Leveraging a patient’s own immune system to fight cancer has long been an objective of researchers and physicians. In chimeric antigen receptor (CAR) T cell therapy, for instance, doctors extract a patient’s T cells and then introduce genetic material that trains those cells to recognize tumor antigens that they might normally overlook. Upon reintroduction into the body, the goal is for the CAR T cells to locate and kill the cancerous cells. While the strategy can be a game-changer—one trial sent 22 of 27 patients with a severe form of B cell lymphoma into full or partial remission—CAR T-cells are time consuming to produce and sometimes cause nasty side effects, not unlike traditional cancer treatments that indiscriminately harm noncancerous cells.
Another immunotherapy strategy using antibodies against cancer-specific peptides can effectively target tumor cells for destruction, but these targets are often inside cells and not accessible at the cell surface.
In three studies published March 1, researchers engineered so-called bispecific antibodies to use one of their arms to tightly bind even very low levels of cancer-related peptides and with the other recruit a T cell to destroy the peptide-presenting cell. In mice, these antibodies kill cancerous cells specifically, both in solid tumors and in T cell leukemia, while preserving most healthy cells, the studies showed.
“The unifying principle here is adapting immune therapies, on the one hand with greater precision and, on the other hand, going after targets that are traditionally considered to be untargetable,” says Jonathan Schatz, an oncologist and researcher at the University of Miami who was not involved in the work. “It boils down to new ways of personalizing immunotherapy.”
Bispecific antibodies grab hold of rare antigens
These two objectives were behind the work of a group led by oncologist Bert Vogelstein and cancer biologist Kenneth Kinzler of the Ludwig Center at Johns Hopkins Medicine. The gene that encodes the tumor suppressor p53 is one of the most commonly mutated in cancer, but there’s not a good therapeutic to target the mutant protein. Cancer cells present fragments of this protein complexed with the major histocompatibility complex (MHC), human leukocyte antigen (HLA), but there are typically just a handful of these HLA-p53 complexes on the cell surface.
For their study in Science, the researchers generated one antibody arm highly specific to the HLA-p53 combo—so specific that it could distinguish a mutant p53 that differed from the wildtype p53 by one amino acid. This specificity meant this arm found the small number of these complexes on a tumor cell’s surface and also bound tightly and didn’t fall off. They linked this antibody fragment to another arm that binds CD3, a T cell surface receptor and activator.
The resulting bispecific antibody recruited T cells to destroy cancerous cells with the mutant p53 protein on their surface, both in cell culture and in mice. The researchers injected immunodeficient mice with both human T cells and p53 mutant tumor cells. Once the tumors were established, the team infused the mice with bispecific antibodies. The tumors stopped growing and regressed in mice that got the therapeutic, but not in controls that received an antibody that’s not supposed to bind the target. In Science Immunology, they showed a similar effectiveness for a bispecific antibody targeted to proteins made from mutant RAS oncogenes in cell culture and in mice.
“This particular technology, these bispecific antibodies by themselves, that’s not new. A regular antibody can grab with one arm a tumor antigen that is expressed at high [levels] on the tumor cell and with the other arm, it can grab the T cell,” says Theresa Whiteside, who studies immunotherapy at the University of Pittsburgh and was not involved in the work. The US Food and Drug Administration approved the first bispecific antibody, which targets CD19 and CD3 for acute B cell lymphoblastic leukemia, in 2015 and has since approved at least five others; many more are in some stage of clinical trials. But the use of these antibody fragments “with high affinity provides a very sensitive means of detecting a peptide presented by MHC in a tiny amount,” she adds.
Bispecific antibodies instead of CAR T cells
The most successful use of CAR T cells has been in treating B-cell–related blood cancers, but casualties of the treatment are healthy B cells. Wiping out T cells has much more dire immune consequences than depleting B cells, so the immunotherapy hasn’t been successfully used in treating T-cell–derived cancers. “The problem . . . is that if you engineer a T cell to attack a T cell antigen, there’s this phenomenon called fratricide, where the engineered T cells just kill each other before they ever do much to the tumor,” says Schatz.
In their study in Science Translational Medicine, the Hopkins collaborators tested whether or not bispecific antibodies would avoid killing healthy T cells. One way to classify T cells is by which one of 30 varieties of so-called T cell receptor β chains they present on their cell surface. A cancer will be made of clones with the same β chain, so the researchers made a bispecific antibody with the CD3 effector binding fragment on one side and an arm targeted to the relevant β chain antigen on the other side. In culture and in mice, the bispecific antibody made a bridge between the two T cells, facilitating the destruction of malignant cells and healthy cells with the same β chain antigen, but leaving the other, healthy T cells intact.
In theory this approach should be able to take out all of the leukemic T cells, says Brian Lichty, who studies anti-tumor immune responses at McMaster University in Canada and did not participate in the work. There are some open questions pertaining to both applications of the approach, including how widely applicable the development of antibodies to certain HLA-mutant peptide combinations will be and whether or not there will be immune consequences of wiping out this fraction of the T cells, he adds. “At least until their immune system recovers, they could be transiently susceptible to certain infections, but there’s no way of knowing until you test this clinically.”
“It’s going to take another few years to translate all this preclinical development into clinical trials . . . because we need to optimize the bispecific antibodies to make sure they’re safe for humans,” Shibin Zhou, a cancer biologist at the Ludwig Center and a coauthor of the studies, tells The Scientist. To facilitate the commercialization of this and other cancer-related tech, the researchers have founded, serve on the boards of, and consult with several companies, including Exact Sciences, Thrive Earlier Detection, and Personal Genome Diagnostics.
J. Douglass et al., “Bispecific antibodies targeting mutant RAS neoantigens,” Science Immunology, doi:10.1126/sciimmunol.abd5515, 2021.
E.H.-C. Hsiue et al., “Targeting a neoantigen derived from a common TP53 mutation,” Science, doi:10.1126/science.abc8697, 2021.
S. Paul et al., “TCR beta chain-directed bispecific antibodies for the treatment of T-cell cancers,” Science Translational Medicine, doi:10.1126/scitranslmed.abd3595, 2021.