Translating the Microbiome

Over the past two decades scientists have been working steadily to unravel the complex interplay of the microbiome, human health, and disease.

Our understanding of the microbiome—or the community of microorganisms that reside in the human body—has been greatly advanced thanks to the improvements in next-generation sequencing techniques, with methods such as amplicon and shotgun metagenomic sequencing allowing scientists to sequence microbial genetic material directly from a sample without having to culture and grow microbes in the lab. These methods have generated orders of magnitude more data than were available two decades ago. The explosion of data has led to a clearer understanding on the link between diet, the microbiome, and health; how early exposures shape future health outcomes; and even associations between disease states and microbiome composition.

“Researchers have made great strides over the past 15 years in trying to understand how organisms may be contributing to disease,” says Cynthia Sears, professor of medicine and oncology at Johns Hopkins University. She adds that for a subject like the microbiome, which is complex, diverse, and even varies between people, “that is a pretty short timeline of discovery.”

The next step is to bring this knowledge into the clinic. The race is on to develop prognostics, diagnostics, and even therapeutics with microbiome data. Many predict that next year will see the first FDA-approved microbiome-based therapeutics targeted against C. difficile, and others are rapidly developing diagnostics against a range of diseases, including cancer and irritable bowel syndrome. While there is much promise in the field, there are also considerable challenges.

Your microbes are what you eat

One of the most active areas of research in the microbiome field is the effort to better understand the link between nutrition, gut microbes, and health. In 2011, a seminal study was published by researchers from the University of Pennsylvania that showed diet was directly related to gut microbiota composition. Researchers found that people could be grouped into categories, called enterotypes, based on the species and relative proportions of microbes in their gut. In addition, they showed that a shift in diet, from low fat/high fiber to high fat/low fiber, led to a significant change in the composition of their gut microbes. In some cases the changes happened within as little as 24 hours.

A few years later, researchers from the University of Pittsburgh furthered this line of research linking diet to colon cancer risk in African Americans. They wanted to understand why African Americans living in the U.S. had higher rates of colon cancer than Africans living in rural South Africa despite both groups being of African descent. In the study, the researchers replaced high fiber/low fat South African diet with the high fat/low fiber American diet. After only two weeks, they observed marked changes in inflammation and metabolic risk factors in the South African cohort. One change in particular was worrisome. The South African group seemed to lose a specific type of gut microbe that synthesizes the chemical butyrate, known to break down fiber and confer a lower risk of colon cancer. The butyrate levels in the South Africans plunged by as much as 2.5 times. The results demonstrated how changes in diet could impact disease risk through the microbiome.

“These studies showed that changes brought on by diet can be quite quick and have a big impact on your metabolism,” says Sears. “We all have the capacity to improve health through our diet,” she adds.

Another active area of investigation is early exposure to germs and development of disease later in life. The incidence of asthma and allergies has risen sharply over the past several decades as the population has shifted to an industrialized lifestyle, with better sanitation and medicines like antibiotics. Some postulate that “cleaner” living has reduced childrens’ exposure to germs and thus negatively impacted our immune systems.

Studies of children in Amish and Hutterite communities provide some of the best evidence for this idea. The two groups share a genetic ancestry, but the Amish have continued their traditional farming practices while the Hutterites have adopted a more modern and industrialized lifestyle. It has long been known that children who grow up on traditional farms, like the Amish, have a much lower incidence of asthma than the general population. In fact, asthma is four times lower among the Amish compared to their Hutterite counterparts. In addition, high concentrations of bacterial endotoxins, which are inflammation-producing sugars shed from bacterial cells, have been found in Amish households. And in studies using mice, inhalation of these endotoxins seemed sufficient to ward off asthma and allergies. Now, next-generation sequencing studies in urban American children have found an explanation in the microbiome.

In a study published in 2017, researchers collected and analyzed stool samples from 300 newborns in Detroit, Michigan, and followed the babies for several years. They were looking for signatures in the microbiome that could be linked to later developing asthma or allergies. After four years of follow up they found three distinct microbial signatures, each incurring a different risk profile for later developing asthma. The highest risk group had a low abundance of certain bacterial strains, a high abundance of fungi, and microbial metabolic outputs that cause inflammation.

Studies like this are just the beginning. In January of 2020, researchers at the Chinese University of Hong Kong announced one of the largest studies to date that will investigate these links. They launched a new study that aims to recruit 100,000 mother-baby pairs in the Greater Bay area in China to be followed for more than seven years. They hope to better characterize healthy and disease-promoting microbiomes. They also hope to develop biomarkers and identify risk factors for diseases like irritable bowel syndrome, obesity, and other immune system-related conditions.

