revealed that signaling in OPS-γδ cells is uniform across the population, avoiding the "tonic signaling" and subsequent exhaustion often seen in high-expression CAR-T cells.

Key Results: Superior Cytotoxicity and Persistence

OPS-γδ cells demonstrated significantly enhanced performance compared to unmodified γδ T cells:

  • Enhanced Killing: OPS-γδ cells showed potent, antigen-specific cytotoxicity against GD2+ tumor cells compared to unmodified controls.
  • Superior Persistence: In mouse models, stIL15-OPS-γδ cells persisted in the blood and maintained tumor control significantly longer than unmodified cells, even without exogenous cytokine support.
  • Phenotypic Optimization: Transduction triggered a shift to a CD16a+/CD16b- phenotype, the optimal configuration for potent ADCC-mediated tumor killing.
  • Synergy with Zoledronic Acid: Combining OPS-γδ therapy with zoledronic acid (a common bone drug) sensitized osteosarcoma cells, further boosting efficacy in 3D models.
Enhanced cytotoxicity (C) and expansion capacity (D) of OPS-γδ cells compared to unmodified controls.

Limitations and Future Directions

While promising, several areas require further investigation. Primary in vivo studies were conducted in immunodeficient NSG mice; future work in humanized models is needed to fully assess the recruitment of a complete human immune system. Additionally, while we focused on GD2, the platform is modular; opsonins can be swapped to target other antigens (e.g., CD20, CEACAM5) to mitigate the risk of tumor antigen escape. Finally, achieving deep penetration into large, dense solid tumors remains a universal challenge for cellular therapy that may require additional homing modifications.

In conclusion, stIL15-OPS-γδ cells represent a novel allogeneic platform that combines direct cytolysis with bystander activation to effect solid tumor control.

Summary updated

Abstract

Abstract T cell-based cancer immunotherapy has typically relied on membrane-bound cytotoxicity enhancers such as chimeric antigen receptors expressed in autologous αβT cells. These approaches are limited by tonic signalling of synthetic constructs and costs associated with manufacture of bespoke patient products. γδT cells are an emerging alternative chassis for cellular therapy, possessing innate anti-tumour activity, potent antibody-dependent cytotoxicity (ADCC) and minimal alloreactivity. We present an immunotherapeutic platform technology built around the Vγ9Vδ2 γδT cell chassis, harnessing specific characteristics of this cell type and offering an allo-compatible means of delivering cellular therapy that recruits bystander immunity. We engineered γδT cells to secrete synthetic opsonins and stabilized IL15 (stIL15). Using GD2 as a model antigen we show how opsonin-secreting Vγ9Vδ2 (OPS-γδ) have enhanced cytotoxicity and also confer this benefit on lymphoid and myeloid bystander cells. Reflecting the secreted nature of the engineered efficacy modules, the entire product rather than just the gene-modified fraction exhibited enhanced activation and cytotoxic profiles, superior persistence and proliferative capacity even upon repeated tumour challenge. Secretion of stIL15 abrogated the need for exogenous cytokine supplementation during expansion and further mediated functional licensing of bystander NK cells. Compared to unmodified γδT cells, stIL15-OPS-γδ cells exhibited superior in-vivo control of subcutaneous tumour and persistence in the blood. stIL15-OPS-γδ cells were further efficacious in 3D patient-derived osteosarcoma models, where efficacy could be boosted with the addition of immunomodulatory aminobisphosphonate drug, zoledronic acid. Together the data identify stIL15-OPS-γδ cells as a novel allogeneic platform combining direct cytolysis with bystander activation to effect solid tumour control. One Sentence Summary Armoured, opsonin-secreting OPS-γδ cell immunotherapy is built on the innate strengths of the Vγ9Vδ2 cell chassis for allogeneic solid tumour targeting.
BibTeX
@misc{fowler2022ops,
  author = {Daniel Fowler and Marta Barisa and Alba Southern and Callum Baird Nattress and Elizabeth Hawkins and Eleni Vassalou and Angeliki Kanouta and John R. Counsell and Enrique Miranda and Petra Vlckova and Benjamin Draper and Christopher J. Tape and Kerry Chester and John Anderson and Jonathan Fisher},
  title = {{OPS-γδ: allogeneic opsonin-secreting γδT cell immunotherapy for solid tumours mediates direct and bystander immunity}},
  year = {2022},
  doi = {10.1101/2022.10.23.513387},
  howpublished = {bioRxiv},
}
Figure