The roles of stem cell memory T cells in hematological malignancies

Memory T cells (including CD4⁺ and CD8⁺ memory T cells) include several subtypes: stem cell memory (T_SCM), central memory (T_CM), transitional memory (T_TM) (described only in CD4⁺ memory T cells), effector memory (T_EM), and terminal effector (T_TE) T cells [16, 22]. T_SCM cells were first observed in a murine model of GVHD by Zhang et al., who reported a new subset of post-mitotic CD44^loCD62L^hiCD8⁺ T cells expressing Sca-1 (stem cell antigen 1), CD122, and Bcl-2. This population of T cells was able to generate and sustain all allogeneic T cell subsets in GVHD reactions. These alloreactive CD8⁺ T cells were demonstrated to have enhanced self-renewal capacity and multipotency. These cells are capable of differentiating into T_CM, T_EM, and T_TE cells [14, 21]. In humans, an example came from the identification of a population of naive yellow fever (YF)-specific CD8⁺ T cells after vaccination. These cells were stably maintained for more than 25 years and were capable of ex vivo self-renewal. In-depth analysis of markers and genome-wide mRNA profiling have shown that these cells are distinct from genuine naive cells from unvaccinated donors and resemble the recently described stem cell-like memory subset T_SCM [23]. Moreover, epigenetic analysis has also revealed that histone modifications and gene expression signatures could distinguish T_SCM from other CD8⁺ T cell subsets [24]. Increasing data have supported the notion that the human T_SCM subset is a clearly distinct subset in between the naive T cell (T_N) and T_CM subsets. Human T_SCM cells have been described as a long-lived memory T cell population which are CD45RO⁻, CCR7⁺, CD45RA⁺, CD62L⁺, CD27⁺, CD28⁺, and IL-7Rα⁺. These markers are characteristic of naive T cells. The immunophenotypic markers expressed in different T cell subtypes (from T_N to T_TE cells) are summarized in Table 1. T_SCM cells express increased levels of CD95, IL-2Rβ, CXCR3, and LFA-1 and exhibit numerous functional attributes distinct from memory cells. However, human T_SCM cells constitute only approximately 2–4 % of the total CD4⁺ and CD8⁺ T cell population in the periphery and can be identified by polychromatic flow cytometry based on the simultaneous expression of several naive markers together with the memory marker CD95 [25]. A linear T cell differentiation model and the minimum set of markers used for identifying and sorting T_SCM are depicted in Fig. 1 [25].

Self-renewing memory T cells may be regulated by shared signaling pathways such as those involved in hematopoietic stem cells or memory B cells. The Wnt-β-catenin pathway is an evolutionarily conserved pathway that regulates hematopoietic stem cell self-renewal and multipotency by limiting stem cell proliferation and differentiation. Similarly, a key role for Wnt signaling during the maintenance of “stemness” in CD8⁺ T_SCM cells was demonstrated by Gattinoni et al. It was shown that disrupting the Wnt/β-catenin pathway by glycogen synthase-3β (GSK-3β) inhibitors promoted the generation of CD44^lowCD62L^highSca-1^highCD122^highBcl-2^high self-renewing multipotent CD8⁺ T_SCM cells with proliferative and antitumor capacities that exceeded those of the T_CM and T_EM subsets [11, 26, 27]. In addition, antigen-specific T_SCM cells were shown to preferentially reside in the lymph nodes (LNs) and less so in the spleen and bone marrow [28].

There are numerous factors that act as modulators regulating the maturation and activation of CD8⁺ T cells, for example, suppressor of cytokine signaling (SOCS) is one of the key modulators [29]. Moreover, it has been reported that activation of naive T cells with anti-CD3 and anti-CD28 antibody-conjugated beads in the presence of low doses of IL-7 and IL-15 promotes the generation of CD45RA⁺CD62L⁺CCR7⁺CD95⁺ T_SCM cells [30].

It is well known that antigen-specific T cells are crucial components for antitumor or antivirus immunity in patients with hematological malignancies, particularly in patients after HSCT. It is possible that the number of antigen-specific T_SCM cells may be the determining factor of immunity. However, there have been few reports on antigen-specific T_SCM cells. Low frequency of these cells limits detailed characterization. For example, <1 % of total human T cells are defined as CD8⁺CD45RA⁺CCR7⁺CD127⁺CD95⁺ viral-specific T_SCM cells. Human CMV-specific T_SCM cells can be detected at frequencies similar to those observed in other subsets, with frequency around ∼1/10,000 T cells [31, 32].

Antigen-specific T_SCM cells represent a long-lasting component of the cellular immune response to viruses and tumor-associated antigens (TAAs). For virus-specific T_SCM cells, research has first focused on human immunodeficiency virus type 1 (HIV-1)-specific CD8⁺ T_SCM cells. It is known that HIV-specific CD8⁺ T cells can influence HIV-1 disease progression during untreated HIV-1 infections, and recent data have shown that HIV-1-specific CD8⁺ T_SCM cells are detectable in all stages of HIV-1 infection. These cells were found to be increased in number in patients receiving suppressive antiretroviral therapy when compared with those untreated patients [33]. It was found that CD4⁺ T_SCM cells were susceptible to HIV infection; thus, HIV-1 virus may exploit the stem cell characteristics of cellular immune memory T cells and lead to long-term viral persistence [34]. Similar findings were demonstrated in a study of human T cell leukemia virus type 1 (HTLV-1)-infected CD4⁺ T_SCM cells in patients with adult T cell leukemia (ATL). This report first demonstrated an association between T cell malignancy and T_SCM cells. T_SCM cells from ATL patients were capable of sustaining themselves in a less proliferative mode, yet they were able to differentiate into other memory T cell populations during the rapidly propagating phase. These cells have been identified at the hierarchical apex capable of reconstituting identical ATL clones [35]. A decrease in the infection of CD4⁺ T_SCM cells was found to preserve CD4⁺ T cell homeostasis and prevents disease progression despite significant viremia in both HIV-1 and HTLV-1 infections [36].

