Risk factors for colorectal cancer in patients with inflammatory bowel disease undergoing surveillance: a prospective cohort study

Both ulcerative colitis and Crohn*s colitis are associated with an increased
risk of developing colorectal cancer (CRC). It is assumed that the increased
risk is caused by the exposure of the colonic mucosa to chronic inflammation,
which leads to colorectal cancer via the inflammation-dysplasia-colorectal
cancer sequence. Currently, colitis patients are advised to undergo colonic
surveillance to detect dysplasia or asymptomatic cancer. Solid evidence for the
effectiveness of this strategy is lacking however, as there are no prospective
randomized controlled trials available.
Although the increased risk of CRC in colitis patients is well established,
several studies show that the risk varies widely between patients, depending
on the presence of risk factors such as extent, duration and severity of
inflammation, a concomitant diagnosis of primary sclerosing cholangitis (PSC)
or a positive family history of CRC. Recently, several of these risk factors
were implemented in the updated British guidelines for surveillance. The new
guideline recommends stratification of patients in a high, medium or low risk
group depending on the presence of risk factors and to adjust the surveillance
interval accordingly. However, the risk factors implemented in the guideline
are solely based on retrospective case control studies, and may be subjected to
major limitations such as publication or selection bias. Prospective data
regarding the phenotype and genotype associated with an increased risk of CRC
is important to further optimize surveillance in the future. A promising
variable in this respect might be mucosal healing. From a recent study it
appears that a macroscopically normal mucosa is associated with a risk of CRC
comparable to the background population. Since medication is the cornerstone in
achieving mucosal healing in IBD patients, the effect of medication use on the
risk of developing IBD-associated CRC is of interest. Indeed several
case-control studies have shown that regular use of 5-aminosalicylic acid
(5-ASA) is able to reduce the risk of IACC. Recent data from mouse models
suggests that anti-TNF* compounds might have a chemopreventive effect as well,
although clinical data is virtually non existent.
Effectiveness of surveillance relies on the ability to accurately detect
pre-neoplastic lesions with high positive and negative predictive values for
progression to CRC. In clinical practice, this has been proven to be a huge
challenge. First, endoscopical detection of pre-neoplastic lesions or dysplasia
is difficult and typically requires the taking of random biopsies throughout
the colon as dysplasia can be present in flat mucosa. From various studies, we
know that most gastroenterologists do not adhere to recommended protocols or
guidelines, which results in a decreased sensitivity for dysplasia. Second,
(mis) interpretation of pathologists when grading dysplasia has been shown to
seriously hamper the usefulness of this tool. Finally, no consensus exists
regarding the consequences of the presence of dysplasia, especially in case of
low grade dysplasia. Therefore, the search for straightforward new markers
reliably identifying patients at risk for malignant transformation to CRC is of
great importance for improving the effectiveness of surveillance. Several
studies on this subject showed promising results, in particular for the markers
P53 and K-RAS in colonic biopsies. However, most studies performed thus far
have compared expression of tumormarkers between IBD-patients with and without
colorectal cancer in a cross-sectional design. Whether expression of tumor
markers can accurately predict development of dysplasia or CRC during follow-up
in a setting of periodic surveillance is currently unknown. Furthermore, it
can be hypothesized that tumor markers from blood and faeces can detect
patients at risk of developing CRC as well, potentially obviating the use of
time-consuming and invasive procedures like endoscopies.
As described above, the increased risk for CRC in IBD patients is assumed to be
caused by exposure of the colonic mucosa to chronic inflammation. Mucosal T
cells are key players in maintaining barrier function and controlling the
delicate balance between immune activation and immune tolerance and aberrant
function of gut T cells are thought to play an important role in (progression
of) IBD pathogenesis. The gut environment tightly regulates differentiation,
activation and function of mucosal T cells resulting in a unique pallet of
(regulatory) T cells. Besides immune regulation by the well-described CD4+
Foxp3 regulatory T cells (Treg), a novel mechanism used by CD4+ T cells to
avoid excessive activation in the gut in mice was recently described, resulting
in intestinal CD4+CD8aa T cells with regulatory properties. A recent study
characterized these CD4+CD8aa T cells with regulatory properties in the lamina
propria of the human intestine, and IBD patients were found to have decreased
amounts of these lymphocytes in both the gut and periphery. Defining the
precise phenotypic and functional properties of effector and regulatory T cells
(including specifically the emerging T regulatory subsets) involved in IBD is
seen as an important challenge in IBD research, especially the translation of
concepts from experimental models to the clinical setting (one of the
priorities as defined by the Crohn*s and Colitis Association of America). Our
hypothesis is that CD4+CD8aa Treg are the intestinal *tissue equivalent* of the
systemically well described FoxP3 Treg and play a crucial role in gut immune
homeostasis. Analyzing the different regulatory T cell populations present in
the blood and tissue of IBD patients therefore has the potential to provide
clues for designing therapies that restore immune tolerance and prevent
development of chronic inflammation resulting in CRC in IBD patients.

