Gary R. Lichtenstein (ed.)Medical Therapy of Ulcerative Colitis10.1007/978-1-4939-1677-1_26

26. Chemoprevention in Ulcerative Colitis

Fernando Velayos¹

(1)

Department of Gastroenterology, University of California, San Francisco, 2330 Post St., Suite 610, San Francisco, CA 94115, USA

Fernando Velayos

Email: fernando.velayos@ucsf.edu

Keywords

ChemopreventionUlcerative colitisColorectal cancerFamilial adenomatous polyposis (FAP)Hereditary nonpolyposis colorectal cancer (HNPCC)

Background

Colorectal cancer (CRC) is a feared complication of chronic ulcerative colitis (UC). The cumulative probability of developing colorectal cancer (CRC) among persons with UC is significantly higher than in the general population [1]. UC is the third highest-risk condition for colorectal cancer after two genetic syndromes, familial adenomatous polyposis syndrome (FAP) and hereditary nonpolyposis colorectal cancer (HNPCC) [2]. Data from a comprehensive meta-analysis suggest that the probability of CRC in IBD is 2 % after 10 years of disease, 8 % after 20 years, and 18 % after 30 years [3, 4]. In comparison, the cumulative lifetime probability of developing CRC for the general population in the United States is approximately 5 % [3, 4].

Described in terms of relative risk, UC increases the risk of CRC five- to sixfold relative to the general population [1]. The risk of developing CRC increases as a greater proportion of the colon is involved by inflammation [1]. Interestingly, not all populations of IBD patients have an increased risk of CRC. In a Danish population-based cohort, the risk was 0.4 % after 10 years, 1.1 % after 20 years, and 2.1 % after 30 years of disease [5]. These probabilities are comparable to those of an American population-based cohort where the cumulative probability of developing CRC was 0 % at 5 years, 0.4 % at 15 years, and 2.0 % at 25 years after a CUC diagnosis [6]. Also notable is that high-risk populations may be experiencing a reduction in cancer risk and mortality over time. A UK surveillance colonoscopy cohort reported a reduction in cancer risk over the past 30 years (1970–2000) and quantified the risk as a 2.5 % at 20 years, 7.6 % at 30 years, and 10.8 % after 40 years of disease [7]. A cohort in Sweden found a still elevated, but declining, trend in CRC incidence and mortality over a 35-year time period (1960–2004) [8].

It is not known why the risk of CRC is lower in certain populations than others or if the risk of CRC and mortality from CRC are declining in all populations. However, these promising data suggest that the risk of CRC in IBD is modifiable. Greater rates of proctocolectomy for medical treatment failures, better rates of surveillance colonoscopy with proctocolectomy for dysplasia, and finally greater use of potentially chemopreventive drugs used for IBD, vitamins, and other therapies used by UC patients could all be important factors that reduce the risk of CRC in a given population [9, 10].

To mitigate the risk of colorectal cancer, many patients and their physicians choose to follow a diagnostic program of screening and surveillance colonoscopy with a goal of detecting dysplasia and other early neoplastic lesions at a curable stage [11]. The rationale for undergoing surveillance colonoscopy is based on the premise that determining which patients are likely to progress to cancer in the near term can be reliably predicted by the presence or absence of histological dysplasia on colonoscopy. Surveillance colonoscopy is a secondary prevention strategy, whereby the risk of colorectal cancer is mitigated by either (1) identification and removal of dysplastic polypoid tissue or (2) detection of flat or subtle dysplasia followed by proctocolectomy. In other words, intervention occurs after dysplasia occurs.

This strategy of secondary prevention is quite successful in colon cancer screening in the general population. However, dysplasia in inflammatory bowel disease is different. There is greater incidence of flat or subtle dysplasia that may be missed during colonoscopy, and important lesions can be obscured in the setting of inflammatory polyps [12, 13]. Thus instead of relying exclusively on secondary prevention (detecting and intervening after dysplasia has occurred), there has been significant interest in exploring strategies for primary prevention (intervening early to prevent the development of dysplasia in the first place). Chemoprevention is a primary prevention strategy.

