Learning points

• Understanding the relevance of colonoscopic surveillance in IBD patients
• Surveillance interval recommendations
• Methods for colonoscopic surveillance
• Management of dysplasia
• The future of colonoscopic surveillance

Key words

1. Colorectal Cancer
2. Inflammatory Bowel Disease
3. Dysplasia
4. Surveillance
5. Image enhanced endoscopy

Abbreviations

CRC – colorectal cancer, IBD – inflammatory bowel disease, UC – ulcerative colitis, PSC – primary sclerosing cholangitis, HD – high definition (colonoscopy), SD – standard definition (colonoscopy), WLE – white light endoscopy, IEE – image enhanced endoscopy, CE – chromoendoscopy, VCE – virtual chromoendoscopy,  EMR – endoscopic mucosal resection, ESD – endoscopic submucosal dissection

Background

Colorectal cancer (CRC) is the fourth most common cancer in the UK. When diagnosed at an early stage, one-year survival is extremely favourable at 98% compared with 44% in those diagnosed late¹. Patients with inflammatory bowel disease (IBD) are at an increased risk of CRC which accounts for 10%-15% of deaths in IBD². Colonoscopic surveillance can detect precursor lesions or early neoplasia, removal of which may prevent development of CRC³.

The British Society of Gastroenterology (BSG) has published guidelines on stratifying IBD patients’ CRC risk⁴ . Variation in colonoscopic surveillance strategies exists and regular endoscopic surveillance places a significant burden on services and more importantly on patients⁵. The future of IBD-CRC surveillance may need to incorporate non-endoscopic methods to assess risk to ensure the right patients are getting the right procedure at the right time.

Why should IBD patients undergo colonoscopic surveillance?

IBD patients are at an increased risk of developing CRC; recent large population studies showed overall adjusted hazard ratios of 1.66 (95% CI 1.57–1.76) and 1.40 [95% CI 1.27, 1.53] in UC and Crohn’s patients respectively compared with the general population^6,7 . CRC is likely to be detected earlier in IBD patients undergoing regular surveillance translating into a better prognosis³.

The key factor driving carcinogenesis in IBD is chronic mucosal inflammation. IBD-CRC progresses through the dysplasia-carcinoma sequence, although unlike sporadic CRC the transition from normal tissue to low and high-grade dysplasia and finally CRC is not always stepwise² . Dysplasia is often detected in flat mucosal abnormalities that are less easily vualised than adenomatous polyps (see examples figure 1). In the UK, the 3-year post colonoscopy CRC (PCCRC) rate is more than three times higher in IBD patients than non-IBD patients⁸. Despite improvements in PCCRC overall, this key performance indicator has remained stubbornly high for IBD patients. This finding is in keeping both with the difficulty in detecting IBD-dysplasia and the accelerated carcinogenesis of IBD-CRC⁹.

These features suggest that a more tailored approach to surveillance in IBD-CRC is required.

When should surveillance be carried out?

Recognising risk factors for CRC in IBD patients is crucial in guiding surveillance recommendations. Disease severity and extent have consistently been shown to be independent risk factors for CRC in UC; extensive disease carries the highest risk of CRC, and those with proctitis have no increased risk compared with the general population¹⁰. Other risk factors include young age at onset, primary sclerosing cholangitis (PSC), colonic strictures, pseudo-polyposis, previous IBD-associated dysplasia and a family history of CRC⁴.

BSG guidelines⁴ risk-stratify into sub-groups which determines surveillance intervals (figure 2). Age, sex and certain life-style factors that may also play a role in the risk of CRC development¹¹ are not included in this risk assessment.

Surveillance procedures should ideally be undertaken during periods of quiescent disease with effective bowel cleansing regimens to minimise the risk of a missed cancer diagnosis^8,12.

How should surveillance colonoscopy be conducted?

Historically dysplasia was thought to be macroscopically invisible. As a result, guidelines recommended an extensive biopsy protocol in which multiple random biopsies from each colonic segment. With improved endoscopic resolution, most cases of dysplasia in IBD are now visible endoscopically¹³. The SCENIC consensus guidelines published in 2015 strongly recommend high-definition (HD) over standard-definition (SD) colonoscopy when using white light. The addition of image-enhanced endoscopy, particular in the era of HD-colonoscopy, has proven more controversial¹⁴.

