Noncolorectal malignancies in inflammatory bowel disease: more than meets the eye.

In patients with Crohn's disease, the risk of small bowel adenocarcinoma is 20-40 times higher than the low background risk of the general population. In the subset of patients with longstanding small bowel lesions, the absolute risk of small bowel adenocarcinoma exceeds 1 per 100 patient-years after 25 years of follow-up and becomes equivalent to the risk of colorectal cancer. Growing evidence suggests that the pathogenesis of small bowel adenocarcinoma arising in inflammatory lesions of Crohn's disease is similar to that of colorectal cancer complicating chronic colonic inflammation (inflammation-dysplasia-cancer sequence). However, contrasting with the established endoscopic detection of colonic advanced neoplasias in patients with longstanding extensive colitis, there is no consensus at this time how to face the excess-risk of small bowel adenocarcinoma in patients at high risk. There are no specific clinical or imaging alert signs and endoscopic surveillance of the totality of the inflamed small bowel mucosa would suppose to perform repeated enteroscopies, with the potential limiting factor of stenosis. Very preliminary data suggest that chemoprevention with salicylates could be an alternative way for reducing the risk. Data from referral centers and from the CESAME cohort suggest that intestinal lymphomas may arise in the chronically inflamed segments in patients with inflammatory bowel disease (IBD). Regarding nonintestinal lymphomas, it is now established that IBD patients treated with thiopurines have an excess risk of lymphomas, exhibiting in most cases pathological features of lymphomas associated with immunosuppression, including the frequent presence of EBV in neoplastic tissues. There is growing evidence that treatment with thiopurines is responsible by itself for this excess risk. IBD patients receiving immunomodulators, especially young men, are also at risk (0.4 for 10,000 patient-years in the CESAME study) for developing fatal early post-mononucleosis lymphomas, like in Purtilo's syndrome, maybe in association with a background genetic susceptibility. Finally, patients receiving thiopurines and/or TNF-inhibitors are at risk for developing fatal hepatosplenic T cell lymphomas, but this risk is low (no case in the CESAME study). Whether patients receiving a monotherapy with methotrexate and/or TNF inhibitors are at increased risk of lymphomas is not known. Concordant data suggest that women receiving immunosuppressive therapy are at increased risk for developing uterine cervix dysplasia and require closer surveillance. But it is not established whether the risk of uterine cervix cancer and basal and squamous cell skin cancers (that may be associated with chronic human papillomavirus infection) is increased in patients receiving immunomodulators.