Does Matthias Gromeier know a better way to treat cancer?

When the professor of neurosurgery modified the poliovirus, he helped bolster immunotherapy.
June 11, 2018

The honor of pressing the button fell to Matthias Gromeier. The button activated a pump that was connected to a syringe containing a genetically modified form of poliovirus, infused in a liquid. Once Gromeier initiated the process, the infusion would travel through a long plastic tube and into a small hole drilled into twenty-one-year-old Stephanie Lipscomb’s skull. It would take six-and-a-half hours for the entire virus to reach its target: a large brain tumor, called a glioblastoma, that had outmuscled chemotherapy and radiation and was threatening to kill the South Carolina woman. 

No one watching the process at Duke’s Neuroscience Intensive Care Unit knew whether the treatment would help destroy Lipscomb’s tumor. It had never been tried on a human patient before. Gromeier, a professor of neurosurgery, felt confident that, at minimum, the virus would cause no harm: He had engineered it himself, as a postdoc in the 1990s, altering the structure so it couldn’t cause the paralyzing disease poliomyelitis. He had spent years convincing the federal government that human trials would be safe. “We just didn’t know what to expect in terms of ‘Could it help her?’ ” he says.

Later, as the infusion ended, Gromeier stood at the foot of Lipscomb’s bed. He was joined by Annick Desjardins, an associate professor of neurosurgery, who at the time was running the clinical trial. “I was scared,” Desjardins says of Lipscomb’s decision to become Patient No. 1. “I was excited to be able to offer her a new trial. But at the same time, it would have been so much easier to offer her something that we had already used. We were both going into the unknown.”

There was so much unknown in May 2012. No one knew the correct dose to give or what to expect after the treatment. They didn’t even know how the modified poliovirus might attack a tumor. It would take another five years for Gromeier and his colleagues to publish a journal article explaining the biological mechanism, and an additional year for the clinical trials to expand beyond Duke.

A quarter-century after Gromeier first re-engineered the poliovirus—after enduring skeptical colleagues, fearful regulators, and research breakthroughs followed by setbacks—cancer experts are now paying close attention to his work. “It’s a very bold approach,” says Stephen J. Russell, a professor of molecular medicine at Minnesota’s Mayo Clinic College of Medicine & Science, who describes Gromeier as “a strong, innovative scientist with great depth.” Russell says it’s too soon to know whether the research will lead to an effective next-generation cancer treatment. Still, he says, “I’m watching with interest.”

Some of those directly involved in the research, while acknowledging how much work remains, are nonetheless speaking with a heady optimism. “Even for somebody who is very seasoned like myself, I have never been as excited as I am about this strategy and this therapy,” says Mitchel Berger, director of the Brain Tumor Research Center at the University of California, San Francisco, one of the new clinical-trial sites. (Berger is also a scientific adviser to Istari Oncology, a private research start-up cofounded by Gromeier.) “The data that’s come out so far—it’s just extraordinarily impressive. I can’t imagine it not making a huge impact in this field.” 

Gromeier is a soft-spoken fifty-two-year-old with a salt-and-pepper Van Dyke beard and enormous blue eyes. Originally from Germany, he earned his medical degree from the University of Hamburg, but already knew he didn’t want to treat patients. Earlier, during his compulsory military service, Gromeier had worked at a large breast-cancer center. “Breast cancer, back then especially, was a losing battle,” he says. “It wouldn’t fulfill my life to prescribe chemotherapy so patients suffered and it didn’t work.” If he wanted to make a difference, he decided, he needed to go into research.

He originally wanted to study HIV. It was the late 1980s, before protease inhibitors were saving lives, and the disease was still considered a death sentence. But Gromeier couldn’t find a physician with an HIV lab who would welcome him as a student. “The only lab that would take me was a tired, not very successful polio lab,” he says.

After medical school, Gromeier came to the United States for a postdoctoral fellowship at New York’s Stony Brook University. There he studied with Eckard Wimmer, one of the world’s leading virologists and a polio specialist. Gromeier spent much of the 1990s studying polio pathogenesis—that is, how the virus causes the infectious disease associated with iron lungs and Franklin D. Roosevelt. (Polio was nearly eradicated by a vaccine developed in the 1950s, but still exists in a handful of developing countries.)

Around 1993, while at Stony Brook, Gromeier engineered the virus he’d later use in the Duke cancer trials, swapping out a critical part of the structure with the equivalent part of the human rhinovirus, which causes the common cold. “This was made for very nerdy basic research—curiosity, let’s see what happens, that kind of thing,” he says. It wasn’t until he began testing it on animals—first mice, and later monkeys—that he understood the importance of his recombined virus. It had lost the ability to cause poliomyelitis, which had implications for understanding how it and other viruses work, and how the brain responds to them.