Image-enhanced endoscopy (IEE) has been extensively explored in IBD surveillance due to the predominance of subtler, flat lesions. Chromoendoscopy (CE) utilises a blue dye (methylene blue or indigo carmine) while virtual chromoendoscopy (VCE) is an electronic imaging technique. Both provide contrast enhanced images of the mucosal surface¹⁵.

There are a large amount of data comparing IEE to WLE alone; however, only the more recent studies included HD-colonoscopy. The SCENIC group recommended the use of CE and considered HD-chromoendoscopy the ideal surveillance method¹⁴. The universal adoption of CE is limited by lack of availability, the quality of bowel preparation required and the cumbersome method of application. The European Society of Gastrointestinal Endoscopy (ESGE) updated their guidance in 2019 to recommend that either CE or VCE could be utilised in IBD surveillance¹⁶. On the basis of more recent data^17,18. Rabinowitz et al ¹⁹ suggest that CE, VCE and WLE are all acceptable modalities, provided HD-scopes are available to an endoscopist adequately trained in dysplasia detection using the modality of choice.

The utility of non-targeted biopsies in the era of advanced endoscopic imaging has long been contested²⁰. In one randomised controlled trial of 246 patients with UC >7 years duration, the proportional yield of dysplasia detection was significantly higher in the targeted biopsy group (6.9% [24 of 350]) than in the random biopsy group (0.5% [18 of 3725]; P < .001)²¹. However, despite their low yield, random biopsies may be useful in high-risk patients¹⁹. The ESGE recommend that when CE or VCE is employed, in experienced hands during quiescent disease activity, non-targeted biopsies can be forsaken but should be considered in patients with a personal history of colonic neoplasia, tubular-appearing colon, strictures or PSC¹⁶.

Flat lesions and background inflammation/pseudo-polyposis make the optical diagnosis of IBD dysplasia uniquely challenging. Although there is no data to support this, it seems logical for endoscopists with expertise in IBD dysplasia detection to perform surveillance on dedicated lists. As there are no validated training programs ESGE advise self-learning via on-site training of at least one-week’s duration with an expert and at least 20 procedures with histological feedback from targeted biopsies. The expert panel defined proficiency as a dysplasia detection rate of >10%²².

Management of dysplasia

Endoscopic resectability of a pre-neoplastic lesion is the most important predictor in determining the risk of progression to CRC; the distinction between a sporadic adenoma or dysplasia associated lesion/mass is less relevant¹². The SCENIC group developed the Modified Paris classification of visible lesions¹⁴ (figure 3) which is used to guide endoscopic management. With advancements in endoscopic therapy, including endoscopic submucosal dissection (ESD), fewer patients need surgical intervention. In cases of invisible dysplasia, the procedure should be repeated by an endoscopist with expertise in IBD surveillance. If no visible lesion is detected management should be on a case-by-case basis¹⁹. The Ulcerative Colitis-Cancer Risk Estimator (UC-CaRE) is a useful providing an estimated risk of progression from low grade dysplasia to CRC²³. This also supports patients to have a more active role in shared decision making.

What next?

Surveillance techniques are rapidly evolving as technology changes with most recent data suggesting that VCE, CE and HD-WLE are currently all acceptable methods. Recommendations for standardised IBD colonoscopy reporting have been made and their adoption should improve overall quality²⁴. Artificial intelligence is likely to become part of our armamentarium in the near future. As IBD therapy improves, the burden of regular colonoscopic surveillance is growing. Further work is needed to develop strategies to target surveillance to patients at highest risk of dysplasia and cancer and minimise surveillance in low-risk and co-morbid patients in whom dysplasia or CRC is unlikely to occur in their estimated lifetime.  Molecular biomarkers are emerging as a novel method to optimise risk stratification to be more personalised²⁵. Comprehensive management should involve a multi-disciplinary team strategy as well as empowering IBD patients to have a prominent role in decision-making regarding their care.

Summary

Table 1: Summary of IBD CRC Surveillance; onset, interval, modalities, management and follow-up based on recent guidance and expert opinion as discussed in main body of text^4,16,19,22.

Figure 1: Examples of endoscopic images of IBD-associated dysplasia under standard white light endoscopy (top panel) and narrow band imaging (bottom panel).

Figure 2: Surveillance recommendations for patients with Colitis. Adapted from updated BSG IBD Guidelines⁴.

Figure 3: Modified Paris Classification of Visible lesions¹⁴.

Author Biographies