NEW YORK—Chimeric antigen receptor (CAR) T cells have “remarkable” activity, according to a speaker at the NCCN 9th Annual Congress: Hematologic Malignancies.
“[T]his chimera binds like an antibody, but it acts like a T cell, so it combines the best of both worlds,” said Jae H. Park, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York.
He then traced the evolution of CAR T-cell design, discussed clinical trials using CD19-targed T cells, and described how investigators are working at building a better T cell.
Researchers found that T-cell activation and proliferation require signaling through a costimulatory receptor, such as CD28, 4-1BB, or OX-40. Without costimulation, the T cell becomes unresponsive or undergoes apoptosis.
So based on this observation, Dr Park said, several research groups created second- and third-generation CARs to incorporate the costimulatory signal.
The first generation was typically fused to the CD8 domain. Second-generation CARs include a costimulatory signaling domain, such as CD28, 4-1BB, or OX40. And the third generation contains signaling domains from 2 costimulatory receptors, CD28 with 4-1BB and CD28 with OX40.
The built-in costimulatory signal proved superior to the first-generation CAR T cells.
In NOD/SCID mice inoculated with NALM-6 lymphoma cells, Dr Park said, about 50% more were “cured,” in terms of survival, using a CD80 costimulatory ligand with CD19-targeted T cells compared to those without the ligand.
Clinical trials using second-generation CD19-targeted T cells in relapsed B-cell acute lymphoblastic leukemia (ALL) at MSKCC produced an overall complete response (CR) rate of 88% in a median of 22.1 days. And 72% of the CRs were minimal residual disease (MRD) negative.
So the CAR T cells produce a “very rapid and deep remission,” Dr Park said.
CAR T-cell therapy, however, comes with adverse events, most notably, cytokine release syndrome (CRS), which results from T-cell activation. CRS causes fevers, hypotension, and neurologic toxicities including mental status changes, obtundation, and seizures.
“CRS is not unique to CAR T-cell therapy,” Dr Park said. “Any therapy that activates T cells can have this type of side effect.”
Dr Park noted that CRS is associated with disease burden at the time of treatment. “The larger the disease burden pre T-cell therapy,” he said, “the more likely [patients are] to develop CRS.”
In the MSKCC trial, no patient with very low disease burden—5% blasts in the bone marrow—developed CRS.
However, there is also a correlation between tumor burden and T-cell expansion, he added. T cells expand much better with a larger disease burden, because there is a greater antigen load.
The investigators found that serum C-reactive protein can serve as a surrogate marker for the severity of CRS. Patients with levels above 20 mg/dL are more likely to experience CRS.
And Dr Park pointed out that CRS symptoms respond pretty rapidly to steroids or interleukin-6 receptor blockade.
CAR T-cell therapy has also been used to treat chronic lymphocytic leukemia, but with much more modest response rates than in ALL. Both University of Pennsylvania and MSKCC trials in CLL have produced overall response rates around 40%.
Building a better T cell
Dr Park described efforts underway to develop the fourth-generation “armored” CAR T cells to overcome the hostile tumor microenvironment, which contains multiple inhibitory factors designed to suppress effector T cells.
Armored T cells can actually secrete some of the inflammatory cytokines to change the tumor microenvironment and overcome the inhibitory effect.
Dr Park described a potential scenario: The armored CAR T cells secrete IL-12, enhance the central memory phenotype, enhance cytotoxicity, enhance persistence, modify the endogenous immune system and T-cell activation, and reactivate tumor-infiltrating lymphocytes.
He said future studies will focus on translation of these armored CAR T cells to the clinical setting in both hematologic and solid tumor malignancies.