ABSTRACT
Surveillance colonoscopy (SC) is considered important for the early detection and treatment of colorectal cancer (CRC) in patients with ulcerative colitis (UC). Here, we investigated whether current guidelines are appropriate in preventing UC patients from being diagnosed with CRC at an incurable stage. Among 1583 patients under treatment for UC, 27 patients were diagnosed with CRC. Of these, we excluded two patients who had not undergone colonoscopy before CRC diagnosis. We then divided the remaining patients into three groups based on colonoscopy interval (A, 1 year or less; B, between 1 and 2 years; and C, 2 years or longer). Fifteen patients had tubular adenocarcinomas, and 10 had other types (8 poorly differentiated adenocarcinomas, 1 mucinous adenocarcinoma, 1 endocrine cell carcinoma). Five (20%) of 25 patients developed CRC within 8 years after the onset of UC, of which one case was detected at stage IV. Six patients were classified into group A, 8 into group B, and 11 into group C. On distribution by histologic type, tubular adenocarcinomas were detected in stages 0 - II in 100% in group A, 100% in group B, and 57.1% in group C. In contrast, other types of carcinomas were detected in stage 0 - II in 100% in group A, 40% in group B, and 0% in group C. Current guideline recommendations for SC are not sufficient for the detection of early stage CRC in patients with UC. SC should be commenced earlier than recommended in the current guidelines and repeated annually.
Key Words: surveillance colonoscopy, ulcerative colitis, colorectal cancer
INTRODUCTION
Colorectal cancer (CRC) was first recognized as a complication of ulcerative colitis (UC) by Crohn and Rosenberg in 1925.1) A meta-analysis of 116 studies reported that the overall prevalence of CRC in any UC patient was 3.7%, and that cumulative probabilities were 2% at 10 years, 8% at 20 years, and 18% at 30 years.2) Another study with a 15-year follow-up period reported that CRC was diagnosed in 9 (1.3%) of 681 patients with inflammatory bowel disease (IBD), including UC and Crohn’s disease patients.3) Accordingly, UC patients are advised to undergo surveillance colonoscopy (SC) aimed at detecting dysplasia or asymptomatic early CRC at a surgically curable stage.4-6)
With regard to surveillance, several institutions in Western countries have independently issued and updated surveillance guidelines aimed at detecting CRC in UC patients, including the British Society of Gastroenterology (BSG), American Gastroenterological Association (AGA), and European Crohn’s Colitis Organization (ECCO).7-9) These guidelines all indicate that SC is required for extensive colitis, i.e. pancolitis and left-sided colitis, but not for proctitis. The BSG and ECCO recommend starting SC approximately 10 years after the onset of colitic symptoms,7,9) while the AGA recommends starting no later than 8 years after the onset of symptoms.8) Guidelines issued by an institution in Japan recommended that SC should be started at 8–10 years after disease onset.10) Overall, these guidelines recommend that SC should be started after 8–10 years of disease in the case of extensive UC. Recommended intervals after the first SC differ, however: the AGA recommends the next SC within 1 to 2 years after the initial SC, and then every 1 to 3 years;8) the Japanese guideline recommends annual or biannual SC after the initial SC;10) while the BSG and ECCO guidelines both recommend yearly, 3-yearly or 5-yearly SC depending on the risk of CRC in individual patients.7,9) This variation in recommended intervals indicates a lack of consensus on the best interval between SC in these patients.7-10)
Recent population-based and referral center cohorts have shown that while the incidence and prevalence UC in Asia is rising, they remain lower than in the West.11) The incidence rate of CRC among UC patients in Japan was reported to differ from those in Western countries.2,3,12) Of note, UC phenotype or disease course might differ according to ethnicity and region, and guidelines from Western countries might not be appropriate for detecting CRC at a curable stage in UC patients in Japan. To date, however, no well-designed clinical study has evaluated SC protocols in Japan or overseas, and the legitimacy of the current SC guidelines for UC patients requires verification.
Here, we retrospectively evaluated patients treated for UC-associated CRC in a network of hospitals in Japan, and aimed to determine whether the current guidelines are appropriate to prevent the development of incurable advanced colorectal cancer in these patients.
