I nterval Col orectal Cancers

i n I nfl ammatory Bowel

Di sease: The Grim Statistics and True Stories
Silvia Sanduleanu, MD, PhD
a,
*,
Matthew D. Rutter, MBBS, MD, FRCP (London, Edinburgh)
b,c
The past decade has witnessed considerable progress in the management of inflam-
matory bowel disease (IBD), including improvements in the quality and effectiveness
of colonoscopic surveillance.
13
Patients with ulcerative colitis (UC) or Crohns colitis
have a greater risk of colorectal cancers (CRC), which may develop earlier and prog-
ress more rapidly than sporadic CRCs. Although most societies nowendorse intensive
colonoscopic surveillance to reduce the CRC risk,
46
the efficacy of this strategy re-
mains controversial. Several recent studies have cast doubt about the limited effective-
ness of colonoscopy at reducing the incidence of sporadic CRC in the general
population, especially in the proximal part of the colon,
7,8
resulting in the occurrence
of interval CRCs. Little is known, however, about the magnitude of this problem in pa-
tients with IBD and the most common explanations. Similar to the sporadic interval
CRCs, two factors contribute to interval CRCs in IBD: clinician-dependent factors,
such as missed, incompletely resected lesions or suboptimal surveillance; and
a
Division of Gastroenterology and Hepatology, Department of Internal Medicine, GROW,
School for Oncology and Developmental Biology, Maastricht University Medical Center, Postbox
5800, 6202 AZ, Maastricht, The Netherlands;
b
Department of Gastroenterology, University
Hospital of North Tees, Hardwick, Stockton-on-Tees, Cleveland, TS19 8PE, UK;
c
DurhamUniver-
sity School of Medicine, Pharmacy and Health Queens Campus, University Boulevard,
Stockton-on-Tees, Cleveland, TS17 6BH, UK
* Corresponding author.
E-mail address: [email protected]
KEYWORDS

Colorectal cancer

Interval colorectal cancer

Inflammatory bowel disease

Ulcerative colitis

Crohns disease

Surveillance

Colonoscopy
KEY POINTS
Interval colorectal cancers (CRCs) may account for approximately half of all CRCs identi-
fied during IBD surveillance, which highlights the need for improvements.
The cause of interval CRCs is multifactorial, with procedural factors likely to play an impor-
tant role.
Molecular events promoted by inflamed mucosa may augment the cancer risk and
perhaps explain some interval CRCs.
Gastrointest Endoscopy Clin N Am 24 (2014) 337348
http://dx.doi.org/10.1016/j.giec.2014.03.001 giendo.theclinics.com
1052-5157/14/$ see front matter 2014 Elsevier Inc. All rights reserved.
molecular features of the inflamed mucosa underlying the development of cancer. The
endoscopic knowledge, equipment, and techniques have evolved in recent years,
contributing to a paradigmshift in the diagnosis and endoscopic resection of CRCpre-
cursors. The nonpolypoid (flat or depressed) colorectal neoplasms (NP-CRNs) play a
significant role in the genesis of interval CRCs.
9
Such subtle-appearing lesions are
indeed more likely missed or incompletely resected endoscopically than their polypoid
counterparts, and a subgroup of them harbor an aggressive biologic behavior.
This article provides insight into the magnitude and most common factors underlying
the cause of interval CRCs during surveillance for IBD. Milestones of the literature
regardingCRCriskinpatientswithIBDarereviewed. Specifically examinedtotheoccur-
rence of interval CRCs are the contribution of missed, incompletely resected lesions; the
adherence to surveillance; and distinct biologic features of the inflamed mucosa. Key
principles are presented for ensuring the quality of IBD surveillance practice.
INCIDENCE OF CRC AND INTERVAL CRC
A casual glance at the overall incidence of CRC in patients with IBD reveals discrepant
outcomes, with a few studies showing similar CRC rates in patients with IBD versus
the general population,
10,11
whereas others show greater rates.
1214
In a nationwide
cohort of close to 50,000 Danish patients with IBD who were followed over
three decades (19792008), CRC was identified in 338 (0.71%) cases (268 in patients
with UC and 70 in patients with Crohns disease).
