Fighting exhaustion – even cells need help sometimes

Immune infiltration into tumors is largely driven by T cells, which provide an important line of defense against fast-growing cancer cells. This infiltration leads to inflammation, as cytotoxic T cells release cytolytic molecules to kill cancer cells, as well as recruit other cell types such as myeloid cells, B cells and NK cells. As this activity is prolonged and the elimination of target cells fails, chronic inflammation in tumors eventually leads to, you guessed it, exhausted T cells.

T cell exhaustion is well documented in all kinds of diseases, and is an integral part of the cancer-fighting process that we are only beginning to fully understand – a process that immunotherapy, particularly immune checkpoint blockades, aims to help modulate. If we can boost these exhausted T cells out of their current state and back into robust fighters, we predict that they will overcome barriers to combatting cancer and rid the body of disease.

While immune checkpoint blockade therapies have helped improve patient outcomes across certain cancers, breast cancer patients have not benefitted much from this novel therapeutic approach – likely influenced by variation in immune infiltrate activity, including T cell exhaustion, across subtypes. In fact, this complexity of the interaction between tumor and immune cells and the classification of different subtypes has made it difficult to rely on any one therapy for breast cancer. For example, two checkpoint inhibitors targeting the PD-1/PD-L1 pathway, atezolizumab and pembrolizumab, can increase progression-free survival in patients with triple-negative breast cancer but are ineffective for patients with luminal HER2-negative breast cancer subtypes.

Unfortunately, with little understanding of the mechanisms of resistance or response to immunotherapy, designing new immunotherapies and effectively using existing ones is a challenge the cancer community is trying to address.

Exposing T cell exhaustion

A foundational study out of ETH Zurich leveraged mass cytometry and Imaging Mass Cytometry™ to investigate the dynamics involved in T cell exhaustion as it applies to breast cancer. This work provides pivotal information that could facilitate the discovery of new therapeutic targets, and proposes a new and effective biomarker combination for patient stratification.

Figure 1: Sample selection and experimental approach

Single-cell technologies such as mass cytometry and Imaging Mass Cytometry provide a strong advantage for the analysis of immune cells and their behavior. Mass cytometry offers the ability to classify tumors by cell type, identifying cell subsets, cell frequencies and function in the same dataset. Multiplexed tissue imaging generates valuable insights into where these cells are located and how they move depending on their interactions within the tumor.

T cell exhaustion is characterized by expression of inhibitory checkpoint receptors such as PD-1, CTLA-4, Tim-3 and LAG-3. The process was originally associated with reduced effector functions, but now is known to correlate with continued proliferation and a highly inflammatory state. Myeloid cells, tumor-associated macrophages, B cells, NK cells and dendritic cells also play important roles in tumor infiltration, and are involved in pro- and antitumor effects including promoting angiogenesis and tissue remodeling.

The research group initially analyzed breast cancer samples by mass cytometry, finding that luminal breast tumors could be classified into three groups based on infiltrating T cell and myeloid cell phenotypes and their relationships with tumor cells, two of which are further investigated. One group showed a strong exhausted T cell phenotype, accounting for around 13% of tumors and indicating potential for immune checkpoint inhibitor therapy. A second group accounted for 48% of tumors and contained a high number of T cells but did not display signs of immune exhaustion.

Of note – this data shows the importance of informing both T cell phenotype and frequency. While the T cell frequencies between the two environments are relatively similar, significant distinguishing factors between the three groups can be gleaned from characterizing the detailed T cell phenotypes and tumor-immune associations.

Explaining cell communication and proximity

Figure 2: UMAP plot of scRNA-seq data colored by cell type

These groups were then further analyzed by RNA sequencing and Imaging Mass Cytometry to look at phenotypes, frequencies, communication patterns and spatial context. These technologies enable comprehensive single-cell analysis at the transcript and protein levels for a systematic analysis of tumor immune microenvironments (TIMEs) with and without evidence of T cell exhaustion. For imaging, a 42-marker protein panel and an RNA plus protein panel (12 RNA markers and 26 protein markers) were developed based on the RNA-seq data.

While transcriptomic analysis can predict intercellular communication, it is blind to spatial proximity, which is a key requirement for cellular interaction. Integrating IMC™ data added spatial mapping of cells predicted to interact, demonstrating whether these cells were physically proximal in the tissue and whether spatial patterns differ in the exhausted immune environments (IE1) and the non-exhausted environments (IE2).

A richly comprehensive view of the makeup of T cell exhausted versus non-exhausted environments reveals clear profiles and distinct signatures for each group. The IE1 exhausted immune environments display hallmarks of tumor reactivity and proliferation, and coincide with elevated major histocompatibility class I (MHC-I) expression on tumor cells. MHC-1 expression is indicative of tumor-immune interactions, as immune cells target tumor cells for destruction.

Exhausted environments also showed signs of exhaustion through an altered cytolytic profile and evidence of chronic inflammation with the presence of natural killer T (NKT) cells and immature tertiary lymphoid structures (TLSs), but protein expression levels showed existing capacity for tumor cell killing. Interestingly, mature TLSs are seen in non-exhausted environments.

Figure 3: Example IMC image showing staining patterns for the indicated markers (left). Single-cell masks colored by cell type (right). IMC staining patterns and single-cell masks were compared for all 77 images with similar results.

Biomarkers, signaling and therapeutic response

These findings suggest that PD-1, CXCL13 and MHC-I could serve as a new biomarker combination for patient stratification of the exhausted phenotype, which is predicted to be more responsive to immunotherapy.

Additionally, a map of cellular interactions within the breast TIME was generated, predicting strongly elevated immunomodulatory, chemotactic and cytokine signaling in tumors enriched in exhausted T cell environments. IMC data showed a more than five-fold higher proportion of cytokine-expressing cells in IE1 than in IE2 tumors especially for T, NK, stromal and myeloid cells. This data shows that spatial cytokine expression patterns differ in exhausted and non-exhausted environments, and that these expression patterns are linked to immune cell type distribution.

All in all, the study demonstrates fundamental differences between immune environments with and without signs of T cell exhaustion and that these differences can be explained by different immune escape mechanisms, including avoidance of tumor-specific T cell activation in non-exhausted environments and progressive T cell dysfunction through chronic inflammatory signaling in exhausted environments.

Unless explicitly and expressly stated otherwise, all products are provided for Research Use Only, not for use in diagnostic procedures. Find more information here.