METHODS
With approval by the ethics committee of each institution, we retrospectively examined 1583 UC patients experienced at our hospital (Nagoya University Hospital) and three related hospitals (Toyohashi National Hospital, Toyohashi City Hospital, and Yamashita Hospital) from October, 1998 to July, 2011. A total of 27 patients under treatment for UC were diagnosed with CRC, all by colonoscopy. During these colonoscopies, targeted biopsies of suspicious lesions for cancer or dysplasia were performed. Of these 27 patients, we excluded 2 who had not undergone colonoscopy before CRC diagnosis, and investigated the diagnostic course of CRCs associated with UC in the remaining 25 (Table 1). We defined ‘colonoscopy interval’ as the duration between the time when CRC was diagnosed and the time when the last colonoscopy had been performed before CRC diagnosis. For these 25 cases, the former endoscopic findings had shown no suspicious lesions for cancer or dysplasia except for one patient, case no. 5, who had been diagnosed with dysplasia 6 months previously. Fourteen patients had relapsing-remitting disease and 11 had chronic continuous disease. With regard to disease extension, 18 cases were of the pancolitis type, 7 were left-sided colitis, and none were of the proctitis type. We divided the patients into three groups based on CS interval (A, 1 year or less; B, between 1 and 2 years; and C, 2 years or longer) (Table 2), and examined incidence by interval. Patients were staged according to the Union for International Cancer Control’s (UICC) TNM classification.13)
Table 1.
Case no. |
Sex | Age at onset of UC (years) |
Age at cancer diagnosis (years) |
UC duration at cancer diagnosis |
Extent of UC |
Clinical course | Colono-scopy interval | Location | Histologic type | TNM classification (UICC2002) |
---|---|---|---|---|---|---|---|---|---|---|
1 | F | 22 | 30 (1999) |
7y6m | pancolitis | chronic continuing | 5m | R | por | I (T1N0M0) |
2 | F | 49 | 71 (2000) |
22y | pancolitis | relapsing-remitting | 14y | C | por | III B (T4N1M0) |
3 | F | 31 | 46 (2000) |
13y8m | pancolitis | relapsing-remitting | 1m | R | tub | I (T1N0M0) |
4 | F | 36 | 56 (2002) |
20y10m | pancolitis | relapsing-remitting | 4y | R | por | III B (T4N1M0) |
5 | F | 19 | 43 (2003) |
23y10m | pancolitis | chronic continuing | 6m | C | tub | I (T2N0M0) |
6 | F | 16 | 27 (2004) |
11y10m | pancolitis | chronic continuing | 2y | S | ecc | IV (TXNXM1) |
7 | M | 26 | 36 (2005) |
10y10m | pancolitis | chronic continuing | 3y | S | tub | 0 (TisN0M0) |
8 | M | 41 | 70 (2008) |
3y6m | left-sided | relapsing-remitting | 1y | D | tub | 0 (TisN0M0) |
9 | M | 17 | 30 (2009) |
12y10m | pancolitis | relapsing-remitting | 1y5m | T | por | II A (T3N0M0) |
10 | F | 29 | 67 (2009) |
39y3m | pancolitis | chronic continuing | 2m | R | tub | I (T2N0M0) |
11 | F | 30 | 42 (2010) |
11y11m | left-sided | chronic continuing | 1y2m | R | por | III A (T2N1M0) |
12 | F | 54 | 69 (2010) |
15y1m | pancolitis | chronic continuing | 2y | S | tub | I (T1N0M0) |
13 | F | 44 | 49 (2008) |
4y7m | left-sided | relapsing-remitting | 1y | S | tub | I (T1N0M0) |
14 | F | 22 | 49 (2009) |
27y3m | left-sided | chronic continuing | 1y2m | T | tub | I (T1N0M0) |
15 | M | 63 | 71 (2010) |
7y11m | pancolitis | chronic continuing | 1y1m | R | por | I (T1N0M0) |
16 | M | 35 | 47 (2010) |
12y | pancolitis | chronic continuing | 2y | R,D | por | IV (T4N2M1) |
17 | F | 56 | 70 (2003) |
14y3m | pancolitis | relapsing-remitting | – | A | tub | IV (T4N2M1) |
18 | F | 26 | 41 (1999) |
13y7m | pancolitis | chronic continuing | 11y | R | tub | III C (T4N2M0) |
19 | M | 17 | 23 (2004) |
6y | pancolitis | relapsing-remitting | – | R | muc | IV (T4N2M1) |
20 | M | 29 | 38 (2009) |
9y6m | pancolitis | relapsing-remitting | 2y1m | R | tub | 0 (TisN0M0) |
21 | M | 27 | 60 (2004) |
33y3m | pancolitis | relapsing-remitting | – | S,T | por | IV (T4N2M1) |
22 | M | 18 | 47 (2011) |
29y1m | pancolitis | relapsing-remitting | 1y1m | R | tub | 0 (TisN0M0) |
23 | F | 35 | 42 (2001) |
8y1m | left-sided | relapsing-remitting | 5y1m | V | tub | IIA (T3N0M0) |
24 | M | 31 | 60 (2003) |
29y10m | left-sided | relapsing-remitting | 14y | R | tub | IIIA (T3N1M0) |
25 | F | 28 | 58 (2008) |
30y | left-sided | relapsing-remitting | – | S | tub | IIA (T3N0M0) |
Table 2.