10
The overall CRC risk among pa-
tients with UC in this study was similar to that of the general population (relative
risk, 1.07; 95% confidence interval, 0.951.21). In contrast, a North American study
15
conducted from 1998 through 2010 found that the incidence of CRC in patients with
Crohns disease or UC was 60% higher than in the general population.
The Danish study found a marked decline in the overall relative risk of CRC among
patients with UC over the past decades, from 1.34 (95% confidence interval, 1.13
1.58) in 1979 to 1988 to 0.57 (95% confidence interval, 0.410.80) in 1999 to
2008,
10
possibly reflecting refinements in the anti-inflammatory arsenal (ie, immuno-
suppressive therapy, biologicals), but perhaps also caused by a gradual adoption of
CRC screening and surveillance. Conversely, the North American study
15
found a
fairly stable CRC rate in patients with IBD over time. Controversies surrounding the
time-trends in CRC risk are not surprising, and likely reflect the cumulative effect of
several factors, such as advancements in endoscope technology, a greater aware-
ness, and improvements in the quality of colonoscopic performance.
As a common denominator, such epidemiologic studies lack relevant information
about the disease duration, degree and extent of inflammation, presence of risk fac-
tors (ie, primary sclerosing cholangitis, personal or family history of CRC), and patients
compliance with the recommended follow-up. Although clinical studies provide such
details, most have focused on the optimal frequency of surveillance, paying less atten-
tion to the quality of examination. A systematic characterization of the lesions pheno-
type, in particular the location, size, shape, and histology, is often lacking.
Very few data are available about the occurrence of interval cancers during surveil-
lance for IBD. The first paper dates back to 1982.
16
In this surgical review of 676 pa-
tients with UC undergoing long-term follow-up, a total of 35 CRCs were identified.
Twelve of these were diagnosed because of symptoms, 10 as incidental findings at
proctocolectomy, and 13 CRCs were diagnosed during the follow-up at least 1 year
after the initial UC diagnosis. This latter subgroup was referred to as interval
CRCs. In a St Marks study reviewing the UC surveillance program over approxi-
mately three decades, a total of 74 patients (12.3% of the total population) developed
Sanduleanu & Rutter 338
neoplasms, including 30 CRCs.
17
The authors defined interval CRCs as cancers pre-
senting after a negative index-colonoscopy or advanced (Dukes C/disseminated)
cancers detected at surveillance. During a median follow-up of 1.5 years, nine pa-
tients were identified with Dukes C cancers and four patients with disseminated can-
cers (4 of these 13 cases were diagnosed within 12 months). In three cases, CRC was
diagnosed at colonoscopy because of symptoms; one of these was attributable to
noncompliance. Of note, more than half (16 out of the 30) of the CRCs identified
with this program were interval cancers, raising concerns about the effectiveness of
colonoscopic cancer prevention. A statistically significant reduction in CRC rates
over time was observed in this study (r 5 0.40; P 5 .04), especially in the proximal
colon.
From these data, we can conclude that there is sparse understanding of the magni-
tude and clinical significance of interval CRCs in patients with IBD. Indeed, a wide vari-
ation exists with regard to the terminology used in endoscopy and pathology
diagnostic protocols across countries, IBD centers, and studies.
Standardization of the nomenclature and clinical protocols, and uniformity in report-
ing on interval CRCs during IBD surveillance, would help to define quality targets. As a
first step, a universal terminology is required for dysplasia and interval cancers. Pre-
viously used terms, such as flat dysplasia or dysplasia associated lesion or mass,
need to be revisited. A rigorous description of the endoscopic shape and histologic
features of the detected lesions is required, using international classifications (ie,
Paris-Japanese endoscopic classifications
18,19
and the World Health Organization
histopathologic classifications
20,21
). Interval cancers should be considered those inva-
sive cancers diagnosed after a negative screening examination, but before the next
recommended follow-up colonoscopy, as endorsed by the current international IBD
surveillance guidelines.
