Senescent T Cells as a Resistance Mechanism to Lung Cancer Immunotherapy

In this issue of Clinical Cancer Research, Ferrara and colleagues (1) found that a high abundance of peripheral blood circulating senescent T cells is associated with a worse clinical response to PD-1 inhibitors in patients with advanced non–small cell lung cancer (NSCLC). Immunosenescence is associated with physiologic aging. However, high levels of senescent T cells can be found in young patients with chronic antigen-presentation exposure such as autoimmune disease, chronic viral infection, or carcinogenesis (2). In this regard, anergy, exhaustion, and senescence are considered mechanisms of peripheral T-cell tolerance which are found in patients with cancer, and are potentially involved in tumor immune escape and cancer immunotherapy resistance (Fig. 1). Each of these tolerogenic states is characterized by different functional impairments. Anergy is generally described as the induced hyporesponsive state (i.e., low IL2 production) of T cells upon antigen presentation in the context of insufficient costimulatory signals. Exhaustion is considered the result of maintained T-cell activation at a site of chronic inflammation, characterized by cell-cycle arrest and the reduction of cytokine expression. Senescence is characterized by telomere shortening which determines cell-cycle arrest and is phenotypically associated with loss of CD28 expression. In contrast to anergy and exhaustion, senescence is associated with an increase in the secretion of cytokines such as TNFα and IFNγ, and it is unclear whether it is associated with loss of T-cell killing activity. Ferrara and colleagues have used the combination of CD28⁻CD57⁺KLRG-1⁺ markers to define what they have termed senescent immune phenotype (SIP) in this article. CD57 is a terminally sulphated carbohydrate determinant and its expression has been found to progressively increase while CD28 decreases in chronically antigen-stimulated T cells. Decreases in CD28 and increases in CD57 have both been associated with a senescent T-cell profile. KLRG-1, in contrast, has been previously associated with a subset of nonproliferative T cells expressed by both exhausted and senescent T cells, but not anergic T cells (Fig. 1). Using this description, the authors have identified that one-fifth of patients with advanced NSCLC in this study showed a high proportion of circulating T cells with a SIP. Consistent with previous studies describing the senescent T-cell functional profile (3), the SIP T cells showed lower Ki-67 expression compared with non-SIP circulating T cells, but also higher TNFα/IFNγ secretion. The authors found no clear explanation for the high frequency of senescent T cells in these patients with lung cancer after studying age, cytomegalovirus (CMV) infection history, and previous chemotherapy among other clinical variables implicated in immunosenescence promotion. They found, however, that a higher frequency of circulating SIP⁺ T cells was associated with a worse overall response rate and progression-free survival in a cohort of 37 patients with NSCLC treated with immune checkpoint inhibitor (ICI). This result was validated in an additional cohort of 46 patients with NSCLC. Furthermore, the authors have described a higher abundance of circulating Th1, Tc1, and OX40 Treg lymphocytes in SIP⁺ patients with NSCLC and this was interpreted as the signature of a chronic systemic proinflammatory state in line with the proinflammatory profile of SIP T cells. Taken together, these data suggest that senescent T cells are involved in the resistance to ICI in patients with NSCLC. Fundamental questions, however, need to be addressed to understand whether senescent T cells are the consequence or the cause of this resistance. As this body of work is based on peripheral blood mononuclear cell analysis, it remains to be clarified whether circulating senescent T cells in patients with NSCLC are reflecting the tumor-infiltrating T cells composition or, in contrast, it is an expression of the systemic immunity of the patient. Immunosenescence is considered a systemic immune process, and most studies have reported the presence of these T cells in peripheral blood. However, less is known about the presence of senescent T cells in the tumor microenvironment and the contribution of tumorogenesis to their development. Similarly, it would be important to determine whether senescent T cells are enriched in tumor-specific T cells; or rather are a pool of T cells recognizing nontumoral antigens, considered “bystander T cells” from the cancer immunotherapy point of view. Ferrara and colleagues, hypothesize that the development of these senescent T cells could be associated with chronic viral infections or aging, but neither of these two factors has been associated with the abundance of SIP T cells in this particular study. It is important to highlight that CMV immunization was explored in half of all the patients with NSCLC and that other chronic infections, potentially involved in the promotion of this SIP+ T-cell population, were not assessed. The identification of the mechanisms involved in the development of SIP T cells, their antigen specificity, and to what extent they are infiltrating the lung cancer microenvironment are critical questions to understand the relevance of this T-cell subset in lung cancer immunotherapy. In recent years, an increasing number of studies has explored the heterogeneity of tumor-infiltrating lymphocytes (TIL) expanding our capacity to understand the different programs and mechanism contributing to the lack of effector and proliferative functions of TIL (4). For instance, although senescence and exhaustion share overlapping characteristics, they appear to be two distinct process independently regulated. In the context of chronically activated T cells, nuclear factor TOX is induced for sustaining T-cell response and mediates transcriptional and epigenetic reprogramming driving exhaustion. In contrast, senescence-related pathways leading to cell-cycle arrest can be induced by different mechanisms including DNA damage, p38 activation through chronic IFNγ signaling or glucose deprivation (5). This collective body of work may pave the way for the development of better T cell–targeted immune therapies, looking to enhance the activation of those TIL that can be recovered, but also to prevent the mechanisms that drive the dysfunction of these effector lymphocytes. Little is known, nevertheless, about the specific factors contributing to T-cell senescence in patients with cancer and the mechanism by which these senescent T cells may contribute to antitumor immune impairment. The results by Ferrara and colleagues encourage further analysis of this T-cell subset in patients with NSCLC and may serve as a hypothesis generating study to further evaluate the involvement of senescent T cells in cancer immune escape and the development of novel strategies to enhance cancer immunotherapy.

No disclosures were reported.

M.F. Sanmamed is supported by a Miguel Servet contract from Instituto de Salud Carlos III, Fondo de Investigacion Sanitaria (Spain). I. Eguren-Santamaria is supported by an AECC Clinical Junior grant.