A yearslong effort to take the guesswork out of creating personalized cancer vaccines advanced significantly last week with the publication of a scientific paper by researchers at the La Jolla Institute for Immunology.
Using a functional technique first tried in patients in 2018, a team led by immunologist Stephen Schoenberger and computational biologist Bjoern Peters claims that their method can isolate valid tumor targets called neoantigens “at rates that are about 10-fold higher” than computer-based modeling can currently deliver.
It’s a significant finding as cancer vaccines, if loaded with the correct snippets of proteins matching tumor features, can cause a patient’s own immune system to manufacture disease-fighters called T-cells made to home in on cancer cells directly.
The LJI approach is clever in that it uses a patient’s own blood to determine which targets are already viable among the thousands that algorithms predict might be effective. Analysis involves comparing the DNA of non-cancerous cells to those collected from tumors to determine which mutations might be recognized by the immune system.
“We simply asked the question from the other end, ‘are there T-cells that see mutations already in the patient’s blood primed by their own immune system?” Schoenberger said.
After using custom computer programs to generate a list of every likely target, and ranking them by likely prevalence in tumors, the team manufactured proteins called peptides that incorporate the neoantigens that computer analysis predicted might work. An extra step incubated each target with T-cells gleaned from each patient’s blood. If cells started producing telltale chemicals when exposed to a neoantigen, researchers could infer that they were created to detect that particular target.
The results of this analysis went further than the team expected, finding that T-cells responded not just to antigens thought to be present in cancers such as melanoma, which have a high amount of mutation, but also in types thought to be “cold” in that there were fewer changes to detect.
“It suggests that immunotherapy that is designed to activate those cells is something that can be applied to more tumors: melanoma and lung cancer and high mutational burden tumors but also for colon cancer and ovarian cancer and pancreatic cancer and on and on,” Schoenberger said.
Immunologist John Connolly, chief science officer for the Parker Institute for Cancer Immunotherapy, reviewed the paper, published in the journal Science Translational Medicine, and said he found its conclusions convincing.
Taking a functional approach to validating which neoantigens are recognized by a patient’s immune system, he said, is a novel enough approach that it is likely to be used more broadly among the ever-growing list of researchers exploring cancer vaccines.
“I think it’s a real innovation that, now that it’s published, people will and should consider adopting into their existing platforms,” Connolly said.
The notion that the immune system has recognized tumors and has even made T-cells to kill them may seem confusing for those without doctorates in immunology. Why, many may wonder, doesn’t the immune system just kill the tumors it has detected?
Why is a vaccine even necessary?
The answer, Schoenberger said, lies in the differences between the innate and adaptive portions of the immune system. Programming T-cells is part of the adaptive system, but it is the innate piece of the puzzle that causes those cells to activate.
A vaccine, then, must not only provide the right set of instructions, but must also prompt action. The body must be made to believe an attack is under way.
“What we try to do with a cancer vaccine is we introduce the information on a specific set of mutations in the context of a viral infection,” Schoenberger said. “We use something called an adjuvant, something that will activate the innate arm of the immune system and, combining these, we get cells, many more cells, that are now armed to eradicate the cells expressing the antigens that they’re able to recognize.”
The approach LJI used also showed an ability to find neoantigens that activate not only CD8 T-cells, the kind that most cancer vaccines using computer-only prediction methods target, but also CD4 T-cells, which, the researcher said, have thus far been beyond the reach of predictive analysis alone.
Schoenberger said the next path of exploration is to explore ways that antigens that activate CD4 and CD8 T-cells could be linked. Previous work has shown that these two types of cells can work together, delivering a bigger punch.
“It’s a concept that we’re eager to test in the next version of the vaccine,” Schoenberger said.
Cancer vaccines are not yet a silver bullet, as is illustrated by the 2022 passing of Tamara Strauss, the first of 13 patients to receive a vaccine during the LJI trial. Though her pancreatic tumor was held in check for several years, progress was not enough to result in a cure. Full information on the outcome of the trial’s first 13 patients is not yet available but there are plans to publish those results soon.
The rapid pace of advances in cancer vaccines, Connolly said, suggests that they may soon be a first-line treatment for a wide range of malignancies.
“I would say that, within 10 years, we’ll see a sea change, and that’s really because of the great work that you see coming from so many labs,” Connolly said.