POTENTIAL ETIOLOGIC FACTORS OF INTERVAL CRCS
Similarly to sporadic CRCs, most interval CRCs in IBD probably can be explained by
clinician-dependent factors, such as missed, incompletely resected lesions or devia-
tion fromsurveillance protocols. The understanding of the underpinnings of such inter-
val CRCs is of importance because it may permit identification of modifiable factors,
for example gaps in knowledge and training on the recognition of nonpolypoid neo-
plasms and their endoscopic resection. In this case, tailored educational programs
would improve the awareness and help to shape practical skills, to ultimately safe-
guard the quality of colonoscopy. Furthermore, it is important to understand whether
certain molecular features of the inflamed mucosa could augment the risk of cancer
progression. Such information may help to develop personalized (ie, molecular-
based) surveillance strategies.
Missed Lesions
Two recent studies exploring the cause of sporadic interval CRCs in the general
population found missed lesions represent by far the most important contributor
(>50% of all interval CRCs).
22,23
Undoubtedly, missed lesions are likely to account
for a significant proportion of interval CRCs in IBD, although a thorough analysis using
structured algorithms
24
has not yet been performed. A recent population-based anal-
ysis by Wang and colleagues,
25
using SEER cancer registry data from 55,008 older
patients with CRC, found rates of early/missed CRCs were three-fold greater in IBD
than in patients without IBD (15.1%for Crohns disease, 15.8%for UC vs 5.8%for pa-
tients without IBD; P<.001). Early/missed CRCs were defined as CRCs identified
Interval CRC in IBD 339
within 6 to 36 months after a colonoscopic examination that did not detect cancer.
This study was based on administrative data, and therefore lacked detail about the
completeness of colonoscopy, bowel preparation, extent of colitis, characteristics
of mucosal lesions identified at the baseline examination, and resection outcomes.
Such observations underscore the importance of meticulous inspection of the entire
colonic mucosa, which should be ideally clean and free of inflammation, and the
need for formal training of the endoscopist in the recognition of IBD neoplasms. Pres-
ence of active or chronic background inflammation and the diversity in endoscopic
appearance of dysplasia by IBD may, however, increase the complexity of diagnosis.
Fig. 1 illustrates a lateral spreading tumor of granular subtype, which could have been
missed at a previous examination.
A substantial number of studies demonstrated that indigo carmine or methylene
blueguided chromoendoscopy (CE) improves the diagnostic yield of dysplasia and
Fig. 1. (AD) Lateral spreading tumor of granular type, located in the descending colon of a
patient with a Crohns pancolitis. (E) Histopathology revealed low-grade dysplasia (Hema-
toxylin and eosin, original magnification 20). (F) Colonoscopic examination 8 months
earlier showed Mayo 2 inflammation only at the same anatomic site (arrows), suggesting
this lesion could have been missed.
Sanduleanu & Rutter 340
invasive CRC during IBD surveillance. This is not surprising, because a significant pro-
portion
2628
of dysplastic lesions in patients with IBD appear to have a flat appear-
ance, as illustrated in Table 1. Pancolonic CE delineates the borders and permits a
detailed analysis of the epithelial surface, thus facilitating the diagnosis of subtle le-
sions and their endoscopic resection. A few meta-analyses now demonstrate CE
with targeted biopsies is superior to white-light colonoscopy with random biopsies
in the detection of dysplasia and invasive CRCs.
2931
CE yielded a 7% increase in
the detection of any dysplasia.
31
Compared with white-light colonoscopy with random
biopsies, the likelihood to detect any dysplasia with CE and targeted biopsies was 8.9-
fold greater, and 5.2-fold greater for detecting nonpolypoid dysplasia. In a Mainz study
of 165 patients with long-standing UC who were randomized to undergo standard co-
lonoscopy using white light versus CE (0.1% methylene blue), significantly more intra-
epithelial neoplasms were detected in the CE group (32 vs 10; P 5.003). CE detected
more intraepithelial neoplasms in flat mucosa than white-light endoscopy (24 vs 4;
P 5 .0007), and more invasive cancers (3 vs 1).
