Uncovering CD4⁺ T-cell exhaustion in B-ALL

In this issue of Blood, Tracy et al¹ describe a key subset of exhausted CD4⁺ T cells in Philadelphia chromosome positive (Ph⁺) B-cell acute lymphoblastic leukemia (B-ALL) that modulates antileukemic effects. The authors use a nonirradiated immunocompetent murine model of Ph⁺ B-ALL and primary patient samples to dissect the CD4 T-cell immune response and to provide a rationale for combination therapies of tyrosine kinase inhibitors that target the BCR-ABL oncoprotein and programmed death-ligand1 (PD-L1) blockade to reverse the exhausted state of CD4⁺ T cells and enhance the clearance of leukemia.

T-cell activation is orchestrated by intrinsic positive and negative T-cell regulators, with inhibitory molecules setting thresholds to avoid uncontrolled immune activation. However, within tumors, continuous antigen exposure in the setting of an immunosuppressive microenvironment shifts the T-cell activation toward an inhibitory exhausted state that results in loss of antitumor function. Although CD8⁺ T-cell exhaustion in chronic infections and cancers has been explored in detail,² the characteristics and pathological impact of CD4⁺ T-cell exhaustion in cancer, and specifically in hematologic malignancies, is now becoming apparent. The expanded cytotoxic role of CD4⁺ T cells predicts a pattern of T-cell exhaustion comparable to that of the CD8⁺ exhausted T-cell phenotype. Indeed, recent discovery of CD4⁺ T cells with exhaustion properties in murine melanoma models and the association of phenotypically exhausted CD4⁺ T cells with inferior outcomes in leukemias support this observation.^3,4

Studies of CD8⁺ T-cell exhaustion have previously shown that identification of T-cell exhaustion extends beyond phenotypic characterization of co-inhibitory molecules and now includes heterogeneous populations with diverse functional, transcriptional, and epigenetic profiles,² most of which have not yet been clearly defined for exhausted CD4⁺ T cells. To help clarify these issues, Tracy et al described the transcriptomic profile of CD4⁺ T-cell exhaustion in a murine model of Ph⁺ B-ALL that was characterized by the expression of programmed cell death protein 1 (PD-1), lymphocyte activating 3, and T cell immunoglobulin and mucin domain-containing protein 3 (TIM3), whereas functional studies confirm loss of tumor necrosis factor-α with preserved interferon-γ (IFN-γ) production. When drawing parallels to CD8⁺ exhausted T cells, these functional characteristics in CD4⁺ T cells phenocopy so-called terminally exhausted CD8⁺ T cells, an effector cytokine (particularly INF-γ)-producing population without self-renewing properties. Evidence suggests that these cells develop from progenitor exhausted T cells, in which cytokine function is downregulated but proliferative self-renewal capacity remains intact.^5,6 Interestingly, this progenitor exhausted T-cell population responds to checkpoint inhibition by conversion into functional tumor-directed effector cells, which contrast with terminally exhausted T cells that are mostly unaffected by this treatment modality.⁵ Expression patterns of the transcription factor TCF1 (encoded by TCF7) or SLAMF6 distinguish terminally differentiated from progenitor or stem-like exhausted CD8⁺ T cells in which these transcription programs are upregulated.⁵ Whether CD4⁺ T cells undergo a similar transitional program from progenitor to terminally exhausted T-cell state is currently unknown. The article by Tracy et al suggests that at least in the B-ALL murine leukemia model and in samples from human patients with primary leukemia at diagnosis, the phenotypic, functional, and transcriptional profile of CD4⁺ T cells resembles the terminally exhausted phenotype of CD8⁺ T cells with upregulation of PD-1, TIM3, and IFN-γ and downregulation of TCF1. Interestingly, in contrast to CD8⁺ terminally exhausted T cells, this CD4⁺ T-cell population demonstrates response to PD-1 blockade with reversal of exhaustion transcriptional features and expansion of leukemia-specific T-cell clones despite expression of multiple inhibitory receptors. These results highlight the importance of performing additional transcriptional and epigenetic studies to better understand the pathologic impact of cell states, including phenotypically exhausted CD4⁺ T cells.

Despite the observation of leukemic control through a combination of tyrosine kinase (TK) inhibitors and checkpoint blockade in this murine Ph⁺ B-ALL model, the responses to checkpoint blockade alone, particularly for B-ALL, have been disappointing in clinical trials.⁷ The intense lymphodepleting regimen that patients with B-ALL undergo during treatment may deplete this exhausted responder T-cell population, leading to the lack of response in this tumor type. Interestingly, checkpoint inhibitors have been successfully used in the treatment of relapsed B-cell lymphomas.⁸ One could speculate that in contrast to leukemia, the nodular conformation of the lymphoblasts in lymphoma might provide a homeostatic tumor microenvironment that partially protects exhausted T cells from the lymphodepleting effects of chemotherapeutic agents, which would allow response to checkpoint blockade. Exploring the T-cell exhaustion profile after lymphodepleting chemotherapy in leukemias and lymphomas may provide additional mechanistic insight into the tumor microenvironment that may account for these clinical differences.

An indication that checkpoint blockade might provide clinical benefit in the context of B-ALL when appropriate responder T-cell populations are present is demonstrated by preclinical studies and the early results of clinical trials of combinatory approaches of checkpoint blockade with T-cell–directed immunotherapies, such as the bispecific T-cell engager blinatumomab and CD19-directed chimeric antigen receptor T-cell therapies.^9,10 The combination of TK inhibitors and checkpoint blockade (as suggested in the study by Tracy et al) has proven successful in melanomas and renal cell carcinomas, but whether the effects are additive or synergistic is yet to be determined. Most likely, as discussed in the article by Tracy et al, the benefits of nilotinib are multifactorial. One of the factors is slowing disease progression, which allows an exhausted responder population to develop. Another is direct tumor lysis, which results in local inflammation (with increased antigen availability that results in an immune-stimulatory milieu). And a third factor is the potentially direct CD4 cytotoxic, helper, and CD8 cytotoxic T-cell effects of the TK inhibitor. An in-depth characterization of each of these effects will be necessary to discover the mechanistic impacts of TK inhibition on the function of cytotoxic CD4⁺ T cells.

In summary, the study by Tracy et al provides novel insights into the mechanisms by which CD4⁺ T-cell exhaustion develops in Ph⁺ B-ALL and describes the immunomodulating effects of combinatory therapy with nilotinib and PD-1 on this population. The results of this work elucidate a mechanistic pathway that could play a role in improving the efficacy of cancer immunotherapies in ALLs.