CD-associated diverted anorectal cancer is a rare tumor. Carcinogenic rates in previous studies were 0.63% [10] and 3.1% [11]. The meta-analysis reported a relative morbidity risk for CD-associated CRC of 2.5 (range, 1.3–4.7) [12], 4.5 (range, 1.3–14.9) for large intestinal lesions, and 1.1 (range, 0.8–1.5) for ileal lesions. The relative morbidity risk of CRC was 2.4 in a meta-analysis of 60,122 patients with CD reported by Alexander et al. [13]. CD was considered a high-risk factor for cancer [14]. Besides, the following conditions have been reported as characteristics of CD-associated CRC: patients were approximately 10 years younger than those with typical CRC, had a long duration of CD (10–20 years), in many cases, cancer developed in the right colon, the histologic type was mucinous adenocarcinoma, and cancer was likely to develop in lesions caused by CD [15–17]. Bettner et al. reported that while the risk of CRC in inflammatory bowel disease (IBD) populations have been well described, there is a lack of data on the risk of CRC in diverted IBD bowel segments [1]. We believe that our cases can be included in this category. Few studies focusing on diverted CRC have been published, and based on these reports, only a small percentage of cases have been identified [1, 18–22]. These are summarized in Table 4.
Table 4
Previous reports on diverted colorectal adenocarcinoma in patients with Crohn's disease
Patient
|
Author and year
|
Age, sex
|
Stoma to cancer
|
Biopsy before surgery
|
Histology
|
Stage
|
Treatment
|
Recurrence
|
Outcome
|
1
|
Bettner 2018 [1]
|
41 M
|
7 years
|
N/A
|
LGD
|
N/A
|
APR
|
|
Alive at 11 years
|
2
|
Ogawa 2013 [20]
|
68 F
|
|
N/A
|
muc
|
IV (TxNxMx)
|
Not resected
|
|
Dead at 5 months
|
3
|
Ogawa 2013 [20]
|
34 F
|
|
No malignancy
|
muc
|
III (T4N2M0)
|
APR
|
No
|
Alive at 63 months
|
4
|
Ogawa 2013 [20]
|
33 F
|
|
Adenocarcinoma
|
muc, sig
|
III (T4N0M0)
|
Pelvic exenteration
|
Distant
|
Dead at 14 months
|
5
|
Iesalnieks 2010 [22]
|
|
|
|
muc
|
|
APR
|
|
Unknown
|
6
|
Cirincione 2000 [19]
|
36 F
|
2 years
|
Squamous cell carcinoma diagnosed at post-surgery
|
SCC
|
|
APR
|
distant
|
Alive at 24 months
|
7
|
Cirincione 2000 [19]
|
35 F
|
9 years
|
Adenocarcinoma
|
por
|
|
APR
|
Distant
|
Dead at 36 months
|
8
|
Cirincione 2000 [19]
|
54 M
|
3 years
|
Adenocarcinoma
|
m/d
|
|
APR
|
Pelvis
|
Alive
|
9
|
Church 1984 [18]
|
65 F
|
|
Squamous cell carcinoma
|
SCC
|
|
Radiotherapy
|
Inguinal lymph node
|
Dead
|
APR, abdominoperineal resection; F, Female; LGD, low-grade dysplasia; N/A, not available; muc, mucinous adenocarcinoma; m/d, moderately differentiated adenocarcinoma; M, Male; por, poorly differentiated adenocarcinoma; SCC, squamous cell carcinoma; sig, signet ring cell carcinoma |
None of the references reported on a comparison between CD-associated diverted and non-diverted anorectal cancer. In our study, there were no significant differences in clinical features and histological findings between the diverted and non-diverted anorectal cancer groups. However, the recurrence and overall survival rates were significantly different. Two cases had stage IV cancer in the diverted anorectal cancer group, although this may be because the diagnosis was made after a very advanced stage, and early diagnosis was required.
Basseri et al. [23] argued that, although there was a strong link between IBD and CRC, an active surveillance colonoscopy approach remained controversial. The Crohn's and Colitis Foundation of America guidelines recommend that at least one-third of the colon 8–10 years after the onset of CD be observed for colonoscopy screening. If surveillance colonoscopy is negative for dysplasia or cancer, an examination is recommended every 1–2 years after that [24]. Currently, there are no consensus guidelines to inform surveillance endoscopy of diverted segments [1]. The insertion of an endoscope was impossible due to the stricture of the diverted large intestine, except for one patient diagnosed at stage I. Pelvic magnetic resonance imaging and EUA are performed in patients who are not colonoscopy candidates due to stricture. However, Devon et al. suggested that cancer might not be diagnosed even after multiple examinations [21]. Sjodahl et al. reported that patients with a diverted colon and rectum with a rectal stump and chronic anal fistula should be under careful surveillance due to their significantly increased risk of cancer [25]. One patient at stage I underwent an annual surveillance colonoscopy and showed disease-free survival. Thus, surveillance colonoscopy or EUA is essential for early-stage cancer diagnosis. The establishment of better surveillance methods for similar cases is urgently needed.