Group A | Group B | Group C | Total | ||
---|---|---|---|---|---|
No. of patients Male/female ratio |
6 1:5 |
8 4:4 |
11 5:6 |
25 10:15 |
P=0.433 |
Age at cancer diagnosis (yr) (mean±SD) | 50.8±15.2 | 47.8±15.9 | 50.5±15.3 | 49.7±14.8 | P=0.912 |
UC duration at cancer diagnosis (mo) (mean±SD) | 185±167 | 191±93 | 216±116 | 201±119 | P=0.855 |
Extent | |||||
Pancolitis left-sided |
4 2 |
7 1 |
8 3 |
19 6 |
P=0.659 |
Clinical course | |||||
chronic continuing relapsing-remitting |
3 3 |
6 2 |
2 9 |
11 14 |
P=0.044 |
RESULTS
Among the 25 patients with UC-associated CRC, cancers were detected at stage 0 in 4, stage I in 8, stage II in 3, stage III in 5, and stage IV in 5 (Tables 1, 3). CRC was detected within 8 years after UC onset in five patients, one of whom was at stage IV (Table 1). Histologically, 15 CRCs were tubular adenocarcinomas and 10 were other types of carcinoma (8 poorly differentiated adenocarcinomas, 1 mucinous adenocarcinoma, and 1 endocrine cell carcinoma) (Table 1). Six patients were classified into group A, consisting of one with stage 0 disease and five in stage I (Table 4A). Eight patients were classified into group B, consisting of one patient in stage 0, three in stage I, one each in stages II and III, and two in stage IV. Eleven patients were classified into group C, consisting of two patients each in stages 0 and II, four in stage III, and three in stage IV. Cancers in Stage 0, I, or II, which are generally treated by radical excision, were diagnosed in 100% of patients in group A, versus only 62.5% in group B and 36.4% in group C (P=0.037). Further, on distribution by histologic type, tubular adenocarcinomas were detected in stages 0-II in 100% of patients in groups A and B versus 57.1% in group C (Table 4B). Moreover, other types of carcinoma were detected in stages 0-II in 100% of patients in group A, versus 40% in group B and 0% in group C (Table 4C). Tubular adenocarcinomas were detected at a significantly earlier stage (stages 0-II) than other types of carcinoma (Table 5, P=0.012).
Table 3.
Stage 0 | Stage I | Stage II | Stage III | Stage IV | ||
---|---|---|---|---|---|---|
No. of patients Male/female ratio |
4 4:0 |
8 1:7 |
3 1:2 |
5 1:4 |
5 3:2 |
P=0.026 |
Age at cancer diagnosis (yr) (mean±SD) | 47.8±15.6 | 53.0±14.6 | 43.3±14.0 | 54.0±12.7 | 45.4±20.4 | P=0.795 |
UC duration at cancer diagnosis (mo) (mean±SD) |
159±132 | 209±143 | 203±139 | 236±86 | 185±125 | P=0.920 |
Extent | ||||||
pancolitis left-sided |
3 1 |
7 1 |
1 2 |
3 2 |
5 0 |
P=0.232 |
Clinical course | ||||||
chronic continuing relapsing-remitting |
1 3 |
6 2 |
0 3 |
2 3 |
2 3 |
P=0.207 |
Table 4.