Gromeier also studied receptors, proteins on the surface of cells that viruses evolve to recognize. (Think of a lock-and-key system: If a virus matches the receptor, it can enter the cell.) It turns out that the poliovirus binds to a receptor called CD155, which is found on many solid tumor cells. Through a succession of experiments, starting in the mid-’90s, Gromeier concluded that the modified poliovirus could potentially target cancer. It helped, he believes, that he was not a cancer expert.

“This was an ignorance-is-bliss type situation,” he says. “I didn’t have preconceived notions of what cancer was and how to fight it. Sometimes for researchers, if you’re very, very well read, very exposed to current opinion, it can mean that you’re biased and have a closed mind.”

Gromeier’s early thinking focused on what he now calls a “simple paradigm”: the ability of his modified poliovirus to infect and kill tumor cells. “That’s relatively rare,” he says. “Viruses have evolved over millennia to do certain things, and killing tumors is not generally part of their natural program.”

While the virus does kill tumor cells, this turned out to be an oversimplified version of how it attacks cancer. “I was very naïve,” he says.

Gromeier came to Duke in 1999, drawn in large part by the strength of its brain-tumor research. He had always seen brain cancer as a prospective target for his work. “Poliovirus is the virus that’s most capable of causing the most damage in the brain,” he says. It invades the central nervous system and can paralyze the muscles we need to walk, swallow, and even breathe. “It may sound counterintuitive,” he says, “but I saw this as a sign that it might be a good agent to be used in the brain.”

“Counterintuitive” was an understatement. To some experts, it was outrageous. “Early on, when we first heard about it, it was ‘What the hell?’ ” says Mayo Clinic’s Russell. Even local colleagues were skeptical. “I thought this was nuts,” says Henry Friedman, deputy director of Duke’s Preston Robert Tisch Brain Tumor Center. “A virus, however modified, that produces motor-neuron attack and paralysis—it was one of the real plagues of mankind, and I just didn’t understand how one could exploit it.”

Given the horror of poliomyelitis, the U.S. Food and Drug Administration would need some serious convincing to allow the virus to be tested in humans. “They were so afraid that there was going to be toxicity,” says Darell Bigner, director emeritus of the Tisch Brain Tumor Center. “They were afraid, at the simplest level, that it was going to lose its genetic stability and revert to a wild-type poliovirus and cause poliomyelitis. And they were also afraid that the mutations might cause some very new, horrible type of virus, causing diseases we’ve never seen before.”

Before the FDA would approve human clinical trials, it required researchers to inject the virus into the brains of thirty-nine macaques. (This was done at an outside contractor’s lab.) What’s more, the federal agency required Gromeier to explain the mechanism that assured the virus’ safety in human subjects. “If you come to the FDA with something outrageous, they give you an impossible list of tasks to do,” Gromeier says. “And they did that to me.” All told, it took almost a decade of work before the FDA approved human trials.

This is not unusual in the history of medicine, Gromeier notes. “People are reluctant to embrace something that’s not yet standard practice,” he says. “And what we did was daring; there’s no way around it. It’s important in science to not just do those things that everyone feels are safe. You have to push the envelope.”

Gromeier invokes one of his heroes, Albert Sabin, whose oral polio vaccine helped bring about a global near-eradication. “He faced terrible opposition to his idea because he wanted to give live poliovirus to children,” Gromeier says. “He had to do his clinical work in the Soviet Union because he couldn’t do it in the United States. And see how that all ended up?”

FDA green-lighted the first clinical safety trial in 2011. The authorization cleared a tremendous obstacle. But that didn’t guarantee Gromeier and his colleagues a straightline path.

Even before she became Patient No. 1, Stephanie Lipscomb knew she wanted to be a pediatric oncology nurse. A student at University of South Carolina Upstate, from a religious family, she had chosen her career after babysitting a sick child, according to a profile in People magazine. It seemed unlikely she’d achieve her goal: In April 2012, ten months after undergoing chemotherapy and radiation, Lipscomb had a seizure that signaled the return of her tumor. Most patients with recurrent glioblastoma live for less than a year after it returns.

Offered the chance for the experimental treatment, she showed little hesitation, recalls Desjardins, the neuro-oncologist who had treated Lipscomb’s cancer and would later run the clinical trial. “She said, ‘Yes, absolutely. This is my job. I always wanted to bring new therapies and help other people that live with cancer, and I’m doing it.’

