Local γδ T cells: translating promise to practice in cancer immunotherapy

Abstract

Rapid bench-to-bedside translation of basic immunology to cancer immunotherapy has revolutionised the clinical practice of oncology over the last decade. Immune checkpoint inhibitors targeting αβ T cells now offer durable remissions and even cures for some patients with hitherto treatment-refractory metastatic cancers. Unfortunately, these treatments only benefit a minority of patients and efforts to improve efficacy through combination therapies utilising αβ T cells have seen diminishing returns. Alongside αβ T cells and B cells, γδ T cells are a third lineage of adaptive lymphocytes. Less is known about these cells, and they remain relatively untested in cancer immunotherapy. Whilst preclinical evidence supports their utility, the few early-phase trials involving γδ T cells have failed to demonstrate convincing efficacy in solid cancers. Here we review recent progress in our understanding of how these cells are regulated, especially locally within tissues, and the potential for translation. In particular, we focus on the latest advances in the field of butyrophilin (BTN) and BTN-like (BTNL) regulation of γδ T cells and speculate on how these advances may address the limitations of historical approaches in utilising these cells, as well as how they may inform novel approaches in deploying these cells for cancer immunotherapy.

Fig. 1. Expression of BTN3A isoforms in normal tissues and solid cancers.

Gene expression profiles of each BTN3A isoform were extracted from the OncoDB database (https://oncodb.org/cgi-bin/genomic_normal_expression_search.cgi) and plotted as log₂ fold change between median expression levels in each cancer and corresponding normal tissue. Red fill denotes increased expression in a specific cancer type compared with corresponding normal tissue. Blue fill denotes increased expression in normal tissue compared with corresponding cancer. Size of circles are proportional to log₂ fold change. Black border denotes unadjusted P value < 0.05.

Fig. 2. Translating tissue biology of Vδ1 to effective cancer immunotherapy.

a Adoptively transferred, tissue-derived Vδ1 T cells (left panel) may preferentially traffic to and accumulate in target organs dependent on tissue-specific BTNL expression (middle panel). This could reduce detrimental activation in uninvolved organs and thus improve therapeutic windows (middle panel). Within target organs, the hypothetical logic-gate functionality of the TCR may provide further fine-tuning of Vδ1 T-cell activation to target neoplastic cells whilst sparing healthy cells. b Clinically relevant TAAs, such as human epidermal growth factor receptor 2 (HER2), which are expressed at low levels on most healthy epithelial cells and only modestly upregulated on cancer cells (e.g., gastric cancer, treatment-resistant HER2⁺ breast cancers) can be difficult to target with tolerable safety windows using a “single argument” approach (e.g., a monoclonal antibody). The hypothetical logic-gate functionality of the γδ TCR to permit activation based on both the absence of normality and the presence of stress can be exploited for increased therapeutic windows. Bispecific engagers, which have excellent tissue penetration, can recognise TAAs via a monoclonal antibody (mAb) domain whilst engaging γδ T cells via a TCR stress ligand. Binding of bispecific engagers to physiologically expressed TAAs on healthy cells would not be sufficient to trigger γδ T-cell activation as these cells still express normality-associated tissue-specific BTNL molecules (left panel). On the other hand, the downregulation of BTNL molecules on neoplastic cells in combination with bispecific engagement permits activation of γδ T cells within tumours (middle and left panel). Moreover, the modular nature of bispecific antibodies allows for bespoke tuning by targeting of other activating axes on γδ T cells (e.g., innate NK receptors).