Section 6 -

The Pathogenesis of Colorectal Carcinoma in Inflammatory Bowel Disease Thomas C Smyrk Department of Pathology, Mayo Clinic Rochester, USA

Inflammatory bowel disease confers increased risk for colorectal carcinoma. The risk is best documented for ulcerative colitis (UC), but Crohn's disease is unequivocally a cancer-associated condition as well. A meta-analysis covering 116 published studies put the overall prevalence of colorectal cancer in patients with UC at 3.7%, and documented increasing risk with increased duration of pancolitic disease: the cumulative incidence was 2% at 10 years, 9% at 20 years and 18% after 30 years of disease. [1] One recent single-institution study found lower risk and deserves mention here: prospectively collected data from the surveillance program operating at over a 30-year period St Mark's Hospital showed a cumulative cancer incidence of 2.5% at 20 years, 7.6% at 30 years and 10.8% after 40 years of disease. [2] The cancer incidence was constant over the duration of disease and there was a significant reduction in cancer incidence after 1975, perhaps reflecting better control of inflammation by disease-modifying drugs. A similar temporal trend has been observed in Scandinavia [3] and Olmsted County, Minnesota . [4] In fact, no Olmsted County resident diagnosed with UC after 1980 has developed carcinoma (201 patients). In addition to duration of disease, putative risk factors for colorectal cancer in UC are anatomic extent of disease, disease activity, presence of primary sclerosing cholangitis and a family history of colorectal cancer. Ekbom et al found a relative risk for carcinoma of 1.7 in patients with proctitis only, 2.8 in patients with disease extending to the hepatic flexure, and 14.8 in patients with disease beyond the hepatic flexure. [5] A case-control study from St Mark's Hospital found a correlation between inflammation scores (both endoscopic and histologic) and the risk for colorectal neoplasia. [6] On multivariate analysis, only the histologic inflammation score had a significant relationship to cancer occurrence; a 1-unit change in the semiquantitiative score (0-4) increased the odds of colorectal neoplasia by a factor of 4.69. Backwash ileitis, which can be considered another marker of disease severity, has also been implicated as a risk factor for carcinoma. [7] Patients with both UC and primary sclerosing cholangitis are approximately four times as likely to develop carcinoma as those without PSC, according to a recent meta-analysis of 11 studies. [8] In our own unpublished study of 111 patients with UC-carcinoma, 27% also had primary sclerosing cholangitis, compared to 2% of controls matched for age, sex and extent of disease. Family history of colorectal cancer was associated with a relative risk for carcinoma of 2.5 in a population-based cohort study of nearly 20,000 patients with UC or Crohn's disease. [9] In our own unpublished study, UC patients with carcinoma were more likely to have a sibling with any type of malignancy than control patients (16% vs 8%). All of the above highlights two contributors to carcinoma in UC: chronic inflammation and, as suggested by the association with family history, innate patient characteristics. Many common cancers develop in the setting of chronic inflammation: hepatocellular carcinoma (in chronic viral hepatitis), gastric cancer (in chronic gastritis) and colorectal cancer (inflammatory bowel disease) are examples. Persistent inflammation drives increased cellular turnover, particularly in the epithelium. Free radicals generated by inflammatory responses can injure cells by gene mutation or by post-translational modification, leading to disruption of protective functions such as DNA repair, cell-cycle checkpoints and apoptosis. p53 appears to be a critical mediator of response to stress; nearly every study of carcinogenesis in UC makes disruption of p53 function an important early step. In fact, it has been convincingly established that there are extensive clonal abnormalities of p53 (among other genes) even in the non-dysplastic mucosa of colitic patients with dysplasia or carcinoma. [10] The critical role of the p53 network as a response to inflammatory stress was beautifully outlined by Staib et al. [11] The authors compiled the gene expression profiles associated with four different components of the inflammatory response: nitric oxide, hydrogen peroxide, DNA replication arrest and hypoxia. Overall, 1396 genes changed in a p53-dependent manner. There appeared to be unique profiles associated with each type of injury, with ony 14 genes altered in response to all four conditions. It seems clear that interference with p53-mediated response to inflammatory stress can have