SEARCH

SEARCH BY CITATION

Keywords:

  • inflammatory bowel diseases;
  • colorectal neoplasm;
  • ulcerative colitis;
  • Crohn's disease;
  • histopathology;
  • risk factors

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Background: The histological variability in colitis-associated colorectal cancer (CRC in inflammatory bowel disease [IBD]) and the association to clinical factors is unknown.

Methods: In population-based material including 67 patients with CRC in IBD, histopathology of the cancers and tissue samples from different colorectal localizations were reevaluated, and relationships to clinical factors analyzed.

Results: Forty-three of 60 patients (75%) showed dysplasia in the colorectum apart from the cancer, while 17 (25%) had no dysplasia at cancer diagnosis. Mean age at onset of IBD was 22 years in patients with and 34 years in patients without dysplasia (P = 0.01). The mean duration of colitis-CRC interval was 21 years in patients with and 16 years in patients without dysplasia (P = 0.02). The latter group included all patients with a colitis-CRC interval <10 years. Active inflammation was more likely to occur in patients with dysplasia (odds ratio [OR] 4.2). The 2 groups were not discriminated by gender, family history of CRC or IBD, diagnosis of PSC, medical treatment, active symptoms, or histological features like type of cancer and differentiation. In multiple logistic regression analysis the age at onset of IBD was the strongest predictive variable for dysplasia at cancer diagnosis (P = 0.025).

Conclusions: Widespread neoplasia occurs in the majority of cases with CRC in IBD and is associated with early onset of IBD. Localized neoplasia occurs in about a quarter of the patients and shows an association with late-onset IBD. The 2 groups probably represent different pathogenetic entities of neoplasia in IBD. This might have consequences for surveillance strategies.

(Inflamm Bowel Dis 2008)

Inflammatory bowel disease (IBD) is associated with an increased risk of colorectal cancer (CRC).1, 2 Colonoscopic screening programs in IBD have not shown the desired efficacy in early detection of cancer in IBD.3 Previous studies have focused on identifying risk factors for colitis-associated cancer.4–10 Possible subgroups of CRC in IBD that might need different surveillance strategies have not been identified yet. Few studies have analyzed the clinical and histological variability within the group of CRC in IBD.11–13 The duration of IBD prior to the diagnosis of cancer varies considerably.14 Twenty percent of CRC in IBD occurs within 10 years from diagnosis of IBD.14 Early CRC is associated with late onset of IBD.15 A varying, but relatively high, proportion of mucinous carcinomas and well-differentiated carcinomas in IBD have been reported.16, 17 Widespread dysplasia, i.e., in mucosa localized distant from the cancer, occurs in 75% of CRC in IBD. In 25% of the patients, dysplasia in the colorectal mucosa apart from the cancer is lacking.17–21 Relationships between histological and clinical factors within the group of CRC in IBD have, to our knowledge, not been investigated.

The aim of the present study was to analyze the histological variability of colitis-associated CRC, including tissue samples from different localizations of the colorectum in a cohort that is representative for this group of patients in Norway. We also wanted to analyze whether the histological factors are associated with clinical parameters as a basis for finding better surveillance strategies for IBD patients.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Recruitment of Patients

The study population was recruited as reported previously.14 In short, 74 patients with CRC in IBD from 3 university hospitals in Oslo were identified by matching 9090 patients with possible IBD against the CRC files of the Cancer Registry of Norway and by reevaluating clinical data and histological sections. Six patients were excluded because CRC was diagnosed prior to the confirmation of IBD and 1 because follow-up data were missing. Thus, 67 patients with confirmed CRC in either ulcerative colitis (UC) or Crohn's disease (CD), including 12 patients with primary sclerosing cholangitis, were included for further analysis.

Data Collection and Processing

The clinical data were collected as described previously.14 For the present study the analysis of the formalin-fixed, paraffin-embedded material included subclassification of the adenocarcinomas and the examination of histological changes in the mucosa distant from the tumor. All histological slides that did not contain adenocarcinoma were defined as distant to the tumor. Colonoscopic biopsies from the time of the diagnosis of the CRC and tissue slides from the surgical specimens were analyzed. The median time between endoscopy and the surgical resection was 13 days (range 1–182 days). When original slides were not available for reevaluation, new slides were cut from the existing paraffin blocks. All slides were evaluated by 1 experienced pathologist in a blinded fashion. When the reevaluation was divergent from the one documented in the original pathology report, a consensus was achieved by consulting a third experienced pathologist.