A) All cases | ||||
---|---|---|---|---|
* P=0.037 (χ2 test) | ||||
Group A | Group B | Group C | Total | |
Stage 0 | 1 | 1 | 2 | 4 |
Stage I | 5 | 3 | 8 | |
Stage II | 1 | 2 | 3 | |
Stage III | 1 | 4 | 5 | |
Stage IV | 2 | 3 | 5 | |
discovery rate of Stage 0-II |
100%* | 62.5%* | 36%* | 60% |
B) Tubular adenocarcinoma cases | ||||
---|---|---|---|---|
* P=0.117 (χ2 test) | ||||
Group A | Group B | Group C | Total | |
Stage 0 | 1 | 1 | 2 | 4 |
Stage I | 4 | 2 | 6 | |
Stage II | 2 | 2 | ||
Stage III | 2 | 2 | ||
Stage IV | 1 | 1 | ||
discovery rate of Stage 0-II |
100%* | 100%* | 57%* | 80% |
C) Other types of carcinoma | ||||
---|---|---|---|---|
* P=0.117 (χ2 test) | ||||
Group A | Group B | Group C | Total | |
Stage 0 | 0 | |||
Stage I | 1 | 1 | 2 | |
Stage II | 1 | 1 | ||
Stage III | 1 | 2 | 3 | |
Stage IV | 2 | 2 | 4 | |
discovery rate of Stage 0-II |
100%* | 40%* | 0%* | 30% |
Table 5.
Tubular adenocarcinoma |
Other type of carcinoma |
Total | |
---|---|---|---|
* P=0.012 (χ2 test) | |||
Stage 0 | 4 | 0 | 4 |
Stage I | 6 | 2 | 8 |
Stage II | 2 | 1 | 3 |
Stage III | 2 | 3 | 5 |
Stage IV | 1 | 4 | 5 |
discovery rate of Stage 0-II |
80%* | 30%* | 60% |
DISCUSSION
We investigated 25 patients with UC who developed CRC. Fifteen cases were tubular adenocarcinoma and 10 were other types of carcinoma. Five (20%) of 25 patients developed CRC within 8 years after the onset of UC. Tubular adenocarcinomas were detected in stages 0-II in 100% of patients under surveillance at intervals of 2 years or less. For other types of carcinoma, in contrast, although CRCs were detected in stage 0-II in 100% of patients under annual surveillance, they were detected in stages 0-II in only 40% of those under surveillance intervals of 1 to 2 years. This indicates that an SC interval of 2 years was sufficient to detect CRCs at stages 0-II in UC patients who developed tubular adenocarcinomas, but was not sufficient for those who developed other types of carcinoma.
An association between UC and CRC was first reported in the 1920s.1) Several later studies also described this association, and it is now generally accepted.2,3) The incidence of CRC in UC patients increases with the prolongation of disease course. This cumulative increase in probability appears to justify the start of SC at some point in the course of UC, with repeat SC at appropriate intervals. However, an optimal starting point and intervals have yet to be generally defined. Earlier and more frequent SC increases the probability of detecting CRC at an early stage, but also increases medical workload, and the risk of complications and inconvenience to the UC patient. Eaden et al. reported that the incidence of CRC increased 10 years after the onset of UC,2) and current guidelines recommend starting SC at 8–10 years after the start of colitis or at 10 years after the onset of colitic symptoms.7-10) Specifically, the AGA recommends starting SC at 8–10 years after the onset of colitis;8) the BSG and ECCO recommend that all patients with extensive UC undergo SC approximately 10 years after the onset of colitic symptoms;7,9) and the Japanese guideline recommends starting SC at 8–10 years after UC onset.10) In our study, however, 5 (20%) of 25 patients developed CRC within 8 years after the onset of UC, one case of which was found at stage IV. This indicates that starting SC at 8 years after the onset of UC is not sufficient for all UC patients.
