Biomarkers for inflammation and surveillance strategies in inflammatory bowel disease

Chronic inflammation of the colonic mucosa, as observed in patients with inflammatory bowel disease (IBD), is associated with an increased risk of colorectal cancer (CRC). Endoscopic surveillance aimed at the detection of dysplasia and asymptomatic CRC is therefore recommended in order to mitigate this risk. In part 1 of this thesis we critically explored various aspects of endoscopic surveillance in patients with IBD. We showed that the longer surveillance intervals as recommended in the new British guidelines for surveillance are preferable, as they offer a significant reduction in colonoscopic workload of 22%. A potential drawback of a longer interval between two surveillance procedures is the increased risk of CRC developing between two surveillance colonoscopies, also known as interval CRC.We showed that the overall incidence of these interval CRCs was low among IBD patients undergoing regular endoscopic surveillance at the appropriate interval supporting the implementation of longer surveillance intervals. Traditionally, any visible lesion containing histologically confirmed dysplasia was known as a ”dysplasia associated lesion or mass” (DALM) which was associated with a high risk of CRC. Recent studies showed that the sub-group of lesions with an endoscopic appearance resembling that of sporadic adenomas (subcategorized as adenoma-like DALM) also found in patients without IBD have a lower risk of CRC.In contrast to these studies, we found that for IBD patients with an adenoma at baseline, the incidence of CRC during follow-up was higher compared to IBD patients without adenoma and non-IBD patients with an adenoma.In addition, we showed that there is a low agreement among endoscopists when classifying lesions as either adenoma-like DALM or non-adenoma-like DALM, suggesting that this classification is ineffective. Several studies have demonstrated that dye-based chromoendoscopy aimed at highlighting subtle changes in pit pattern can increase dysplasia detection rates compared to white-light endoscopy with random biopsies.Therefore, chromoendoscopy is now recommended as the preferred surveillance method in most international guidelines. In contrast to these previous studies,we showed that the dysplasia detection rate did not increase after the implementation of chromoendoscopy as the primary surveillance method. As IBD is characterized by relapsing episodes of inflammation, constant monitoring for the presence of inflammation is vital in evaluating response to treatment. In part 2 of this thesis we aimed to study the value of fecal calprotectin, a non-invasive marker for inflammation. We reported that fecal hemoglobin had a similar diagnostic accuracy for the detection of inflammation as calprotectin in patients with CD and UC. We also showed that low calprotectin levels could predict a sustained clinical remission during follow-up in IBD patients with complete mucosal healing. These results support the concept of trying to achieve “extreme deep remission” defined as clinical and endoscopic remission in combination with a low calprotectin level. Finally, we reported that low fecal calprotectin could accurately identify IBD patients without active inflammation in which CRC surveillance is most effective. Using a high cut-off level for calprotectin of 539 mg/kg, the number of ineffective procedures could be reduced from 14% to 3%.