Predictive power

One very active area of investigation is the prognostic value of the microbiome for new personalized cancer treatments like immunotherapies. Despite being one of the most promising new approaches to tackle cancer, immunotherapies don’t work for everyone. Factors in the cancer’s genome, like the mutation burden, or even the host’s immune system can interfere with the success of the treatments, and many scientists are looking for ways to expand the population for which immunotherapies are effective.

Various studies have shown that the composition of the microbiome can have an impact on the efficacy of treatment, and there are dozens of clinical trials recruiting patients now to look for a link between microbiome and immunotherapy. Many hope that altering the microbiome concentration could be a way to improve current therapies. Researchers at the University Health Network in Toronto, for example, have launched a clinical trial that is evaluating the efficacy of gut microbes taken from healthy donors and given to cancer patients. The microbes are delivered in oral pill form, which is a safer alternative to fecal microbiota transplants alongside immunotherapy. If the trial is successful, it could mean that many more patients will be eligible for immunotherapy. It could also indicate that themicrobiome could be a tool for pronostic applications.

In 2020, research was published in the journal Nature showing the cancer diagnostic potential of the microbiome. A group of researchers from University of California San Diego made a surprising discovery: blood and tissue harbor their own communities of microbes. The finding overturned a long held belief that the blood is a sterile environment, consisting only of blood cells, platelets, and plasma. Furthermore, they showed that the specific composition of microbes could be linked to the tumor type a patient may harbor.

The team, led by Rob Knight from the University of California San Diego, collected tens of thousands of tumor and blood samples from patients with 33 different types of cancer. Using next-generation sequencing and machine-learning algorithms to analyze the DNA, they found that they could distinguish between healthy individuals and those with cancer just by looking at the composition of the microbial DNA in the blood. Furthermore, using the DNA signatures in the blood they could also tell the difference between tumor types. Some relationships were expected, such as the presence of human papillomavirus (HPV) and cervical, head, and neck cancers and Helicobacter pylori and stomach cancer, but the specificity of the community found beside one of these microbes were totally new associations.

“It was a fantastic discovery,” according to Sandrine Miller-Montgomery, co-author of the study and CEO of Micronoma, a microbiome-driven liquid biopsy start-up that was launched by the team who published the report. “The microbiome in the breast tissue was completely different from the one in the colon tissue in the same way that the microbiome in the ocean differs from the microbiome in a prairie, and this was transpiring in the blood of the patients, making it possible to develop a minimally invasive clinical diagnostic tool,” she said. Sears calls the paper “a tour-de-force in trying to define the microbiome of different cancer types.”

And research from other groups supports the findings as well. In January of 2022, the microbiome was added to the famous list of cancer hallmarks. The Hallmarks of Cancer, originally published in 2000, was meant to highlight the various things that need to happen for a normal cell to turn into a cancer cell. Other hallmarks have to do with genetic mutations, tissue invasion, and metastasis, for example. The addition of the microbiome to this list is a sign that has come of age as a significant contributor to disease.

And now the race is on to bring these advances into the clinic. In 2019, Knight, Miller-Mongomery, and their colleagues founded a startup called Micronoma, which hopes to develop blood-based assays that detect cancer microbial signatures in the blood. First up in their pipeline is a blood test for lung cancer. Lung cancer is the leading cause of cancer deaths and one of the reasons is that it is often caught too late, when it is difficult to treat. Micronoma hopes to develop a blood-based assay for early detection for this type of cancer and then later move on to others.

“I think it is a really cool approach,” says Sven Borchmann, cancer researcher at the University of Cologne in Germany. Borchmann’s lab focuses on developing liquid biopsies. Borchmann published a study in 2021 detailing new links between bacteria, viruses, and cancer. Borchmann found many microbial species that were not known to colonize humans in his cohort of cancer patients, for example those with chronic lymphocytic leukemia.

Borchmann says there are still many challenges in using this data as a diagnostic. Chief among them are contamination and sensitivity. The test has to be extremely sensitive to pick up the minuscule amounts of circulating genetic material in the blood. He thinks instead that it would be a good tool to measure remission or even to match tumor types to personalized therapies.