With regard to SC interval, the AGA recommends annual or biannual SC following the first SC for UC patients, and after two negative examinations (no dysplasia or cancer), further SC should be performed every 1 to 3 years.8) In contrast, the BSG and ECCO recommend SC should be conducted yearly, 3-yearly or 5-yearly according to the duration and extent of UC and additional risk factors.7,9) Strikingly, these guidelines recommend that low-risk groups undergo SC every 5 years.7,9) The UICC TNM classification system, the most commonly used staging algorithm for colon cancer, categorizes patients on the basis of three variables (tumor, node, and metastasis);13) when CRC patients are staged according to this classification, it is critical to find CRC at stage 0, I, or II, not only to facilitate curative resection (stage 1, 99.5%; stage II, 97.0%) but also a good 5-year survival rate (stage 0, 94.3%; stage 1, 90.6%; stage II, 81.2%).14) According to the TNM system, although tubular adenocarcinomas were detected in stages 0-II in 100% of our group B patients, other types of carcinomas were detected in stage 0-II in only 40% of this group. Given our current inability to predict which UC patients will develop adenocarcinomas or other types of carcinomas, our data indicate that SC at intervals longer than 1 year is inappropriate for some UC patients.
Reported risk factors for CRC include early-age onset of UC,15,16) family history of CRC,17,18) and primary sclerosing cholangitis.19,20) In our 25 patients, 4 were diagnosed with UC before age 20 years, whereas 12 were diagnosed at 30 years or older. Only 3 of the 25 patients had a family history of CRC, and none had primary sclerosing cholangitis. Although preliminary, these findings suggest that the known risk factors for CRC do not assist in deciding which patients should receive more frequent SC.
Several limitations of this study warrant mention. First, sample size was small. Second, the study was conducted under a retrospective design. A conclusive understanding of the appropriate starting time and interval for SC in UC patients will require prospective studies in larger UC populations. Finally, despite evidence that cancers can be detected at an earlier stage in patients undergoing SC, confirmation of an effect on survival in patients with extensive colitis requires further study.
In conclusion, this study shows that SC at a two-year interval is appropriate for the early detection of tubular adenocarcinomas, but that other types of carcinoma require annual SC. Because we cannot distinguish UC patients by CRC type, any protocol which starts SC later than 8 years after UC onset and repeats it at longer than 1-year intervals is inappropriate. None of the current guidelines is appropriate for detecting all CRCs in UC patients at their respective curable stages. Studies aimed at distinguishing UC patients who will likely develop a specific type of CRC will aid in the establishment of appropriate guidelines for SC for UC patients.
Conflict of Interest and Source of Funding
None declared.
REFERENCES
- 1).Crohn BB, Rosenberg H: The sigmoidoscopic picture of chronic ulcerative colitis (non-specific). Am J Med Sci, 1925; 170: 220–228
- 2).Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut, 2001; 48: 526–535. [DOI] [PMC free article] [PubMed]
- 3).Katsanos KH, Tatsioni A, Pedersen N, Shuhaibar M, Ramirez VH, Politi P, Rombrechts E, Pierik M, Clofent J, Beltrami M, Bodini P, Freitas J, Mouzas I, Fornaciari G, Moum B, Lakatos PL, Vermeire S, Langholz E, Odes S, Morain CO, Stockbrügger R, Munkholm P, Tsianos EV. Cancer in inflammatory bowel disease 15 years after diagnosis in a population-based European Collaborative follow-up study. J Crohns Colitis, 2011; 5: 430–442. [DOI] [PubMed]
