The surfaces of our bodies are battlegrounds.The mixture of bacteria on our bodies contains the same breadth of evolutionary variety as humans and rice, and the battle between species for nutrients and space is a fierce one. On top of that, diverse phages – the viruses of bacteria – are parasites on this swarm, competing with each for bacterial prey. Together these warring factions make up the human microbiome.

Over the long term, the battle is a deadlock: the numbers of dominant bacteria and their less successful counterparts are surprisingly stable. For some of us, this stability is good news; for others, it can pose serious health risks. Studying the differences between our microbiomes, we can now predict and explain various health risks and outcomes, from obesity risks to how well we respond to therapeutic drugs. Therefore doctors and scientists are increasingly interested in reshaping our microbiomes, an ambition that's not merely human vanity or a fad for the health-conscious, but a real therapeutic strategy to mitigate real risks.

In recent months, researchers have teased apart links between what's happening on the microbial battleground and outcomes on a different front: cancer. These studies underscore how urgent it is that we develop precise tools to control the makeup of our microbiomes.

Our bodies’ normal functioning depends on balance, all the way down to the cellular level. In 1972, molecular biologist François Jacob poetically asserted that the "dream" of every cell is to become two cells. Yet each type of cell in our body (skin, neuron, immune cell) is under strict orders to strike a balance, between that dream of proliferation and restraint. It's when cells experience specific mutations in their DNA that they then break the rules, grow uncontrollably, and lead to a collection of diseases we call cancer. 

The term cancer is a catchall that masks a lot of biological diversity: different types of cancer arise from different cell types; patients with the same kind of cancer progress differently; and a single tumor in a single patient contains a mixture of different cancer cells. The human body is so complex that the number of ways it can fail is astronomically large.

Fortunately, diverse kinds of cancer can sometimes be traced back to single causes. In the past, this tracing has helped develop successful health interventions, from genetic counseling and testing to guide individualized treatment, to public health campaigns against known cancer-causing toxins, and vaccines to protect against cancer-causing viruses.

The arduous and necessary journey from observations to true causal relationships and treatments is the purview of experimental biologists who, in controlled environments, put theories to the test. Recently, two separate long-standing associations between the microbiome and cancer – one with therapy outcomes and one with underlying causes – have gained more causal evidence. Each points, tentatively, towards new therapeutic strategies that require manipulating gut bacteria in different ways.

Cancer treatments face a daunting task: how to discriminate between rogue cancer cells and healthy human ones. One successful strategy, termed immunotherapy, recruits our own immune system, which consists of multiple cell types already skilled in surveilling our bodies for abnormal cell activities. In immunotherapy, these surveillance cells are given a license to kill cancerous cells by way of drugs that activate them. For some melanoma patients these drugs are the proverbial silver bullet: their cancer is kept at bay for years without relapse. But this is the case for only one in four melanoma patients, while the rest have no response at all to treatment – a painful mystery that we now understand, and may be able to correct.

Suspects at the scene of the crime

Three stunning, back-to-back papers in the journal Science have converged on the role of particular microbiome members – dominant in some people and less abundant in others – in successful response to these drugs. The teams of researchers, from the University of Texas MD Anderson Cancer Center, the University of Chicago, and the Gustave Roussy Cancer Campus in France, have together painted a intriguing new picture. It appears that some species of bacteria in the gut play a unique role in “priming” immune cells for activation by chemotherapy drugs. While we know that gut bacteria play active roles in tuning immune cell activity both up and down in other contexts, this large effect in the context of cancer therapy has shocked many researchers. Encouragingly, the finding has prompted new clinical trials to test the combined effects of immunotherapy with a probiotic cocktail, containing the key bacterial species.

Probiotics refers to the direct transfer of live, beneficial bacteria into your system, a form of manipulating the microbiome by swarming the battlefield with fresh troops. But restoring balance sometimes demands manipulations in the opposite direction as well, as when a minor species becomes too dominant, abundant or strays out of its usual territory. It turns out these cases are not hard to find. Researchers in colon cancer biology have compiled a shortlist of otherwise common gut bacteria spotted at the scene of the cancerous crime (albeit with no discernible motive). Colon cancers arise from once-healthy cells in one of the most bacterially-dense areas of the body, where a layer of mucus separates bacteria from our body’s surface cells. Despite this barrier, particular bacterial species are known to home in on colorectal tumors, breaching the tumors’ exterior surfaces, and living comfortably inside. Suspicious behavior, to be sure, but only circumstantial.

