SCUBA (Stratifying Crohn's Using Biomarker Assessment)
Research type
Research Study
Full title
Stratifying Crohn's Using Biomarker Assessment (SCUBA).
IRAS ID
246230
Contact name
Daniel Gaya
Contact email
Sponsor organisation
NHS Greater Glasgow and Clyde
Clinicaltrials.gov Identifier
Duration of Study in the UK
0 years, 11 months, 26 days
Research summary
Summary of Research
Crohn’s disease (CD) is a relapsing and remitting condition requiring lifelong monitoring.
Stool sampling for disease monitoring in inflammatory bowel disease (IBD) is non-invasive, cost-effective and acceptable to patients. Faecal calprotectin (FC) and quantitative Faecal Immunochemical Testing (qFIT) are both stool-based tests.
FC is a surrogate marker of neutrophil influx into the gut lumen. It accurately predicts mucosal healing (MH) at colonoscopy, and thus FC is already widely used in clinical practice in disease monitoring in CD patients.
qFIT, testing stool for haemoglobin, has been used in the Scottish Bowel Cancer Screening Programme since November 2017. A ‘negative’ qFIT is also a useful ‘rule-out’ test for significant colorectal pathology (including colorectal cancer, high-risk adenomas and IBD) in primary care. qFIT has been shown to predict MH in both CD and ulcerative colitis (UC), and has been used to predict relapse in patients with UC but not CD. There is no UK study to date comparing the ability of qFIT and FC to predict flare in CD.
qFIT is a cheaper, more stable test with a quicker turn-around time than FC. It is also less labour intensive for the lab.
CRP is a cheap and easily available biomarker but is insensitive, non-specific and inferior to FC at predicting relapse in CD. CRP can also be elevated by infective complications or other concomitant inflammatory disease which makes its interpretation difficult. New data from our group has suggested that low serum zinc has higher predictive accuracy at determining risk of future flare than both FC and CRP; we are unsure if this is due to higher faecal losses in ‘grumbling’ CD patients.
This observational, prospective cohort study will recruit patients with luminal (affecting small and/or large bowel) CD in clinical remission (i.e. asymptomatic). A stool sample will be used to check a qFIT and faecal zinc in addition to the routinely monitored FC. At the time of routine blood collection, an additional sample will be taken to check plasma zinc. CRP is already checked routinely.
Patients will be followed up for one year, or until flare/relapse - this information will be accessed remotely through electronic patient records.
The ability of qFIT, serum zinc and faecal zinc to predict relapse in CD will be compared to FC using area under the ROC curve (AUC).
Summary of Results
276 patients with Crohn’s disease (CD) were recruited. 232 provided at least one usable sample for the biomarkers being studied. All 232 included patients either reached the study end point or completed 12 months of follow-up.
The average age of participants was 34 years. Around one in ten were current smokers and a similar proportion had smoked in the past. Patients had been living with CD for a median of just over 10 years. More than a third had previously undergone surgery related to CD. Almost half had disease affecting both the small and large bowel and disease behaviour was mixed between inflammatory, stricturing and penetrating disease.
After one year of follow-up 46 patients had experienced a relapse (defined as need for escalation or change of CD related treatment due to suboptimal control, steroids, surgery or hospitalisation for CD related complication). The most frequent outcomes of relapse were a change in medical therapy or treatment with steroids.
There were no differences between patients who relapsed and those who did not in terms of age, sex, smoking history, disease duration, previous surgery, disease type or current CD medications. However, several blood and stool markers differed between the two groups. Patients who relapsed had higher faecal calprotectin (FCP), quantitative faecal immunochemical test (qFIT) and platelet counts at baseline. There were no differences in baseline C reactive protein, haemoglobin or albumin levels between the groups.
FCP and qFIT both showed good utility for prediction of relapse at 1 year with no significant difference in performance. Using both tests together did not meaningfully improve prediction compared with either test alone. An FCP level above 180 was associated with a higher chance of relapse, while a low level was reassuring and strongly associated with remaining well. Similar findings were seen for positive qFIT levels defined as above 9.
When taking into account patient characteristics and routine blood tests, both raised FCP and raised qFIT remained strong predictors of relapse.
When patients were grouped based on whether both faecal tests were raised, those with high levels of both markers relapsed much sooner than all other groups. Patients with low levels of both markers had the lowest risk of relapse during follow-up.
The accuracy of the stool tests varied depending on where disease was located. Both tests performed poorly in patients with disease limited to the end of the small bowel. In colonic disease, qFIT performed particularly well and FCP also showed good accuracy. In patients with disease affecting both the small and large bowel, FCP remained reliable while FIT was slightly less informative.
About one quarter of patients had low blood zinc levels at baseline. Zinc deficiency was not linked to disease location, disease behaviour or risk of relapse. There was no relationship between zinc levels measured in blood and those measured in stool. Neither blood nor stool zinc levels were useful in predicting relapse over one year in this cohort of patients.REC name
West of Scotland REC 4
REC reference
20/WS/0028
Date of REC Opinion
13 Feb 2020
REC opinion
Favourable Opinion