van Schaik, Fiona D.M. MD*; Mooiweer, Erik MD*; van der Have, Mike MD*; Belderbos, Tim D.G. MD*; ten Kate, Fiebo J.W. MD, PhD†; Offerhaus, G. Johan A. MD, PhD†; Schipper, Marguerite E.I. MD, PhD†; Dijkstra, Gerard MD, PhD‡; Pierik, Marieke MD, PhD§; Stokkers, Pieter C.F. MD, PhD‖; Ponsioen, Cyriel MD, PhD‖; de Jong, Dirk J. MD, PhD¶; Hommes, Daniel W. MD, PhD**; van Bodegraven, Ad A. MD, PhD††; Siersema, Peter D. MD, PhD*; van Oijen, Martijn G.H. PhD*; Oldenburg, Bas MD, PhD*; on behalf of the Dutch Initiative on Crohn Colitis (ICC)
It is commonly accepted that both Crohn's colitis and ulcerative colitis (UC) are associated with an increased risk of colorectal cancer (CRC).1 The cumulative risk was found to be 18% after 30 years of disease duration in patients with UC and similar risks have been reported in patients with Crohn's colitis.2,3 Colitis-associated CRC is thought to be preceded by dysplasia, which can be found in flat mucosa or in endoscopically visible lesions. For the latter the term dysplasia-associated lesion or mass (DALM) was coined in 1981.4 DALMs are heterogeneous in their endoscopic appearance and are therefore subdivided into adenoma-like and nonadenoma-like lesions. The nonadenoma-like subtype refers to all irregular, diffuse masses or plaque lesions that generally cannot be removed by endoscopic resection. These lesions were reported to be frequently accompanied by synchronous malignancy, and guidelines therefore generally recommend colectomy, although data supporting these guidelines are limited.5,6 The adenoma-like DALM refers to all discrete, either pedunculated or sessile polyps that resemble sporadic adenomas in patients without IBD. Previous studies reported that CRC risk is low in these patients and that polypectomy with regular surveillance is an adequate treatment strategy.7,8 Although these initial studies were based on small patient numbers with limited follow-up, two larger studies published more recently seemed to confirm these data.9,10 However, little is known about the additional risk of adenomas in IBD patients as compared to IBD patients without an adenoma and patients with an adenoma but without IBD.
The aim of this study was therefore to describe the risk of subsequent colorectal neoplasia in a large cohort of IBD patients with an adenoma, and to compare this risk to that in IBD subjects without adenomas as well as non-IBD patients with adenomas.
PATIENTS AND METHODS
The nationwide pathology archive (PALGA) that contains all pathology reports from the Netherlands dating back to 1971 was used to identify three cohorts of patients: 1) IBD patients with an adenoma (IBD + adenoma); 2) IBD patients without an adenoma (IBD-nonadenoma); and 3) subjects with an adenoma but without IBD (nonIBD + adenoma). A PALGA search was performed in seven university medical centers for the period between 1995 and 2005.
IBD-nonadenoma patients were matched to IBD + adenoma patients for age at inclusion and center. Year of inclusion was matched to the year of adenoma diagnosis in the IBD + adenoma patients. The latter was performed in order to obtain an equal follow-up time between these groups. Exclusion criteria were a history of any dysplasia (low-grade dysplasia [LGD] or high-grade dysplasia [HGD]) or CRC, a diagnosis of HGD or CRC at the moment of adenoma diagnosis, a history of (sub)total colectomy, no endoscopic or surgical follow-up, and incomplete or missing data. Furthermore, patients with Crohn's disease (CD) without involvement of the colon were excluded.
