Combating the immunosuppressive tumor microenvironment
Tumors rarely exist in isolation but are instead surrounded by a supportive network of cells and surrounding tissue known as the tumor microenvironment (TME). In many cancers, both tumor cells and surrounding non-malignant cells contribute to creating an immunosuppressive environment that protects the tumor from immune attack. One of the key molecules involved in this process is programmed death-ligand 1 (PD-L1) which directly dampens anti-tumor T-cell responses.
For more than a decade, our research group has explored ways to overcome this suppressive state by harnessing a subset of naturally occurring “anti-regulatory” T cells. Rather than solely targeting tumor cells, these T cells recognize antigens derived from immunosuppressive molecules expressed by both malignant and non-malignant cells within the TME. Among the most compelling examples are PD-L1-specific T cells. We have previously shown that these cells exist in both healthy donors and cancer patients and can eliminate PD-L1-expressing tumor and immune cells. Encouragingly, their therapeutic potential has also translated into the clinic, where PD-L1- and indoleamine 2,3-dioxygenase (IDO)-based peptide vaccination combined with anti-PD-1 therapy produced promising responses in patients with metastatic melanoma.
These findings raised an exciting question for us: could we expand the therapeutic potential of PD-L1-specific T cells towards a dedicated cellular therapy?
That became the starting point for this study.
First steps: reconstructing a T-cell receptor specific for a PD-L1-derived epitope
To translate PD-L1-specific T-cell immunity into a cellular therapy approach, we first sought to identify the T-cell receptor (TCR) responsible for recognizing PD-L1-derived epitopes. Specifically, we isolated a monoclonal T-cell population specific for the HLA-A2-restricted, PD-L1-derived peptide PDL101 from a patient with breast cancer and reconstructed the cognate TCR.
With the TCR sequence in hand, we used a non-viral CRISPR-Cas9-based strategy to replace the endogenous TCR of primary T cells with the newly identified PDL101-specific TCR, thereby redirecting their specificity (Figure 1). Seeing the engineered cells stain positively with fluorescent tetramers for the first time was a particularly exciting moment in the project and a clear indication that the new TCR had been successfully introduced.
Encouragingly, our initial functional assays confirmed that the engineered PDL101-TCR-T cells were fully functional and antigen-responsive. The cells showed potent proinflammatory responses when exposed to their antigen and efficiently killed target cells presenting the PDL101 epitope, demonstrating strong and specific antigen-dependent cytotoxicity.
Targeting both tumor cells and the tumor microenvironment
A key feature of PD-L1 as a target is that it is not restricted to cancer cells but is also expressed by multiple immunosuppressive cell populations within the TME.
This opened the possibility of a dual-targeting strategy with the engineered PDL101-TCR-T cells (Figure 2).
In our experiments, the engineered T cells selectively recognized and killed a PD-L1-positive cancer cell line, while sparing its PD-L1-negative sister cell line. Importantly, they also reacted to suppressive myeloid populations resembling tumor-associated macrophages (TAMs) and myeloid-derived suppressor cells (MDSCs)/M2-like cells.
These findings were exciting because they suggested that these T cells may not only attack the tumor directly but also help dismantle the surrounding immunosuppressive environment. Rather than targeting a single malignant population, PD-L1-specific TCR-T cells could “inflame” the tumor microenvironment and thereby promote broader anti-tumor immune responses.
Translational considerations: balancing efficacy and safety
During CRISPR-Cas9 editing, we observed a relatively low integration and expression rate of the transgenic TCR compared to traditional viral transduction. This led us to ask whether the engineered T cells could be boosted by vaccination. Indeed, the proportion of PDL1-specific T cells increased markedly after repeated stimulations with peptide-loaded dendritic cells in vitro. While this was only a proof-of-concept, it suggests a potential strategy of combining adoptive TCR-T-cell therapy with peptide vaccination to expand and sustain transferred cells in patients.
As with all therapies targeting broadly expressed immune molecules, safety remains an important consideration. PD-L1 is not exclusive to tumors, and previous PD-L1-targeted CAR-T approaches have raised concerns about on-target, off-tumor toxicity. Our approach differs in an important way: instead of using an artificially engineered CAR, we use a naturally derived, MHC-restricted TCR specificity.
We speculate that this physiological recognition mechanism may provide a higher activation threshold and potentially a more favorable safety profile, although this will need careful evaluation in future preclinical studies.
Looking ahead
In summary, this study represents the first steps toward translating the biology of anti-regulatory PD-L1-specific T cells into a defined cellular therapy. By combining naturally occurring TCR specificity with precise CRISPR-based engineering, we showed that PD-L1-specific TCR-T cells can target both tumor cells and immunosuppressive cells relevant within the TME.
This platform lays the groundwork for further research into immune modulatory cellular therapies that are designed not only to attack cancer directly, but also to reshape the environment that allows tumors to persist.