Recently, a group from the Dana-Farber Cancer Institute demonstrated that these tumor-bacteria associations are stable over time, and actually benefit the tumor. When human colorectal tumors (including their associated bacteria) were repeatedly transferred to mice and between mice, the bacteria remained associated. Further, patients whose colorectal tumors had spread to the liver still had the bacteria in the spun-off metastases. To ask whether the presence of bacteria is beneficial to the tumor, Bullman and colleagues deployed one of the best tools we have for controlling bacteria: antibiotics. Giving antibiotics to the mice burdened with a human tumor helped deplete the cancer, leading to improved rates of survival, suggesting that these tumors are reliant on their bacterial hangers-on.

A clearer understanding of the bacterial cause now poses a problem: do we have the right tools with which to intervene?

Weapons of mass bacterial destruction

Antibiotics are one of the best tools we have for controlling bacteria – not because they’re so great, but because they’re basically all we have. The Dana-Farber group acknowledged that antibiotics are likely not the best solution for treating these colorectal tumors, because of well known and long-lasting negative effects of antibiotics on the microbiome. Although the species in our microbiome are diverse, most are sensitive to antibiotics, which have historically been designed to concentrate on what bacteria have in common – their most ancient, most fundamental genes. These genes don't vary much between species because they are evolutionarily constrained, meaning they cannot sustain mutations without killing the bacteria. They can't easily mutate to gain resistance against the antibiotic.

And yet bacteria can gain antibiotic resistance in other, often unanticipated ways. Unfortunately, the effect is that antibiotics can wipe our body's battlegrounds clean, clearing a field for those resistant mutants to emerge and thrive. In contrast, if a targeted approach were possible, like snipers from on high, the battle would continue on and naturally keep resistant mutants in the minority. Targeted approaches are therefore consistent with the goals of antibiotic stewardship, limiting the use of these extreme tactics, and making sure antibiotics work in the long run.

In some ways the challenge of targeting a specific bacterial species is as daunting as targeting cancer cells in a sea of healthy cells; selectively killing bad bacteria in a sea of good ones will require a strategic shift in how we design antibiotics. There are some cases where drugs can be specifically designed to target the unique characteristics of pathogenic or carcinogenic bacteria, and when possible, these may fit the bill.

A fundamentally different approach is to enlist the natural predators of bacteria: their phage (viruses). Most phage are naturally exclusive about their prey, targeting their particular nuances and ignoring other species of bacteria entirely. In addition to their targeted killing, phage have the benefit of numbers. Phage are by far the most abundant biological entity on Earth, and for every bacterial species there are many different phages that will happily kill it. Abundant and rapidly replicating, phage generate their own diversity; luckily for us, that diversity serves our shared goal: killing some bacteria.

In the pre-antibiotic era, phages were the secret ingredient in many bacterial infection treatments, but in Western medicine they lost popularity with the discovery of cheap, conceptually simple antibiotic drugs. But phage therapy is currently a demonstrably safe, effective, and is a rigorously tested practice in Eastern Europe, particularly in Georgia and Poland. Now phages are regaining Westerns clinicians' interest, especially as they face infections resistant to multiple drugs, and phage researchers have launched databases so that clinicians can easily employ phage as a last resort. Multi-drug resistant bacterial infection is perhaps the most clear-cut case for phage intervention, but the experiences of other clinical practices around the world, coupled with growing research of how cancers relate to our microbiomes, should encourage us to grab ahold of whatever tools we have.

The challenge ahead will be discriminating at finer and finer resolution between good and bad bacteria, in some cases within the same species. Evidence from high-risk colon cancer patients suggests that screening for gene-level distinctions, within bacterial species, could effectively guide treatment decisions. Precision medicine, the philosophy that therapies should be tailored to the specific circumstances, genetic and environmental, of each patient, is gaining influence around the world. As it does so, it's changing the direction of drug development and research into the battlegrounds of our bodies. The microbiome is a now a major player in what makes a patient's cancer unique to treat, and research and development should be rushing to fill the gap in our ability to effectively manipulate it for patient outcomes.