NEW YORK — A new immunotherapy approach that targets cancer driver gene mutations could be used to develop a suite of treatments.
Most proteins altered in cancer are not expressed on the cell surface and rather do their work inside cells, which keeps them apart from antibodies. This presents a challenge for cancer immunotherapies aimed at genetic alterations, Bert Vogelstein, co-director of the Ludwig Center at the Johns Hopkins Kimmel Cancer Center, said during an online presentation last week at the Advances in Genome Biology and Technology general meeting.
But when proteins are degraded, their peptides can sometimes be presented on the cell surface in complex with HLA molecules, which immune cells like T cells can see. For their new immunotherapy approach, Vogelstein and his colleagues turned to bispecific antibodies.
With bispecific antibodies, one end of the antibody latches on to a bit of altered protein called a mutation-associated neoantigen, or MANA, which is produced and displayed by a cancer cell, while the other end binds to a T-cell, bringing the cancer and immune cells together.
This in turn could allow the immune cell to attack the cancer cell and kill it without harming non-cancerous cells. As they also reported last week in a series of papers appearing in Science and related journals, the researchers developed bispecific antibodies aimed at common cancer alterations in TP53 and RAS.
These bispecific antibodies could someday be the basis to treat a range of cancer mutations. "We can develop a panel of antibodies against the most frequently mutated oncogenes or tumor suppressor genes against the most popular HLA subtypes," co-author Shibin Zhou, an associate professor of oncology and the director of experimental therapeutics at the Ludwig Center, said in an interview. He added that if he and his colleagues were to develop antibodies that targeted the 10 most common mutations that occur in cancer, they could potentially treat up to a million patients a year around the world.
For their Science paper, the Hopkins team developed a bispecific antibody aimed at a common mutation within TP53, which is itself the most commonly mutated cancer gene. The researchers in particular focused on the R175H mutation in which an arginine at position 175 is replaced with histidine.
The researchers first confirmed that peptides generated by R175H-mutated TP53 are expressed in a complex with HLA-A*02:01 — the HLA allele that is found among more than 40 percent of white Americans — at low levels on the surfaces of cancer cells. There were an average of 2.4, 1.3 and 1.5 copies of these complexes on the cell surface, in three human cancer cell lines with the TP53 mutation, respectively.
Antibodies aimed at detecting these oncopeptide-HLA complexes need to be highly specific, as cancer proteins can differ from normal ones by only one amino acid. The researchers searched through phage-display libraries to uncover T cell receptor mimic single-chain variable fragments (scFvs) that bind the R175H-mutated TP53-HLA-A*02:01 complex. They also searched for scFvs that bind wild-type TP53 to rule them out and ensure the scFvs only bind that specific mutated complex.
They transformed the scFvs into the therapeutic format — a bispecific single-chain diabody in which one arm binds the R175H-mutated TP53-HLA-A*02:01 complex and the other CD3, which is part of the T cell receptor complex. When the researchers tested this bispecific antibody in cell lines, they found it led to T cell-based lysis of cancer cells and, when they tested in mice, there was a decrease in tumor burden.
Further, when they used CRISPR to disrupt the parts of the bispecific antibody targeting either the HLA type or the specific mutations, it no longer recruited T cells, suggesting it is highly specific.
The researchers similarly developed bispecific antibodies targeting RAS mutations, namely RAS G12V and Q61H/L/R mutations that are presented by HLA-A3 and HLA-A1, as they described in Science Immunology. In cell lines, these bispecific antibodies could also induce T cell activation and the targeted killing of cancer cells.
They also adapted their bispecific antibody approach for T cell cancers by instead targeting cancer-associated T cell receptors, as they additionally reported in Science Translational Medicine.
These findings suggested to the researchers that many versions of bispecific antibodies could be developed to target a range of cancer mutations and HLA types and used off the shelf to treat cancer patients. "What we are after are what we call public neoantigens," Zhou said. "Those are the neoantigens that are shared by many patients."
In that sense, Vogelstein noted during his talk, this is a precision medicine approach, as it treats patients on the basis of their specific mutations and HLA types, though it is not individualized.
Zhou added that they are working to extend their panel of antibodies. In particular, they are focusing on other common cancer driver mutations, but are also prioritizing more deadly cancers for which there are limited treatment options, like pancreatic cancer and ovarian cancer.
They are targeting driver mutations in particular for their antibody therapy panel as those mutations are less likely to be lost by cancer cells, making treatment resistance less likely, Zhou added. Of course, he noted that cancer cells could still become resistant to bispecific antibody treatment in other ways, such as by no longer expressing HLA. That is an issue they would have to later overcome, he said.
In a related commentary appearing in Science, the University of Texas at Arlington's Jon Weidanz noted that single-chain diabodies are typically cleared quickly from the body, suggesting the treatment might have to be continuously delivered through an implanted pump. That is an issue Zhou also said they are also working to optimize.