A definitive blood test for post-infectious irritable bowel syndrome?

KEYWORDS:

• Irritable Bowel Syndrome
• blood test
• cytolethal distending toxin
• vinculin
• post-infectious

1. Introduction

Irritable bowel syndrome (IBS) is one of the most common conditions in gastroenterology [1], and is also one of the most challenging for clinicians. This is in part due to the lack of specific biological markers, as well as the lack of hallmark findings on endoscopy. In attempting to elucidate the pathophysiology of IBS, experts have investigated several avenues including the role of stress [2], alterations in the brain-gut axis [3], changes in bile acid metabolism [4], and altered intestinal motor function [5], amongst others. However, two bacterial concepts have proven to have great significance in IBS. These include the findings of altered small bowel microbiota in IBS [6], and that acute gastroenteritis can precipitate the development of IBS [7].

1.1. Burden of IBS

IBS represents a significant health burden. In clinical practice, up to 50% of visits to gastroenterologists may be due to IBS [8] and, based on the American Gastroenterological Association burden of illness assessment from over a decade ago, this condition represents a tremendous economic burden [9]. This was due in part to the lack of specific diagnostic tools, which would be of tremendous benefit to patients and would also substantially reduce the healthcare costs associated with IBS.

1.2. Diagnosing IBS

For more than three decades, the diagnosis of IBS has been based on a set of symptom criteria, most notably those developed by the Rome committee, including the recent Rome IV criteria [10]. There are number of significant problems with the symptom criteria-based approach to IBS. First, the Rome criteria are vague and nonspecific. This leads to a great number of false positives in gastrointestinal disease conditions – for example, more than 70% of patients with inflammatory bowel disease would meet the symptom criteria for IBS using the Rome criteria [11]. As a result, IBS has become a ‘diagnosis of exclusion’ and, at the discretion of the clinician, subjects may undergo endoscopy, celiac testing, blood panels, stool testing, radiological studies and other invasive, time consuming and expensive testing prior to being diagnosed with IBS [12].

Compounding this direct expense, the Rome criteria do not provide the patient with confidence in a diagnosis, and patients may be motivated to seek further opinions. This pattern of testing and doctor shopping substantiates data that it can take up to 6 years to secure a diagnosis of IBS [13].

1.3. Microbes and IBS

In the last 20 years, there has been a growing understanding that microbes play a role in IBS. It has been demonstrated that subjects with IBS have small intestinal bacterial overgrowth (SIBO) [14]. The concept that the pathophysiology of IBS has a microbial basis was further supported by the recent approval of rifaximin for diarrhea-predominant IBS (D-IBS) by the US FDA [15], but the root cause of these changes remained uncertain.

Concurrent to the evolving SIBO hypothesis in IBS, data emerged suggesting that acute gastroenteritis was a definitive precipitant of IBS. In fact, based on multiple large-scale prospective and longitudinal studies, gastroenteritis is the only proven root cause of IBS [16]. A recent study used a mathematical model to calculate the number of IBS patients in the USA that could result from acute gastroenteritis, using data from the CDC and the breadth of published work at the time on rates of IBS from gastroenteritis and remission rates [17]. Based on the study assumptions, a steady state would be reached at 15 years with a point prevalence rate for IBS of 9.1% of the US population derived solely from acute gastroenteritis. Although this study is a hypothetical analysis, it raises the possibility that the majority of IBS results from acute gastroenteritis.

To better understand the pathophysiology of IBS, an animal model was developed based on infection with Campylobacter jejuni 81–176 [18]. Rats were initially infected with C. jejuni, allowed to recover from the acute illness and then assessed 90 days after C. jejuni clearance from the stool. In this postinfectious period, rats were found to have developed altered stool form, increased rectal lymphocytes and SIBO when compared to controls, all of which are seen in human subjects with postinfectious IBS [18]. These findings, for the first time, unified the concepts of SIBO and the role of bacterial pathogens in postinfectious IBS.

2. Autoimmunity and IBS

This validated animal model has allowed the discovery of pathways linking gastroenteritis to the development of IBS. The four major bacterial pathogens that cause gastroenteritis all produce cytolethal distending toxin (Cdt), a heterotrimeric complex of three subunits, CdtA, CdtB, and CdtC, of which CdtB is the active toxin. In a follow-up study, rats infected with a mutant C. jejuni strain with an insertional deletion of CdtB were found to exhibit a near-normal postinfectious phenotype [19], suggesting that CdtB plays a key role in the pathophysiology of postinfectious IBS.

