A team of researchers has developed a new CAR-T cell approach to target a genetic change that commonly occurs in all types of cancer. This genetic alteration, termed loss of heterozygosity (LOH), occurs in cancer when a gene loses one of its two polymorphic copies, or alleles. As such, the missing gene copy can be used as a signal to direct CAR-T cells to selectively destroy cancer cells. The findings were published in the Proceedings of the National Academy of Sciences (2021; doi: 10.1073/pnas.2022410118) by a research team from the Ludwig Center, Lustgarten Laboratory, and the Bloomberg~Kimmel Institute for Cancer Immunotherapy at the Johns Hopkins Kimmel Cancer Center.
A fundamental challenge in developing new therapies for cancer is selectively targeting all neoplastic cells while sparing essential normal cells, thereby avoiding unmanageable toxicities for patients. The genetic changes present in cancer offer one mechanism of unequivocal specificity in distinguishing cancer cells from their normal counterparts.
"As a genetic alteration observed in 90 percent of cancers, LOH is an attractive therapeutic target," said Ken Kinzler, PhD, leader of the study and Co-Director of the Ludwig Center at Johns Hopkins. "All normal cells in our body have two copies of each gene, one inherited from each of our parents. In cancer cells, however, many genes can lose one of their two original copies. If the two alleles happen to be polymorphic, this phenomenon of losing one allele is called loss of heterozygosity. When the polymorphism translates into amino acid differences in the protein products, we can potentially use antibodies to detect which gene copy is still expressed in cancer cells and which is lost."
Conceptually, however, leveraging the loss of a gene copy therapeutically is challenging because it involves targeting a protein that is no longer expressed in cancer cells but is still expressed in normal cells.
"Recent advances in CAR-T cell engineering have opened the door for targeting LOH," stated Shibin Zhou, PhD, Associate Professor of Oncology and study co-leader.
Chimeric antigen receptors (CARs) are synthetic protein molecules that use an antibody component to redirect T cells to destroy cancer cells expressing the target of that antibody, i.e., the antigen. Inhibitory CARs (iCARs) act in the opposite manner by turning off T cells when cells express the targeted antigen.
"We saw an opportunity to combine CARs and iCARs to target LOH by creating a NOT logic gate that will only activate T cells when a targeted gene allele has undergone LOH and is absent in a tumor," Zhou noted.
This new approach, named neoplasm-targeting allele-sensing CAR (NASCAR), works by targeting an iCAR to the gene allele lost in a cancer and a CAR to the allele that is still expressed. When both CAR and iCAR in a T cell are activated by a normal cell expressing both alleles, the iCAR is dominant and inhibits the CAR signal. The T cell remains unactivated and the normal cell is unharmed. However, when the allele targeted by the iCAR is lost in a cancer cell by LOH, the CAR activity is not blocked by the iCAR. The T cell then becomes activated and destroys the cancer cell.
HLA as Proof-of-Concept Model
In their study, the research team tested the NASCAR approach by targeting human leukocyte antigen (HLA) loss of heterozygosity. HLA molecules are used by our body's immune system to recognize and attack virally infected cells and even cancer cells.
"We chose to target HLA LOH for our proof-of-concept models for a number of reasons. First, it occurs quite often across many types of cancers and could offer a viable therapeutic target for cancers that currently lack good treatment options, like pancreatic cancer," stated Michael Hwang, PhD, first author on the study and former graduate student at the Johns Hopkins University School of Medicine. "Second, HLA molecules are highly diverse cell surface proteins, so they are ideal targets for allele-specific antibodies that can be used in CAR and iCAR designs."
HLA LOH is also increasingly observed clinically as a resistance mechanism to immunotherapy treatments, such as checkpoint blockade. The NASCAR approach applied to HLA LOH may therefore also be used to help patients whose cancers develop resistance to other types of immunotherapy.
The research team documented the optimization of the NASCAR approach, demonstrating that the structure and expression level of the iCAR component were essential to the LOH selectivity of the engineered T cells. The team then tested the optimized NASCAR-T cells against models of HLA LOH in three different cancer cell lines: lymphoma, lung cancer, and pancreatic cancer. In a tissue culture dish, the NASCAR-T cells could selectively kill the cancer cells exhibiting HLA LOH but not the non-LOH control cells for each cancer type. Finally, the NASCAR-T cells were used to treat a pancreatic cancer mouse model, curing all mice implanted with tumors exhibiting HLA LOH, while showing minimal activity against non-LOH control tumors.
In the future, the research team plans to develop more potent pairs of CARs and iCARs for the NASCAR approach.
"Specificity is key for targeting LOH. When the CAR target is expressed on many normal cell types, the iCAR needs to inhibit 100 percent of CAR activity. Before going to the clinic, we want to be sure we have the most potent iCARs possible," described Brian Mog, an MD/PhD candidate at the Johns Hopkins University School of Medicine and co-first author of the study. The team is also planning to identify other targetable genes that commonly undergo LOH in cancer to expand the potential therapeutic indications for this innovative immunotherapy strategy.
The NASCAR approach exemplifies the research paradigm of the team led by Kinzler and Bert Vogelstein, MD, Co-Director of the Ludwig Center at Johns Hopkins. After decades of investigation defining the genetic alterations that contribute to the development and progression of cancer, this research group seeks to leverage these same genetic changes as highly specific therapeutic targets across all types of cancer.