Demographic and clinical data were collected from medical charts for all three cohorts. For the IBD + adenoma and IBD-nonadenoma cohorts, clinical data included date of IBD diagnosis, type of IBD, extent of IBD, medication use, and the presence of primary sclerosing cholangitis (PSC). Disease extent was defined as the maximum extent according to either histology or endoscopy reports. In UC and indeterminate colitis (IC) or IBD-unclassified (IBD-U) patients disease extent was defined as either left-sided or extensive (inflammation distal or proximal to the splenic flexure, respectively). In patients with Crohn's colitis, involvement of three or more anatomical parts of the colon was considered extensive disease, whereas involvement of one or two sections was considered limited disease. In all three cohorts a family history of CRC was documented. For the IBD + adenoma and nonIBD + adenoma cohorts, histopathology and endoscopy reports were reviewed to identify the date of first adenoma diagnosis and to collect information about size, location within or outside an area of previous inflammation, and endoscopic appearance. We selected patients based on their histological adenoma diagnosis. However, in the endoscopy reports we encountered a wide variety of endoscopic descriptions of adenomas. Therefore, for practical purposes, adenomas were classified as either “typical” or “atypical.” Typical adenomas included all lesions described as discrete solitary, sessile, or pedunculated polyps resembling sporadic adenomas. All other endoscopic lesions, including adenomatous fields (i.e., areas of multiple, clustered polyps), lesions with an irregular surface and lesions endoscopically described as postinflammatory polyps (but histologically classified as adenoma) were characterized as atypical adenomas. Dysplasia was classified as either low-grade or high-grade, according to the criteria and definitions articulated by Riddell et al.11
Histopathology and endoscopy reports were reviewed to detect whether patients developed LGD, HGD, or CRC during follow-up. Advanced neoplasia (AN) was defined as a finding of HGD or CRC. Duration of follow-up was measured in months and defined as time from the first adenoma diagnosis (IBD + adenoma and nonIBD + adenoma cohorts) or from the moment of inclusion in the study (IBD-nonadenoma cohort) to one of the following endpoints: 1) end of follow-up; 2) end of study period (1st of December, 2010); 3) death; or 4) subtotal or total colectomy.
All analyses were restricted to the period beyond the first 6 months of follow-up to exclude patients with prevalent AN.
Baseline characteristics were analyzed with standard descriptive statistics and compared between the three groups. Continuous variables were analyzed using analysis of variance (ANOVA) or Kruskal–Wallis analysis, where appropriate. Categorical variables were analyzed using Pearson's chi-squared or Fisher's exact test, where appropriate. Five-year cumulative incidences of AN were calculated using Kaplan–Meier survival analysis and comparisons between groups were made using log-rank testing. Patients who did not develop AN during follow-up were censored at the moment of last colonoscopy or colectomy. Factors associated with the development of AN during follow-up in IBD patients (IBD + adenoma and IBD-nonadenoma patients) were assessed in a Cox proportional hazard model. P < 0.05 was considered statistically significant. Statistical analysis was performed using SPSS 15.0 for Windows (Chicago, IL).
This study was carried out with the approval of and in accordance with the ethical guidelines of the Research Review Committee of our institution.12
Our search yielded 617 IBD + adenoma patients, 472 IBD-nonadenoma patients, and 902 nonIBD + adenoma patients. Review of medical charts, endoscopy, pathology, and surgery reports yielded 110 IBD + adenoma, 179 nonIBD + adenoma and 123 IBD-nonadenoma patients eligible for enrolment. The reasons for exclusion are shown in the flowchart (Fig. 1). Clinical characteristics of the three patient groups are given in Table 1.
A total of 216 adenomas were identified in 179 non-IBD + adenoma patients and 133 adenomas in 110 IBD + adenoma patients. Adenoma characteristics are given in Table 2. Adenomas were characterized as typical in 82 (75%) and atypical in 28 IBD + adenoma patients (25%). The latter consisted of 15 lesions endoscopically classified as post-inflammatory polyps (54%) and 13 nonadenoma-like lesions (46%). In nonIBD + adenoma patients eight adenomas (4%) were characterized as atypical and 171 (96%) as typical.