The morphology of the mucosal tissue was described as either flat, adenoma/polypoid, or as a tumor mass. Flat mucosa was defined as tissue separate from hyperplastic polyps, adenomas, tumor masses, or other elevated lesions. We did not use the definition “dysplasia associated lesion or mass” (DALM) because a part of the material originated from a period before this diagnosis was established. In concordance with the World Health Organization (WHO),22 the histological type of adenocarcinoma was recorded as ordinary adenocarcinoma, mucinous, or signet-ring cell carcinoma. The cellular differentiation was graded as poor, moderate, or well. The histological stage of the cancer was recorded as defined by the TNM classification,23 rating the depth of infiltration into the gut wall (T1-T4), spread to regional lymph nodes (N0-N2), and diagnosis of distant metastasis (M0-M1), summarized as Stage I (T1 N0 M0; T2 N0 M0), Stage II (T3 N0 M0; T4 N0 M0), Stage III (any T, N1-2, M0), and Stage IV (any T, any N, M1). The presence of Crohn's-like lymphocytic infiltration of the cancers was determined as described by Jass,24 but was not graded in subgroups. Grade of dysplasia was categorized as indefinite, low-grade, high-grade dysplasia, or intramucosal carcinoma, according to Riddell et al.25 The highest grade of dysplasia within the examined slide was recorded and the highest grade of dysplasia in each patient was selected for statistical analysis and recoded as: 1 = no dysplasia, 2 = indefinite/low-grade dysplasia, and 3 = high-grade dysplasia/intramucosal carcinoma. We also noted the presence of villous hypermucinous mucosa as described by Morson and Jass: villous surface with a high number of goblet cells, but without nuclear atypia.26, 27 Inflammation was recorded as chronic and/or active as published by Rutter et al.28 Thus, active inflammation was categorized as mild (cryptitis but no crypt abscesses), moderate (a few crypt abscesses), or severe (numerous crypt abscesses in the epithelium). The duration of extensive colitis prior to CRC was determined by review of the clinical journals and defined as the time interval from the date when extensive distribution was first registered during course of disease to the date of the diagnosis of CRC.

Completeness of Data

A total of 915 paraffin blocks from 67 patients with CRC in IBD were analyzed. In 60 patients colorectal mucosa distant from the tumor was available for review (median 10 blocks, range 1–30 blocks, sum 641 blocks). In 7 patients only tissue from the tumor had been sampled. The diagnosis of IBD was confirmed in earlier biopsies in these patients. Histological staging of the CRC was possible for 61 patients. In 6 patients resection of the bowel was not performed, mainly because of disseminated cancer or multimorbidity of the patient. The date of onset of IBD symptoms was known in 58 patients. In the other 9 patients the age at onset of IBD symptoms and the colitis-CRC interval could not be calculated.

Data Analysis and Statistics

Statistical analyses were performed using SPSS 14.0 for Windows (Chicago, IL). Explanatory factors of distant dysplasia at cancer diagnosis were investigated by univariate analysis and multiple logistic regressions. Groups were compared using independent t-tests or 1-way ANOVA for continuous outcome variables and Pearson chi-square or Fisher's Exact test for categorical outcome variables. Bonferroni correction was used for comparisons of groups with more than 2 categories. All mean values are presented with standard error of the mean (SEM). In multiple logistic regression analysis a forward variable selection procedure was used to find the model that best predicted distant dysplasia at cancer diagnosis.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Histological Characterization of the Adenocarcinomas in the Cohort

