Adenocarcinoma of the ileal pouch mucosa: case report and literature review

Proctocolectomy with IPAA has become the surgical standard of care for patients with UC and FAP, often indicated for dysplasia or carcinoma of the colon or rectum [1, 2]. To date, there are 27 reported cases of adenocarcinoma within the pouch itself [6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26] including our newly described case (Table 1), in addition to nearly 50 reported cases of adenocarcinoma at or near the anastomosis [5, 10]. These reports suggest that true pouch cancers most often arise in patients with a history of neoplasia, typically many years following IPAA, and without postoperative endoscopic pouch surveillance. Furthermore, many of these patients present with advanced stage and appear to have poor survival.

Despite these results, the risk of cancer following IPAA is probably low based on the current literature. A review of more than 1000 patients who underwent IPAA found no cases of malignancy near the pouch [27] whereas a more recent report of 1200 patients found an ATZ/pouch neoplasia incidence of 1.3 % [10]. A Cox model in a group of more than 3200 UC patients showed a cumulative incidence of ATZ/pouch cancer at 5, 10, 15, 20, and 25 years of 0.2, 0.4, 0.8, 2.4, and 3.4 %, respectively [28]. The pouch cancer rate in FAP although low is believed to be slightly higher than that in UC [29]. However, the reliability of current data is limited by the lack of standardization of pouch surveillance and variable definitions of pouch cancer, as well as the growing number of IPAAs being performed.

There are a number of known risk factors for the development of colorectal carcinoma in patients with UC, and many of these have also been suggested as risk factors for ATZ and pouch dysplasia including previous UC-associated dysplasia or cancer [6, 30], a long duration of UC [5], a long pouch duration [31], the presence of type C mucosa of the pouch [32], pancolitis [14], and concurrent PSC [20, 33]. Yet many of these risk factors are reported in series with small sample size or have not been validated. Recently however, a large series showed that a history of previous dysplasia or colon cancer is associated with an approximate 4- and 25-fold increase in risk, respectively, of developing either ATZ or pouch neoplasia [10]. Indeed in our review, 17/23 patients had a history of dysplasia or cancer. Less certain is the association between PSC and pouch cancer, as seen in our recent patient. A previous study in patients with UC and PSC who had an ileal pouch showed an increased risk of pouch neoplasia [33] (as well as pouchitis [34]) whereas another did not [28].

A number of authors have suggested that residual rectal mucosa could be the primary cause in the development of ATZ cancers [7, 11, 35], but in true ileal pouch cancers, this seems implausible. Another theory for cancer development in any cancer following IPAA is the inflammation-dysplasia-cancer sequence similarly seen in colorectal cancers [31, 36]. There is a concern that “backwash ileitis” [16, 31] or chronic pouchitis [14, 18] seen in many UC patients can initiate this cascade, although this has recently been disputed by some authors [10, 28]. A third theory is of primary cancer recurrence [10], which seems unlikely, given the long interval seen in a number of these patients (mean 8.9 years after IPAA in our review). Lastly, ileal adenomatous malignant transformation likely represents the underlying etiology in the majority of FAP patients presented within this review (particularly those without adequate endoscopic surveillance), more so than primary small bowel malignancy in patients not carrying this genetic predisposition [37].

An interesting hypothesis is that patients with defects in DNA repair mechanisms, and susceptible genes may remain at high risk for neoplasia despite the diseased colon being removed [10]. Whether true pouch cancer patients have a separate oncologic pathogenesis or prognosis than cancers of the ATZ is unclear. Over the past 25 years, these ATZ/pouch cancers as a whole have been described classically as occurring after a long history of antecedent ulcerative colitis. A review by Liu et al. found that the mean interval from UC diagnosis to ATZ/pouch cancer diagnosis was 23.1 years, with the shortest being 10.3 years [5]. In our review of true pouch cancers, the mean and shortest intervals were even longer at 28.2 and 20 years, respectively.

The mean time from pouch creation to adenocarcinoma was 8.9 years (range 2–25 years), and patients with a history of neoplasia in their original proctocolectomy were more likely to develop pouch cancer. However, the mean time to adenocarcinoma diagnosis was only 4.8 years in patients without neoplasia in their original proctocolectomy specimen compared to 8.2 years in those with prior neoplasia. This further clouds our understanding of pouch cancer pathogenesis, albeit in a small sample, but does continue to highlight the need for long-term follow-up in all patients.

