Probiotics punch up CAR-T therapy for solid tumors in mice

Probiotics could be the key to making universal chimeric antigen receptor T cell (CAR-T) therapies work for solid tumors, new research has suggested.  

In an article published Oct. 12 in Science, bioengineers from Columbia University reported that a system that harnessed engineered probiotics to guide CAR-T cells to breast tumors in mice decreased the size of the tumors and increased survival time. Another form of the treatment utilizing cytokine-producing bacteria made it even more effective at clearing out cancerous cells.  

The researchers are now seeking patents for the therapies—dubbed ProCAR and ProCombo, respectively—and are working on optimizing them for the clinic.    

“There is precedent to indicate that ProCAR could be safely achieved in humans with … efficacy at low bacterial doses,” Boston University scientist Wilson Wong, Ph.D., and Ph.D. student Eric Bressler, neither of whom was involved in the research, said in a perspective article in Science. “The study … is an important proof-of-concept for a potential approach to treating heterogeneous, immunologically cold, and poorly infiltrated solid tumors.” 

In standard CAR-T therapy, a patient’s T cells are genetically engineered to produce a chimeric antigen receptor that binds to specific antigens on cancer cells. But while the technology has made headway in blood cancers—as of October 2023, there are six FDA-approved CAR-T therapies for leukemia, lymphoma and multiple myeloma—it has so far failed to pan out for solid tumors. That’s partly because solid cancers express a more diverse, less specific mix of antigens than liquid ones do, making it a challenge to find a robust target that isn’t also found on healthy tissue. 

Some types of bacteria can offer a solution because they can infiltrate and colonize tumor tissue. The same strain used to develop the tuberculosis vaccine, Mycobacterium bovis, has been used to treat early-stage bladder cancer since the 1970s. More recently, scientists have tried using salmonella to get inside tumors and recruit immune cells to them, though clinical trials on this approach in the early 2000s failed to show efficacy.  

For ProCAR and ProCombo, the researchers used a nonpathogenic strain of E. coli, dubbed E. coli Nissle, which was engineered to release synthetic antigens, or tags, when it colonized the tumor issue; the ProCombo bacteria also released a cytokine that heightened T cell activation. These would be infused directly into the tumor, followed by CAR-T cells that were constructed with CARs that bound to the tags. With a clear target to bind to that wasn’t present on healthy tissue, the systems could bypass the problem of ambiguous antigens.  

E. coli Nissle was a strong contender because of its well-established safety profile, Rosa Vincent, a Ph.D. student at Columbia and first author of the study, told Fierce Biotech Research in an email. It’s also already being studied as an immunotherapy in clinical trials, and the research team had previously characterized it as a delivery vector for cancer therapies in breast cancer, colorectal cancer and liver metastasis, she said.  

“Beyond this, [the strain] is an extracellular bacterium that can deliver large quantities of payload directly within the tumor, and significantly boosts the activity of the CAR-T cells in our system,” Vincent wrote.  

After seeing success in cells, the researchers tested their therapies in mice. They studied around 10 different models, including mice that had been engineered to have human immune systems, with about five mice in each experimental group. As the main tumor type under investigation was triple-negative breast cancer, the mice were female, though future studies will include both males and females, Vincent said. The researchers also ran experiments using mice with leukemia, ovarian or colorectal cancer.  

ProCAR slowed the tumor growth in some mouse models of triple-negative breast cancer. That was also the case for some of the leukemia and colorectal cancer models, which also had increased survival. ProCombo—the therapy where the bacteria also expressed cytokines—was even better at suppressing tumor growth.  

The researchers noticed that besides directing the cells to the tumor tissue, the bacteria seemed to enhance the effectiveness of the CAR-T cells once they got there. Additionally, experiments revealed a possible mechanism: The bacteria stimulated Toll-like receptor, or TLR, pathways, which regulate various aspects of the innate immune response.  

“Together, these observations highlight the use of tumor-colonizing bacteria to mediate [universal] CAR-T cell activity while simultaneously boosting T cell effector functions through natural TLR stimulation,” the scientists wrote in their paper.  

While the E. coli Nissle is a non-pathogenic probiotic, the researchers were careful to assess whether it triggered an unwanted immune response or got into the bloodstream. They didn’t see the bacteria in any major off-target organs, nor did the mice experience noticeable immune-related side effects.  

However, that doesn’t mean the bacteria won’t trigger a reaction in humans, which are more sensitive to bacteria, Vincent noted. Approaches to minimize this will be a major focus of future research, she said.  

“We’re actively exploring various approaches to minimize the immunogenicity of [E. coli Nissle] including genetic attenuations that have enabled intravenous delivery of bacterial therapies in previous clinical trials, and genetic circuits that facilitate safer systemic delivery,” Vincent added.