The global epidemiology of inflammatory bowel disease (IBD) is changing with a rising
burden of disease in several developing countries, including India.[1] Patients with ulcerative or Crohn's colitis have an increased risk of developing
dysplasia and neoplasia in the colon and need regular surveillance. Improvement in
treatment and surveillance strategies has reduced the overall incidence of colorectal
cancer (CRC), yet it is one of the leading causes of surgery and mortality.[2] Conventional adenoma, sessile serrated adenoma, and traditional serrated adenoma
may all be precursors of CRC in IBD.[3] In addition, serrated epithelial change (SEC), a distinct pathologic entity, has
also been implicated as a precursor to CRC in IBD. There is a lack of standard criteria
for diagnosing SEC. One criterion defines SEC as a luminal serrated surface contour
limited to the upper portion of the crypts, without any features of dysplasia on random
mucosal biopsies (endoscopically not visible, definition 1).[4] Another definition does not consider endoscopic appearance (visible or invisible)
and relies on histopathological features. This includes disorganized crypt architecture
with crypts losing perpendicular orientation to the muscularis mucosae, diffuse irregular
serrations, and goblet cell-rich epithelium (definition 2).[4] Information on the SEC in IBD is still emerging and, at times, conflicting.
This issue of the journal reports a retrospective study from Australia on SEC in IBD.[5] The authors identified 26 patients with SEC from the pathology database. A follow-up
colonoscopy or sigmoidoscopy was essential for inclusion in this study. Crohn's disease
was present in 16, while 10 had ulcerative colitis (UC). Overall, 30 SEC were noted
among the 26 subjects (19.2% had multiple lesions). SECs were commonly located in
the caecum and rectum, and 60% were visible endoscopically as polypoid or nodular
lesions. The remaining were detected on random biopsies and were not visible despite
high-definition endoscopy and/or chromoendoscopy being used in all the cases. SEC
was defined as nondysplastic colonic mucosa with goblet cell-rich epithelium and distorted,
serrated architecture involving the upper crypts, without distortion of the basal
crypt architecture. The diagnosis was independent of endoscopic visibility of the
lesion. The median follow-up was 24.5 months, and only one patient was noted to have
dysplasia. This work provides useful insight into the characteristics, burden, and
outcome associated with SEC in IBD. The clear inclusion/exclusion criteria, confirmation
of SEC by two pathologists, and evaluation of follow-up data are the key strengths.
Inclusion of patients with “IBD without SEC” as controls would have provided information
on the added risk of dysplasia due to this entity. Lack of a comparison group is an
important limitation of this study. Endoscopic inflammation was noted in 50% cases,
and the association of SEC with background disease activity and its location in relation
to the inflamed segment was not described. However, considering the infrequent nature
of SEC, the authors must be commended for their efforts in presenting this data.
The literature on SEC in IBD is predominantly retrospective in nature and the majority
has been published from the United States in the past decade. [Table 1] summarizes the key findings on SEC in IBD.[6]
[7]
[8]
[9]
[10]
[11] As expected, due to the colonic location of SEC, most cases are reported in patients
with UC.[6]
[7]
[9]
[11] While the onset of IBD is commonly in the third or fourth decade, SECs have been
reported in the fifth to sixth decade.[6]
[7]
[8]
[9]
[10]
[11] The average time interval between the diagnosis of IBD and detection of SEC is about
15 to 20 years, and a male preponderance has been noted.[6]
[7]
[8]
[9]
[10]
[11] This supports the notion that the SEC may be associated with longstanding inflammation
in the colon. Interestingly, despite image-enhanced endoscopy being freely available
and recommended for surveillance in IBD, this is infrequently used.[8]
[9]
[10] The vast majority of the SEC have been observed in the rectum or left colon.[6]
[7]
[10]
[11] This is in contrast to the paper published in the current issue of this journal,
which found caecum to be the most common site.[5] SEC may be multifocal in about 15 to 20% cases, although it was noted in 59% cases
in one study.[6]
[7]
[10]
[11] No association has been demonstrated between SEC and endoscopic or histological
inflammation.[7]
[11] A disproportionately higher number of cases with primary sclerosing cholangitis
has also been observed among cases with SEC.[7]
[10]
[11] An important consideration is the endoscopic visibility of lesions. It is apparent
from the available data that SEC occurs in both normal-appearing mucosa as well as
in areas with nodularity, etc.[9]
[10]
[11] Hence, restricting this entity to endoscopically normal areas (definition 1) will
result in missed lesions and underestimation of prevalence.