Polypectomy was performed in 146 nonIBD + adenoma patients (82%) and in 68 IBD + adenoma patients (62%) (P <0.01). Median adenoma size was 5 mm (range 2–20) in the IBD + adenoma cohort and 6 mm (range 1–65) in the nonIBD + adenoma cohort (P = 0.11). There were no significant differences between IBD + adenoma and nonIBD + adenoma patients regarding adenoma location or architecture. In 21 IBD + adenoma patients (19%) adenomas were located outside the maximum endoscopic or histological extent of inflammation.
Development of Neoplasia in IBD + Adenoma Patients During Follow-up
In 36 IBD + adenoma patients (33%) adenomas were detected during follow-up, which were endoscopically classified as typical in 29 (81%) and atypical in seven patients (19%). Four other patients (4%) developed flat LGD and in two patients (2%) both flat and raised dysplasia was detected.
AN was diagnosed in 18 IBD + adenoma patients (16%) after a median follow-up of 53 months (range 7–86) (Fig. 1). Twelve patients developed HGD and six developed CRC, of which three were preceded by HGD. In 10 patients HGD was detected in typical adenomas, which were treated by polypectomy in seven and by subtotal colectomy in two patients. In one of these patients, CRC was detected in the resection specimen. One patient with HGD in a large typical adenoma was not treated with endoscopic or surgical resection due to advanced age. HGD was diagnosed in atypical lesions in three patients and in flat mucosa in two. Of these five patients, three were treated by colectomy. HGD was confirmed in the colectomy specimen of one patient while in the other two patients CRC was detected. In the remaining two patients no polypectomy or colectomy was performed due to advanced age and because the diagnosis of HGD could not be confirmed during follow-up colonoscopies. In 89% of patients AN was located in the colonic segment where the baseline adenoma was situated. Two patients, of whom one was originally treated by polypectomy and one was not, developed AN in another colonic segment.
The 5-year cumulative incidence of AN in the IBD + adenoma cohort was 11% (Fig. 2). This was similar in patients treated with polypectomy and patients in whom the adenoma was only biopsied (11% vs. 10%, respectively, P = 0.99). Patients with an atypical adenoma at baseline had a higher risk of developing AN compared to patients with a typical adenoma (Fig. 3). The 5-year cumulative incidence of AN was not different between patients with an adenoma located inside and patients with an adenoma outside an area of inflammation (10% and 19%, respectively, P = 0.91). No differences were found between CD and UC patients regarding development of AN (P = 0.11)
Development of Neoplasia in IBD-nonadenoma Patients During Follow-up
Within the IBD-nonadenoma cohort 15 patients (12%) developed adenomas during follow-up, which were endoscopically classified as typical adenomas in 10 patients and atypical adenomas in five. Two patients developed flat LGD and four developed both flat and raised dysplasia.
A total of seven patients (7%) developed AN after a median duration of follow-up of 80 months (range 18-91 months; Fig. 1). Three patients developed HGD in a typical adenoma. Of these, two were treat by polypectomy. None of these during follow-up. Four patients developed CRC. One of these developed CRC 4 years after an endoscopically resected flat adenoma, whereas three developed CRC without evidence of prior LGD or HGD. The 5-year cumulative incidence of AN was 3% in the IBD-nonadenoma cohort (Fig. 2). This was significantly lower compared to the IBD + adenoma cohort (P <0.01). No differences were found between CD and UC patients regarding development of AN (P = 0.12). The 5-year cumulative incidence of CRC alone was not significantly different between the IBD-nonadenoma and IBD + adenoma groups (2% vs. 4%, respectively, P = 0.23; Fig. 2b).
Factors Associated with Development of AN
The presence of an adenoma in patients with IBD was associated with an increased risk of developing AN during follow-up (hazard ratio [HR] 3.6, 95% confidence interval [CI] 1.5–8.7) (Table 3). In the multivariate analysis, this effect remained borderline significant after adjustment for duration and type of IBD, extent of inflammation, age, gender, concomitant diagnosis of PSC, and medication use (HR 2.8, 95% CI 1.0–8.2). A diagnosis of UC (compared to CD) was associated with an increased risk of developing AN as well (unadjusted HR 6.1, 95% CI 1.4–26.4), which remained significant in the multivariate analysis (adjusted HR 4.5, 95% CI 1.0–25.4). Medication use, including 5-aminosalicylate (5-ASA) and thiopurines, was not associated with the development of AN.
