Bile acid diarrhoea is a chronic disease that impairs quality of life. One in 100 has the condition and many suffer from the disease without knowing. The current test is called SeHCAT and is expensive and time-consuming and is unavailable in many places, including the US. The disease is often misdiagnosed as irritable bowel syndrome and estimated one third of patients with irritable bowel syndrome of the mixed type and the diarrhoea predominant type suffer from bile acid diarrhoea without knowing.

A blood test called 7α-hydroxy-4-cholestene-3-one (C4) could make it much easier to diagnose bile acid diarrhoea.

To establish the new test, the results of both C4 and SeHCAT are compared with the treatment effect of the drug called colesevelam.

We invite patients who are referred for the SeHCAT test to participate in the trial. The SeHCAT test takes two days that are one week apart. The study patients register stool habits with a diary in the week between the SeHCAT visits. Based on the diary results, we screen for eligibility; e.g. a certain degree/severity of diarrhoea is required for participation. We treat eligible study patients (i.e those with diarrhoea) with either colesevelam or placebo (medicine without effect) that is randomly assigned. 170 study patients need to complete the treatment.

We aim to validate (ie. compare) both the C4-test and the SeHCAT test with the colesevelam treatment response as the reference.

ISSUE Bile Acid Diarrhoea (BAD) affects 1% of the general population and many people live with the disease and the impaired quality of life it causes without knowing it. The current scintigraphic seleno-homo-taurocholic acid retention test (SeHCAT) is time-consuming, expensive and causes a diagnostic bottleneck. the SeHCAT test is unavailable in many countries including the US. The biomarker 7α-hydroxy-4-cholesten-3-one (C4) could provide a cheap and available diagnostic option to surmount the issue of the many undiagnosed patients. The pivotal step for this is validation of the C4-test with the placebo-controlled treatment effect.

PURPOSE

1. To determine the efficacy and safety of colesevelam for treating Bile Acid Diarrhoea

   and

2. 2. To correlate both the current scintigraphic 75-Selenium conjugated Tauro-homocholic acid retention test (SeHCAT) and the biochemical marker of BAD 7alpha-hydroxy-4-cholesten-3-one (C4) with colesevelam treatment response, and to validate the cut-off values determined in the 'VABAD' trial

BACKGROUND Chronic diarrhoea affects 4-5% of the Western adult population and often results in specialist referral and endoscopies. Availability of the diagnostic SeHCAT test for BAD is limited, and BAD is often mistaken for the Irritable Bowel Syndrome (IBS). 32% of patients with two of the three IBS subtypes have BAD and the prevalence of BAD in the general population is thus approximately 1%.

DIAGNOSIS of Bile Acid Diarrhoea The current diagnostic test for BAD is the SeHCAT scintigraphic retention test using 75Selenium labelled Homo-tauro-cholic acid. The γ-emission is measured on day 1 and again on day 8. Seven-day SeHCAT retention rates of <5% represent severe BAD, <10% moderate BAD, and <15% mild BAD. Although not properly validated, in observational studies 70-80% of patients with SeHCAT <10% report a good response to cholestyramine. SeHCAT is unavailable in many countries including the the US, and clinicians instead rely on the patient's response to a therapeutic trial. However, the non-specific action and multiple side effects of the first line treatment with cholestyramine gives rise to multiple diagnostic pitfalls.

BIOMARKERS of Bile Acid Diarrhoea The recent increased insight into bile acid homeostasis and regulation has identified two biomarkers of BAD that may replace SeHCAT. 7α-hydroxy-4-cholestene-3-one (C4) is a bile acid precursor that correlates with bile acids synthesis rate. C4 is increased in BAD and has a sensitivity of 87% and specificity of 86% for diagnosing BAD defined by SeHCAT <10%. However, analysis of C4 is technically difficult and is thus primarily used at centres with special interest and no access to SeHCAT.