Table 1
Summary of findings from studies on serrated epithelial changes in inflammatory bowel
disease
|
Reference
Study design
Country
|
No. of participants
Demographic profile
Disease duration
Follow-up period
|
Definition and characteristics of SEC
|
Risk factors of SEC
Dysplasia
Comparison with controls
Other details
|
|
I. UC and CD combined
|
|
Bahceci et al, American Journal of Surgical Pathology. 2024;48(10):1326–1334.
Retrospective case–control study
USA and New Zealand
|
Cases
46 IBD patients with SEC and dysplasia
UC-74%
CD-26%
Mean age, years (range): 58 (26–76)
Males: 61%
Mean disease duration, years (range): 23 (< 1–52)
Mean follow-up period, years (range): 4.5 (< 1–22.8)
|
Controls
45 IBD patients with HP and dysplasia
UC-80%
CD-20%
Mean age, years (range): 55 (22–80)
Males: 60%
Mean disease duration, years (range): 20 (< 1–59)
Mean follow-up period, years (range): 3.4 (< 1–17.8)
|
Definition 1
Location:
Left colon: 48%, rectum: 31%
Multifocal SEC: 59%
Concomitant PSC: 4%
|
No difference between SEC and the control groups regarding age, gender distribution,
IBD phenotype, presence of PSC, and location
SEC vs. controls
Dysplasias in left colon: 41% vs. 26%, p = 0.028
High-risk subtype of dysplasia: 18% vs. 0%, p < 0.001
Advanced neoplasia: 24% vs. 0%,
p < 0.001
Location concordance between dysplasia and SEC or HP: 76% vs. 49%, p = 0.007
|
|
Batts et al, Postgraduate Medicine. 2021 Jan 2;133(1):66–70.
Retrospective case–control study
USA
|
Cases
94 IBD patients with SEC
UC- Major fraction
Mean age, years (SD): 52.8 (12.3)
Males: 58.5%
Mean disease duration, years (SD): 20.7 (10.9)
Follow-up period: 1–4 years
|
Controls
187 IBD patients without SEC
UC- Major fraction
Mean age, years (SD): 52.9 (13.8)
Males: 55.1%
Mean disease duration, years (SD): 18.2 (10.6)
Follow-up period: 1–4 years
|
Definition 1, but also included invisible SSL-like lesions
No data available regarding the characteristics of SEC
|
No difference between the SEC group and the control group with respect to age, gender
distribution, disease duration, disease extent, activity and medications
SEC vs. controls
Dysplasia: 12.8% vs. 4.3%
Odds ratio: 2.53 (p = 0.081)
|
|
Parian et al, Gastrointestinal Endoscopy. 2016 Jul;84(1):87–95.e1.
Retrospective study
USA
|
187 patients with IBD and SEC
UC- 52.4%
CD- 39%
IBDU- 8.6%
Mean age, years (SD): 48.4 (15)
Males: 57.8%
Mean disease duration, years (SD): 16 (10.9)
Median follow-up period, years (range): 2.3 (0–14.3)
|
Definition 2
Prevalence: 3%
Location: Rectum: 37.8%, left colon: 37.1%
Multifocal SEC: 20.3%
Concomitant PSC: 6.4%
Extensive disease: 53.5%
Visible lesion and targeted biopsy: 17.6%
Nontargeted biopsy: 71.7%
|
Synchronous dysplasia: 8%
Metachronous dysplasia: 21%
HGD and cancer: 6% (17/1000 patient-years)
Location concordance between dysplasia and SEC: 68%
Older age at IBD diagnosis, male gender, family history of CRC, and SEC at follow-up
colonoscopy increased the risk of dysplasia
|
|
Johnson et al, Aliment Pharmacol Ther. 2014 Jun;39(12):1408–1417.
Retrospective case–control study
USA
|
Cases
79 IBD patients with SEC
UC-84%
CD-16%
Median age, years (IQR): 57 (48–64.75)
Males: 62%
Median disease duration, years (IQR): 18 (10.5–29)
|
Controls
100 IBD patients without SEC
(Baseline characters were not mentioned as a whole, rather stratified as per the presence
or absence of prior or synchronous CRN)
|
Definition 1
Detection rate: 10/1000 colonoscopies
(95% CI, 8-14)
Extensive disease: 77%
Concomitant PSC: 29%
Targeted biopsy: 38%
|
SEC patients were older, had longer disease duration and more frequent association
with PSC
SEC vs. controls
Cumulative incidence of subsequent CRN at 3 years: 30% vs. 9%, p = 0.047. Significance was lost following stratification for prior or synchronous
SEC
Location concordance between SEC and dysplasia: 45%
|
|
II. UC
|
|
|
Bahceci et al, American Journal of Surgical Pathology. 2024 Jun;48(6):719–725.
