Active Surveillance Characterizes Human Intratumoral T Cell Exhaustion

You, R. et al. (BioRxiv) DOI: 10.1101/2021.05.06.443033



  • Exhausted T cells

  • T cell migration

  • Two-photon imaging


Main Findings

The concept of exhaustion is now central in research focusing on the T cell dysfunction occurring in cancer over time. Exhausted T cells are mainly defined by a progressive but not complete loss of functions (cytotoxic, cytokine production, and homeostatic proliferation), changes in the epigenetic landscape, and the upregulation of inhibitory markers (such as PD1, LAG-3, and CTLA-4). Based on the latest, immunotherapies targeting specific phenotypic exhaustion markers drastically impact cancer treatment strategies. However, their limited success on solid tumours suggests that first other mechanisms are also involved in this chronic T cell dysfunction but also that “exhausted” T cells might have important functions independently from those phenotypic markers. Currently, studies are looking at deciphering the role/functions of those T cells described as “exhausted”. However, the lack of dynamic study on the migratory behaviour, and cellular interactions of this particular population limits our understanding of this state and its potential outcome.

In this non-peer-reviewed preprint, You et al. showed that “exhausted” T cells are not inert but rather still active with a different set of skills notably with an active program allowing them to greater migrate into tumours. In a previous paper, the authors used a murine tumour model where tumour-infiltrating T cells acquired a very motile behaviour over time which was correlating with the development of their exhaustion profile (Boldajipour et al. doi:10.1172/jci.insight.89289). This highlights the relevance of studying the dynamic T cell behaviour in vivo. Here, to study endogenous T cell population in humans, they developed a new technique using tumour biopsy slices that can recapitulate intravital imaging and developed non-stimulatory labelled antibodies to track human CD8+ T cells without impacting their biology. With optimized conditions, they assessed the migration behaviour of T cells in the different preserved tumour microenvironments from human tumour biopsies (collecting from various tumour types). They described a significant speed variation of T cells among the different tumour subregions in colorectal tumours which do not correlate with T cell and antigen-presenting cell (APC) density, APC-T cell interactions but rather inversely correlates with cancer cell density. Then, using a non-biased classification of human tumours, they demonstrated that CD8+ T cells motility was correlated with their immune states. Tumours with enriched phenotypic exhausted T cells displayed greater T cell motility in tumours. Furthermore, bulk-RNAseq datasets further confirm this phenotype by the upregulation of promotility cell migration-related gene pathways in the enriched phenotypic exhausted T cell tumours.

To conclude, in this preprint the authors showed preliminary data underlying the relevance of studying exhausted T cells function through the migratory aspect. They developed two-photon-based imaging allowing the study of endogenous human T cells behaviour in vivo.


  • Biopsies are often very small so they do not recapitulate the vasculature or soluble microenvironment for instance. This can be improved by more refined parameters.

  • They used a speed cut-off for classifying T cells motility from tumours quite low.

  • This preprint brings relevant descriptive preliminary data but they did not mention or perform mechanistic experiments.


This preprint described a new imaging technique allowing the study of endogenous T cell migration in human tumours using developed non-stimulatory labelled antibodies that do not disturb the T cell biology. This is a new step in the field as this technique does not require transferred cells. Therefore, it is more accurate to the real T cell migration in vivo. 

This is still very descriptive but this opens to many relevant fundamental questions to study such as what are the triggers for this different migration program and what are their related functions in tumours? This could also lead to further applications that can be used to classify patients or test the effect of personalized treatment on patient biopsies.


Reviewed by Julie M. Mazet 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. Follow her on Twitter.