26
In these studies, colonoscopies were performed by dedicated colonoscopists with
expertise in multimodal imaging, and under controlled circumstances (ie, clinical tri-
als), and may preclude generalizability. Recognition of the nonpolypoid dysplasia in
a real-world environment remains challenging and requires additional training. In a
study conducted at Maastricht University Medical Center, where the endoscopists
have been trained on the recognition of nonpolypoid neoplasms,
32
the overall detec-
tion rate of sporadic NP-CRNs (defined as lesions of which the height was less than
half of the diameter) was 5.7% (diagnostic subgroup, 4.7%; screening subgroup,
4.5%; surveillance subgroup, 15.6%).
33
The learning-curve in the detection of
NP-CRNs is, however, tedious, with at least 600 colonoscopies being required to
achieve a detection rate of at least 4.5%.
34
It is highly likely that missed lesions have a major contribution to the development of
interval CRCs in patients with IBD, although this needs further investigation. The cur-
rent data highlight the importance of vigilant inspection and a thorough phenotyping of
lesions identified at colonoscopy, including subtle erosions, shallow ulcerations, and
their relationship with inflammation or strictures. Such exquisite detail may improve
the understanding of the link between inflammation, the occurrence of dysplasia,
and interval CRCs. High-quality videos/photodocumentation obtained in a standard-
ized fashion facilitates this process. Challenging cases should be performed by expert
endoscopists.
Incompletely Resected Lesions
Endoscopic resection of neoplasms in the context of colitis is clearly fraught with dif-
ficulties because of the presence of inflammation and scarring. Such conditions chal-
lenge the accurate detection, clear demarcation, and lifting of the lesions. Studies
examining the diagnostic yield of CE during surveillance for IBD provided, however,
limited information about the effectiveness of the endoscopic resection, which re-
quires further investigation. In a long-term follow-up evaluation, Odze and col-
leagues
35
compared the outcome after polypectomy among three subgroups: (1)
patients with UC with adenoma-like dysplastic lesions, (2) patients with UC with spo-
radic adenomas, and (3) a non-UC sporadic adenoma subgroup. Prevalence of polyp
formation on follow-up, albeit high in this study, did not significantly differ across sub-
groups (62.5%, 50%, and 49%, respectively) indicating IBD-associated dysplasia may
be effectively treated endoscopically. Indeed, over the past few years, endoscopic
mucosal resection and endoscopic submucosal dissection resection techniques
proved to be increasingly safe and effective in the Western practice.
3638
A study
Interval CRC in IBD 341
Table 1
Detection of flat dysplastic lesions in patients undergoing surveillance for IBD
Study, Author, Year Design
Endoscopist
Experience
Number of
Patients Dye
Number of Patients
with IEN
Flat Dysplastic
Lesions (Numbers)
Invasive CRCs
(Numbers)
Kiesslich et al,
26
2003 Randomized 1:1, CE
vs standard WLE
Several experienced
endoscopists
165 MB 0.1% CE group: 32 IENs in
13 pts
WLE group: 10 IENs in
6 pts
CE group: 24
WLE group: 4
(P 5 .0007)
CE group: 3 invasive
CRC
WLE group: 1
invasive CRC
Matsumoto et al,
27
2003
Prospective
cohort CE
Single experienced
endoscopist
57 IC 0.2% 12 pts with 117 IENs 27 4 HGD/invasive CRCs
Rutter et al,
28
2004 Prospective cohort
back-to-back
WLE/CE
Single experienced
endoscopist
100 IC 0.1% CE group: 9 pts
WLE group: 2 pts
CE group: 75 flat
topped elevated
WLE group: 0
None
Kiesslich et al, 2007 Randomized 1:1
CE (N 5 80) vs WLE
(N 5 73)
Several experienced
endoscopists
153 MB 0.1% CE group: 19 in 11 pts
CC group: 4 in 4 pts
CE group: 16
CC group: 2
None
Hlavaty et al, 2011 Tandem
colonoscopies
Several experienced
endoscopists
30 IC 0.4% WLE: 2
CE: 4
2 None
Gu nther et al, 2011 Randomized
Group 1: random
biopsies; Group 2
CE 1 biopsies;
Group 3 CE 1
confocal
endomicroscopy
Two experienced
endoscopists
150 IC 0.1% Group 1: no IEN
Group 2: 2 pts
Group 3: 4 pts
Group I (0)
Group II (18)
Group III (10)
1 (Group 3)
Abbreviations: CC, conventional colonoscopy; HGD, high-grade dysplasia; IC, indigo carmine; MB, methylene blue; WLE, white-light endoscopy.
