The Hottest New Cancer Drugs Depend on Gut Microbes

Immunotherapy doesn’t work for everyone, and a series of new studies might explain why.

Few recent developments have created more excitement in the world of cancer research than the rise of immunotherapy. After decades of frustration, scientists have finally found effective ways of turning the immune system against tumors, with spectacular results. Patients with kidney cancers and melanomas that had spread all over their bodies—diseases that would typically carry dire prognoses—have been cured. Immunotherapy, once a poster child for neglect and failure, has finally come of age.

The same could be said for the human microbiome. The trillions of bacteria and other microbes that share our bodies were ignored for centuries. But recent studies have shined a spotlight onto these multitudes, by showing how important they are, not least in their ability to train and calibrate our immune systems.

Today, these two trendy fields are colliding head-on. Working independently, two teams of scientists have shown that gut microbes—at least in mice—can dramatically affect the immune system's ability to deal with cancer. These microbes affect an individual's natural immunity to cancer, and how well they respond to immunotherapy drugs. And certain species of bacteria are especially potent at driving anti-tumor immunity, suggesting new ways of making new cancer drugs that much more potent.

There were signs of this already. In 2013, two groups of scientists showed that three cancer drugs can mobilize the immune system to kill tumors, but only in the presence of the right gut microbes. Laurence Zitvogel at the Gustave-Roussy Cancer Campus, who led one of the two teams, wanted to see if the same was true for a new wave of promising immunotherapy drugs called checkpoint inhibitors.

These drugs work by unshackling T-cells—a class of immune cells that seek and destroy potential threats, cancers included. T-cells are typically muzzled by “checkpoint proteins” on their surface. By inhibiting these checkpoints, drugs like ipilimumab can remove the muzzles, and unleash packs of slavering, destructive T-cells upon the tumors.

But not, it seems, without the right gut bacteria. When Marie Vetizou, a member of Zitvogel's team, treated mice with antibiotics, which raze the gut of its native microbes, ipilimumab lost its sting. It failed to mobilize the usual T-cell army, and failed to keep the rodents' tumors under control. Similarly, the wonder drug didn't work on germ-free rodents that had been raised in the absence of microbes. “It was black and white,” says Zitvogel, presenting her work at the NCRI Cancer Conference in Liverpool. “There’s no longer efficacy of ipilimumab in the absence of the gut microbiota.”

Some bacteria are more important than others. The team found that two species—Bacteroides thetaiotamicron (B-theta, one of the most thoroughly studied gut microbes) and Bacteroides fragilis (B-frag, a potent anti-inflammatory bug)—managed to re-sensitize the mice to ipilimumab. The bacteria weren't even necessary: The mice also started responding to the drug after receiving T-cells that specifically recognized B-frag.

Does any of this matter to actual cancer patients? To find out, the team studied 25 people with advanced melanoma, and showed that ipilimumab changes their microbiome. It shifts the microbes from one distinctive community to another, each dominated by different species of Bacteroides. When the team transplanted the “after” communities into germ-free mice, these rodents responded well to ipilimumab; the “before” communities had no such effect. And once again, B-frag was especially important: The greater its numbers, the smaller the tumors.

But why does ipilimumab stimulate the growth of microbes that, in turn, make it more effective—and how? “Many important questions remain before these findings can be translated to human trials, though,” says Sarkis Mazmanian from the California Institute of Technology, an immunologist who has studied B-frag in the context of autism. “But the link between this cancer immunotherapy and the microbiome is very exciting.”

Meanwhile, a second team led by Thomas Gajewski at the University of Chicago took a different approach and ended up pointing in the same general direction. They first noticed that two strains of lab mice reacted very differently to melanomas. Those purchased from Jackson Laboratory (JAX) developed less aggressive tumors than those from Taconic Biosciences (TAC), because they spontaneously mounted a more vigorous T-cell response.

When the team housed these rodents together, this difference vanished. Why? Because mice will readily to eat each other's poop, picking up gut microbes from their neighbors. By acquiring JAX microbes, the TAC mice became better at controlling cancer. And when student Ayelet Sivan and postdoc Leticia Corrales carried out these transfers deliberately, they found that the JAX stools were as effective at slowing the growth of tumors as an anti-PD-L1 antibody—another kind of checkpoint-inhibitor drug. The combination—feces and drug—was even more effective.

The two strains of mice differed in the numbers of more than 250 microbes, but only one genus—Bifidobacterium (or Bifs, for short)—was consistently associated with anti-tumor T-cells. The numbers of these microbes went up by more than 400 times when the TAC mice ate the cancer-controlling JAX stools. And by deliberately giving Bifs to mice, Gajewski's team boosted their ability to make anti-tumor T-cells, and to control their tumors.

Although both studies were mainly done in mice, they hint at why checkpoint inhibitors work so absurdly well in a minority of patients, but not in everyone. That lucky minority may harbor gut microbes that make these drugs that much more effective.

If that's true, it may be possible to sequence a patient's microbiome and work out how likely they are to respond to ipilimumab and other immunotherapies. If their odds are low, perhaps doctors could prescribe them with B-frag, Bifs, or other immune-stimulating microbes—or even a fecal transplant from a patient who is responding well to the drugs. “This may lead to greater drug efficacy within individuals, or benefits for a larger subset of people afflicted with certain cancers,” says Mazmanian.