Biologics plays a vital role in managing the active CD. However, biologics are not useful for abscess-forming CD lesions, including complex anal or rectal fistula and rectal stricture. In this study, most patients were not treated with biologics before diversion. We did not use biologics after diversion because there was no evidence of sufficiency. Infliximab was administered to only two patients before surgery. The administration of drugs such as biologics for cancer is controversial because they are associated with immunosuppression [26, 27]. For instance, anti-tumor necrosis factor (TNF) treatment without an immunomodulator does not increase cancer risk in patients with IBD [28]. Investigation of cancer progression associated with treatment using biologics is necessary in most cases. After diagnosing cancer, patients with malignancies generally avoid the administration of anti-TNFα antibody agents because tumor growth may be enhanced by TNFα suppression [28]. The right timing for the administration of biologics after cancer healing must be addressed in the future.
Diverted anorectal cancer is generally difficult to treat; surgical resection is the only cure. Diagnoses are typically made at the advanced stage of cancer. In this study, nine patients were diagnosed with advanced cancer (pathological stage II-IV). The prognosis was poor. There was a discrepancy in the staging of clinical and pathological diagnosis; therefore, a negative resected margin was needed. CD-associated with CRC in Japanese patients develops primarily in the rectum and anus. However, the etiology remains unknown. It was challenging to dissect the total mesorectal excision plane because of perirectal inflammation. Recurrence, including intrapelvic recurrence, frequently occurs despite chemotherapy administration after APR. A multidisciplinary treatment combining neoadjuvant chemotherapy, total pelvic exenteration, lateral lymph node dissection (a unique Japanese therapy) during surgery, and perioperative irradiation may be necessary. A previous study (Table 4) reported that four out of nine patients died, showing poor outcomes when anorectal strictures complicate this disease [18–20, 22]. Although there was no evidence for diverted anorectum, the literature suggests that carcinoma may still develop despite perianal healing. Cirincione et al. [19] suggested that surveillance proctoscopy must be performed; otherwise, a preventive APR with permanent fecal diversion should be considered in patients with CD complicated by an anorectal stricture [19]. APR should be performed early to ensure further investigation and prevention of cancer development [22].
For the diversion of anorectum with stricture or inflammation in patients that pose difficulty for screening and surveillance, an early APR may be considered a treatment option that could be done in advance before the patients are diagnosed with advanced cancer based on the symptoms only. A report by van Overstraeten et al. suggested that patients with anorectal CD who need proctectomy should undergo proctocolectomy with ileostomy despite the absence of proximal colonic involvement [29]. Therefore, we opted for APR. These patients passed away from intrapelvic dissemination. Most of the patients with diverted anorectal cancer developed local recurrence. In Europe and the United States, preoperative radiotherapy is a mainstay treatment for rectal cancer, and tumors are often reduced. Japanese standardized treatment for rectal cancer (especially low rectal cancer with T3 depth) is total mesorectal excision with lateral lymph node dissection; however, some CRC research institutions in Japan perform neoadjuvant chemoradiotherapy and report good outcomes [30]. Therefore, in Japan, reducing tumors by preoperative treatment (neoadjuvant chemoradiotherapy) for similar cases should be considered for anorectal cancer. However, the biggest problem is the inability to perform curable surgery because of an intraoperative or postoperative diagnosis, which needs further study.
This study has some limitations. First, it was a retrospective single-institution study and included a limited number of patients. CD therapy in this cohort did not include many patients on biologics or immunomodulator therapy. There are also differences in the treatment strategies for CRCs between Japan, Europe, and America. Second, heterogeneous treatments did not include the use of chemoradiotherapy.
In conclusion, CD-associated anorectal cancer was diagnosed in 10 of 232 (4.3%) patients following fecal diversion, had a poor prognosis after curative resection, and was challenging to diagnose at an early stage. Subsequently, the risk of patients with a diverted CD of the rectum and anus developing cancer cannot be ignored. We recommend annual surveillance colonoscopy or EUA to detect early-stage cancer. In the patients who find it challenging to undergo these modalities, the alternative treatment considered with sufficient informed consent is early APR or TPC to prevent advanced cancer. Thus, further prospective multi-institutional studies with a large population are needed to confirm CD-associated anorectal cancer prognosis following diversion.