How gut bacteria impact cancer treatment: Meet the STAT Madness 2018 editors’ pick

The idea came to her during someone else’s talk.

Dr. Jennifer Wargo listened as a researcher described mice responding differently to certain cancer therapies depending on the composition of bacteria in their guts. She was intrigued.

“I got up to the microphone, and I was like, ‘This is amazing! Have you studied this in patients?’” Wargo recalled. “They said they were planning to do so, but I saw this as a huge opportunity. … So I ran back home to MD Anderson and wrote a protocol.”

Three years later, Wargo, from MD Anderson Cancer Center in Houston, published her team’s findings in Science. On Monday, STAT announced that Wargo’s work was the “editors’ pick” in STAT Madness 2018, a tournament in its second year that highlights top projects in science and medicine.

“It’s a total honor,” said Wargo, adding that a clinical trial is now in the works to see if changing someone’s gut bacteria while they are in the cancer treatment she studied will improve their response to the treatment.

STAT Madness is STAT’s annual two-pronged celebration of biomedical science. In one part of the contest, 64 research projects compete in a crowd-driven bracket, and for six rounds, voted for which projects will advance to the finals. More than 372,000 votes later, East Carolina University won, just beating Children’s National Health System in Washington. The second prong of the contest is the Editors’ Pick.

The judges, a team of editors and reporters at STAT, said picking a winner was difficult — each of the finalists brought something clever, elegant, and potentially game-changing to the table.

One of the finalists for the Editors’ Pick was also Children’s National Health System, whose researchers used facial recognition software to help identify genetic disorders typically found through blood testing. Another, from Ohio Sate University, involved a new technology that can reprogram cells in a number of different situations, including to help heal wounded tissue. Yet another, from the University of Rochester Medical Center, used a mouse-human chimera to find that glial cells — which support and insulate neurons — might play a role in the development of schizophrenia in children.

But the MD Anderson project emerged because of its potential, the judges said. Of nearly 150 entries submitted to the annual contest, Wargo’s stood out for its ingenuity, its progress in humans, and its potential impact on a class of cancer treatment that is a big target for drug development.

A clue into checkpoint inhibitor efficacy

The project has to do with a relatively new kind of cancer drug called checkpoint inhibitors, which are rapidly gaining in popularity — and scrutiny.

Our immune systems need to be able to tell the difference between our own cells and foreign invaders. Cancer cells should be seen as a threat, and eliminated, but they sometimes camouflage themselves as just another part of the body so that the immune system will let them proliferate unchecked. That’s where checkpoint inhibitors come in: They stop our immune systems from turning a blind eye to cells that seem safe. It might mean that some of our own tissues get damaged, but it also allows our bodies start aggressively attacking tumors.

“Immunotherapy has been amazing for cancer patients, but we’re still only succeeding in maybe 20 or 30 percent,” said Dr. Cynthia Sears, a professor of medicine at Johns Hopkins, who was not involved in Wargo’s research.

Wargo wanted to find out if the bacteria in patients’ bodies might help determine how well they responded to these kinds of drugs. Skin cancer has, in some sense, become the poster child for immunotherapy, because its high rate of genetic mutations means it can be especially susceptible to attack from immune cells. So Wargo’s study examined patients whose skin cancer had metastasized.

“In clinic, patients who were coming in, we would approach them and ask if they would sign on to this protocol, and then we would swab their cheek right in clinic,” said Wargo.

Then they handed them a kit so they could send stool samples back in the mail.

After that, the patients came in once every few weeks for infusions of immunotherapy. After months of this routine, the patients had their bodies scanned to assess the state of their tumors, and the researchers divided them into two groups: responders and nonresponders. If your tumors went away, shrank, or hadn’t grown after six months, you were considered a responder. If your tumors kept growing after treatment, you were considered a nonresponder.

The bacteria from the cheek swabs didn’t reveal much.

“There was no difference in the oral microbiome between responders and nonresponders,” said Vancheswaran “Deepak” Gopalakrishnan, a postdoc in Wargo’s lab and the first author on this paper.

But what was striking was that, as in mice, responders tended to have much higher diversity in the kinds of bacteria found in their guts. Wargo and her team even identified certain “beneficial” genera of bacteria that seemed to be associated with responding well to immunotherapy.

But so far, all they had were correlations. They wanted to start looking for causality. So they took some stool from responders and squirted it into the guts of mice with cancer — a liquid bacterial implant. “The tumors actually grew slower in these mice,” said Wargo. “Even before we started the treatment, their immune system was already rejecting the tumor.” The rodents who got feces from nonresponders didn’t fare nearly as well.

“This … is in my view is a massive validation of things that looked pretty strong two or three years ago, but now are being validated in humans,” said Dr. Paul Nghiem, a professor and the head of dermatology at University of Washington School of Medicine, who is not involved in Wargo’s work. “I think it’s safe to say that the cancer immunology community, collectively, is surprised by the magnitude of this effect.”

Changing patient microbiomes might improve response to therapy

Sears, who is an infectious disease expert, pointed out that different papers on this subject, first in mice and now in humans, have zeroed in on different kinds of bacteria as being potentially helpful or harmful when it comes to cancer immunotherapy. So changing the bacteria in a cancer patient’s gut to improve the treatment’s effects might be more complicated than it sounds.

Wargo agreed that that is an issue worth looking into, while other researchers point out that it might not even have to do with the microbes per se. “It may not be the bacteria themselves, whether you have bacterial species A or bacterial species B,” explained Dr. Sandip Patel, deputy director at the San Diego Center for Precision Immunotherapy, who also was not involved in this study. “These could, in fact, be surrogates for the actual products that bacteria make, which either stimulate or help resist cancer.”

While researchers feel it’s important to understand what is making the responders respond, they don’t necessarily think they should stall on trying to improve immunotherapy outcomes in patients by reshaping their microbial communities.

For the clinical trial, Wargo is teaming up with the Parker Institute for Cancer Immunotherapy and Seres Therapeutics to test this idea out in patients with advanced metastatic melanoma who are on checkpoint inhibitors. They’ll look to see if changes to gut bacteria will improve their response to the immunotherapy.

As Sears put it, “Doing something to enhance the power of these therapies is a very exciting prospect.”