Tregs in gliomas – the jury is still out

Regulatory T cells (Tregs) promote tumor growth chiefly by suppressing tumor-specific T cell responses in the tumor tissue. This is the common notion that provides the rational for targeting Tregs in cancer with the aim at enhancing antitumor immune responses. If this rational is true and if we think that there is a natural – albeit insufficient - antitumor immune response, then tumor-infiltrating and/or circulating Tregs should influence the natural course of tumor disease with high numbers predicting poor outcome. Conversely, the number of intratumoral T cells with antitumor properties, particularly CD8+ cytotoxic T lymphocytes, ought to predict a favorable outcome. While most studies analysing tumor-infiltrating T cells in gliomas support the latter, there has been ambiguous data on the relevance of tumor-infiltrating Tregs in human gliomas.^1–6

Tregs are classically viewed synonymous with FoxP3 + CD25 + CD4+ T cells. These cells are usually identified by flow cytometry from peripheral blood or preparations of mononuclear cell suspensions from fresh tumor specimens. In the current study by Thomas and colleagues (this issue) tumor-infiltrating Tregs and CD3+ T cells were measured not by flow cytometry but by epigenetic qPCR and by immunohistochemistry. This approach offers the significant advantage of performing analyses on archival tissue but at a risk of losing specificity. Although FoxP3 transcripts and methylation patterns are generally considered to be specific for Tregs, expression has been detected in CD8+ T cells⁷ and non-lymphoid CNS tissue.⁸ Hence, Treg quantifications based on FoxP3 expression without single-cell resolution have to be viewed with caution. With their methods Thomas and colleagues did not find peripheral or tumor-infiltrating Tregs to be predictive of outcome in their small study cohort.

Does this and the negative results of most previous studies mean that Tregs are not important in glioma biology? FoxP3 + CD4+ Tregs in humans comprise several subsets, which are heterogeneous both with respect to phenotype and function. For instance, FoxP3^loCD45RA + CD25^lo cells are naive or resting Treg (rTreg) cells, which differentiate into highly suppressive FoxP3^hiCD45RA-CD25^hi effector Treg (eTreg) cells upon antigenic stimulation. In contrast, FoxP3^loCD45RA-CD25^lo non-Treg cells do not possess suppressive activity but can secrete pro-inflammatory cytokines.⁹ In addition, there is an increasing arsenal of additional markers subdividing the Treg family including CD127, Helios, CTLA-4 and CD39. Furthermore, the diversity of T cells extends well into the CD8+ compartment including FoxP3 expression.⁷ Future studies ought to incorporate the complexity of Treg development and differentiation including antigen-specificity to answer this question.

Is this a worthwhile effort or just l'art pour l'art? With the increasing arsenal of checkpoint inhibitors spilling over into the glioma arena we need to provide answers to the central questions: Is there a meaningful natural (antigen-specific) anti-glioma T cell immunity and if so is this immunity suppressed by cellular mediators, such as Tregs, resulting in a net poorly immunogenic tumor phenotype? Because only the presence of a meaningful natural anti-glioma T cell immunity would then provide the rational to release the break on this anti-glioma T cell immunity by introducing checkpoint inhibitors or other strategies to inhibit tumor-associated immune suppression (e.g. Tregs) as single agents. This has been demonstrated for melanoma,¹⁰ but solid evidence for gliomas is lacking. Consequently, clinical studies testing checkpoint inhibitors in gliomas clearly should be backed by an in-depth analysis of pre- and post-treatment tumor tissue including analyses of tumor-infiltrating Tregs and the repertoire of antigen-specific effector T cells. In addition to the relevance of Tregs for suppressing the natural anti-glioma immunity, this immune cell population may constitute a significant barrier for antigen-specific peptide vaccines, which are on the verge of entering the clinical arena.^11–15 If tumor-infiltrating Tregs (or even circulating Tregs) counteracted this induced antigen-specific anti-glioma T- and B cell immunity, then this would again constitute a strong rationale for combining peptide vaccines with checkpoint inhibitors, agents depleting Tregs or simply temozolomide chemotherapy upfront. Also here, clinical trials evaluating the efficacy of antigen-specific vaccines ought to incorporate pre- and post-treatment analyses of mechanisms preventing the induction of an effective anti-glioma immune response including Tregs. Importantly, these analyses need to take into account the heterogeneity and tissue-specificity of Tregs, which extends well beyond the classic CD4 + CD25 + FoxP3+ phenotype, which most studies including the study by Thomas and colleagues are restricted to.

Funding

This work was supported by grants of the Helmholtz Foundation (VH-NG-306) to M.P. and the German Research Foundation (SFB938 TPK) to M.P. and W.W.

Conflict of interest statement. The authors hold patents on IDH1R132H as an immunotherapeutic target.