S
a
n
d
u
l
e
a
n
u
&
R
u
t
t
e
r
3
4
2
examining the effectiveness of endoscopic resection of NP-CRNs found that 93% of
those larger than 10 mm were successfully resected.
36
Residual neoplasia was iden-
tified in 10% of cases on the first follow-up examination, although complete resection
was obtained in all cases after one to three follow-up examinations. Likewise, Buchner
and colleagues
37
found that large sessile and NP-CRNs could be managed endoscop-
ically in 91% of cases, with a perforation rate of 0.4% and a bleeding rate of 11%.
Because 9%
23
to 50%
38
of the sporadic interval CRCs are thought to be caused by
an ineffective polyp resection, the precise contribution of this factor to the genesis
of interval CRCs in patients with IBD needs further elucidation.
Adherence to Colonoscopic Surveillance
Adherence to colonoscopic surveillance guidelines is indeed vital, but seems to be
often problematic.
3942
There are several caveats to keep in mind, foremost of which
is the patients understanding of the cancer risk.
43,44
Disease flares and presence of
comorbidity may further reduce the compliance to surveillance. Because the presence
of disease activity challenges the endoscopic and histologic appreciation of dysplasia,
colonoscopic surveillance should be ideally performed in the quiescent phase. How-
ever, surveillance should not be delayed too long, because those with more active
disease carry a greater risk of developing CRC. With regard to bowel preparation, a
low-residue diet the days before the procedure in conjunction with split-dose polyeth-
ylene glycol solutions is often sufficient for adequate cleansing, without inducing
inflammation.
Biologic Features
The precise biologic events underlying chronic inflammation and leading to a faster
progression to CRC are presently unknown and need further exploration. A subset
of dysplastic lesions identified in patients with IBD harbor a villous phenotype, as illus-
trated in Fig. 2. Such macroscopic features have been suggested to represent a red
flag for the presence of invasive CRC, especially of colloid subtype.
45
Other CRCs har-
bor signet ring cells, features associated with a more aggressive biologic behavior.
Fig. 3 illustrates a small signet ring cell carcinoma that displayed clear signs of local
invasion. Approximately 6% of the cancers in patients with IBD are small flat invasive
CRCs, without adjacent adenomatous tissue,
30
suggesting that progression to CRC
may involve a pathway different from the classic adenoma-carcinoma sequence.
Fig. 2. (AC) Lateral spreading tumor of the rectum in a patient with distal ulcerative colitis.
Examination using high-definition endoscopy in conjunction with chromoendoscopy clearly
showed a villous appearance. (D) Histopathology revealed low-grade dysplasia (Hematoxy-
lin and eosin, original magnification 20). (E) Fuller view of lesion with indigo carmine
chromoendoscopy.
Interval CRC in IBD 343
The newly described serrated neoplastic pathway may also explain a subset of interval
CRCs in patients with IBD.
46
Interestingly, a recent study by Voorham and col-
leagues
47
found that sporadic nonpolypoid neoplasms are likely to herald 5q loss,
and less likely MSI and APC mutations, features resembling the carcinogenesis pro-
cess in inflammatory conditions, such as IBD.
In summary, clinician-dependent factors and biologic factors intermingle in the gen-
esis of interval CRCs by IBD. It is important to understand whether presence of NP (flat
or depressed)-CRNs in patients with IBD signifies a diagnostic and therapeutic chal-
lenge alone. The most effective filter of missed or incompletely resected lesions would
then be training for improving the education and endoscopic skills. Clinical decisional
algorithms, including the characterization of shape, epithelial surface of lesions, and
their relation with inflammation,
31
have the potential to steer the diagnostic and ther-
apeutic process and optimize outcomes. If a subset of the NP-CRNs contains molec-
ular features associated with a greater risk of CRC, such patients need to be identified
and closely surveyed to prevent CRC.
