For years, researchers have tried to find biomarkers that could aid the diagnosis of irritable bowel syndrome, and point to its underlying causes. But will one test ever be enough?
It was a hunch almost two decades ago that launched Mark Pimentel's interest in irritable bowel syndrome (IBS). Back then, despite the prevalence of the condition, people with IBS were often told that their symptoms were all in their heads. But Pimentel, then a gastroenterology fellow at Cedars-Sinai, a medical center in Los Angeles, California, focused elsewhere in their bodies, on the rumblings in the abdomen: bloating, distension and flatulence — in short, gas.
Pimentel wondered whether bacteria fermenting food in the gut were producing an excess of gas. In 2000, along with his mentor, Henry Lin, Pimentel reported that almost 80% of people with IBS harboured an overgrowth of bacteria in their small intestines1. For about half of these patients, eradicating the extra bacteria improved symptoms. Around that time, other groups were also making bacterial connections — linking food poisoning to IBS. In the same vein, when Pimentel and his colleagues infected rats with the bacterium Campylobacter jejuni, a common cause of travellers' diarrhoea, they found that the animals developed not only IBS-like symptoms, but also similar bacterial overgrowth2. That's where things began to get interesting, says Pimentel, who now directs the gastrointestinal-motility programme at Cedars-Sinai. “From that point on, we said, 'OK, something in these bugs is causing IBS'.”
The culprit, Pimentel says, is cytolethal distending toxin B (CdtB), a key part of a toxin produced by pathogens that cause gastroenteritis. Last year, Pimentel reported that in a model of IBS, rats infected with C. jejuni formed antibodies to CdtB, which in turn triggered an autoimmune response against cells of the gut wall3. Furthermore, he found that blood-plasma levels of CdtB antibodies could help to distinguish people with diarrhoea-predominant IBS (IBS-D) from those with inflammatory bowel disease or coeliac disease and from healthy controls4. With these results in hand, Pimentel and diagnostics company Commonwealth Laboratories of Salem, Massachusetts, developed a blood test called IBSchek. “We've never had a biomarker that says, if you're positive, you have IBS,” Pimentel says. “This is that test.”
People with IBS need a diagnostic test. Patients get the label only when diseases that have similar symptoms, such as Crohn's disease, coeliac disease or colorectal cancer, are excluded. The current system can leave people in diagnostic limbo — the average time to diagnosis is 6.6 years5, during which they shuttle between doctors and undergo multiple tests, and still end up not quite being able to say that they have a disease (see page S112). Multiple attempts have been made to identify markers that could provide a positive diagnosis, but with limited success. In 2009, a team that included Anthony Lembo, a gastroenterologist at Beth Israel Deaconess Medical Center in Boston, Massachusetts, who went on to be a co-author of the 2015 CdtB-antibody clinical study, launched a 10-biomarker blood test for IBS with pharmaceutical company Prometheus Laboratories. But the test is no longer being sold. The main problem, says Lembo, was that the markers in the panel weren't related to what caused IBS.
Pimentel says that his test does just that — at least for IBS-D, which about one-third of people with the condition have. But not all gastroenterologists are convinced that IBSchek is as widely applicable as its developers say, or that it represents the underlying cause of the disorder. IBS is extremely heterogeneous, and so the test requires more validation in different groups of patients, critics say. Meanwhile, some researchers are focusing on this heterogeneity to identify the underlying physiological differences that could separate IBS into distinct, diagnosable conditions.
The term IBS was adopted in the 1960s, but even then clinicians struggled to define the condition (see page S102). “It had all these weird names and groupings, and there was no consensus about what these people experienced,” says Lembo. Today, the condition is diagnosed on the basis of symptoms using the Rome criteria (see page S107).
The conventional diagnostic categories are functional: constipation-predominant IBS (IBS-C), IBS-D and mixed IBS. “That distinction has nested itself into clinical practice because we have drugs that work specifically for IBS-D and IBS-C,” says Jan Tack, a gastroenterologist at the Univeristy of Leuven, Belgium. “But you could subdivide these people in many, many ways.” For example, he says, IBS cases could be distinguished by whether they begin acutely or appear gradually, whether the predominant issue is pain or bloating, or if psychosocial symptoms such as anxiety are present. Whether and how any of these groupings align with physiological changes is not clear, but they generally warrant different treatment, says Tack.
IBS symptoms seem to be driven by a tangle of factors that include irregularities of the central nervous system, problems in gut permeability, impairment of the immune system or a microbial imbalance in the gut (see page S104). Efforts to identify biomarkers that could track IBS have been largely unsuccessful. Immune markers to detect inflammation, serotonin as a way to reveal an out-of-sync connection between the gut and the brain, and non-molecular measures such as gut hypersensitivity and brain imaging have all been explored. Even within a subgroup such as IBS-D, there seems to be a lot of variation in what might be causing the symptoms (see page S116).