Retrospective case–control study
USA
|
Cases
28 UC patients with SEC
Mean age, years (range): 47 (24–91)
Males: 43%
Mean disease duration, years (range): 13 (3–26)
|
Controls
51 UC patients without SEC
Mean age, years (range): 47 (19–75)
Males: 47%
Mean disease duration, years (range): 13 (1–35)
|
Definition 1
Location:
Left colon: 86%
Multifocal SEC: 14%
Concomitant PSC: 4%
Pancolitis: 75%
|
No potential risk factor could be identified that predicted the development of SEC
No difference between the histological inflammatory scores between the SEC and non-SEC
groups
|
|
Parian et al, Inflammatory Bowel Diseases. 2021 Aug 19;27(9):1475–1481.
Retrospective case–control study
USA
|
Cases
98 UC patients with SEC
Mean age, years (SD): 48.7 (14.4)
Males: 60.2%
Mean disease duration, years (SD): 16.8 (10.7)
Mean follow-up period, years (SD): 2.9 (2.9)
|
Controls
98 matched UC patients without SEC (matched for age, disease duration, and extent)
Mean age, years (SD): 48 (14.8)
Males: 44.9%
Mean disease duration, years (SD): 15.4 (9.3)
Mean follow-up period, years (SD): 3.1 (3.9)
|
Definition 2
Location: Rectum: 52%
Sigmoid colon: 37.8%
Multifocal SEC: 22.4%
Concomitant PSC: 10.2%
Nontargeted biopsy: 66.3%
|
SEC vs. non-SEC
Neoplasia: 26.5% vs. 3.1%, p < 0.001
HGD or CRC: 11.2% vs. 2%, p = 0.02
SEC was a strong predictor of neoplasia in UC patients (odds ratio, 11.4; 95% CI,
2.5–51.98; p < 0.01)
|
Abbreviations: CD, Crohn's disease; CI, confidence interval; CRC, colorectal cancer;
CRN, colorectal neoplasia; HGD, high-grade dysplasia; HP, hyperplastic polyps; IBD,
inflammatory bowel disease; IBDU, inflammatory bowel disease unclassified; IQR, interquartile
range; PSC, primary sclerosing cholangitis; SD, standard deviation; SEC, serrated
epithelial change; SSL, sessile serrated lesion; UC, ulcerative colitis.
The clinical significance of SEC lies in its reported association with dysplasia and
the future risk of CRC. The risk of dysplasia, including high-risk dysplasia, more
than doubles in the presence of SEC.[6]
[8]
[11] A systematic review and meta-analysis including three studies with 195 IBD patients
with SEC found a significantly higher risk of neoplasia compared with the non-SEC
group (relative risk 4.11, 95% confidence interval 2.23–7.58, p < 0.001).[11] Older age, male gender, and family history of CRC increase the risk of dysplasia.[10] A higher risk of both synchronous and metachronous dysplasia has been observed.[10] The concordance between the location of SEC and dysplasia ranges from 45 to 75%.[6]
[9]
[10] This further supports the link between SEC and dysplasia. The future risk of CRC
also appears to be increased, but this outcome is infrequently reported.[9]
[10] The risk of dysplasia in the paper by Vaitiekunas et al cannot be estimated due
to the lack of a control group, and there were no cases of CRC.[5] Only one patient was found to have dysplasia, and the frequency appears to be lower
than the studies included in [Table 1].
While the awareness of SEC is increasing, there are several issues that need to be
addressed. These include identification of risk factors for SEC, the strategy to improve
their endoscopic detection, formulation of standard histopathological criteria for
diagnosis, appropriate follow-up protocol in those with visible and invisible lesions,
etc.[4] Studies are also needed from different parts of the world, as the current data are
generally from the West. The epidemiology of CRC varies globally, and how this impacts
SEC requires confirmation by data from different regions of the world. Prospective,
multicentric, registry-based data may help to understand the characteristics and natural
history of SEC more clearly and guide surveillance strategy.