The preferred pouch creation technique has long been debated [5, 6]. The hand-sewn anastomosis with mucosectomy was the initial technique described [3]; however, since the mid-1980s the double-stapled technique (without mucosectomy) has grown in popularity because of decreased operating time and complication rates as well as a reported better functional outcome [38, 39]. However, many authors have expressed concern over the possibility of an increased risk of malignancy associated with retained rectal mucosa [7, 11, 35]. This being said, even after macroscopically complete mucosectomy, there are retained islets of rectal mucosa in at least 20 % of patients [35, 40]. Following mucosectomy, surgeons may have a false sense of security in the anticipated risk of any cancer development, which could be hazardous given the high number of cancers seen isolated to the ileal pouch as well as all ATZ/pouch cancer groups (18/23 patients in our review and 27/55 patients [5, 10], respectively). It is reasonable then to conclude that the concern over a perceived increased cancer risk following IPAA when not performing a mucosectomy, although logical, is unfounded because of the number of cases existing in the literature following both anastomotic techniques. Nevertheless, we do not wish to suggest that performing a mucosectomy, in this group of ileal pouch mucosal cancers, is a causative factor but rather is more likely a manifestation of a small sample.

There is no consensus, particularly in UC patients, on the necessity of endoscopic pouch surveillance, nor on its optimal interval, duration, or location, and number of biopsies. Image-enhanced endoscopy techniques such as chromoendoscopy have recently been used in the surveillance of pouches (particularly in FAP); however, it is unclear whether these modalities improve pouch cancer detection [25]. A group from the Lahey Clinic [41] doubted the necessity of ileal mucosa endoscopic surveillance in a series of over 700 patients, in which only one patient developed dysplasia. On the other hand, Ault et al. suggested that routine surveillance as often as every 6 months in all patients with long-standing ileal pouches might be beneficial [9]. Others have suggested that pouches with high-grade dysplasia or dysplasia-associated lesions should be universally resected [16], similar to the accepted indication for colectomy in UC patients [42]. Derikx et al. challenged this notion by describing the natural history of pouch dysplasia, where they found that pouch dysplasia has high regression and low progression rates to carcinoma [10]. They went on to suggest, similar to others [5], that surveillance should be undertaken in patients with a history of dysplasia/cancer with particular attention paid to the ATZ, with the caveat that this surveillance may not necessarily lead to increased early cancer detection rates or improved prognosis. Last year however, the European Crohn’s and Colitis Organization stated that there is no clear evidence to support pouch surveillance in UC patients, but that those with dysplasia/cancer in the colectomy specimen, atrophic (type C) pouch mucosa, and PSC are at increased risk for neoplasia [43]. More recently, the Standards Practice Task Force of the American Society of Colon and Rectal Surgeons looked at surveillance in UC patients and stated that outside of the symptoms, routine surveillance of the ileal pouch does not appear to be beneficial although surveillance of the rectal cuff or ATZ may be recommended to detect malignancy [29]. Notably however, in FAP patients, there is a recognized need for thorough pouch surveillance after IPAA, similar to patients with an ileorectal anastomosis [24, 25, 44].

In reviewing cases of ileal pouch cancer, our data suggests that although these lesions appear uncommon, endoscopic pouch surveillance may result in earlier cancer detection in both UC and FAP patients (Table 2). All seven patients who underwent uninterrupted, regular surveillance were diagnosed with stage 1 or 2 disease compared to only four of 17 patients who did not. However many authors do not clearly report on specifics of their pouch surveillance or provide full pathological staging, which may reflect an element of reporting bias. Regardless, only three cases can be definitively described as stage 1, a trend similarly observed in our earlier report looking at all IPAA-associated cancers [6].

One of the strengths of this review is the distinction between true ileal pouch and all other IPAA-associated cancers. However, there are some limitations to this study. Firstly, this series of cases occurred over a long period of time with significant changes in surgical and oncological management. Secondly, it may be argued that drawing substantial conclusions from a sample including both inflammatory bowel disease and FAP patients is lessened given their divergent pathogenesis and treatment. Finally, we describe a small sample size using English language articles.

In summary, the risk for developing a pouch cancer following IPAA appears low. However, the increasing reporting of these cancers are of concern as most patients present at a young age with advanced disease and often many years following surgery. Opinions on surveillance are varied, particularly in UC. We believe a surveillance program after IPAA should include random and directed biopsies of the ATZ and the ileal pouch mucosa, particularly those with a previous history of dysplasia or cancer, those presenting with symptoms, and FAP. The establishment of expert consensus guidelines on surveillance should be considered in the near future.