top of page

Spatiotemporal co-dependency between macrophages and exhausted CD8+ T cells in cancer

Kersten K. et al. (BioRxiv) DOI: 10.1101/2021.09.27.461866

Spatiotemporal co-dependency between macrophages and exhausted CD8+ T cells in cancer


  • T cell exhaustion

  • Tumor associated macrophages

  • Tumor microenvironment

Main Findings

Tumor-associated macrophages (TAMs) and exhausted CD8+ T cells (Texh) are predominant immunosuppressive cell types in the tumor microenvironment (TME), and their frequency in tumors correlates with resistance to immune checkpoint blockade. Transcriptomic, imaging, and immune profiling studies have shown that TAMs and Texh colocalize in many tumor types, yet until this point, a causal relationship between TAM presence and T cell exhaustion in the TME has not been demonstrated. In this preprint, Kersten et al. describe a novel positive feedback loop between tumor-associated macrophages and exhausted CD8+ T cells in the tumor microenvironment.
The authors used mice bearing OVA-expressing B78 melanoma tumors and adoptive transfer of OVA-specific OT-I T cells to study tumor-reactive T cells. Macrophage depletion with anti-CSF1R antibodies decreased markers of exhaustion on tumor-infiltrating T cells, while CD8 depletion decreased antigen presentation markers on TAMs, establishing a correlation between T cell exhaustion and TAM antigen presentation. T cells with prolonged tumor residence (14 vs 4 days) showed increased transcript levels of exhaustion markers. To directly examine interactions between TAMs and Texh, the authors performed imaging studies revealing an immune synapse with clusters of TCRs at the contact points between these two cell types. TCR signaling induced persistent low levels of intracellular Ca2+ flux that nevertheless failed to support proliferation. In the apex of the paper, a spatial transcriptomics technique called Zipseq revealed spatially segregated transcriptomic signatures at the outer rim, middle region, and inner core of the tumor. Moving from the outer to the inner tumor, TAMs became more terminally differentiated with decreased antigen presentation, while CD8+ T cells became more exhausted, increasing CSF1 and CCL4 signaling with macrophages.
Altogether, the authors of this preprint found that TAMs and Texh interact in an antigen-specific immune synapse leading to persistent, low levels of TCR signaling and initiating an exhausted T cell program. Concurrently, exhausted T cells produce chemokines and growth factors to recruit monocytes to the tumor and induce their differentiation into TAMs, altering their antigen presentation capacity. These altered TAM and exhausted T cell phenotypes are found at higher frequencies in the tumor core and enhanced by hypoxia in that region. Together, these findings illustrate a novel circuit promoting two predominant immunosuppressive cell types in the TME.


The preprint could have benefitted from more granular dissection of T cell functional states. Exhausted T cells may be in early stages of exhaustion and responsive to checkpoint blockade, or terminally exhausted and resistant to therapy. Additionally, PD1 is expressed on effector T cells. Figures 1 and 2 utilized a robust Texh signature (PD1, CD38, TOX, Tim3, Lag3) that shifted to a more sparse, nonspecific signature toward the end of the paper, such as PD1 alone in Figure 5.
Likewise, though gene set scores in the scRNAseq dataset in Figure 6 relied on extensive Texh signatures, the most differentially expressed genes in exhausted T cells (Cd3g, Cd8a, Cd8b1, Gzmb, Nkg7, Ccl1) were not exclusively exhaustion markers. This may be due to inherent technical limitations of scRNAseq relating to low cell counts. Overall, a more consistent characterization of exhausted T cell phenotype and more extensive debate on implications for T cell function would have been appreciated.


Immunosuppression in the TME is a major mechanism of immunotherapy resistance. Thus, understanding how TAMs and Texh are promoted and sustained in the TME is critical to enhancing immunotherapy responses. This is an excellent preprint demonstrating, for the first time, a causal link between presence and function of TAMs and Texh in solid tumors. This is highly significant as these are the two most abundant immunosuppressive cell types in the TME, making them attractive therapeutic targets. Indeed, many therapies targeting TAMs or alternative T cell checkpoints are currently in clinical trials. This preprint highlights the need for therapeutic strategies targeting both TAMs and Texh to overcome this immunosuppressive feedback loop and promote the anti-tumor immune response.


Reviewed by Gabrielle Lubitz as part of the cross-institutional journal club of the Immunology Institute of the Icahn School of Medicine, Mount Sinai and the Kennedy Institute of Rheumatology, University of Oxford.

bottom of page