CONCLUDING REMARKS
Interval CRCs may account for approximately 50% of the CRCs identified during IBD
surveillance, favoring the idea that clinical consent should include information about
cancer risk. Improvements in the quality of colonoscopic examinations are vital for
minimizing the CRC risk of patients with IBD. Box 1 summarizes basic concepts for
achieving that goal. Standardization of clinical protocols is required, including the
use of high-definition and high-resolution colonoscopes coupled with the application
of pancolonic CE with targeted biopsies. Surveillance colonoscopy using white light
with random biopsies should be abandoned. Formal training in recognition of NP-
CRNs and proficiency in endoscopic resection techniques should be compulsory
for providers who perform surveillance in patients with IBD. Comprehensive colonos-
copy and pathology data reporting using a standardized nomenclature and interpre-
tation of findings using tailored algorithms may ultimately shed light on the cause of
interval CRCs and the required improvements.
Fig. 3. (A) A 10-mm sized, Paris type IIa1IIc lesion, with a central ulceration that has been
identified at the hepatic flexure of a patient with Crohns colitis. (B) Examination of the sur-
gical specimen showed the small cancer (arrows). (C) Histopathology revealed a poorly
differentiated signet ring cell adenocarcinoma, with signs of lymphangioinvasion. The
lesion was located near a stricture (Hematoxylin and eosin, original magnification 30)
(D, E) and staged pT3N1Mx.
Sanduleanu & Rutter 344
REFERENCES
1. Mowat C, Cole A, Windsor A, et al. Guidelines for the management of inflamma-
tory bowel disease in adults. Gut 2011;60(5):571607.
2. Rutter MD, Riddell RH. Colorectal dysplasia in inflammatory bowel disease:
a clinicopathologic perspective. Clin Gastroenterol Hepatol 2014;12(3):
35967.
3. Murthy SK, Kiesslich R. Evolving endoscopic strategies for detection and treat-
ment of neoplastic lesions in inflammatory bowel disease. Gastrointest Endosc
2013;77(3):3519.
4. Cairns SR, Scholefield JH, Steele RJ, et al. Guidelines for colorectal cancer
screening and surveillance in moderate and high-risk groups (update from
2002). Gut 2010;59(5):66689.
5. Farraye FA, Odze RD, Eaden J, et al. AGA technical review on the diagnosis and
management of colorectal neoplasia in inflammatory bowel disease. Gastroenter-
ology 2010;138(2):74674.
6. Kornbluth A, Sachar DB, Practice Parameters Committee of the American Col-
lege of Gastroenterology. Ulcerative colitis practice guidelines in adults (update):
American College of Gastroenterology, Practice parameters Committee. Am J
Gastroenterol 2004;99(7):137185.
Box 1
Principles for quality colonoscopy during surveillance for IBD: The Six Ts
Timing
Ideally, surveillance should be performed in the quiescent phase.
Tolerance
Tolerance of bowel preparation is key for adequate cleansing; a low-residue diet can be
added to standard split-dose bowel preparations.
Technology
High-definition/high-resolution colonoscopes should be used.
Technique
Pancolonic chromoendoscopy with targeted biopsies should be considered the standard of
care.
Training
The IBD specialist and pathologist needs to be formally trained on detection, classification,
and diagnosis of nonpolypoid (flat or depressed) colorectal neoplasms.
Traits
A thorough registration of patient and lesion characteristics may help identify the most
likely cause:
Patient with IBD: duration and extent of disease, presence of primary sclerosing
cholangitis, family history of CRC, personal history of colorectal polyps (including
postinflammatory polyps), history of medication, compliance and response to therapy
Lesions at baseline examination: location, size, shape, histology, and the relationship
with inflammation and strictures
Interval CRCs: recommended surveillance intervals, time to CRC diagnosis, location of
tumor, histology, and tumor stage
Interval CRC in IBD 345
7. Lakoff J, Paszat LF, Saskin R, et al. Risk of developing proximal versus distal colo-
rectal cancer after a negative colonoscopy: a population-based study. Clin Gas-
troenterol Hepatol 2008;10:111721.
8. Brenner H, Chang-Claude J, Seiler CM, et al. Protection from colorectal cancer
after colonoscopy: a population-based, case-control study. Ann Intern Med
2011;154(1):2230.