- 4).Winawer S, Fletcher R, Rex D, Bond J, Burt R, Ferrucci J, Ganiats T, Levin T, Woolf S, Johnson D, Kirk L, Litin S, Simmang C. Colorectal cancer screening and surveillance: clinical guidelines and rationale. Update based on new evidence. Gastroenterology, 2003; 124: 544–560. [DOI] [PubMed]
- 5).Kornbluth A, Sachar DB. Ulcerative colitis practice guidelines in adults (update):American College of gastroenterology, Practice Parameters Committee. Am J Gastroenterol, 2004; 99: 1371–1385. [DOI] [PubMed]
- 6).Eaden JA, Mayberry JF. Guidelines for screening and surveillance of asymptomatic colorectal cancer in patients with inflammatory bowel disease. Gut, 2002; 51 (Suppl 5): v10–v12 [DOI] [PMC free article] [PubMed]
- 7).Cairns SR, Scholefield JH, Steele RJ, Dunlop MG, Thomas HJ, Evans GD, Eaden JA, Rutter MD, Atkin WP, Saunders BP, Lucassen A, Jenkins P, Fairclough PD, Woodhouse CR. Guidelines for colorectal cancer screening and surveillance in moderate and high risk groups (update from 2002). Gut, 2010; 59: 666–689. [DOI] [PubMed]
- 8).Farraye FA, Odze RD, Eaden J, Itzkowitz SH, McCabe RP, Dassopoulos T, Lewis JD, Ullman TA, James T 3rd, McLeod R, Burgart LJ, Allen J, Brill JV. AGA medical position statement on the diagnosis and management of colorectal neoplasia in inflammatory bowel disease. Gastroenterology, 2010; 138: 738–745. [DOI] [PubMed]
- 9).Annese V, Daperno M, Rutter MD, Amiot A, Bossuyt P, East J, Ferrante M, Götz M, Katsanos KH, Kießlich R, Ordás I, Repici A, Rosa B, Sebastian S, Kucharzik T, Eliakim R. European evidence based consensus for endoscopy in inflammatory bowel disease. J Crohns Colitis, 2013; 7: 982–1018. [DOI] [PubMed]
- 10).Katsuyoshi, Matsuoka, Tsung-Chun Lee, editors. Guidelines for the management of ulcerative colitis in Japan. IBD Research, 2010; 4: 189–239.
- 11).Ananthakrishnan AN. Environmental triggers for inflammatory bowel disease. Curr Gastroenterol Rep, 2013; 15: 302. [DOI] [PMC free article] [PubMed]
- 12).Hirai T, Kato T, Kanemitu Y. Colorectal Cancer in Patients with Inflammatory Bowel Disease in Japan. Stomach and Intestine, 2002; 37: 887–893.
- 13).Sobin LH, Wittekind Ch, editors. TNM classification of malignant tumours. New York: Wiley-Liss, Inc.; 2002.
- 14).Watanabe T, Itabashi M, Shimada Y, Tanaka S, Ito Y, Ajioka Y, Hamaguchi T, Hyodo I, Igarashi M, Ishida H, Ishiguro M, Kanemitsu Y, Kokudo N, Muro K, Ochiai A, Oguchi M, Ohkura Y, Saito Y, Sakai Y, Ueno H, Yoshino T, Fujimori T, Koinuma N, Morita T, Nishimura G, Sakata Y, Takahashi K, Takiuchi H, Tsuruta O, Yamaguchi T, Yoshida M, Yamaguchi N, Kotake K, Sugihara K. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol, 2012; 17: 1–29. [DOI] [PubMed]
- 15).Ekbom A, Helmick C, Zack M, Adami HO. Ulcerative colitis and colorectal cancer. A population-based study. N Engl J Med, 1990; 323: 1228–1233. [DOI] [PubMed]
- 16).Prior P, Gyde SN, Macartney JC, Thompson H, Waterhouse JA, Allan RN. Cancer morbidity in ulcerative colitis. Gut, 1982; 23: 490–497. [DOI] [PMC free article] [PubMed]
- 17).Nuako KW, Ahlquist DA, Mahoney DW, Schaid DJ, Siems DM, Lindor NM. Familial predisposition for colorectal cancer in chronic ulcerative colitis: a case-control study. Gastroenterology, 1998; 115: 1079–1083. [DOI] [PubMed]
- 18).Askling J, Dickman PW, Karlen P, Broström O, Lapidus A, Löfberg R, Ekbom A. Family history as a risk factor for colorectal cancer in inflammatory bowel disease. Gastroenterology, 2001; 120: 1356–1362. [DOI] [PubMed]
- 19).Shetty K, Rybicki L, Brzezinski A, Carey WD, Lashner BA. The risk for cancer or dysplasia in ulcerative colitis patients with primary sclerosing cholangitis. Am J Gastroenterol, 1999; 94: 1643–1649. [DOI] [PubMed]
- 20).Jayaram H, Satsangi J, Chapman RW. Increased colorectal neoplasia in chronic ulcerative colitis complicated by primary sclerosing cholangitis: fact or fiction? Gut, 2001; 48: 430–434. [DOI] [PMC free article] [PubMed]