When Dr. Adam Sonabend treats patients with glioblastoma — one of the deadliest forms of brain cancer — he has long faced a frustrating mystery. Most of his patients don't respond to immunotherapy, a treatment that has revolutionized care for other cancers. But a small number of patients do remarkably well, and no one understood why.
"Mostly, we have been disappointed with the results," said Sonabend, a neurosurgeon and researcher at Northwestern Medicine in Chicago. "But, occasionally we had some remarkable success, and we just didn't know why."
Now, Sonabend and his team have found an answer. Their research, published in the journal Nature Communications, discovered a biological signal that predicts which glioblastoma patients will benefit from immunotherapy. The key is a cell signaling pathway called MAPK/ERK — sometimes described as a chain of molecular switches inside cells.
The scientists used a gene-editing tool called CRISPR to scan hundreds of genetic targets in mouse models of brain cancer. The MAPK pathway emerged as the strongest predictor of how tumors would respond to treatment. When this pathway was more active, tumors became easier for the immune system to recognize and attack.
"What we realized is that when this pathway is activated, the tumor cells overall are more immunogenic," Sonabend explained. "They engage in a pathway called interferon response that is super important for T-cells to recognize the tumors. They're more likely to drive T-cells into the tumor."
In experiments, increasing the pathway's activity improved how mice responded to immunotherapy. Some mice even developed lasting protection against tumor growth — suggesting the treatment had trained their immune systems to keep fighting.
The findings fit with earlier observations. Previous research in Sonabend's lab had noticed that rare genetic mutations activating the MAPK pathway appeared more often in patients who responded well to immunotherapy. The new study confirms why: the pathway essentially flips on the signals that help T-cells spot and attack cancer.
For patients, this could mean a way to know before starting treatment whether immunotherapy is likely to work. "This biomarker might actually be relevant for all kinds of glioblastoma immunotherapy that rely on T-cells," Sonabend said.
The discovery does come with a caution. The MAPK pathway appears necessary for immunotherapy to work, but it's not a guarantee on its own. "It is very clear that this MAP kinase activation is necessary, but not sufficient," Sonabend said. "Because you can have this pathway activated and not respond, but it is very clear that if you don't have this pathway, you're unlikely to respond."
Still, Sonabend's team at Northwestern Memorial Hospital has already begun testing tumor samples for this biomarker — giving patients new hope that doctors may soon be able to tailor treatment more wisely. "We're now routinely testing for this biomarker at Northwestern," he said.