Seventy-seven adenocarcinomas were diagnosed among the 67 patients. Sixty-three (82%) presented as a tumor mass, 12 (16%) as an adenoma/polypoid lesion, and 2 (3%) were found in flat mucosa. One of the 2 patients with adenocarcinoma in flat mucosa had a synchronous tumor mass, and the other had a large villous adenoma with high-grade dysplasia at another site of the colon. Twenty adenocarcinomas (26%) were located in the cecum, 8 (10%) in the ascending colon, 6 (8%) at the hepatic flexure, 12 (16%) in the transverse colon, 4 (5%) at the splenic flexure, 5 (7%) in the descending colon, 8 (10%) in the sigmoid, and 14 (18%) in the rectum (Fig. 1). Sixty-seven of 77 (87%) were classified as ordinary adenocarcinomas, 8 of 77 (10%) as mucinous-, and 2 of 77 (3%) as signet-ring cell carcinomas. Of the ordinary adenocarcinomas, 5 (7%) were well, 50 (75%) moderately, and 12 (18%) poorly differentiated. All mucinous and signet-ring cell carcinomas were defined as poorly differentiated (Table 1). Crohn's-like lymphocytic infiltration (CLLI) was diagnosed in 7 of 77 (9%) carcinomas. In 4 tumors the tissue sample was too small for CLLI to be evaluated. Fifty-nine patients (88%) had 1 and 8 patients (12%) had multiple cancers in the colorectum (6 patients had 2 and 2 patients 3 synchronous CRCs). Five out of 8 patients with synchronous CRC had tumors in both the right and the left side of the large intestine. One patient showed different types of adenocarcinomas and 3 patients had a different grade of differentiation in the synchronous tumors (Table 2). Stage I was diagnosed in 11, Stage II in 17, Stage III in 20, and Stage IV in 13 patients.

thumbnail image

Figure 1. Localization and type of lesion of 77 adenocarcinomas in 67 patients with CRC in IBD.

Download figure to PowerPoint

Table 1. Type of Adenocarcinoma and Grade of Differentiation of All Cancers in the Cohort in Relation to the Morphological Appearance
Type of LesionType of AdenocarcinomaTotal
Ordinary AdenocarcinomaMucinous CarcinomabSignet Ring Cell Carcinomab
WellaModeratelyaPoorlya
  • a

    Grade of tumor differentiation.

  • b

    By definition poorly differentiated.

Tumor mass437128263
Adenoma/polypoid11100012
Flat mucosa020002
Total550128277
Table 2. Clinical Data and the Histological Characterization of the Tumors in Patients with Synchronous CRC in IBD
CasesGenderType of IBDType of LesionSize of LesionLocalization of TumorType of AdenocarcinomaGrade of DifferentiationDepth of Infiltration
  • a

    Polypoid mucosa around all of the circumference of this location, diameter not measured.

  • b

    Polypoid mucosa in a small area of this location, diameter not measured.

1MaleUCTumor mass8×6cmCecumOrdinaryModerateT3
   Tumor mass4×4cmAscendingOrdinaryModerateT1
2MaleUCTumor mass3×3cmCecumOrdinaryModerateT3
   flat mucosaHepatic flexureOrdinaryModerateT2
3MaleUCTumor mass3.5×3.5cmSplenic flexureOrdinaryPoorT3
   Tumor massUnknownHepatic flexureOrdinaryModerateT3
4MaleUCTumor massaDescendingOrdinaryModerate
   Tumor massbHepatic flexureOrdinaryPoorT3
5MaleUCAdenoma3×3×2cmSigmoidOrdinaryModerateT2
   AdenomaUnknownCecumOrdinaryModerateT1
6MaleUCTumor mass2.5×3.5cmSigmoidOrdinaryModerateT3
   Adenoma2.5×3cmAscendingOrdinaryModerateT2
7MaleUCTumor mass15×5cmTranverseMucinousT3
   Tumor mass6.5×5cmSigmoidMucinousT3
   Tumor mass4×4cmTranverseOrdinaryHighT1
8MaleUCTumor mass7.5×7cmRectumOrdinaryPoorT2
   Tumor mass3.5×5cmRectumOrdinaryModerateT3
   Tumor mass3×3cmSigmoidOrdinaryPoorT3

Histological Characterization of the Colorectal Mucosa Distant from the Cancer

Dysplasia in flat colorectal mucosa in at least 1 of the 8 segments of the colorectum was found in 43 of 60 patients: 5 indefinite, 20 low-grade, 15 high-grade dysplasia, and 3 intramucosal carcinomas. In 17 patients (25%) distant dysplasia was not detected at the time of the diagnosis of the cancer. In these patients the pathology reports from the time prior to the diagnosis of CRC were reviewed. Two of 17 patients had once been diagnosed with low-grade dysplasia, but the findings were not confirmed in any of the colonoscopies during follow-up until the diagnosis of cancer (2 years and 10 years follow-up, respectively). A third patient showed alternating none and low-grade dysplasia, and once high-grade dysplasia during 5 years prior to the diagnosis of CRC. This patient, age 20 years at onset of IBD symptoms, was the only one with synchronous CRC in the group of patients without distant dysplasia. In the remaining 14 patients no dysplasia had ever been detected during the course of disease (1.5 colonoscopies [median], including a total of 19.5 biopsies [median]).