Further immunohistochemical analyses revealed a striking reduction in the numbers of the interstitial cells of Cajal (ICC) in the gastrointestinal tracts of C. jejuni-infected rats, particularly the deep muscular plexus [20]. This was significant as ICCs are required for normal intestinal motility [21], and are also reduced in human IBS subjects. Consistent with the fact that an absence or reduced frequency of these contractions is known to induce SIBO [5,21], the reduction in ICCs correlated with the development of SIBO in these rats [20]. Antibodies to CdtB were found to bind elements of the neuromuscular apparatus of the gut, including ICCs and myenteric ganglia [22]. This pattern of staining was also seen in small intestinal sections from rats that were never exposed to C. jejuni, and from human subjects who did not have IBS [22], suggesting that antibodies to the CdtB toxin were cross-reacting with a neuronal protein in the host gut. Through a series of studies, this was identified as the cytoskeletal protein vinculin [22]. Vinculin is a cytoplasmic actin-binding protein and an important component of focal adherens junctions (FAJs) and tight junctions, with roles in structural integrity and epithelial barrier formation [22]. Significantly, C. jejuni-infected rats exhibited reduced vinculin expression in the small intestinal wall, and levels of antibodies to CdtB correlated both with the degree of loss of vinculin and with the development of SIBO in these animals [22].

This series of studies now supports a potential pathophysiological pathway in the development of postinfectious IBS. Acute gastroenteritis with resulting systemic exposure to CdtB leads to positive seroconversion. Subsequently, homology between CdtB and vinculin leads to the generation of autoantibodies to vinculin in a subset of subjects and ongoing reductions in ICC numbers, amongst other effects. These in turn lead to changes in neuromuscular function, ultimately resulting in SIBO and the development of IBS. This forms the basis for IBS-D being an antibiotic-responsive disease.

3. Development of PI-IBS biomarkers

The results from the above-described studies suggested that serum biomarkers for IBS based on previous gastroenteritis as a causative agent could be feasible. To examine this, an enzyme-linked immunosorbent assay (ELISA) was developed to accurately measure the levels of anti-CdtB and anti-vinculin in humans [23]. These biomarkers were tested in a large-scale (180 center) study of subjects with IBS-D, comparing to subjects with celiac disease, Crohn’s disease, ulcerative colitis, and healthy controls [23]. These two biomarkers were found to be highly useful in identifying and distinguishing IBS-D from other causes of diarrhea. Based on a pretest probability of IBS in a gastroenterologist practice, the test characteristics produce a posttest probability of IBS of nearly 95% [23].

Although these biomarkers were highly specific for D-IBS, they had some limitations. Despite the greater than 90% specificity, the sensitivity was below 50% at the selected cutoff levels. These cutoffs were selected for optimization of the likelihood ratio and to improve test accuracy. In other words, a positive test meant there was a high likelihood that the patient had a definitive diagnosis, reducing the need for additional testing.

The second important issue was whether the test would only identify subjects with IBS-D, or would also identify subjects with IBS-M and IBS-C. In a second study, anti-CdtB and anti-vinculin titers were assessed in these three subject groups. The results demonstrated that anti-CdtB and anti-vinculin titers had a statistically significant negative gradient from IBS-D to IBS-M, then to IBS-C and healthy individuals (p < 0.001) [24]. There was no statistical difference between antibody levels in subjects with IBS-C as compared to healthy individuals. Therefore, these biomarkers can be used to identify subjects with IBS-D and IBS-M, but not IBS-C [24].

4. Implications of a plasma biomarker for IBS

The measurement of anti-CdtB and anti-vinculin antibodies as a means of diagnosing IBS represents an important breakthrough for a number of reasons. This represents the first test that makes IBS a ‘diagnosis of inclusion.’ A positive test is seen in 56% of subjects with IBS-D, giving these patients a positive diagnosis. These patients would no longer need the additional testing to rule out other diseases required by the Rome criteria. Furthermore, they could proceed to receiving appropriate therapy for their IBS more rapidly. Another aspect is that these biomarkers represent part of a pathophysiological mechanism in IBS. Previous biomarker tests were based on computer algorithms primarily involving blood markers that would rule in Crohn’s or ulcerative colitis [25]. This new test validates IBS as a disease (and not simply a condition) which anchors patients in their diagnosis.

One final important aspect of these biomarkers for diagnosing IBS-D (and recently IBS-M) is that this could lead to a significant reduction in healthcare costs. In a recent study, it was determined that more routine use of these markers could save more than $500 per patient in healthcare costs. Some patients have described up to $20,000 out-of-pocket after insurance costs for lists of normal tests such as colonoscopy [26].

In summary, the newly discovered plasma biomarkers anti-CdtB and anti-vinculin are diagnostic for IBS-D and now IBS-M (but not IBS-C). The use of this test has led to IBS becoming a diagnosis of inclusion based on a pathophysiological mechanism that includes postinfectious IBS and SIBO. While future studies may use this test to predict therapy, these have yet to be conducted. In the meantime, these biomarkers provide patients with validation of their disease and appear to demonstrate significant cost benefits.

Declaration of interest

M. Pimentel is a consultant for Synthetic Biologics. M. Pimentel and A. Rezaie are members of the advisory board for Commonwealth Laboratories and Valeant Pharmaceuticals. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

This paper was not funded.