Another biomarker Fibroblast Growth Factor 19 (FGF19) is released to the portal circulation from the terminal ileum in response to bile acid absorption. In the liver, FGF19 inhibits bile acid synthesis enticing a negative feedback loop. Fasting values of FGF19 correlate inversely with C4 and with SeHCAT, but the sensitivity of 67% and specificity of 77% for detecting even severe BAD is clinically insufficient.

Performance of the biomarkers in a modern Danish population The investigators studied C4 and FGF19 in two pilot studies in selected populations and further established a Danish national collaboration with four university hospitals and prospectively recruited 71 subjects. The positive test cut-off C4 ≥ 46 ng/mL had 52% sensitivity and 91% specificity for detecting BAD defined by SeHCAT ≤ 10%.

The analysis of these results and the literature suggest that taking account of liver cirrhosis, hyperbilirubinemia, and use of alcohol and statins may increase the diagnostic yield of C4.

TREATMENT OF BILE ACID DIARRHOEA Sequestrants bind bile acids in the intestinal lumen and thus alleviate the diarrhoea symptoms. The first line therapy is cholestyramine, but this powder-formulation is distasteful with a low patient acceptability. Colesevelam tablets are much better tolerated, and therefore placebo-controlled studies of colesevelam for BAD are warranted.

A systematic review found cholestyramine to be effective in 70% of BAD patients. A placebo-controlled trial of colesevelam in patients with suspected type 1 BAD. Diarrhoea intention-to-treat remission rates were 67% for colesevelam and 27% for placebo (p=0.057), but extreme selection criteria slowed recruitment and the study ended prematurely. Of note, only one of 19 dropped out due to side effects to colesevelam. In summary, colesevelam and cholestyramine have similar response rates of 70%. For a clinical trial, colesevelam is superior due to effective blinding and a much lower dropout rate.

METHODS Effect parameters in chronic diarrhoea Symptoms reported by diary The Bristol stool form scale (BSFS) classifies stool from 1 (hard lumps) to 7 (completely watery). Clinical trials often use a seven-day diary. Patients with an organic cause of diarrhoea more often have ≥3 stools per day or a consistently watery stool consistency (BSF≥6) than those who have a functional cause of the diarrhoea.

Criteria for activity and remission of diarrhoea (Hjortswang's criteria) Hjortswang correlated chronic diarrhoea symptoms with impact on quality of life for patients with microscopic colitis, and validated the definition of clinical activity as ≥ 3 stools per day or ≥ 1 watery stool (BSF 6-7) and remission as <3 stools per day and < 1 watery stool as a mean of the seven-day diary.

MEDICINE Colesevelam The hospital pharmacy of the Capital Region over-encapsulates colesevelam tablets with Capsugel® DBcaps®

Placebo Matrix placebo tablets, over-encapsulated as the colesevelam tablets. No side effects are expected.

Dose: One, two or three capsules of 625mg twice daily. The dose is titrated by a central blinded study nurse without contact to the investigators.

POWER CALCULATION Assumed

• remission rate of 67% for colesevelam and 27% for placebo.
• two-sided α = 0.05 and β = 0.20 (ie. 80% power).
• 1:1 allocation. This gives 23 subjects with BAD in each arm (G.Power 3.1: z-test of two independent proportions). 27% of subjects in our prospective pilot study that would be eligible for the SINBAD study had SeHCAT <10%. Thus, (2 x 23 / 27%) = 170 subjects.

BIOCHEMISTRY C4 and bile acid species are analysed with liquid chromatography - tandem mass spectrometry. FGF19 is analysed with ELISA (R&D Systems, MN, USA).

The hospital laboratories of the participating centres perform the routine biochemical analyses.

STUDY PLAN Pre-screening

1. We send a written invitation to eligible patients who already are referred for SeHCAT
2. At the first SeHCAT visit, further information is provided and voluntary informed consent is obtained.

Study visit 1 - Day 1: start of baseline registration The investigator creates an electronic Case Record Form (EasyTrial) including medical and surgical history, medication, and physical status. Baseline blood analyses include ALT, ALP, bilirubin, amylase; and on indication HCG.

