Glioblastoma, the brain cancer that Sen. John McCain has, is invariably — and accurately — described as aggressive and as having a poor prognosis: Not even Sen. Edward Kennedy or Beau Biden, with access to the best care and most cutting-edge therapies in the world, beat it. Recent advances, however, have persuaded some scientists that effective treatments are on the horizon — and some might even help McCain live longer than earlier patients did.
“We’ve learned a tremendous amount about the biology of this cancer,” said Dr. David Reardon, director of neuro-oncology at Boston’s Dana-Farber Cancer Institute. That includes understanding “the mutations and other abnormalities that give tumors an advantage, how they adapt and become resistant [to chemotherapy], and how to get treatments across the blood-brain barrier” that is notorious for keeping out drugs.
As a result, he said, “there are an increasing number of drugs that are making their way toward the clinic.”
Market research company Informa Pharma Intelligence counts 87 experimental therapies, from vaccines and gene therapy to monoclonal antibodies (the workhorses of precision medicine), in clinical trials for glioblastoma. Most of those won’t be approved for years, if ever. But the newest tool in oncologists’ kit is one that McCain might benefit from immediately.
In June, the U.S. Food and Drug Administration approved an imaging agent that turns glioblastoma pink, making it stand out against healthy brain tissue. Normally, glioblastomas are so difficult to see that surgeons cannot “get it all” without also removing normal gray matter, which can leave patients with severe cognitive, speech, or other impairments.
Glioblastomas “have little tentacles growing into the normal brain that are challenging to see even with a microscope,” said Dr. Costas Hadjipanayis, chairman of neurosurgery at Mount Sinai Beth Israel in New York City. When a patient drinks the new agent, called Gleolan, three to five hours before surgery, it “allows us to visualize those fingers of tumor tissue,” he said.
Getting all of a tumor makes a big difference. In a 2016 analysis of 37 studies involving some 41,000 patients, researchers found that when surgeons removed all of a glioblastoma, patients were 38 percent less likely to die within the next year. They were only 16 percent less likely to die by the end of the second year, however, underlining how even “getting it all” is not the same as curing glioblastoma.
“Removing all the visible tumor does not cure this cancer — ever,” said Dr. Mark Gilbert of the National Cancer Institute. Years ago, surgeons removed the entire half of the brain that a glioblastoma was in. Even that “did not markedly increase survival,” he said: Patients still died because malignant cells had already invaded the rest of the brain.
Chemotherapy is of only limited use because glioblastomas shrug off the drugs like Iron Man swats away bullets. These tumors are a crazy quilt of different kinds of cells, said Gilbert. Each kind deploys different molecular tricks — a landmark paper identified four main types — in order to proliferate.
A drug might shrink and contain the tumor for a few months, killing susceptible cells, but the surviving cells are resistant. “You knock out one of the pathways and the others go their merry way,” Gilbert said. “They say, fine, we have free rein,” multiplying and dividing unchecked. That’s why glioblastoma is so aggressive, with a high rate of recurrence and a five-year survival rate of barely 5 percent. Just half of patients are still alive 15 months after their diagnosis.
The exception is if a patient’s glioblastoma is much more uniform than the norm. In that case, it is “addicted” to one specific proliferation pathway that’s vulnerable to a molecularly targeted therapy, Gilbert said. Since McCain will have access to the best care, he will presumably have his tumor genetically profiled (which costs upward of $8,000) to see if it falls into this category. With any luck, he will be in the 2 percent to 3 percent of patients with an extremely uniform glioblastoma that succumbs to a therapy that knocks out its one proliferation pathway.
Otherwise, the only shot at a true cure is to kill the multiple types of cells that make up a glioblastoma, probably sequentially rather than simultaneously. “We recognize that one single approach will not cure glioblastoma,” said Ann Kingston of the National Brain Tumor Society.
But the biological version of whack-a-mole might. That would entail determining the molecular profile of the most abundant cells in the glioblastoma, killing those, then determining the profile of the survivors and going after those. One of the most promising experimental approaches to glioblastoma is therefore to enlist the immune system, which has an unmatched ability to seek out and destroy malignant cells that differ from one another in the way that glioblastoma cells do.
