Skip to Main Content

As cancer-killing CAR-T cells course through the body, they make occasional pit stops at the gut. What they do there — and which gut microbes they meet up with — could potentially change the prospects of these engineered immune cells.

Scientists are just beginning to tease apart the conditions that might lead to a gut microbiome to influence how well CAR-T therapy works.

“The gut microbiota is a little bit like the thermostat of the immune system,” said Andrea Facciabene, a cancer immunotherapy researcher at the University of Pennsylvania who co-authored two studies on the topic presented this month at the American Society of Hematology annual meeting. “This immune system needs the gut microbiota as a source of stimuli to be kept in equilibrium and be fit.”


Anything in a person’s environment — from their dinner choices to their roommates to their pets — can perturb the gut microbiome. Cancer patients risk more severe shocks to the microbiome as antibiotics, chemotherapy, radiation or other treatments can nuke the native, commensal bacteria that reside within us. In the past, researchers have found that antibiotic exposure in particular is associated with worse immunotherapy outcomes, said Melody Smith, a bone marrow transplant and cell therapy physician-scientist at Stanford University.

“But there’s no peer-reviewed literature assessing the association of the intestinal microbiome with CAR-T cell therapy,” said Smith, who set out to study the topic with collaborators at Memorial Sloan Kettering Cancer Center and the University of Pennsylvania, including Facciabene.


In one study, they looked at 228 patients with lymphoma or leukemia who received CAR-T therapy, 47 of whom received one of three specific antibiotics before starting CAR-T treatment. Those patients ended up with worse survival outcomes on average compared to the other 181 patients, Smith said.

Those antibiotics have a tendency to hose many commensal species of bacteria first. It might be that the patients who got these antibiotics did worse because the treatment was preferentially killing these “good” bugs. Antibiotic treatment reduces the diversity of the patients’ microbiomes, Smith added, which may have negative consequences for CAR-T therapy.

“The idea is basically that the loss of these bacteria is having an impact on [the immune system] that’s very important for a favorable response to CAR-T cells,” Smith said.

In another study by Facciabene and colleagues, one which Smith was not involved in, the researchers found that giving mice with blood cancers the antibiotic vancomycin had the opposite effect for CAR-T therapy. Mice given vancomycin ended up with significantly smaller tumors after CAR-T therapy compared to mice who were only given CAR-T.

Then Facciabene did a follow-up experiment, this time with the goal of modeling what might happen in a human gut, rather than a mouse one. So, he took fecal matter from four healthy human donors and transplanted it into mice. Some of those mice again got oral vancomycin and, once again, they did better on CAR-T therapy than mice who didn’t get the antibiotic — with one exception. Mice who received fecal matter from one particular donor did worse when they got the antibiotic compared to mice with the same human donor who didn’t get the antibiotic.

The seemingly contradictory results from Facciabene and Smith’s experiments underscore how early the research is — and how complicated microbiome and immune system interactions can be, said Matthew Frigault, an oncologist at Mass General Cancer Center who studies CAR-T therapy.

“You can end up with conflicting outcomes depending on what you’re starting with. Depending on the patient, the region they’re from, what your family has, what your environment provides, can all result in a different microbiome,” Frigault explained. “Maybe having diversity in the microbiome helps to modulate the immune system in a preferential way. Some antibiotics will kill certain bacteria preferentially, and then who knows what you’re left with?”

A microbiome made more homogenous by antibiotics might be interacting differently with the immune system and thus changing CAR-T cell outcomes, Frigault said. But even the same experiment may end up with different outcomes in different individuals, simply because everyone’s microbiome is unique to start. That complicates scientists’ ability to identify which antibiotics or which bacterial species are beneficial or deleterious to cancer immunotherapy.

“It would be nice if it was so nicely packaged. That the same taxa are being identified in different studies, and it’s a eureka kind of moment,” Smith said wistfully.

But Smith and Facciabene are embracing the challenge, and they have a list of hypotheses they want to test. Perhaps there’s some byproduct that certain, even unrelated, bacteria are making that’s affecting CAR-T cells or other cancer therapy. Or perhaps there’s a certain assemblage of bacteria that can promote immune therapy, if you can just get the right healthy donor to provide the fecal matter.

“What I’m trying to understand is how may I get the microbiome to go in the right way, the way that would be synergistic with the therapy we’re doing,” Facciabene said. “We are just starting to scratch the surface.”

Create a display name to comment

This name will appear with your comment