”Not everyone was so enthusiastic. “Wait a minute,” Desjardins recalls Lipscomb’s mother saying. She was old enough to grasp the fearsome potential of poliomyelitis and worried that her daughter didn’t understand the experimental treatment. Lipscomb did understand, and wanted to proceed.

Gromeier expected to be anxious when the infusion started. “But I wasn’t,” he says. “It was a calm moment.” He credits the patient, who displayed no visible anxiety. “She had such poise,” he says. “She was funny even.”

When Lipscomb returned for a checkup two months later, the seizures had disappeared and she felt better. On her MRI, though, the tumor looked larger. “In cancer, anything larger is bad,” says Desjardins. “Dr. Gromeier came and looked at the MRI with me, and he said, ‘This is exactly what we see in animals. We just need to sit on it and wait.’”

At six months, Gromeier says, “I thought her MRI looked horrible.” (He is quick to add, “I am not an expert on reading MRIs.”) He deferred to Desjardins’ wisdom, and this time it was her turn to reassure. “She described to me some features on the MRI that she said she was encouraged by,” he says “And Dr. Desjardins was right.”

After that the tumor shrank, and today Lipscomb remains in remission. She was married in March and works as a registered nurse, according to her wedding announcement.

The Duke team had a number of impressive outcomes like Lipscomb’s. By that point, Gromeier was coming to understand that poliovirus does more than kill cancer cells directly. It also stimulates an immune response, directing the body’s own immune system to destroy the tumor.

“It looked like it might be a home run in the field of oncology,” says Mayo Clinic’s Russell, who heard Gromeier speak at a 2013 cancer conference in Quebec City. “It felt like a dream come true.”

Then came the setbacks. Some patients, says Desjardins, had tumors that were growing too fast for an activated immune system to make a difference. In others, the virus triggered too much immune response for the patient to tolerate. A sixty-year-old social worker named Donna Clegg was treated with a dose three times stronger than Lipscomb’s. She then experienced inflammation that put pressure on her brain. (“The brain is limited in ability to expand by this enclosed bone box that is the skull,” Desjardins explains.) Clegg became partially paralyzed, decided to withdraw care, and died in 2015.

Raising the dose had backfired. That’s not how cancer treatments usually work. “In cancer, there is a very old, ingrained culture: You want to push the dose as high as you can,” Gromeier says. “Just look at how we use chemo: We basically say, ‘How much can we give the patient before killing him or her?’ and we stay a little bit under that.”

But with the immune system, more is not necessarily better. “Often, you get a worse response,” Gromeier says. Recognizing this, the Duke researchers scaled back the dose.

Different patients offered different lessons. Brendan Steele, an IT manager in his thirties, suffered and survived a brain hemorrhage when his catheter was being removed, leading researchers to suspect that he didn’t receive much of the virus. Seven months later, his tumor returned. “We gave him one dose of chemotherapy and—bing!—the tumor broke down and almost completely disappeared,” says Desjardins. “That’s really, really rare.”

Gromeier speculates that a single dose of chemo (as opposed to a full course) might have triggered an “immunological reset.” Chemotherapies used for brain cancer deplete the body of the white blood cells called lymphocytes, some of which help fight cancer and some of which suppress the body’s immune response. “You could imagine that by wiping out everything, you allow the more desirable anti-tumor lymphocytes space to grow,” he says.

In 2016, Desjardins gave an update, in the form of a poster presentation, at the American Society of Clinical Oncology’s annual meeting in Chicago. She reported that 23.3 percent of the modified-poliovirus recipients were alive after two years, compared to 13.7 percent of a similar group of brain-cancer patients treated conventionally at Duke. There’s considerable more testing to come before the FDA grants its approval for clinical use. But that same year, the agency promised to expedite development, based on the initial results, when it designated the poliovirus treatment a “breakthrough therapy.”

Associate professor of neurosurgery Annick Desjardins, in white coat, ran the poliovirus clinical trial and Smita Nair, right, a professor of surgery, team up with Gromeier to explore the science in mice.



Early on, my comfort zone was the fact that the virus was great at killing tumor cells,” Gromeier says.
“But that’s not the whole story.” As it became evident that his re-engineered virus activated the immune system, he set out to understand how. Immunotherapy—helping the body’s own immune system fight disease— has become one of the hottest topics in cancer research, and Gromeier believed his new therapy might be relevant to the conversation.

He teamed up with Smita Nair, a professor of surgery at Duke, and an expert in how immune cells help our bodies survive infection or cancer. The two had met in 2013 through mutual colleagues. Together, they worked both with human cells and with mice that had been genetically engineered with the human CD155 receptor.