Villous hypermucinous mucosa was found in 10/60 patients. In all patients it was concomitant with other grades of dysplasia: indefinite in 3, low-grade in 1, high-grade in 3, and intramucosal carcinoma in 3 patients. All but 1 patient of the cohort showed chronic inflammation in the colorectum at the time of diagnosis of CRC. Active inflammation was present in 36/60 patients, mild in 3/36, moderate in 14/36, and severe in 19/36. In the remaining 24/60 patients no active inflammation was diagnosed.

Association Between Clinicohistological Factors and Distant Dysplasia

Age at Onset of IBD Symptoms

Mean age at onset of IBD symptoms was 22 years (SEM = 2.1 years) in patients with and 34 years (SEM = 4.5 years) in patients without distant dysplasia at the time of the diagnosis of the cancer (P = 0.01) (Fig. 2). Mean age at onset of IBD symptoms was similar in patients with indefinite/low-grade dysplasia and patients with high-grade dysplasia/intramucosal carcinoma (22 years, SEM = 2.9 years and 23 years, SEM = 3.2 years, respectively, P = 0.7).

thumbnail image

Figure 2. Age at onset of IBD symptoms in patients with and patients without distant dysplasia at diagnosis of CRC.

Download figure to PowerPoint

Duration of the Colitis-CRC Interval

Mean duration of the colitis-CRC interval was 21 years in patients with (SEM = 1.2 years) and 16 years (SEM = 1.9 years) in patients without distant dysplasia (P = 0.02) (Fig. 3). The latter group included the 3 patients with a colitis-CRC interval <10 years from onset of IBD symptoms. Mean duration of the colitis-CRC interval was 23 years (SEM = 1.5 years) in patients with indefinite/low-grade dysplasia and 18 years (SEM = 1.7 years) in patients with high-grade dysplasia/intramucosal carcinoma. By aid of Bonferroni correction we found a significant difference between the group of “no dysplasia” and “indefinite/low-grade dysplasia” (P = 0.013), but not between the group of “no dysplasia” and “high-grade dysplasia/intramucosal carcinoma” (P = 1) or between the group of “indefinite/low-grade dysplasia” and “high-grade dysplasia/intramucosal carcinoma” (P = 0.14).

thumbnail image

Figure 3. Duration of the colitis-CRC interval in patients with and patients without distant dysplasia at diagnosis of CRC.

Download figure to PowerPoint

Active Inflammation

The distribution of active inflammation in relationship to the presence of distant dysplasia at the time of the diagnosis of CRC is shown in Table 3. Of 43 patients with distant dysplasia, 30 had active inflammation and 13 did not. In 17 patients without dysplasia, 6 showed active inflammation and 11 did not. Active inflammation was more likely to occur in the group of patients with distant dysplasia (odds ratio [OR] = 4.2, 95% confidence interval [CI]: 1.3–13.9). Mean age at onset of IBD symptoms was 23 years (SEM = 2.3 years) in patients with and 30 years (SEM = 3.8 years) in patients without active inflammation (P = 0.11). Mean time from onset of IBD symptoms to cancer was 20 years in both groups.

Table 3. Distribution of the Grade of Active Inflammation in the Groups of Patients with and without Distant Dysplasia at the Time of Diagnosis of CRC in IBD
 Grade of Active InflammationTotal
NoneMildModerateSevereSubtotal
No distant dysplasia11141617
Distant dysplasia present13210183043
Total24314193660
Other Clinical or Histological Factors

The 2 groups were not discriminated by gender (P = 0.62), family history of CRC (P = 0.38) or IBD (P = 0.66), a diagnosis of PSC (P = 0.53), treatment with sulfasalazine (P = 0.81), mesalazine (P = 0.13), or prednisolone (P = 0.62) 2 years prior to the diagnosis of CRC, and the presence of active symptoms 2 years prior to CRC (P = 0.21) or histological features like the type of cancer (ordinary adenocarcinoma versus mucinous/signet-ring cell carcinoma, P = 0.17), the grade of differentiation (P = 0.97), the localization (right versus left colon, P = 0.46), or the stage of the cancer (stage I/II versus III/IV, P = 0.5). Seven of 8 patients with synchronous CRC had distant dysplasia, 1 had no dysplasia between the 2 cancers. The distribution of the type of IBD was not analyzed due to the small number of patients with CD (6 patients with CD and 54 with UC).

