Chimeric antigen receptor (CAR) T cells are now a standard-of-care treatment for relapsed and refractory disease in several hematologic malignancies. Pivotal trials in acute lymphoblastic leukemia, large B-cell lymphoma (LBCL), mantle cell lymphoma, and multiple myeloma have shown impressive efficacy that led to approvals of five CAR T-cell products.1–5 Elucidating the critical interactions between genetically engineered immune effector cells, host immunity, and the tumor microenvironment (TME) is essential for the successful clinical development of CAR T-cell therapy in solid tumors and to improve on the response rates in hematologic malignancies.
Based on the remarkable clinical response of one patient with progressive multifocal glioblastoma multiforme (GBM) to IL13Rα2-CAR T cells,6 Dr. Darya Alizadeh and colleagues hypothesized that additional mechanisms were involved beyond direct CAR-targeted T-cell antitumor activity. They set out to evaluate the indirect effects of IL13Rα2-CAR T-cells on the endogenous immune system and the potential role of IFN-γ with a series of well conducted experiments. First, they compared T cells isolated from the patient's blood in the pre- and post-CAR T-cell settings and found enhanced tumor reactivity and tumor-specific killing in endogenous CAR-negative T cells after product infusion. They then tested a mouse counterpart to their human IL13Rα2 CAR in murine immune-competent models of GBM. The murine CAR T cells specifically killed mIL13Rα2-engineered K-Luc (poorly immunogenic) and GL261-Luc (immunogenic, anti-PD1 responsive) cells; this was associated with production of inflammatory cytokines IFN-γ and TNFα. Additionally, they showed potent in vivo CAR T-cell activity against orthotopically engrafted glioma tumors in C57BL/6 immunocompetent mice and found in vivo response was superior in immunocompetent mice, as compared to immunodeficient NSG mice, when the TME exhibited an established immune microenvironment.
The authors demonstrate the capacity of CAR T-cells to induce endogenous immunity and how this process is dependent on the establishment of an immune cell enriched microenvironment. Lymphoid and myeloid subpopulations in the TME were characterized using single-cell RNA-seq and gene set enrichment analysis (GSEA). Significant changes described in both compartments after CAR T-cell therapy include enrichment in activated, memory, or effector T cells in the lymphoid compartment and increase of activated myeloid cells and M1 type macrophages in the myeloid compartment. IFN-γ deficiency impacted in vivo CAR T-cell antitumor activity and resulted in higher M2 type macrophages. Furthermore, lack of host immune cells responsiveness to IFN-signaling results in reduced expression of genes involved in activation and proinflammatory responses. IFN-signaling blockade in macrophages and T cells was linked to reduced activation of host immune cells. These data indicate IFN-γ production by CAR T-cells, host immune cells' response to the acute effects of IFN-γ, and TME reshaping are key to enhance responses. Intriguingly, the authors provided evidence of myeloid mediated induction of endogenous antitumor T-cell responses independent of the CAR target.
In Brief
The implications of these findings must be considered in the context of prior studies. First, the importance of endogenous antitumor immunity in the context of CAR T-cell therapy is unclear. Seminal reports demonstrated that “armored” CARs secreting cytokines (e.g., IL-18) can reshape the immune landscape and induce an antitumor T cell response against non-CAR targets.7 In the context of CD19 directed CAR T, a minority of post treatment LBCL relapses exhibit CD19 loss. One possible explanation is that augmentation of immunity against alternate non-CD19 targets is critical for durable responses and the report from Dr. Alizadeh and colleagues supports this hypothesis. Alternatively, a single CD19 CAR T-cell clone having a survival advantage based upon its genomic insertion disrupting the functional heterozygous TET2 allele, mediated remission of chronic lymphoblastic leukemia.8 This suggests that only CAR-targeted effects are needed to eradicate tumor. Certainly, mechanisms of effectiveness and tumor evasion may differ across diseases. Additional studies must be conducted in larger numbers of patients, both in solid tumors and hematologic malignancies, to better understand whether durable responses to CAR T require myeloid mediated induction of T cell responses against other tumor targets.
Second, in the context of CAR T-cell therapy, conflicting findings on the roles of IFN signaling and myeloid cells within the tumor microenvironment have been reported. The functional status of CAR T-cells is often determined by interferon-γ secretion assays, which are part of the release testing. However, the presence of chronic inflammation and IFN signals within the tumor prior to treatment may have the opposite effect. Dr. Michael Jain and colleagues recently reported that tumor IFN signaling and myeloid cells in pretreatment LBCL tumor samples are associated with decreased CAR T expansion and lack of durable responses.9 This tumor IFN stimulated gene signature (ISG.RS) was previously shown in solid tumors to mark the presence of chronic IFN stimulation leading to T-cell exhaustion and resistance to immune checkpoint blockade.10 In many facets of immunology a balance exists between activation and suppression: the findings of both Drs. Jain and colleagues and Alizadeh and colleagues can likely be reconciled in this context. Acute IFN-γ from CAR T cells clearly exerts an antitumor effect, while chronic IFN signals in the tumor can induce a suppressive environment resistant to more IFN-γ. Ultimately more experimental models and larger translational studies are needed to determine whether suppression mediated by chronic IFN signaling and tumor myeloid cells can be overcome by direct and indirect CAR T-cell effects mediated by IFN-γ or other cytokines.
Competing Interests
Dr. Perez-Perez indicated no relevant conflicts of interest. Dr. Locke is the corresponding author on the Jain et al manuscript referenced above.