T_SCM cells may play a major role in specific antitumor response and long-term immune surveillance directed against tumors [17, 37, 38]. In addition, T_SCM cells have been proposed to be one of the key determinants of immune memory. It may be interesting to monitor the level of T_SCM cells and its significance for immune reconstitution and prognosis of patients with hematological malignancies before and after therapy, particularly HSCT. There have been only a few studies on TAA-specific T_SCM cells. Recently, dynamic changes of T_SCM cells were longitudinally tracked in patients who underwent haploidentical HSCT. These studies demonstrated that donor-derived T_SCM cells were highly enriched early after HSCT. T_SCM cells can differentiate directly from naive precursors infused in the grafts. Through T cell receptor (TCR) gene analysis, T_SCM cells have been found to have diversification in immune memory after allogeneic HSCT [10]. It was also demonstrated that the level of T_SCM cells may be used to evaluate immune reconstitution in patients who received posttransplant cyclophosphamide (pt-Cy) for GVHD prophylaxis. Similarly, donor-derived T_SCM cells were found to be the most abundant circulating T cell population in the early days following haploidentical HSCT and pt-Cy. These donor-derived T_SCM cells preceded the expansion of effector cells. Antigen-specific T_SCM cells generated detectable recall responses; thus, it has been proposed to explore T_SCM cells derived from donor naive precursor cells in the clinical setting to overcome immunodeficiency [12]. With the ability to expand and differentiate into effectors capable of mediating potent xenogeneic GVHD in immunodeficient mice, these donor naive precursor-derived T_SCM cells were noted to be superior to other memory lymphocytes. Furthermore, gene-modified T_SCM cells were found to be the only T cell subset capable of expanding and mediating GVHD in serial transplantations [30]. These findings indicate negative aspects of T_SCM cells for clinical application.

T_SCM cells may be a novel and critical therapeutic resource because these cells have the potential to serve as a stable cellular vehicle. Two gene therapy clinical trials with gene-corrected hematopoietic stem cells provided a glimpse into this possibility. Long-term in vivo T cell reconstitution was characterized in these trials. Specifically, the investigators traced the fate of greater than 1700 individual T cell clones in patients who underwent gene therapy. The studies demonstrated that the T_SCM cells persisted and preserved their precursor potential in humans for up to 12 years after the infusion of gene-corrected stem cells [39]. The demonstration of the safe, functional, and decade-long survival of the engineered T_SCM cells in humans sets the stage for their clinical application. Since T_SCM cells were shown to be capable of reconstituting the full repertoire of memory and effector T cells after HSCT, it is particularly attractive to use them for adoptive immunotherapies. T_SCM cells might overcome current limitations, such as inefficient T cell engraftment, poor persistence, and inability to mediate prolonged immune attacks [10‐12, 40].

Even though potent antitumor activity of T_SCM cells was demonstrated in preclinical animal tumor models [26, 27], it is currently not feasible to treat patients with naturally occurring T_SCM cells because it is a scarce and small proportion of circulating lymphocytes. Therefore, strategies that can generate, expand, and enable the redirection of T_SCM cells against cancer cells need to be defined. Cieri and colleagues have recently described that a large number of T_SCM cells were generated by priming T cells with low doses of IL-7 and IL-15. It is therefore possible to generate, expand, and genetically engineer T_SCM cells in vitro from naive precursors. Furthermore, the in vitro-generated T_SCM cells displayed enhanced proliferative capacity upon adoptive transfer into immunodeficient mice, a finding consistent with those using naturally occurring T_SCM cells [11, 30]. T_SCM cells were also expanded from naive precursors by inhibiting Akt signaling during ex vivo priming and expansion. The Akt-inhibited minor histocompatibility antigen (MiHA)-specific CD8⁺ T cells had superior expansion capacity in vitro and induced superior antitumor activity in multiple myeloma-bearing immunodeficient mice. These findings provided a rationale for clinically exploiting ex vivo-generated, Akt-inhibited, MiHA-specific CD8⁺ T cells or TAA-specific CD8⁺ T cells for adoptive immunotherapy [41, 42]. Schmueck-Henneresse et al. also described a simplified culture protocol allowing for fast expansion of virus-specific T_SCM cells from a mixed T_N/T_SCM pool of peripheral blood lymphocytes. This may be the basis for novel cell therapeutic options for life-threatening viral infections [31]. Among the known memory T cell subpopulations, the T_SCM cell subset has profound implications for the design and development of effective vaccines as well as T cell-based therapies [13, 26, 28]. As immunotherapy plays increasingly important roles in cancer management, further exploration of T_SCM cells and their regulation may facilitate clinical development of humoral (monoclonal antibodies and inhibitors of B cell receptor signaling) and cellular (CART) immunotherapies [40, 43‐47].