All but 2 patients, 1 with CD and 1 with UC, either presented with substantial or pancolitis at the diagnosis of IBD, or developed extensive disease during the course of IBD. The mean duration of extensive colitis prior to CRC was 13 years (SEM = 1 year) in 36 patients with distant dysplasia and 10 years (SEM = 2 years) in 12 patients without distant dysplasia (P = 0.23). In 10 patients the date of the diagnosis of extensive colitis prior to the cancer could not be determined.

Multiple Logistic Regression Analysis

Multiple logistic regression analyses were performed with distant dysplasia as the dependent variable and different subsets of age at onset of IBD symptoms, the duration of the colitis-CRC interval, and active inflammation at cancer diagnosis as possible predictors of distant dysplasia. By using the forward selection procedure, age at onset of IBD symptoms as a single factor was the strongest predictive factor for distant dysplasia. The OR for distant dysplasia for a 10-year increase in age at onset of IBD symptoms was 0.61, 95% CI: 0.58–0.64, P = 0.025. In this model, duration of the colitis-CRC interval was close to statistically significant associated (P = 0.09) while active inflammation was not associated with distant dysplasia (P = 0.16).

Factors Investigating a Possible Detection or Intervention Bias

The mean number of colonoscopies per patient per year from diagnosis of IBD to cancer was 0.4 in patients with indefinite/low-grade dysplasia as well as in patients with high-grade dysplasia/intramucosal carcinoma, and 0.7 in patients without dysplasia at cancer diagnosis (P = 0.4).

On average, 13 tissue sections distant from the tumor were reviewed in patients with distant dysplasia and 10 sections in patients without dysplasia (P = 0.06).

Dysplasia surrounded by active inflammation in the same section was found in 129/641 (20%) sections: 89/641 (14%) with mild or moderately active inflammation and 40/641 (6%) with severely active inflammation.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

This study partly confirms the paradigm of the pathobiology of CRC in IBD, that neoplasia is often multifocal, widespread, and may occur in flat mucosa. Twelve percent of our patients experienced synchronous cancers with a heterogeneous histological appearance of the cancers. A relatively large proportion was diagnosed with dysplasia distant from the tumor. The cancers were evenly distributed throughout the colorectum. Infiltrating adenocarcinoma in flat mucosa of the colorectum occurred in 3% of our patients.

Our study, however, partly challenges the view that this model can be applied to all patients with CRC in IBD. In our cohort, 17/60 patients (25%) did not show dysplasia in colorectal mucosa distant from the cancer at the time of the diagnosis of CRC. We were concerned that these findings were influenced by a detection bias, as slightly more histological sections were reviewed in patients with distant dysplasia. Almost all patients without distant dysplasia at cancer diagnosis, however, had also been free of dysplasia prior to the cancer. Only 1 patient without distant dysplasia at cancer diagnosis had repeatedly been diagnosed with dysplasia at earlier examinations. This patient developed synchronous CRC and the absence of dysplasia in mucosa between the 2 tumors might be due to a sampling error. The findings in other studies are similar to our results. The frequency of patients without distant dysplasia was reported as 27% by Ransohoff et al,20 26% by Taylor et al,19 and 18% by Connell et al.18 We conclude that the absence of dysplasia distant from the cancer at the time of the diagnosis of CRC in our material is not only due to sampling errors, but in fact represents a specific subgroup of CRC in IBD.

Ransohoff et al20 and Taylor et al19 did not analyze and Connell et al18 did not find differences in the clinical or pathological records between the 2 groups of patients. In our cohort, age at onset of IBD was significantly associated with distant dysplasia. Low age at onset of IBD symptoms, average 22 years, predicted the presence of distant dysplasia, and high age, average 34 years, the absence of distant dysplasia at cancer diagnosis. Consequently, the mean duration of the colitis-CRC interval was shortest in patients without distant dysplasia. This group showed a considerable variation in the colitis-CRC interval but included all patients who experienced CRC in less than 10 years from onset of IBD. It may be suggested that additional risk factors related to age are important for the development of malignancy in these patients. The group of patients with distant low-grade dysplasia showed a longer colitis-CRC interval compared with patients with distant high-grade dysplasia. Although not statistically significant, this might reflect the well-established malignant potential of high-grade dysplasia in IBD.3