The National Cancer Institute defines chemoprevention as the use of drugs, vitamins, or other agents to try to reduce the risk of or delay the development or recurrence of cancer. This chapter will review the relationship between drugs used to treat UC, vitamins, and other therapies used by UC patients and the risk of the development of colorectal cancer. This chapter will also review the most recent recommendations and guidelines from the American Gastroenterological Association with regard to chemoprevention.

Medications Used to Treat UC

Chronic inflammation has a key role in the development of CRC in UC through years of repeated cycles of cellular damage and repair. Thus, it is logical that medications used to treat ulcerative colitis may reduce chronic inflammation and have important chemopreventive properties.

5-Aminosalicylates

There has been significant attention regarding the potential chemopreventive properties of 5-aminosalicylates. 5-aminosalicylates are among the oldest therapies used to treat ulcerative colitis. Sulfasalazine, composed of a sulfa moiety bonded to mesalamine, has been used in UC since the 1940s [14]. Modern mesalamine derivatives, which lack the sulfa moiety (contributes to GI side effects of sulfasalazine), have been available and approved by the US Food and Drug Administration (FDA) since the late 1980s after the first pivotal randomized control trials using these newer derivatives were published [15].

Interest in the chemopreventive properties of 5-ASA agents started in the mid-1990s, coincident with publications that aspirin and NSAIDs reduced the risk of colorectal cancer in the general population. In 1991, Thun et al. published an observational study showing regular aspirin users had an approximate 40 % reduction in fatal colon cancer [16]. In 1993, Greenberg et al. published an observational study showing that aspirin users had a 48 % reduction in adenomas [17]. A variety of clinical and experimental studies demonstrated that 5-aminosalicylates share several important antiinflammatory and anticancer properties with aspirin and nonsteroidal antiinflammatory medications (NSAIDs). These properties include increased apoptosis, decreased cell proliferation, reduced production of oxidative radicals such as prostaglandins and leukotrienes, and finally improved cellular repair [18].

The first major observational population-based study examining the association between 5-ASA use and cancer risk was published in 1994 by Pinczowski et al. [19]. The authors conducted a nested case-control study in a population of Swedish patients in the Uppsala region with ulcerative colitis treated between 1965 and 1983. They noted that those who used more than 3 months of continuous sulfasalazine had a 60 % reduction in the risk of colorectal cancer compared to those with less than 3 months. It was noted that most subjects used more than 3 months of continuous therapy or in fact were long-term and regular users with high rates of adherence to sulfasalazine [19].

In their study, Pinczowski et al. described the important next step to advance the 5-ASA hypothesis and wrote: “clinical trials should be initiated in patients with inactive ulcerative colitis to determine if continuous treatment with sulfasalazine… has an impact on dysplasia or malignant transformation” [19]. This statement is similar to that from Greenberg et al. the year prior regarding aspirin use in the general population and colorectal adenomas: “this study supports the hypothesis that aspirin has an anti-neoplastic effect in the large bowel. Nevertheless, the question of whether aspirin should be used to prevent large bowel tumors would be best answered by a randomized controlled clinical trial specifically designed to address this issue.”

This historical perspective is relevant in that a randomized trial of aspirin to prevent colorectal adenomas in patients without IBD was published a decade later and showed a 39 % reduction in advanced neoplasia risk. For ulcerative colitis, no such trial has been conducted given complexities in design as well as sample size. Thus, the available data for 5-ASA chemoprevention and the 5-ASA hypothesis remains based on experimental and human observational studies, but not randomized trials. These observational studies are notable for a variety of clinical designs and quality, as well as heterogeneous population of patients.

A systematic synthesis of these early trials was published in a meta-analysis. The pooled data supported the hypothesis that 5-aminosalicylates reduce the risk of colorectal neoplasia in patients with ulcerative colitis [10]. A total of nine studies (three cohort, six case-control) containing 334 cases of colorectal cancer, 140 cases of dysplasia, and a total of 1,932 subjects were analyzed. Pooled analysis showed a protective association with regular 5-ASA therapy and a risk reduction of 49 %. This estimate is consistent with the range of risk reduction observed with aspirin and nonsteroidal use in the general population and colorectal neoplasia risk.