1. To confirm established and identify new risk factors for colorectal cancer
in a prospective cohort of IBD patients undergoing regular surveillance.
Dysplasia or colorectal cancer will be the primary outcome.
2. To provide evidence that mucosal healing results in a significant reduction
of colorectal dysplasia/neoplasia in IBD patients and that this is associated
with 5-ASA or anti-TNF maintenance therapy.
3. Study the expression of several tumor markers in biopsies, blood and faeces
at baseline and determine whether expression of these markers can predict
dysplasia or colorectal cancer development during follow-up.
4. Determine phenotypical, functional and molecular characteristics of human
intestinal and peripheral blood (regulatory) T cell subsets involved in chronic
inflammation, dysplasia and development of CRC in IBD patients, including
investigating potential associations between development of CRC and presence or
absence of specific (regulatory) T cell subsets.

Patients with a confirmed diagnosis of ulcerative colitis, Crohn's colitis or
indeterminate colitis with an indication for surveillance according to the
current guidelines will be asked to participate in this study by their treating
physician . If patients are interested, patients will receive oral and written
information about the study by the investigator. If patients are interested in
participating in the study, a colonoscopy will be performed after signing of
the informed consent. During the endoscopy, 12 extra biopsies will be taken for
analysis of tumormarker expression. Before the colonoscopy, 20 ml of blood will
be drawn. Furthermore, patients will be asked to fill out a questionnaire
regarding diet, potential risk factors for CRC and medication use. Patients wil
be asked to collect a sample of feces and send this to the laboratory.
After the first colonoscopy, patients will be stratified in a high, medium and
low risk group according the the current British guidelines for surveillance
depending on the presence of clinical or endoscopical risk factors in
accordance with the guidelines. During a follow-up time of 5 years all patients
will receive surveillance employing the intervals described by the British
guidelines. During each follow-up surveillance colonoscopy 20 ml of blood, 12
extra biopsies and a sample of feces will be collected. Furthermore, patients
will have to fill out a questionnaire regarding medication use.

- Patients will be contacted by phone before inclusion to provide information
about the study.
- All surveillance colonoscopies in the current study are part of the regular
CRC surveillance program. Therefore, these colonoscopies pose no additional
burdon or risk.
- During all surveillance colonoscopies within the study period, 12 additional
biopsies will be taken, which prolongs the examination with a few minutes and
gives a smal risk of bleeding.
- Before all surveillance colonoscopies within the study period, 20 ml of blood
will be drawn.
- Before the first surveillance colonoscopy, patients will be asked to fill out
a questionnaire regarding diet, potential risk factors for CRC and medication
use. Patients will be asked to fill out the questionnaire regarding medication
use again at each surveillance colonoscopy.
- Before all surveillance colonoscopies within the study period, patients will
be asked to collect a sample of faeces at home and bring this along to the
colonoscopy.