Development of Neoplasia in NonIBD + Adenoma Patients During Follow-up
In 77 nonIBD + adenoma patients (42%) adenomas were detected during follow-up. A total of eight patients (4%) developed AN after a median follow-up of 53 months (range 17-99 months, Fig. 1). In five patients HGD was detected in an adenoma that was treated endoscopically in all five patients. None of these patients developed CRC during follow-up. Three nonIBD + adenoma patients developed CRC, which were all treated by surgical resection.
The 5-year cumulative incidence of AN in the non-IBD + adenoma cohort was 5% (Fig. 2). This was significantly lower compared to the IBD + adenoma cohort (P <0.01), as well as to the subset of IBD + adenoma patients with a typical adenoma resembling sporadic adenomas (P = 0.03). The 5-year cumulative incidence of CRC alone in the nonIBD + adenoma cohort did not differ from the IBD + adenoma cohort (1% vs. 3% respectively, P = 0.06, Fig. 2b).
This study demonstrates an increased risk of AN in IBD patients with an adenoma when compared with non-IBD patients with adenomas and IBD patients without adenomas. Notably, atypical as well as typical adenomas in patients with IBD were found to be associated with an increased risk of AN. This finding is in contrast with most previously published studies, which consistently reported a low risk of AN in IBD patients with polypoid dysplasia.8,10,13,14
The high incidence of AN among IBD + adenoma patients during follow-up in our study was mainly caused by the development of HGD rather than the development of CRC. HGD developed in 12% of IBD + adenoma patients as compared to 2% and 3% of IBD-nonadenoma and nonIBD + adenoma patients, whereas CRC developed in 6% of IBD + adenoma patients compared to 3% and 2% of IBD-nonadenoma and nonIBD + adenoma patients, respectively. Although previous studies primarily focused on CRC incidence, the reported incidences of HGD after a diagnosis of adenoma were low in these studies and varied between 0% and 5%.8–10,15 Since HGD will undoubtedly progress to CRC when left untreated, we opted to use a composite endpoint including both HGD and CRC.16–18 Furthermore, we report a substantially higher CRC incidence in IBD + adenoma patients than reported in several previous studies.8,10,13
One might assume that the inclusion of both patients treated by polypectomy and patients in whom adenomas were not endoscopically removed accounts for the higher risk of AN in our cohort. Eighty-one percent of all nonIBD + adenoma patients were treated by polypectomy, compared to only 62% of IBD + adenoma patients. An explanation for the relatively low polypectomy rate in the IBD + adenoma cohort might be that postinflammatory-like and small flat lesions are more frequently encountered and not deemed serious enough for polypectomy. Our rates are in line with recent data from Vieth et al,9 who found a polypectomy rate of 59% in IBD patients with adenomas, although we could not confirm the higher incidence of neoplasia and CRC among patients in whom no polypectomy was performed as reported in that study. It is therefore conceivable that AN developed either as a result of field cancerization or due to residual dysplastic tissue.