It is unlikely that dysplasia is biased by intervention in our cohort. There was no significant difference in the mean number of colonoscopies per year between the dysplasia and the nondysplasia group. A challenge in the evaluation of dysplasia are epithelial changes that appear to exceed the limits of ordinary regeneration but are insufficient for an unequivocal diagnosis of dysplasia (indefinite dysplasia).25 Five patients in our cohort were diagnosed with a subgroup of indefinite dysplasia, termed “indefinite, probably positive for dysplasia” by Riddell et al,25 as we evaluated the histological findings to be more likely, but not unequivocally, dysplastic and less likely secondary to other factors like mucosal regeneration or processing techniques. We therefore concatenated indefinite dysplasia with low-grade dysplasia for statistical analysis. However, by omitting these patients with indefinite dysplasia from analysis the main results did not change significantly (data not shown).

Another concern in the evaluation of dysplasia is concomitant active inflammation. In the present study, though, the number of histological sections in which the diagnosis of dysplasia was challenged by the presence of severe active inflammation was low (6%). Active inflammation has been recognized as a risk factor for CRC in IBD.28, 29 In our cohort, active inflammation at cancer diagnosis was more likely present in patients with distant dysplasia and showed a tendency to young age at onset of IBD. Despite being a cross-sectional variable in our cohort, the strong association indicates that active inflammation has impact on patients with distant dysplasia but less on patients without distant dysplasia.

In multiple logistic regression analysis, duration of the colitis-CRC interval and diagnosis of active inflammation were no longer significantly predictive for distant dysplasia. Age at onset of IBD was the strongest predictive variable for distant dysplasia.

As a consequence, a pattern of 2 distinct subgroups of CRC in IBD emerges. One group consisting of patients with early onset of IBD, associated with distant dysplasia, represents the traditional model of malignant development in IBD, characterized by widespread neoplasia and active inflammation as well as high-grade dysplasia as a risk factor for malignancy.25, 28–31 This group comprises the majority of cases in our cohort. The other group, defined by a higher age at onset of IBD, with neoplasia unrelated to active inflammation and limited to a localized area of the colorectum, is characterized by an age-related risk, similar to sporadic CRC. Our hypothesis is that patients with late onset of IBD, similar to the general population, accumulate malignant risk factors until the onset of IBD. After the onset of disease, immunogenic factors related to IBD accelerate carcinogenesis in these patients, resulting in dysplasia and cancer at a single location before long-standing inflammation eventually is able to cause multifocal dysplasia. This would explain the short time from onset of IBD to cancer in this group of patients and possibly the fact that some patients are being diagnosed with CRC even before the onset of IBD.

A limitation of the present study is the retrospective design. In a prospective study, one would have sampled more biopsies in a standardized pattern throughout the colorectum. On the other hand, a prospective study would not be feasible due to the large timescale of malignant development in IBD. The strength of the study is based on the fact that our cohort is most likely representative of the general IBD population. Every person with residency in Norway is uniquely identified by an 11-digit national identity number. By this number the Cancer Registry of Norway has recorded all patients who have been diagnosed with any cancer, anywhere in Norway, since the 1950s. All study patients were recruited by the Cancer Registry of Norway from a large cohort of patients that had at least had 1 consultation during their course of IBD in 1 of 3 university hospitals in Oslo. As there are virtually no private care facilities in Norway, the number of patients with IBD who have not experienced at least 1 consultation in a hospital during their course of disease is assumed to be small. Thus, our study patients were recruited unrelated to the activity of the disease, the participation in surveillance programs, or migration within Norway. This might be the reason why the number of mucinous and well-differentiated adenocarcinomas showed a much lower frequency in our study material compared to other cohorts.17, 18

The results of the present study might have consequences for surveillance strategies. As dysplasia is not always endoscopically visible, it takes 33 or more random biopsies to have 90% confidence of finding dysplasia.32 Detection of dysplasia is enhanced by chromoendoscopy.33, 34 Therefore, at least 33 random biopsies or the use of chromoendoscopy or both may be considered for cancer screening in IBD, at least in patients with onset of IBD symptoms at a higher age. Otherwise, one might overlook the localized neoplasia in these patients. One should also bear in mind that CRC might occur in less than 10 years from onset of disease in this group of IBD patients.