Sears also worries about sensitivity with such tests and thinks this work and others like it are a very positive step forward but there are still many challenges in bringing microbiome-based assays to the clinic. In order to push the field forward, Sears advocates for larger and more rigorous studies. She says that studies today suffer from poor study designs that gather data at a single point in time, rather than through many time points over many years. She also says that the sample collection techniques vary greatly from study to study. In addition, DNA sequencing methods, sample storage procedures, and DNA extraction protocols are not standardized and often not fully reported in the studies. These discrepancies and omissions make it difficult to interpret and validate results across studies, she says.

Researchers at Harvard University showed just how variable the research could be when they sent microbiome samples to 15 different high-quality laboratories for analysis and found major discrepancies in results. They blamed the variation on the differing DNA extraction, sample prep, and bioinformatics techniques used. “The bottom line is that very good labs can analyze samples and come up with different results because of their processes”, she says. In order to correct these issues, Sears and others advocate for more transparency and standardization across the field.

Adding rigor and reproducibility

Late last year, a consortium of microbiome researchers published a paper in Nature that outlined a checklist of reporting guidelines for microbiome research, called the STORMS checklist. It is a 17-item list that hopes to encourage more transparency and peer-review in the publishing of these papers.

Levi Waldron, epidemiologist at the CUNY Graduate School of Public Health and Health Policy in New York and senior author of the paper, says that microbiome research lacks standardization and oftentimes the reporting on the methods and analysis vary from one lab to another. “The field is still so heterogenous that it can be hard to find the most basic things in a paper,” he says. Waldron says that reviewing the papers is difficult because it is hard to check all the details that should be reported to make it a replicable paper. Waldron co-founded a consortium of researchers in 2020 that are working to standardize microbiome research in hopes of making the studies easier to evaluate and replicate.

“I am very supportive of that paper,” says Sears. She thinks that in order for the microbiome to make its way into the clinic it will need to tighten its reporting. She compares it to guidelines used in clinical trials called Consolidated Standards of Reporting Trials or CONSORT. The guidelines are a 25-item checklist of reporting guidelines that requires researchers to report on how the trial was designed, analyzed, and interpreted. Much like the STORMS checklist, it is meant to ensure transparency and it facilitate peer review.

The checklist is meant to “encourage good habits in the field” and urges scientists to report on the DNA extraction kit that was used, for example, and also on statistical methods used to avoid confounding factors. The STORMS consortium is actively working with journals to enact these guidelines for all studies. Some of the Nature journals have adopted these guidelines and others are considering them.

Despite the lack of guidelines and standardization, microbiome-based therapeutics are pushing ahead. There are at least three companies developing microbiome-based treatments for the intractable intestinal infection, C. difficile. The infection, most often occurring after use of antibiotics in hospitalized patients, affects roughly half a million people every year and is extremely difficult to treat. Antibiotics aren’t very effective against the infection and many patients end up with chronic and severe illness.

Fecal microbiota transplants have been very effective in fighting C. difficile in patients who don’t respond to other treatments. And now, the FDA allows fecal transplants for patients with recurrent C. difficile but the method still comes with many limitations, such as difficulty in quality control.

In September, Seres Therapeutics completed its submission process for an application to the FDA for a product called SER-109, which is an oral therapeutic that consists of purified Firmicutes spores. Firmicutes is a butyrate-producing bacteria that resides in the gut and is known as a “healthy microbe.” The product is designed to repair the disrupted microbiome in patients with C. difficile.

“C. diff takes advantage of the disrupted microbiome to cause disease,” according to Casey Theriot, C. difficile researcher and associate professor at North Carolina State University. “If you are going to create a therapeutic that targets and leverages the microbiome, I think you start here.” she says.

Theriot is hopeful these therapies will take off but also cautions that there is still much work to do. She cautions that next-generation sequencing tools are great at telling scientists which microbes are present, but it doesn’t necessarily tell you what they are doing. “If you’re looking at the metagenome of a gut, it’ll just tell you which bacteria and genes are present, but it won’t tell you if which genes will eventually be made into proteins and or molecules,” she says.

Theriot thinks it’s critical to use metabolomics, in addition to sequencing platforms, in order to understand the mechanism of how certain bacteria, or lack thereof, contributes to disease.

Tools like this, she says, will help scientists parse the complicated interplay between bacterial composition, metabolism, and disease and aid the development of therapeutics in more complicated diseases like irritable bowel syndrome.

Overcoming these challenges will be essential for the microbiome to come of age in clinical applications, but the field is closer than ever to realizing its potential. “I think we are at that moment, where perhaps we can make all this information much more usable in the world of medicine, if we just take a little more time and care to document exactly what we’re doing,” says Sears.

Monique Brouillette is a freelance journalist who covers science and health.