Another possible explanation for the high incidence of AN might be the design of our study. We selected cases using a pathology database, whereas in previous studies selection was based on the endoscopic identification of solitary polyps.8,10,13 Therefore, our study comprised a heterogeneous group of endoscopic lesions including a large subgroup of “nonadenoma-like” lesions, which have previously been associated with a high risk of synchronous CRC.4,19–21 Current guidelines therefore generally recommend colectomy for these types of lesions.5,6 In our IBD + adenoma cohort, 25% of patients had an atypical adenoma, including both nonadenoma-like lesions and lesions considered to be postinflammatory polyps during endoscopy. The cumulative incidence of AN was substantially higher in these patients: eight patients with an atypical adenoma (29%) developed AN compared to 10 patients (12%) with a typical adenoma. Interestingly, in 15 patients histologically diagnosed adenomas were endoscopically characterized as postinflammatory polyps. Of these, 30% developed AN during follow-up, which highlights the difficulty of differentiation between raised dysplasia and postinflammatory polyps. Inexperience of the endoscopists who assessed the types of polyps may have contributed to this misclassification. Another explanation may be the fact that postinflammatory polyps are proxies for long-standing and severe inflammation and may therefore be associated with an increased risk of CRC.22,23
Of note, IBD + adenoma patients with typical adenomas displayed a significantly increased 5-year cumulative incidence of advanced neoplasia as well: 6% compared to 5% in nonIBD + adenoma patients (P = 0.03, log rank test). This conflicts with data from previous studies that reported no difference in development of subsequent neoplasia between these patient groups.7,13 Our results stress the importance of complete removal and close follow-up of either type of adenoma.
The IBD + adenoma cohort contained patients with adenomas detected both inside and outside colonic areas (previously) involved in inflammation. Since by definition IBD-associated dysplasia only develops in areas of chronic inflammation, one should classify adenomas outside an area of previous inflammation as sporadic. Remarkably, the incidence of AN was not different between IBD patients with adenomas detected within or outside an area of previous inflammation.
The comparison of the IBD + adenoma and IBD-nonadenoma cohorts enabled us to assess the additional risk of adenomas in patients with colitis. IBD + adenoma patients had a higher risk of AN compared to IBD-nonadenoma patients. The presence of an adenoma seemed an independent predictor of AN when corrected for several known risk factors of IBD-associated neoplasia, althoughborderline significance (HR 3.0, 95% CI 1.0–8.5).
Our study has some limitations. First, due to the retrospective design of the study we relied on the descriptions provided in endoscopy reports with regard to endoscopic characteristics and treatment of the adenomas. Second, IBD + adenoma patients and IBD-nonadenoma patients were solely matched on age at inclusion, center, and date of inclusion according to the date of adenoma diagnosis in the IBD + adenoma cohort. We do not think that this introduced important bias, however, since these cohorts were quite well matched with regard to duration and extent of IBD. Third, a significant proportion of patients were excluded due to the lack of follow-up or missing endoscopy or pathology reports. Since patients in the nonIBD + adenoma and IBD-nonadenoma cohort were more frequently excluded than patients in the IBD + adenoma cohort, this might have resulted in a higher risk of AN in these two cohorts, thereby introducing a selection bias. Obviously, this would only strengthen our conclusion that the risk of AN is increased in the IBD + adenoma patients. And last, although patients underwent a complete colonoscopy at inclusion in this study and at least one colonoscopy or a colectomy during follow-up, surveillance colonoscopies with random biopsy sampling were performed in only 52% of IBD + adenoma patients and 55% of IBD-nonadenoma patients. This might have resulted in an underestimation of the presence of AN. However, since most cases of AN can be identified endoscopically, we do not feel that this had a major influence on our results.
In conclusion, this study shows that IBD patients with a histological diagnosis of an adenoma have an increased risk of developing AN compared to nonIBD + adenoma patients and IBD-nonadenoma patients. The presence of atypical adenomas in particular was associated with this increased risk, although patients with typical adenomas were found to carry an additional risk as well. Thus, complete removal of adenomas and subsequent strict surveillance is warranted in IBD patients with atypical as well as typical colonic adenomas.
1. Bernstein CN, Blanchard JF, Kliewer E, et al.. Cancer risk in patients with inflammatory bowel disease: a population-based study. Cancer. 2001;91:854–862.
2. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut. 2001;48:526–535.
3. Gillen CD, Walmsley RS, Prior P, et al.. Ulcerative colitis and Crohn's disease: a comparison of the colorectal cancer risk in extensive colitis. Gut. 1994;35:1590–1592.