In conclusion, widespread and multifocal neoplasia occurs in the majority of cases with CRC in IBD and is associated with early onset of IBD. However, localized neoplasia may occur in about a quarter of the patients and shows an association with later onset of IBD. The 2 groups may represent different pathogenetic entities of neoplasia in IBD. Our results might have consequences for surveillance strategies.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank the Department of Pathology at Aker University hospital, Oslo, for providing paraffin-embedded tissue of the patients recruited at Aker University Hospital, Blefjell sykehus HF, Diakonhjemmets sykehus AS, Glittreklinikken, Haukeland Universitetssykehus/Helse Bergen HF, Helse Fonna HF, Helse Nordmøre og Romsdal HF, Helse Sunnmøre – Ålesund Sykehus, Lovisenberg Diakonale Sykehus AS, Nordlandssykehuset HF, Ringerike sykehus HF, Ski sykehus HF, St Olavs Hospital HF, Stavanger Universitetssjukehus/Helse Stavanger HF, Sykehuset Asker-og Bærum HF, Sykehuset Buskerud HF, Sykehuset Innlandet HF, Sykehuset Telemark HF, Sykehuset Østfold HF, Sørlandet sykehus HF, Ullevål Universitetssykehus HF, Universitetssykehuset Nord-Norge HF UNN, Volvat Medisinske Senter Vestfold AS, as well as Dr. Truls H Brekke, Dr. Kai Heier, Dr. Odd Bjarne Karlsen, Dr. Tom Vøyvik, Dr. Bjarne Ørbeck-Nilsen for providing patient data, Dr. Lars Gustav Lyckander for valuable help in evaluating the histological material, head engineer Anita Fjellum for construction of the database, and cand.scient.Gunn Seim-Ekeland for assistance in collection of material. The study was approved by the regional committee for medical research ethics of Eastern Norway, Oslo, Norway.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut. 2001; 48: 526535.
  • 2
    Jess T, Gamborg M, Matzen P, et al. Increased risk of intestinal cancer in Crohn's disease: a meta-analysis of population-based cohort studies. Am J Gastroenterol. 2005; 100: 27242729.
    Direct Link:
  • 3
    Collins PD, Mpofu C, Watson AJ, et al. Strategies for detecting colon cancer and/or dysplasia in patients with inflammatory bowel disease. Cochrane Database Syst Rev. 2006:CD000279.
  • 4
    Ekbom A, Helmick C, Zack M, et al. Ulcerative colitis and colorectal cancer. A population-based study. N Engl J Med. 1990; 323: 12281233.
  • 5
    Ekbom A, Helmick C, Zack, M, et al. Increased risk of large-bowel cancer in Crohn's disease with colonic involvement. Lancet. 1990; 336: 357359.
  • 6
    Langholz E, Munkholm P, Davidsen M, et al. Colorectal cancer risk and mortality in patients with ulcerative colitis. Gastroenterology. 1992; 103: 14441451.
  • 7
    Rutegard JN, Ahsgren LR, Janunger KG. Ulcerative colitis. Colorectal cancer risk in an unselected population. Ann Surg. 1988; 208: 721724.
  • 8
    Lashner BA, Silverstein MD, Hanauer SB. Hazard rates for dysplasia and cancer in ulcerative colitis. Results from a surveillance program. Dig Dis Sci. 1989; 34: 15361541.
  • 9
    Jess T, Loftus EV, Jr., Velayos FS, et al. Risk of intestinal cancer in inflammatory bowel disease: a population-based study from Olmsted county, Minnesota. Gastroenterology. 2006; 130: 10391046.
  • 10
    Bernstein CN, Blanchard JF, Kliewer E, et al. Cancer risk in patients with inflammatory bowel disease: a population-based study. Cancer. 2001; 91: 854862.
  • 11
    Rubio CA. Serrated neoplasias and de novo carcinomas in ulcerative colitis: a histological study in colectomy specimens. J Gastroenterol Hepatol. 2007; 22: 10241031.
  • 12
    Sugita A, Sachar DB, Bodian C, et al. Colorectal cancer in ulcerative colitis. Influence of anatomical extent and age at onset on colitis-cancer interval. Gut. 1991; 32: 167169.
  • 13