But researchers are now going beyond IBS-C and IBS–D, albeit slowly, and identifying subtypes that have a clear relationship with the underlying biology — and hence with biomarkers, says Magnus Simrén, a gastroenterologist at the University of Gothenburg in Sweden. For instance, IBS-D can arise when bile acid, which is produced by the liver to aid digestion, fails to be absorbed in the small intestine and ends up in the colon. This bile-acid malabsorption can be identified with the SeHCAT test, which is approved in Europe, but not in the United States. The condition responds well to targeted drugs, making a precise diagnosis worthwhile. “This is the way we can go today,” says Simrén, “to define subgroups of IBS patients with biomarkers.”
But Pimentel contends that his IBSchek is more widely relevant. Current estimates suggest that 10–20% of IBS-D cases follow gastroenteritis, but this is based on studies that asked people to recall past incidences of stomach upset. “Two days of mild food poisoning — you may not even remember that 20 years later, now that you have IBS,” Pimentel says. He suspects that 50–70% of IBS-D cases are post-infectious.
Many IBS researchers acknowledge that current estimates of post-infective IBS may be low, but they disagree that it could be as high as Pimentel's estimate. “To flip the coin completely and say that the majority must be post-infectious — we would need much better data for that,” says Tack.
Pimentel's hypothesis that an antibody generated by a bacterial toxin has autoimmune properties is sound, and well-supported by animal studies, Tack says. But he is not yet convinced that this antibody causes IBS, saying that there is no independent evidence that the participants in Pimentel's 2015 study really did have post-infectious IBS. The CdtB antibody, he says, might be a kind of bystander marker: present, but not necessarily causative. People with IBS-D may have increased gut permeability, a condition that allows all kinds of macromolecules to get into the submucosal tissue more easily and trigger an antibody response — no infection required. And Simrén notes that the IBSchek antibodies were found in people who had had the condition for a long time; whether the markers can be detected when patients first present with symptoms is still to be demonstrated.
Another issue with IBSchek is that the test is negative for about half of people who do have IBS, says Lin Chang, a gastroenterologist at the University of California, Los Angeles. What's more, she says, although IBSchek was able to differentiate IBS from ulcerative colitis, it was less successful with Crohn's disease and coeliac disease. Researchers such as Chang and Tack say that wider testing is needed before large-scale commercial deployment of IBSchek. Preferably, this should involve independent researchers working in a clinical setting with large groups of new patients presenting with diarrhoea and abdominal pain, and in geographically diverse populations. “We need to see more extensive data,” Tack says. “To tell a clinician now, 'Use this test and it will solve some of your problems.' I don't think we're there yet.”
Pimentel dismisses this concern. The antibody study, he points out, was conducted in patients from 180 medical centres and was larger than studies of many widely used diagnostics. Moreover, he notes, there are several studies under way in different populations, and at least one, conducted by military researchers, has corroborated his findings6.
Because IBS is such a heterogeneous condition, it will inevitably take large studies with thousands of people to understand the mechanisms of the condition — and hence how to diagnose it. In June, the US National Institute of Diabetes and Digestive and Kidney Diseases will lay out such a road map in their Functional Bowel Disorders Workshop, says Chang. The pan-European network GENIEUR was launched in 2012 to identify the factors that contribute to IBS.
Beate Niesler, a molecular geneticist at the University Hospital of Heidelberg, Germany, who heads GENIEUR along with Simrén, says that the fact that each IBS drug only helps a small number of people indicates that “there are obvious sub-groupings of patients”. There are several early clues that genetics and epigenetics could help to parse these groups. Associations with IBS have been found in genes related to serotonin signalling and immune, neuronal and intestinal barrier function. But most of the studies were small and the results may have been skewed by the different capabilities of each centre, Niesler says. She believes that GENIEUR, which spans 27 countries, can tease out and validate the various pathways by matching detailed data of clinical symptoms and phenotype with in-depth molecular analysis. The consortium has spent the past few years hammering out protocols, and now plans to conduct gene and RNA sequencing, analyse epigenetic markers and sequence the gut microbiome of 5,000 patients.
Almost all IBS researchers agree on the need to achieve a clearer understanding of the biological basis of this condition. This would allow patients to be given a firm diagnosis and, in turn, treatment. Pimentel is pushing towards this goal using the markers that he says reflect the underlying cause of IBS. “Now that we know that this antibody is so important,” he says, the question is, “if we can get rid of it, can we cure IBS?”
This article is part of the Nature Outlook: IBS