A couple years back, some cardiovascular specialists at Boston’s Brigham and Women’s Hospital stumbled across a surprising result.
Novartis had tasked them with carrying out the long-range Phase III trial for canakinumab, an experimental anti-inflammatory drug the company was trying to market for cardiovascular disease. The main results were mixed — modest reductions in fatal cardiovascular events, clear side effects — but an additional, exploratory analysis, turned heads over in oncology: Across 10,061 patients, those who had received canakinumab were less likely to develop lung cancer; 33% less likely for the highest dosage.
“The hazard ratio was striking,” Kwok-Kin Wong, a lung cancer specialist at NYU’s Grossman School of Medicine told Endpoints News, referring to the 33% figure and an even higher number for a different subgroup.
The question: Why?
New research published today in Cancer Research, from one of the earliest RAS-mutant pancreatic cancer researchers, may give a peak into that question, while offering the potential to open up one of the most intractable cancers to the latest wave of targeted therapies.
A team led by NYU’s Dafna Bar-Sagi discovered that RAS-positive pancreatic cancer cells emit an immune-signaling protein called interleukin-1β – also the protein that canakinumab, the Novartis drug, inhibits. That protein turns down the immune system near the tumor in a variety of ways, they found, most notably by helping build more collagen fibers that essentially guard the cancer from T cells that would ordinarily try to invade and kill it.
“It’s a wall the tumor build around itself, and does not allow the immune system in,” Bar-Sagi told Endpoints.
These walls, she said, accounted not only for the tumor’s ability to elude the body’s natural immune system, but also the all-but-complete failure of the immune-unleashing PD-1 checkpoint therapies in pancreatic cancer.
“The PD-1 checkpoint therapy won’t have an effect if the [immune] cell can’t get into the tumor,” she said. Her paper tested if an anti-interleukin drug like canakinumab could be used in combination with a PD-1 drug, the former clearing the way for the latter, finding a 32% reduction in mice. (Novartis did not fund the study but did provide the interleukin and checkpoint drugs).
Gregory Beatty, an assistant professor of immunology at the University of Pennsylvania School of Medicine who was not involved with the study, said the study was “exciting and applicable to not only pancreas cancer but potentially other malignancies.”
“Approaches to overcome immune resistance in pancreas cancer are of utmost importance and IL-1b is a novel target,” Beatty said via email.
Kwok-Kin Wong — who was not involved with the study but works at the same institution — said the findings were important for a range of cancers, including by offering a mechanism for the 2017 findings in lung cancer. Lung cancer specialists could look for the same process.
“This paper will spur a lot of interest,” Wong said, noting many companies already have an interleukin antibody and a PD-1 on hand. “There’s a lot of therapeutic implications.”
Bar-Sagi was one of the earliest researchers into RAS mutations. She joined James Feramisco’s Cold Spring Harbor Lab as a postdoc in 1986, shortly after the oncogene was first discovered, and quickly co-authored in Science one of the first papers suggesting how the RAS gives rise to tumors.
Over the years, though, no one fully identified interleukin-1β in the tumors, even as understanding of RAS progressed to a point at which drug developers could bring targeted therapies into the clinic.
The reason for this was simple, Bar-Sagi said, and it has major implications for any future research: The protein doesn’t show up in tumor cell cultures.
Building off of research at her lab, Washington University of St. Louis and elsewhere, Bar-Sagi devised a way of testing it in vivo. They discovered that the tumor relies on a previously undiscovered ligand binding to its TLR4 receptors. By giving antibiotics to clear bacteria in some mice, they found that a warped microbiome may promote the binding, contributing to the development of cancer.
Only once they inserted the ligand into cell cultures, the cells began producing interleukin-1β . And when they gave RAS-positive mice a combination of a PD-1 and an interleukin-1β antibody, they found a 32% reduction in tumors.
Bar-Sagi said it’s a breakthrough that continues to point to the importance of not just the tumor but also the tumor’s effect on the micro-environment around it and the immune system there. But to explore that, she said, researchers will have to turn away from Petri dishes and find new methods of studying tumors — pancreatic, lung or otherwise — as they exist within the body.
“Many of these effects are not going to be recognized or will be very hard to study using the conventional study methods,” Bar-Sagi said. “And that’s definitely a shift.”