9. Sanduleanu S, Masclee AM, Meijer GA. Interval cancers after colonoscopy:
insights and recommendations. Nat Rev Gastroenterol Hepatol 2012;9(9):5504.
10. Jess T, Simonsen J, Jrgensen KT, et al. Decreasing risk of colorectal cancer in
patients with inflammatory bowel disease over 30 years. Gastroenterology 2012;
143(2):37581.
11. Kappelman MD, Farkas DK, Long MD, et al. Risk of cancer in patients with in-
flammatory bowel diseases: a nationwide population-based cohort study with
30 years of follow-up evaluation. Clin Gastroenterol Hepatol 2014;12(2):
26573.e1.
12. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative
colitis: a meta-analysis. Gut 2001;48(4):52635.
13. Ekbom A, Helmick C, Zack M, et al. Ulcerative colitis and colorectal cancer. A
population-based study. N Engl J Med 1990;323(18):122833.
14. Mellemkjaer L, Olsen JH, Frisch M, et al. Cancer in patients with ulcerative colitis.
Int J Cancer 1995;60(3):3303.
15. Herrinton LJ, Liu L, Levin TR, et al. Incidence and mortality of colorectal adeno-
carcinomas in persons with inflammatory bowel disease from 1998 to 2010.
Gastroenterology 2012;143(2):3829.
16. Prior P, Gyde SN, Macartney JC, et al. Cancer morbidity in ulcerative colitis. Gut
1982;23(6):4907.
17. Rutter MD, Saunders BP, Wilkinson KH, et al. Thirty-year analysis of a colono-
scopic surveillance program for neoplasia in ulcerative colitis. Gastroenterology
2006;130(4):10308.
18. Kudo S, Lambert R, Allen JI, et al. Nonpolypoid neoplastic lesions of the colo-
rectal mucosa. Gastrointest Endosc 2008;68(Suppl 4):S347.
19. Soetikno R, Friedland S, Kaltenbach T, et al. Nonpolypoid (flat and depressed)
colorectal neoplasms. Gastroenterology 2006;130:56676.
20. Hamilton S, Aaltonen LA. WHO classification of tumours. Pathology and genetics:
tumours of the digestive system. Lyon (France): IARC; 2000.
21. Bosman FT, Carneiro F, Hruban RH, et al. WHO classification of tumours. Pathol-
ogy and genetics. Tumours of the digestive system. 4th edition. Berlin: Springer-
Verlag; 2010.
22. Robertson DJ, Lieberman DA, Winawer SJ, et al. Colorectal cancers soon after
colonoscopy: a pooled multicohort analysis. Gut 2013. [Epub ahead of print].
23. Le Clercq CM, Bouwens MW, Rondagh EJ, et al. Postcolonoscopy colorectal can-
cers are preventable: a population-based study. Gut 2013. [Epub ahead of print].
24. Pabby A, Schoen RE, Weissfeld JL, et al. Analysis of colorectal cancer occur-
rence during surveillance colonoscopy in the dietary Polyp Prevention Trial. Gas-
trointest Endosc 2005;61(3):38591.
25. Wang YR, Cangemi JR, Loftus EV, et al. Rate of early/missed colorectal cancers
after colonoscopy in older patients with and without inflammatory bowel disease
in the United States. Am J Gastroenterol 2013;108(3):4449.
26. Kiesslich R, Fritsch J, Holtmann M, et al. Methylene blue-aided chromoendo-
scopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative
colitis. Gastroenterology 2003;124(4):8808.
Sanduleanu & Rutter 346
27. Matsumoto T, Nakamura S, Jo Y, et al. Chromoscopy might improve diagnostic
accuracy in cancer surveillance for ulcerative colitis. Am J Gastroenterol 2003;
98:182733.
28. Rutter MD, Saunders BP, Schofield G, et al. Pancolonic indigo carmine dye
spraying for the detection of dysplasia in ulcerative colitis. Gut 2004;53(2):
25660.
29. Wu L, Li P, Wu J, et al. The diagnostic accuracy of chromoendoscopy for
dysplasia in ulcerative colitis: meta-analysis of six randomized controlled trials.