The investigator notifies the referring doctor.

The participants start Study Diary 1 (screening+baseline).

Study visit 2 - Day 8: Screening results - randomisation This is concurrent with the second SeHCAT visit. All subjects meet in the fasting for blood sampling. All subjects answer the baseline questionnaires (SHS, GSRS, SF36v2).

The screening diary is assessed:

Subjects without diarrhoea: cannot continue to randomisation.

1. SeHCAT results are registered once available
2. Result of the diagnostic work-up is registered at a six-month telephone follow-up.

Subjects with diarrhoea: continue to randomisation. The subject

• is given a uniquely numbered treatment package
• starts Study Diary 2
• is telephoned by the central study nurse for the planning of dose titration

Telephone consultation - Day 9 (treatment day 2):

The study nurse takes a history of adverse events and titres the dose

Telephone consultation - day 13 (± 1 day): End of Run-in-period As on day 9.

Ad hoc telephone consultations Any change in dose mandates a follow-up telephone consult after 2-3 days.

Study visit 3 - Day 21 (+ 1-3 days) Intervention-End The intervention has ended on day 20.

The investigator:

• Collects Study Diary 2
• Takes a history of AEs
• Collects surplus medication to assess compliance
• Distributes questionnaires
• Orders follow-up blood analysis
• Schedules study-end telephone consultation

Telephone consultation - Study day 26 (± 2 day): Clinical Study end AE registration continues three days after the end of treatment.

The investigator:

• Assess biochemical AEs

  o Informs of results

• Takes a history of AEs

• Takes action if needed

Six-month follow-up The questionnaires are redistributed. The patient's Medical file is checked and a telephone interview is done to note any conclusive diagnosis on the patients diarrhoea and (if any) anti-diarrhoea medication

Inclusion Criteria:

• Patient referred to Clinical Physiological/Nuclear Medicine departments for SeHCAT
• Suspected Bile acid diarrhoea
• Age > 18 years and under 80 years
• women of fertile age must use safe contraception during the treatment part of the study
• Ability to give informed consent after written and oral information in Danish language

Exclusion Criteria:

• Inflammatory bowel disease, including microscopic colitis

• Investigator assessed debilitating chronic disease e.g. World Health Organisation performance score 3-5

• Prior treatment with colesevelam

• Treatment with laxatives or anti-diarrhoeal drugs during the study

  • Except for stable dose the last four weeks of psyllium husk and opioids for pain

• Breastfeeding women

• Crucial medication that cannot be separated appropriately from colesevelam

  • i.e. taken one hour before or 4 hours after colesevelam

• Oral anticoagulation, both warfarin, and new oral anticoagulation

• Treatment with cyclosporine within two months

• Bowel obstruction (subileus or ileus)

• Biliary obstruction

• Short bowel syndrome

• Bowel ostomy

• Allergy to colesevelam or its constituents

• Allergy to placebo constituents (excluding lactose)

• Investigator assessed high risk of non-compliance

• If on statin/fibrate medication, unwilling to pause medication between study visits 1 and 2

An electronic Data Management Plan has been created on dmponline.dcc.ac.uk with details on management of raw data, refined data, metadata, organising, storing, backup, sharing, and curation of the data The biological material is saved anonymously and destroyed 15 years after study end.

Positive and negative results will be published in an international peer-reviewed journal. We regard the study protocol, consent form, questionnaires, statistical analysis plan, raw data, refined data, statistical code, and metadata as long-term valuable data to be shared after de-identification in the repository of the Danish National Archives to be available after the primary article is published. Data access is unrestricted and indefinite and with investigator assistance for the first 36 months.

De-identified data will be made available without restrictions after articles from the study have been published

Data access is unrestricted and indefinite and with investigator assistance for the first 36 months. Publications need proper citing.

Study subjects and all investigators are blinded. As effect and side effects of the active drug may demask the blinding, a central study nurse is the primary contact for the subjects during the intervention periode regarding everything except suspected/possible serious events.