Glioblastoma is the most formidable of primary brain tumors, owing to its aggressive nature and the limited efficacy of the best available treatment, which comprises maximal safe surgical resection, radiotherapy, and alkylating chemotherapy.
Most patients die within a year after diagnosis, which underscores the urgent need for innovative therapeutic strategies. Of interest, then, is an interim analysis in this issue of the Journal: Choi et al.1 report the outcomes of treating three patients with a secreting chimeric antigen receptor (CAR) T cell (see Key Concepts).
What Is a CAR T Cell?
CAR T-cell technology has reshaped our therapeutic approach to combating hematologic cancers. It harnesses the patient’s own T cells, genetically equipping them with a CAR that overrides the native recognition of antigens by T-cell receptors and enables T cells to recognize and bind a cancer antigen on the surface of a cancer cell — and then kill the cancer cell.
Food and Drug Administration–approved CAR T-cell therapies, including tisagenlecleucel (Kymriah) and axicabtagene ciloleucel (Yescarta), have shown remarkable efficacy against acute lymphoblastic leukemia and diffuse large B-cell lymphoma. However, extending this success to solid tumors, such as glioblastoma, has been challenging because of the tumor microenvironment, cellular and genetic heterogeneity, and immunosuppressive elements.
Read More
Some tumors have unique cancer-specific proteins, which are attractive targets for T cells. In glioblastoma, a tumor-specific protein is epidermal growth factor receptor (EGFR) variant III (EGFRvIII), a truncated and constitutively active version of EGFR. However, clinical trials have shown that tumors targeted by T cells engineered to recognize EGFRvIII can develop resistance; in particular, tumors with high expression of nonvariant EGFR tend to become resistant to treatment.2
How Does a Secreting CAR T Cell Work?
A CAR is a synthetic protein made up of an extracellular antigen-recognition domain, typically derived from the single-chain variable fragment of an antibody, and an intracellular signaling domain that activates the T cell when binding an antigen. In addition to transducing autologous T cells with DNA encoding the CAR, Choi et al. also transduced these same T cells with DNA encoding a bispecific antibody that recognizes EGFR (expressed by glioblastoma) and CD3 (expressed by the T cell). They call this a T-cell–engaging antibody molecule (TEAM-E); it is a type of bispecific T-cell engager. TEAM-E is secreted by the T cell and is akin to a tether, because it holds the T cell and the glioblastoma cell in close proximity to each other (Figure 1). This strategy is supported by a preclinical study3 showing that the transduced T cells recognize EGFRvIII on the tumor and then locally secrete the TEAM-E, which enhances tumor cell–T cell interactions. In this preclinical study, the authors showed that CARv3-TEAM-E activity was restricted to the brain. The CARv3-TEAM-E cell targets two antigens (EGFR and EGFRvIII) and represents an approach amenable to overcoming the heterogeneity and complex pathologic features of glioblastoma. Its other advantage is the local delivery of secreted elements (by the T cell) into the tumor microenvironment, thereby minimizing systemic toxic effects.
Does CARv3-TEAM-E Work?
Yes and no. Choi et al. observed dramatic radiographic tumor regression shortly after a single infusion of CARv3-TEAM-E T cells. Regression was noted in all three patients, but the response was transient in two of them. In the third, the response was durable. The safety profile was notable, with manageable adverse events and no dose-limiting toxic effects. The early responses support further research into cell-based therapies in treating advanced glioblastoma in the brain parenchyma through intraventricular administration. Although such administration involves a surgical procedure (placement of an injection port called an Ommaya reservoir), it is likely to permit direct traffic of the therapy to the tumor and avoid systemic activity that could result from peripheral administration, especially if the targets of the CAR T cells are expressed outside the central nervous system. Given the devastating nature of glioblastoma, this invasive mode of administration is warranted.
What’s Next?
The transient response in two of the three patients reported by Choi et al. underscores the need for additional research to increase and prolong the efficacy of this promising therapy. Research into enhancing the longevity of the CAR T cell and integrating combination treatments to counter the immunosuppressive nature of the tumor microenvironment are warranted. Indeed, glioblastomas have highly immunosuppressive myeloid compartments that contribute to immune evasion and tumor growth.4,5 Overcoming the challenge of brain-tumor heterogeneity is also important. The field is pivoting toward multitargeted CAR T-cell strategies, as exemplified by an innovative clinical trial of CAR T cells targeting four different antigens to treat children and young adults with brain cancer in Seattle (ClinicalTrials.gov number, NCT05768880). As for CARv3-TEAM-E T cells in the treatment of glioblastoma, the results reported by Choi et al. support a cautious optimism. However, it is only with data collected through the treatment of additional patients that the effect and scalability of this therapy can be evaluated.
References
For references, check the original article at: nejm.org
