Interpreting post-colonoscopy colorectal cancer rates in inflammatory bowel disease

People with colonic inflammatory bowel disease (IBD) have an increased risk of developing colorectal cancer (CRC). Colonoscopic surveillance is advocated in this population, with the aim of preventing death from CRC. This seems intuitive for a high risk group, but there is only limited observational evidence for the efficacy of surveillance [1]. A randomized controlled trial would now be extremely difficult to conduct as one group would be randomized to no colonoscopy surveillance. Furthermore, the absolute risk of CRC is low and it can take years for CRC to develop.

A post-colonoscopy colorectal cancer (PCCRC) is one that develops following a colonoscopy that did not detect CRC. The rate of these cancers, as a proportion of the total number of CRCs, detected over a 3-year period has been recommended as a key indicator of the quality of a colonoscopy service [2]. Colonic IBD has been consistently shown to be one of the strongest factors associated with PCCRC, and the 3-year PCCRC rate in those with IBD is between 30 % and 50 % [3] [4]. There is debate about whether this metric is appropriate for those undergoing regular surveillance, as frequent colonoscopy tests could make PCCRC inevitable.

In the current issue of Endoscopy, Troelsen et al. [5] present Danish population-based data on PCCRC in IBD over a 20-year period to 2015. This large cohort study of nearly 400 000 individuals investigated two outcomes: the cumulative incidence proportion (CIP) of PCCRC within 3 years of a negative colonoscopy, and the 3-year PCCRC rate in IBD patients.

“This study again shows that the 3-year PCCRC rate in those with IBD is much higher than in the general population at 24.3 % vs. 7.5 %. This means that, of all CRCs diagnosed in those with IBD, 24.3 % will have had a colonoscopy where there may have been an opportunity to diagnose CRC at an earlier stage or prevent it altogether.”

They found that the proportion with a PCCRC within 3 years of a first negative colonoscopy was low at ~2 in every 1000 procedures for those with IBD. The adjusted 3-year risk of PCCRC was similar in the IBD and non-IBD population after one or multiple colonoscopies. One interpretation is that the association between IBD and PCCRC is explained by IBD patients having multiple surveillance tests. These increase the likelihood of a negative colonoscopy before a CRC diagnosis. However, frequent tests in the non-IBD population are also a marker of conditions with an increased risk of PCCRC. These include post-CRC or post-polypectomy, having a hereditary CRC risk or colonic symptoms such as diverticular disease. These groups of patients could be more predisposed to missed, incompletely resected, or rapidly growing cancers. Either way, the implication for colonoscopists is to recognize the increased association with PCCRC for patients undergoing frequent examinations.

This study again shows that the 3-year PCCRC rate in those with IBD is much higher than in the general population at 24.3 % vs. 7.5 %. This means that, of all CRCs diagnosed in those with IBD, 24.3 % will have had a colonoscopy where there may have been an opportunity to diagnose CRC at an earlier stage or prevent it altogether.

To understand PCCRC in surveillance populations, the cancers can be further classified into interval or non-interval CRC, depending on when the CRC occurred in relation to a planned surveillance test [2]. The 2018 World Endoscopy Organization PCCRC guidelines define these categories and they are summarized in [Table 1] with case examples specific to IBD.

Table 1 The categories of post-colonoscopy colorectal cancer (PCCRC) based on World Endoscopy Organization definitions, with example scenarios in inflammatory bowel disease.

	Interval type	Non interval types
		Type A	Type B	Type C
	Detected before the recommended interval	Detected at the recommended interval	Detected after the recommended interval	No interval recommended
Clinical scenario	Quiescent pan-UC	Pan-UC & post-inflammatory polyps	Pan-UC & PSC	Proctitis
Recommendation	Surveillance in 5 years	Surveillance in 3 years	Surveillance in 1 year	No surveillance
Outcome	4 years later develops anemia; colonoscopy reveals CRC	CRC is found at the repeat test	Patient misses this and presents 3 years later with CRC	Attends 7 years later with weight loss; CT reveals CRC
Possible verdict	The recommended surveillance interval may be too long	This could be classed as a surveillance “success,” but the recommended surveillance interval may still be too long	Reinforces the importance of adherence to recommended surveillance intervals	Review whether surveillance might have been appropriate

UC, ulcerative colitis; PSC, primary sclerosing cholangitis; CRC, colorectal cancer; CT, computed tomography.

