Ablation of the microbiome with antibiotics upregulated immune checkpoint protein PD-1 in pancreatic ductal adenocarcinoma (PDA) mouse models, reported a study in April (Cancer Discov 2018;8(4):403-416). Because of the study results, the research team is now planning a small clinical trial to determine whether treating pancreatic cancer patients with an antibiotic regimen increases sensitivity to immune checkpoint inhibitor anti-PD-1.

Recent studies have indicated the microbiome modulates tumor response to immunotherapies, such as anti-PD-1 in patients with melanoma (Science 2018;359:97-103, Science 2018;359:104-108). However, the relationship between the microbiome and responses in pancreatic cancer has not been well-established. A study by Geller, et al, first showed a possible relationship between chemotherapeutic agent gemcitabine and the microbiome when they found bacteria in the Gammaproteobacteria class could metabolize gemcitabine into its inactive form (Science 2017;357:1156-1160). The ability to metabolize gemcitabine suggests the microbiome may be responsible for the resistance to gemcitabine therapy that develops in pancreatic cancer patients.

Demand for Better Therapies

PDA is the most common type of pancreatic cancer and the third most common cause of cancer-related death in the U.S.; only 8 percent of pancreatic cancer patients survive past 5 years. Despite having clinical activity in several solid tumor types, immune checkpoint inhibitors like anti-PD-1 have been unsuccessful in treating this aggressive disease.

"We know the immunotherapy results in pancreatic cancer have been very limited," said Florencia McAllister, MD, Assistant Professor in the Department of Gastrointestinal Medical Oncology at The University of Texas MD Anderson Cancer Center, Houston. Florencia and colleagues wrote a related editorial about the study (Cancer Discov 2018;8(4):386-388).

Establishing a Link

The main reason pancreatic cancer does not respond to immunotherapy is the pancreatic tumor microenvironment is highly immunosuppressive. So, to identify the microbiome's role in immunotherapy response and immunosuppression in PDA, the researchers conducted a series of mouse studies.

First, they showed in healthy mice that the fluorescently labeled bacteria migrate from the gut to the pancreas, indicating that the gut microbiome can influence the pancreatic microenvironment. An oral regimen of antibiotics was given to ablate gut bacteria and the outcome was cutting tumor burden by about half in mice, suggesting bacteria may promote the progression of pancreatic cancer. In an additional experiment, gut bacteria in cancerous mice were again ablated with antibiotics and then repopulated with feces from either healthy or cancerous mice. The mice repopulated with feces from cancerous mice had accelerated tumor growth whereas those repopulated with feces from healthy mice did not.

To determine the microbiome's effect on immunity, the microbiome was ablated in mice and immune cells were measured. The result was an increase in intratumoral T cells, a reduction in myeloid-derived suppressor cells, and increased expression of PD-1 in cancerous mice, suggesting that bacteria contribute to immunosuppression in the tumor microenvironment. Administration of oral antibiotics and anti-PD-1 therapy enhanced CD4+ and CD8+ T-cell activation in mice, which are associated with improved survival in human disease.

To characterize the human pancreatic microbiome, researchers sequenced the PDA tumors from 12 patients. Overall, the bacterial composition of patient PDA tumor samples was more diverse than that of healthy pancreas samples. They found 13 phyla, the most common being Proteobacteria (45%), Bacteroidetes (31%), and Firmicutes (22%), which were also found in all samples. Actinobacteria (1%) and genera Pseudomonas and Elizabethkingia were also found in all samples.

The researchers also acquired fecal samples from 32 patients with PDA and 31 healthy individuals and compared the bacterial composition between both groups. They found that the bacterial composition of fecal samples from PDA patients was more diverse than samples from healthy individuals. Specifically, Proteobacteria, Synergistetes, and Euryarchaeota were significantly more abundant in patients with PDA than in healthy individuals. Proteobacteria, which are gram-negative, made up approximately 8 percent of bacteria in the gut yet nearly 50 percent in the pancreas microbiome.

Researchers analyzed the gut microbiome in PDA patients and found significant differences in bacterial composition between PDA patients with stage I or II and stage IV, suggesting bacterial communities differ between early and advanced PDA.

"What we have found, and what others have found, is that the microbiome can modulate immunity," said senior author George Miller, MD, HL Pachter Professor in the Departments of Surgery and Cell Biology at New York University School of Medicine. "We also show that in pancreas cancer if you ablate certain bacteria you get activation of the macrophages and T cells in the tumor, and as a consequence, T cells upregulate PD-1."

He added, "Immunotherapy can potentially work in diseases that it hasn't worked if we address the big elephant in the room, and the elephant is the microbiome.

"In conclusion, the experimental data from Pushalkar and colleagues provide compelling preclinical evidence for the role of the gut and tumor microbiota in the local and systemic activation of the immune system, which not only affects the natural history of the tumor, but also converts an immunotherapy-refractory tumor into a significantly more responsive one," wrote McAllister and authors in the related editorial.

Cautionary Notes & Next Steps

The ablation of the microbiome via antibiotics may increase immunotherapy sensitivity for mice and possibly patients too with PDA, but that concept does not necessarily apply to other tumor types.

"In some cases, it has been reported that the use of antibiotics may decrease the response, so it certainly has to be taken with caution in terms of which tumor you are talking about," said McAllister. She does not see antibiotics as being routinely administered with immune checkpoint inhibitors; the practice may apply only to pancreatic cancer.

McAllister said future studies will have to uncover which bacteria in particular are responsible for immunosuppression in pancreatic cancer so that those can be ablated instead of the whole microbiome. "In patients you can't just give antibiotics, because we know these patients have a lot of other issues, and they sometimes even need antibiotics."

The research team is now planning to launch a proof-of-concept clinical trial to see if the findings in mice translate to humans. Miller said the trial will enroll about 20 patients with resectable pancreatic cancer who will be treated with antibiotics ciprofloxacin and metronidazole followed by an anti-PD-1. The antibiotic course will last about 5 weeks and be given before surgery.

"We don't think that this is necessarily a cure for pancreas cancer," said Miller about the therapy combination. "What we think is that we will see some signal of T-cell activation in humans when we treat with antibiotics."

Christina Bennett is a contributing writer.