Since then, additional studies have been published showing heterogeneous results. An example of two recent large-scale studies includes a Canadian and a French study. The Canadian study included 8,744 subjects with IBD who are part of the population-based epidemiologic database in Manitoba. The authors found no difference in the incidence of CRC among 5-ASA users nor differences between cases and controls with regard to duration of prior 5-ASA use [20]. In contrast, a nested case-control study within the CESAME population-based cohort found a 54 % reduction in CRC risk among 5-ASA users [21]. As such, the 5-ASA hypothesis is unlikely to be definitively answered with further observational studies.

Thus the question is how to synthesize and apply seemingly conflicting data to patient care. We can summarize what is known and not known and make educated guesses based on this information. It is known that chronic inflammation of the colon causes DNA damage and is a risk factor for colorectal dysplasia in ulcerative colitis [18, 22]. It is known that 5-ASA is an effective maintenance therapy for ulcerative colitis and has in vitro and in vivo antineoplastic effects [18, 23]. It is not proven but likely that maintenance 5-ASA therapy manages remission in UC through the reduction of chronic inflammation. It is known that the question of whether 5-ASA should be used to prevent large bowel tumors would be best answered by a randomized controlled clinical trial specifically designed to address this issue, and this is unlikely to occur in the near future if at all.

We can analyze this experimental and clinical data to provide some reasonable conclusions and also refer to the most recent AGA evidence ratings on this topic. For those patients with an indication for maintenance 5-ASA therapy, it is reasonable to suggest that maintenance 5-ASA therapy may in fact have a secondary benefit of reducing the risk of long-term dysplasia and colorectal cancer. Patients should be aware that this does not eliminate the need for regular surveillance nor changes the recommended frequency of surveillance [11]. Whether patients on immunomodulators or anti-TNF therapy but not on a 5-ASA therapy should have such therapy added is not answered with the available data. The AGA evidence rating for 5-ASA as a chemopreventive agent in UC is B (moderate certainty that the magnitude of net benefits is moderate) [24].

Immunomodulators

The class of immunomodulator agents includes azathioprine, its metabolite 6-mercaptopurine, and methotrexate. Given the lack of data for the use of methotrexate in ulcerative colitis, most immunomodulator use in UC and chemopreventive studies has assessed only azathioprine and 6-mercaptopurine.

Both azathioprine and 6-mercaptopurine do not have any obvious or inherent anticancer properties in colonocytes. The biologic effect appears to be on lymphocytes. Azathioprine and 6-mercaptopurine inhibit DNA and RNA synthesis and proliferation of lymphocytes. AZA and 6-MP have antineoplastic effects in certain leukemias. They have pro-neoplastic effects in certain lymphoproliferative tumors and are associated with a fourfold increased risk of non-Hodgkin’s lymphoma in persons with inflammatory bowel disease [25].

There have been at least four studies that have examined the association between azathioprine or 6-MP use and colorectal cancer risk in ulcerative colitis. One study showed a threefold increased risk of cancer; however, this study included high-risk patients for colorectal cancer who had primary sclerosing cholangitis and whose indication for azathioprine or 6-MP was posttransplant immunosuppression [12]. Of the remaining studies, two studies (one unpublished [26], one published [27]) showed a 70 % reduction in colorectal cancer risk [22, 26], whereas one showed no effect [27].

Given the lack of experimental data showing a chemopreventive effect for immunomodulators, the mixed results from clinical data, and the elevated relative risk of lymphoma reported with AZA/6-MP, the AGA evidence rating for immunomodulator use as a chemopreventive agent in UC is I (no recommendation, insufficient evidence to recommend for or against) [24].

As with 5-ASA, how should the above data be synthesized and used clinically? Given that AZA/6-MP use has a specific role within the algorithm for management of ulcerative colitis, it appears reasonable to emphasize to patients that the same therapy that is maintaining the ulcerative colitis in remission is likely having a secondary benefit of reducing long-term colorectal cancer risk. It is important to again emphasize that the only recommended modality for reducing long-term colorectal cancer risk is screening and surveillance colonoscopy [11].

Only gold members can continue reading. Log In or Register to continue