4. Blackstone MO, Riddell RH, Rogers BH, et al.. Dysplasia-associated lesion or mass (DALM) detected by colonoscopy in long-standing ulcerative colitis: an indication for colectomy. Gastroenterology. 1981;80:366–374.
5. Farraye FA, Odze RD, Eaden J, et al.. AGA medical position statement on the diagnosis and management of colorectal neoplasia in inflammatory bowel disease. Gastroenterology. 2010;138:738–745.
6. 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;59:666–689.
7. Engelsgjerd M, Farraye FA, Odze RD. Polypectomy may be adequate treatment for adenoma-like dysplastic lesions in chronic ulcerative colitis. Gastroenterology. 1999;117:1288–1294.
8. Rubin PH, Friedman S, Harpaz N, et al.. Colonoscopic polypectomy in chronic colitis: conservative management after endoscopic resection of dysplastic polyps. Gastroenterology. 1999;117:1295–1300.
9. Vieth M, Behrens H, Stolte M. Sporadic adenoma in ulcerative colitis: endoscopic resection is an adequate treatment. Gut. 2006;55:1151–1155.
10. Kisiel JB, Loftus EV Jr, Harmsen WS, et al.. Outcome of sporadic adenomas and adenoma-like dysplasia in patients with ulcerative colitis undergoing polypectomy. Inflamm Bowel Dis. 2012;18:226–235.
11. Riddell RH, Goldman H, Ransohoff DF, et al.. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol. 1983;14:931–968.
12. van Diest PJ. No consent should be needed for using leftover body material for scientific purposes. BMJ. 2002;325:648–651.
13. Odze RD, Farraye FA, Hecht JL, et al.. Long-term follow-up after polypectomy treatment for adenoma-like dysplastic lesions in ulcerative colitis. Clin Gastroenterol Hepatol. 2004;2:534–541.
14. Torres C, Antonioli D, Odze RD. Polypoid dysplasia and adenomas in inflammatory bowel disease: a clinical, pathologic, and follow-up study of 89 polyps from 59 patients. Am J Surg Pathol. 1998;22:275–284.
15. Jess T, Loftus EV Jr, Velayos FS, et al.. Incidence and prognosis of colorectal dysplasia in inflammatory bowel disease: a population-based study from Olmsted County, Minnesota. Inflamm Bowel Dis. 2006;12:669–676.
16. Rutter MD, Saunders BP, Wilkinson KH, et al.. Thirty-year analysis of a colonoscopic surveillance program for neoplasia in ulcerative colitis. Gastroenterology. 2006;130:1030–1038.
17. Fuson JA, Farmer RG, Hawk A, et al.. Endoscopic surveillance for cancer in chronic ulcerative colitis. Am J Gastroenterol. 1980;73:120–126.
18. Lennard-Jones JE, Morson BC, Ritchie JK, et al.. Cancer surveillance in ulcerative colitis. Experience over 15 years. Lancet. 1983;2:149–152.
19. Rubio CA, Johansson C, Slezak P, et al.. Villous dysplasia. An ominous histologic sign in colitic patients. Dis Colon Rectum. 1984;27:283–287.
20. Butt JH, Konishi F, Morson BC, et al.. Macroscopic lesions in dysplasia and carcinoma complicating ulcerative colitis. Dig Dis Sci. 1983;28:18–26.
21. Rosenstock E, Farmer RG, Petras R, et al.. Surveillance for colonic carcinoma in ulcerative colitis. Gastroenterology. 1985;89:1342–1346.
22. Rutter MD, Saunders BP, Wilkinson KH, et al.. Cancer surveillance in longstanding ulcerative colitis: endoscopic appearances help predict cancer risk. Gut. 2004;53:1813–1816.
23. Velayos FS, Loftus EV Jr, Jess T, et al.. Predictive and protective factors associated with colorectal cancer in ulcerative colitis: A case-control study. Gastroenterology. 2006;130:1941–1949.
© Crohn's & Colitis Foundation of America, Inc.