Population-based databases often lack the granularity to categorize PCCRC in this way. Understanding the reasons for these cancers is important as it facilitates targeted improvement measures to try to reduce PCCRC numbers.

For sporadic colorectal cancer, a detailed review of PCCRC cases showed that 89 % may be avoidable [6]. Following this, a recent UK study has analyzed IBD-related CRC cases at a tertiary referral center [7]. Here, 27 % of patients were diagnosed with CRC who would not have been included in a surveillance program because of disease extent or duration of disease. Of those who were eligible for surveillance, only 12 % would be classified as having World Endoscopy Organization type A, non-interval CRC, and were detected at a timely surveillance test. Of those eligible for surveillance, 64 % were undergoing no surveillance at all.

The Troelsen study has shown that the absolute risk of CRC or PCCRC is low, but a large proportion of CRCs occur in those undergoing regular surveillance tests. Colonoscopies are a burden for patients and healthcare providers and are not without risk. In order to utilize this resource efficiently, we need to target those who might benefit most from surveillance and, importantly, identify those who can safely cease surveillance. Adjuncts to colonoscopy such as blood or stool oncological markers may help identify these groups. Better quality colonoscopy and adherence to surveillance guidelines in appropriate people should improve the performance of the test. It has been shown that ~70 % of dysplasia detected in IBD patients is endoscopically resectable, and resection reduces the 10-year risk of advanced dysplasia [8]. Therefore, it stands to reason that increased detection and resection should reduce the rate of IBD-related PCCRC.

To bring the IBD-related PCCRC rates down, we need a better understanding of the precise etiology of the disease. Some PCCRCs will be detected appropriately during surveillance, but for others there may have been opportunities for earlier diagnosis or even prevention at the index test.

Publication History

Article published online:
20 July 2021

• References

• 1 Bye WA, Nguyen TM, Parker CE. et al. Strategies for detecting colon cancer in patients with inflammatory bowel disease. Cochrane Database Syst Rev 2017; 9: CD000279
• 2 Rutter MD, Beintaris I, Valori R. et al. World Endoscopy Organization consensus statements on post-colonoscopy and post-imaging colorectal cancer. Gastroenterology 2018; 155: 909-925.e3
• 3 Burr NE, Derbyshire E, Taylor J. et al. Variation in post-colonoscopy colorectal cancer across colonoscopy providers in English National Health Service: population based cohort study. BMJ 2019; 367: 16090
• 4 Stjärngrim J, Ekbom A, Hammar U. et al. Rates and characteristics of postcolonoscopy colorectal cancer in the Swedish IBD population: what are the differences from a non-IBD population?. Gut 2019; 68: 1588-1596
• 5 Schønfeldt Troelsen F, Sorensen H, Pedersen L. et al. Risk of a post-colonoscopy colorectal cancer diagnosis in patients with inflammatory bowel disease: a population-based cohort study. Endoscopy 2021; 53: 1023-1033
• 6 Anderson R, Burr NE, Valori R. Causes of post-colonoscopy colorectal cancers based on World Endoscopy Organization system of analysis. Gastroenterology 2020; 158: 1287-1299.e2
• 7 Gordon C, Chee D, Hamilton B. et al. Root-cause analyses of missed opportunities for the diagnosis of colorectal cancer in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2021; 53: 291-301
• 8 Cremer A, Demetter P, De Vos M. et al. Risk of development of more-advanced lesions in patients with inflammatory bowel diseases and dysplasia. Clin Gastroenterol Hepatol 2020; 18: 1528-1536.e5