We humans have two categories of immune cells. “Innate immune cells” are the first responders; they rush to the scene of an injury or infection, within hours, to assess the damage and take quick action. Based on the signals they receive, they then signal the next line of defense: “adaptive immune cells.” (One well-known type of adaptive cell is the T-cell.) Once adaptive cells respond to an abnormality, they create a memory, so they can return if it recurs.

What Nair and Gromeier figured out is that, when the poliovirus uses the CD155 receptor to enter and kill a tumor, the cancer cells release proteins called antigens. The killing of the tumor signals the innate cells to travel to the tumor site, where the innate cells recognize that there’s something both abnormal and dangerous going on. They then present the antigens to the T-cells. “The T-cells realize, ‘These antigens are not part of the body. This is something wrong,’ ” Nair says. The T-cells go on the attack, seeking out and annihilating the cancer cells.

But there’s more: CD155 receptors are also found on two types of innate immune cells, dendritic cells and macrophages. Gromeier knew this already, because a friend of his had discovered it two decades early. Until then, he says, “I thought it was a sideshow.” But it turned out to be key to poliovirus’ potential value as an anti-cancer therapy.

The CD155 receptor, he and Nair learned, allows the virus to enter and activate the innate cells. Dendritic cells activated in this manner become fully capable of presenting antigens to T-cells. “We were not anticipating that,” Nair says. “Research always takes you down a path, and you follow your path.”

When they saw that poliovirus directly infected immune cells, “that’s when everything started coming together and we had a proper story,” Nair says. The team published the findings in September 2017 in the journal Science Translational Medicine.

With the safety phase completed, and breakthrough status from the FDA, the pace of clinical trials has now notched up. In 2016, to attract outside investors and facilitate commercial development, Gromeier and Bigner cofounded the private firm Istari Oncology. Duke’s Desjardins (now a co-investigator) and Friedman, along with Berger from the University of California, San Francisco, all have stakes in the company. “We stand to gain financially from the success of this intervention,” notes Friedman, the chief medical officer.

The profit-making end makes Gromeier uncomfortable. “I’m being shielded from a lot of this,” he says. “I’m involved with Istari but do not go to meetings typically. They very much like to have scientists do science and business-minded folks do business-type things.”

Meanwhile, Duke has launched a safety trial for children with recurrent brain tumors. Scientists are also looking at other types of cancers for which the treatment might be effective, including breast cancer and melanoma. And work has begun on a multi-institutional brain-cancer trial that will measure survival after two years. Half the sixty- two patients will receive the modified poliovirus alone, and the other half will receive the virus plus a single dose of chemotherapy. The other sites are Massachusetts General Hospital; the University of California, San Francisco;and Boca Raton Regional Hospital in Florida.

After this trial, it will take a larger multi-institutional study for the treatment to win FDA approval. Bigner expects that trial to take place in both the United States and Europe.

William Curry, a neurosurgeon at Massachusetts General and the principal investigator at that site, notes that the polio research is part of a larger effort to attack cancer with viruses. For example, a modified herpes virus is now being used to treat advanced melanoma. The early poliovirus data, he says, offer him “genuine enthusiasm” about the approach.

“There’s nothing out there that, by itself, is going to be a cure for all patients with glioblastoma,” Curry says. “There are some [Duke] patients that responded, and many that don’t have a really durable response. But any advance that we can make, therapeutically, in a patient with a recurrent glioblastoma—I haven’t really seen that ever happen in this disease in my career.”

For Gromeier, this moment represents a convergence of his own career and one of the most promising cancer-research trends. “When we started out, we were in Neverland,” he says. “And as we moved forward, we coalesced with what the cancer immunotherapy field wants. That’s very important in science, because you can’t just pigheadedly stick with what you invented. In our case, we were adopted by this much bigger field.”

During his first decade at Duke, Gromeier incrementally began thinking of himself less as a virologist and more as a cancer researcher. “Now I’m in the midst of a second transition,” he says, “from cancer to immunotherapy and immunology.”

He’s not anticipating a third transition. “Pigeonholing is extreme in science,” he says. “You’re supposed to do what you know best and stick to your guns.” Given the funding environment, which rewards specialization, Gromeier plans to continue on his current course until he retires. “It is my personal conviction that the only hope to obtain meaningful, dramatic progress against cancer will be through immunotherapy,” he says. “There’s no alternative.”

If his work eventually leads to a game-changing FDA-approved cancer treatment, that will still be many years in the future. Gromeier doesn’t mind the wait. “This is how things work: incredibly slowly, and you need to have this long horizon if you want to make a dent,” he says. “If you’re interested in cancer, you better make that a lifelong commitment.”

Yeoman is a freelance journalist living in Durham. His work has appeared in The Washington Post, The Nation, Popular Science, and National Wildlife, among other publications.