Colorectal Dis 2012;14(4):41620.
30. Subramanian V, Mannath J, Ragunath K, et al. Meta-analysis: the diagnostic yield
for detecting dysplasia in patients with colonic inflammatory bowel disease.
Aliment Pharmacol Ther 2011;33(3):30412.
31. Soetikno R, Subramanian V, Kaltenbach T, et al. The detection of nonpolypoid (flat
and depressed) colorectal neoplasms in patients with inflammatory bowel dis-
ease. Gastroenterology 2013;144(7):134952.
32. Sanduleanu S, Rondagh EJ, Masclee AM. Development of expertise in the detec-
tion and classification of nonpolypoid colorectal neoplasia: experience-based
data at an academic GI unit. Gastrointest Endosc Clin N Am 2010;20:44960.
33. Rondagh EJ, Bouwens MW, Riedl RG, et al. Endoscopic appearance of proximal
colorectal neoplasms and potential implications for colonoscopy in cancer pre-
vention. Gastrointest Endosc 2013;75:121825.
34. McGill SK, Kaltenbach T, Friedland S, et al. The learning curve for detection of
non-polypoid (flat and depressed) colorectal neoplasms. Gut 2013. [Epub ahead
of print].
35. Odze RD, Farraye FA, Hecht JL, et al. Long-term follow-up after polypectomy for
adenoma-like dysplastic lesions in ulcerative colitis. Clin Gastroenterol Hepatol
2004;2(7):53441.
36. Kaltenbach T, Friedland S, Maheshwari A, et al. Short- and long-term outcomes of
standardized EMR of nonpolypoid (flat and depressed) colorectal lesions > or 5
1cm (with video). Gastrointest Endosc 2007;65(6):85765.
37. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of defiant
colorectal lesions directed to an endoscopy referral center. Gastrointest Endosc
2012;76(2):25563.
38. Huang Y, Gong W, Su B, et al. Risk and cause of interval colorectal cancer after
colonoscopic polypectomy. Digestion 2012;86:14854.
39. Gearry RB, Wakeman CJ, Barclay ML, et al. Surveillance for dysplasia in patients
with inflammatory bowel disease: a national survey of colonoscopic practice in
New Zealand. Dis Colon Rectum 2004;47(3):31422.
40. van Rijn AF, Fockens P, Siersema PD, et al. Adherence to surveillance guidelines
for dysplasia and colorectal carcinoma in ulcerative and Crohns colitis patients in
the Netherlands. World J Gastroenterol 2009;15(2):22630.
41. Spiegel BM, Ho W, Esrailian E, et al. Controversies in ulcerative colitis: a survey
comparing decision making of experts versus community gastroenterologists.
Clin Gastroenterol Hepatol 2009;7(2):16874.
42. Velayos FS, Liu L, Lewis JD, et al. Prevalence of colorectal cancer surveillance for
ulcerative colitis in an integrated health care delivery system. Gastroenterology
2010;139(5):15118.
43. Robinson RJ, Hart AR, Mayberry JF. Cancer surveillance in ulcerative colitis: a
survey of patients knowledge. Endoscopy 1996;28(9):7612.
44. Eaden J, Abrams K, Shears J, et al. Randomized controlled trial comparing the
efficacy of a video and information leaflet versus information leaflet alone on
Interval CRC in IBD 347
patient knowledge about surveillance and cancer risk in ulcerative colitis. In-
flamm Bowel Dis 2002;8(6):40712.
45. Rubio CA. Serrated neoplasia and de novo carcinomas in ulcerative colitis: a his-
tological study in colectomy specimens. J Gastroenterol Hepatol 2007;22(7):
102431.
46. Srivastava A, Redson M, Farraye FA, et al. Hyperplastic/serrated polyposis in in-
flammatory bowel disease: a case series of a previously undescribed entity. Am J
Surg Pathol 2008;32(2):296303.
47. Voorham QJ, Carvalho B, Spiertz AJ, et al. Chromosome 5q loss in colorectal flat
adenomas. Clin Cancer Res 2012;18(17):45609.
Sanduleanu & Rutter 348