Last December, researchers at City of Hope hospital in Duarte, Calif., reported that they had gotten the much-publicized immunotherapy called CAR-T cells to attack one patient’s glioblastoma. CAR-Ts are immune cells that are removed from a patient and then genetically engineered to attack a particular molecule on tumor cells, and hence the tumors themselves. The 50-year-old patient had suffered a recurrence of glioblastoma after surgery, radiation, and the chemotherapy drug temozolomide. When he received multiple injections of CAR-Ts into his brain, his tumor regressed and was kept at bay for nearly eight months.
That suggested that “a better treatment may be attainable,” Christine Brown, who led the study, said at the time. “We can take a patient who has actively growing, advanced, metastatic multifocal glioblastoma, and we can see regression.”
In an indication of how initial hopes can be dashed, however, the man’s tumors returned in different parts of the brain “even more aggressively than before,” said a City of Hope spokeswoman. “In the end, we were unable to keep up with the rapid spread of his disease.” He died of glioblastoma in April.
That might be a warning sign of a big concern about CAR-Ts — they target a single molecule on tumor cells. Given the heterogeneity of glioblastoma cells, that might be way too few, said Dr. Donald O’Rourke of the Perelman School of Medicine at the University of Pennsylvania. Targeting a single molecule, he said, will almost certainly leave tumor cells that have different surface molecules untouched and unleashed.
Last week, O’Rourke and colleagues reported more sobering news about CAR-Ts and glioblastoma. CAR-Ts engineered to target a more common molecule than the City of Hope approach (it’s called EGFRvIII) penetrated glioblastomas in the Penn study’s 10 patients. But the tumors fought back. They created an “immunosuppressive microenvironment,” O’Rourke said, quashing the CAR-Ts.
NCI’s Gilbert, while “very excited” about approaches that target the EGFR molecule, estimated that only 5 percent to 40 percent of a glioblastoma’s cells sport that molecule on their surface.
“Even within the same tumor you have these different molecules,” O’Rourke said. “It’s kind of daunting and is going to be a problem.” He suspects, however, that if CAR-Ts are combined with existing immunotherapies, such as Yervoy and other so-called checkpoint inhibitors, they might be more effective. “But we have a lot of work to do,” he said.
Scientists hope that a different kind of immune therapy will hit more targets, and therefore more of the different kinds of cells in a glioblastoma, than CAR-Ts. Called a neoantigen vaccine, it consists of molecules (actually, pieces of them) found on the surface of glioblastoma cells. One patient’s surface molecules are different from other patients’, so neoantigen vaccines are the epitome of personalized therapy. The patient’s immune system recognizes the molecules as invaders and mounts an attack on the corresponding molecules on the tumor. The approach recently showed success in a small study of melanoma patients, and is being tested in glioblastoma patients in a clinical trial at Dana-Farber.
The hope is that a vaccine against one batch of molecules, and then another vaccine against the molecules that appear next, might keep pace with glioblastoma cells’ protean tendencies. But it takes weeks to make these individualized vaccines, time that patients might not have.
Even further along is a vaccine that also targets molecules on a patient’s unique glioblastoma but in a somewhat different way from the Dana-Farber approach. Manufacturer Northwest Biotherapeutics has recruited 331 glioblastoma patients into a late-stage clinical trial, and at this year’s meeting of the American Society of Clinical Oncology reported that 40 percent of patients treated with “DCVax-L” were still alive at 35 months, compared with 15 percent of patients receiving standard care; 24 percent of patients who got the vaccine were alive after four years, compared with 8 percent otherwise.
It’s a measure of how desperate the search for a glioblastoma cure is that one drug company has gone back to the original “war on cancer” launched by President Nixon 50 years ago to resurrect a compound that wasn’t fully investigated then (because of patent issues that are not longer an obstacle). It’s an old-line chemotherapy drug that damages cells’ DNA. Unfortunately, cells can often repair that damage, eventually making the drug ineffective.
DelMar Pharmaceuticals’s version attacks DNA at a place the repair machinery doesn’t reach, explained CEO Jeffrey Bacha. In a mid-stage clinical trial, patients whose glioblastoma had returned after surgery, radiation, and standard chemotherapy survived a median of 8.3 months, compared with about three months in the comparison group.