The immune system detects and fights the body’s own diseased cells. Yet, cancer cells fool it by activating checkpoints that impede its attack. 

One type of immunotherapy seeks to thwart this strategy by blocking receptors expressed by cancer cells that make them appear healthy. While these immune checkpoint inhibitor drugs have shown significant benefits for numerous patients, not everyone responds successfully to the treatment. The explanation for this variable response remains unclear, but multiple investigations point to the patient’s gut bacteria as one of the factors (1).

While exploring the mechanisms that gut microbes use to influence how patients respond to immunotherapy, a team led by Arlene Sharpe, Dennis Kasper, and Gordon Freeman at Harvard Medical School found a bacterial species that downregulates the expression of two molecules on immune cells in a mouse cancer model. One of these molecules, the repulsive guidance molecule B (RGMb), had not previously been associated with immunotherapy resistance, so the findings, published in Nature, may open the door to developing new strategies to complement existing treatments (2).

“It’s pretty novel that they’re showing that this molecule has an inhibitory effect on the [immune system],” said Rossanna Pezo, a medical oncologist at Sunnybrook Research Institute who did not participate in this study but does collaborate with one of the authors.  RGMb should definitely be further studied to see if it can be targeted to fight immunotherapy resistance, which is a really common problem, she added.

In both animal models and clinical settings, individuals who respond favorably to immunotherapy tend to harbor more of specific gut microbes compared to those who do not (3,4). “But the whole field was pretty much lacking in insightful mechanisms as to why the microbiome influenced immunotherapy,” said coauthor Kasper, a microbiologist and immunologist at Harvard Medical School. 

To fill this gap, Kasper and his colleagues used mouse models of colon cancer with responses to checkpoint inhibitors that depend on the presence of gut microbes. Specifically, in mice treated with these drugs, tumor volumes decreased significantly when the mice had a healthy human or mouse microbiome, but tumors remained the same size when germ-free or antibiotic-treated mice received the same drugs. 

Mice with microbes from healthy humans showed reduced expression of the programmed cell death 1 ligand 2 (PD-L2) in antigen-presenting cells, which are responsible for displaying foreign or potentially harmful proteins to T cells for recognition. PD-L2 binds to programmed death protein 1 (PD-1), a receptor found on the surface of T cells, and this interaction inhibits T cell function. Thus, higher levels of PD-L2 expression helps tumor cells evade the immune response. 

By using antibiotics that selectively target different bacterial groups in the healthy human microbiome alongside selective media and sequencing techniques, the researchers identified Coprobacillus cateniformis as one potential contributor to this beneficial effect. Furthermore, mice solely colonized with this species showed reduced expression of PD-L2. 

The whole field was pretty much lacking in insightful mechanisms as to why the microbiome influenced immunotherapy.
- Dennis Kasper, Harvard Medical School

The team observed that blocking both PD-1 and PD-L2 achieved better outcomes than blocking PD-1 alone, which suggested that the beneficial effect associated with PD-L2 bacteria-mediated downregulation might depend on a different receptor. Since members of the team had previously identified another PD-L2 ligand, RGMb, in the context of respiratory tolerance, they tested its involvement in the response to immunotherapy (5). 

They found that blocking either RGMb or its interaction with PD-L2, combined with checkpoint inhibitor drugs, resulted in a better antitumor response. Furthermore, gut microbiota appeared to modulate RGMb expression. It was six times higher in T cells of germ-free mice compared to mice harboring conventional healthy gut microbes.

The findings suggest that RGMb is a potential new immunotherapy target, said Roberta Zappasodi, a cancer immunologist at Weill Cornell Medicine who did not participate in this study. Zappasodi also sees value in profiling both RGMb and PD-L2 expression levels in patients showing immunotherapy resistance to assess whether that’s the basis of their treatment failure. 

Finally, coauthor Freeman emphasized, “This mechanism of resistance to immunotherapy [is likely not] the only mechanism, but this is one that can be targeted for therapeutic benefit.” 

References

1. Lu, Y. et al. Gut microbiota influence immunotherapy responses: mechanisms and therapeutic strategies. J Hematol Oncol  15, 47 (2022). 
2. Park, J.S. et al. Targeting PD-L2–RGMb overcomes microbiome-related immunotherapy resistance. Nature  617, 377-385 (2023).
3. Vétizou, M. et al. Anticancer immunotherapy by CTLA-4 blockade relies on the gut microbiota. Science  350, 1079-84 (2015). 
4. Baruch, E.N. et al. Fecal microbiota transplant promotes response in immunotherapy-refractory melanoma patients. Science  371, 602-609 (2021). 
5. Xiao, Y. et al. RGMb is a novel binding partner for PD-L2 and its engagement with PD-L2 promotes respiratory tolerance. J Exp Med  211, 943-59 (2014).