    Gyde SN, Prior P, Allan RN, et al. Colorectal cancer in ulcerative colitis: a cohort study of primary referrals from three centres. Gut. 1988; 29: 206217.
  • 14
    Brackmann SA, Andersen SN, Aamodt G, Langmark F, Clausen OPF, Aadland E, Fausa O, Rydning A, Vatn MH. Time to CRC in IBD. Relationship between clinical parameters and the colitis-CRC interval in a cohort of patients with CRC in IBD. Scand J Gastroenterol. 2008; 43: accepted for publication.
  • 15
    Karvellas CJ, Fedorak RN, Hanson J, et al. Increased risk of colorectal cancer in ulcerative colitis patients diagnosed after 40 years of age. Can J Gastroenterol. 2007; 21: 443446.
  • 16
    Choi PM, Zelig MP. Similarity of colorectal cancer in Crohn's disease and ulcerative colitis: implications for carcinogenesis and prevention. Gut. 1994; 35: 950954.
  • 17
    Svrcek M, Cosnes J, Beaugerie L, et al. Colorectal neoplasia in Crohn's colitis: a retrospective comparative study with ulcerative colitis. Histopathology. 2007; 50: 574583.
  • 18
    Connell WR, Talbot IC, Harpaz N, et al. Clinicopathological characteristics of colorectal carcinoma complicating ulcerative colitis. Gut. 1994; 35: 14191423.
  • 19
    Taylor BA, Pemberton JH, Carpenter HA, et al. Dysplasia in chronic ulcerative colitis: implications for colonoscopic surveillance. Dis Colon Rectum. 1992; 35: 950956.
  • 20
    Ransohoff DF, Riddell RH, Levin B. Ulcerative colitis and colonic cancer. Problems in assessing the diagnostic usefulness of mucosal dysplasia. Dis Colon Rectum. 1985; 28: 383388.
  • 21
    von Herbay A, Herfarth C, Otto HF. Cancer and dysplasia in ulcerative colitis: a histologic study of 301 surgical specimen. Z Gastroenterol. 1994; 32: 382388.
  • 22
    Aaltonen LA, Hamilton SR, World Health Organization, et al. Pathology and genetics of tumours of the digestive system. World Health Organization classification of tumours. Lyon, Oxford: IARC Press, Oxford University Press (distributor); 2000.
  • 23
    Sobin LH, Wittekind C, International Union against Cancer. TNM classification of malignant tumours. 6th ed. New York: Wiley-Liss; 2002.
  • 24
    Jass JR. Lymphocytic infiltration and survival in rectal cancer. J Clin Pathol. 1986; 39: 585589.
  • 25
    Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol. 1983; 14: 931968.
  • 26
    Morson BC, Jass JR, Sobin LH. Precancerous lesions of the gastrointestinal tract: a histological classification. London; Philadelphia: Baillière Tindall; 1985.
  • 27
    Andersen SN, Lovig T, Clausen OP, et al. Villous, hypermucinous mucosa in long standing ulcerative colitis shows high frequency of K-ras mutations. Gut. 1999; 45: 686692.
  • 28
    Rutter M, Sauders B, Wilkinson K, et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology. 2004; 126: 451459.
  • 29
    Gupta RB, Harpaz N, Itzkowitz S, et al. Histologic inflammation is a risk factor for progression to colorectal neoplasia in ulcerative colitis: a cohort study. Gastroenterology. 2007; 133: 10991105; quiz 1340–1341.
  • 30
    Rabinovitch PS, Dziadon S, Brentnall TA, et al. Pancolonic chromosomal instability precedes dysplasia and cancer in ulcerative colitis. Cancer Res. 1999; 59: 51485153.
  • 31
    Issa JP, Ahuja N, Toyota M, et al. Accelerated age-related CpG island methylation in ulcerative colitis. Cancer Res. 2001; 61: 35733577.
  • 32
    Rubin CE, Haggitt RC, Burmer GC, et al. DNA aneuploidy in colonic biopsies predicts future development of dysplasia in ulcerative colitis. Gastroenterology. 1992; 103: 16111620.
  • 33
    Kiesslich R, Fritsch J, Holtmann M, et al. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology. 2003; 124: 880888.
  • 34
    Rutter MD, Saunders BP, Schofield G, et al. Pancolonic indigo carmine dye spraying for the detection of dysplasia in ulcerative colitis. Gut. 2004; 53: 256260.