complex and significant downstream effects. How do dysplasia and carcinoma develop in this "initiated" background? As in sporadic colorectal cancer, three overlapping pathways may be in play: the tumor suppressor pathway, the mutator pathway and the methylator pathway. All have been implicated in UC-related carcinogenesis. Maia et al [12] argue that the suppressor pathway is the main contributor to colitis-associated cancers. The authors studied samples from inactive chronic colitis without dysplasia, with dysplasia and with carcinoma using a broad array of techniques: microsatellite instability, loss of heterozygosity (LOH) and allelic imbalance (AI), methylation analysis of MLH1 and CSPG2, and mutation analyses of p53 and APC. No microsatellite instability was detected in any sample. There was no hypermethylation of MLH1, and only a single example of CSPG2 methylation. LOH and AI were frequently observed, particularly at loci on chromosomes 5, 9 and 18, close to the genes for APC, p16 and DCC. This emphasis on mutational changes is consistent with the work of Chen et al, who used DNA fingerprinting to demonstrate mutational changes in 10-20% of sites examined, even in non-dysplastic mucosa; half of those changes were clonally expanded. [10] The contribution of the mutator pathway to UC carcinogenesis is under debate, with estimates on the prevalence of microsatellite instability high (MSI-H) carcinoma in UC ranging from 1% to 45%. A recent large, multi-institutional series found a prevalence of 19/124 (15%), similar to the rate in sporadic carcinoma. [13] The specific mutational profile, however, was different in UC-carcinoma, with TGFbetaR2 and ACVR2 mutations being less common than in sporadic MSI-H, and ICA1 mutations being more common. Tahara et al [14] argue for a more important role for microsatellite instability. The authors used five microsatellite markers and defined MSI-H as alterations in 2 or more loci. They found MSI-H phenotype in 8 of 12 UC-cancers, 4/6 UC-high grade dysplasias, 2/6 low grade dysplasias and 15/59 samples of inflamed UC mucosa without dysplasia. Four patients with MIS-H carcinoma were shown to have microsatellite instability in non-dysplastic biopsies obtained 2-12 years before the cancer diagnosis. The contribution of hypermethylation to UC carcinogenesis also has its advocates. Issa et al proposed the concept of accelerated age-related CpG island methylation, perhaps resulting from chronic inflammation in the setting of increased cell turnover, and suggested that UC can be viewed as causing premature aging of colorectal epithelial cells. [15] Since then, preferential hypermethylation of numerous genes has been documented in dysplastic UC epithelium, or even in non-dysplastic epithelium. The Issa paper specifically mentioned estrogen receptor, CSPG2, MYOD and p16. Others have confirmed methylation of estrogen receptor in non-neoplastic epithelium as a marker of neoplasia risk. [16] Hypermethylation of HPP1 has also been described as an early event in UC neoplasia. [13, 17] Methylation of CDH1 (the E-cadherin gene) may also be an early event, [18] and may be a way to affect the wnt pathway without mutations in APC. [19] Hyperphosphorylation is another post-translational modification that could play a role in carcinogenesis. Ying et al demonstrated hyperphosphorylation of the retinoblastoma protein in both mouse and human colitis, with subsequent release of E2F1 and downstream genes associated with proliferation (PCNA and cyclin D1) and apoptosis inhibition (phosphor-Akt). [20] The chronic inflammation of UC inflicts cellular damage via free radical injury and generates abnormal cell populations well before the appearance of dysplasia. Interference with normal functioning of the p53 tumor suppressor network is a critical early event. Mutations and post-translational modifications probably both play a role in promoting the development of dysplasia and carcinoma, but the nature and timing of the specific abnormalities remains under investigation. References Eaden JA, Abrams KR, Mayberry JF. The true risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gastroenterology 2001;48:526-535. Rutter MD, Saunders BP, Wilkinson KH et al. Thirty-year analysis of a colonoscopic surveillance program for neoplasia in ulcerative colitis. Gastroenterology 2006;130:1030-1038. Rubio CA, Befrits R, Ljung T, Jaramillo E, Slezak P. Colorectal carcinoma in ulcerative colitis is decreasing in Scandinavian countries. Anticancer Res 2001;21:2921-2924. Jess T, Loftus EV, Velayos FS et al. Risk of intestinal cancer in inflammatory bowel disease: a population-based study from Olmsted County , Minnesota . Gastroenterology 2006;130:1039-1046. Ekbom A, Helmick C, Zack M et al. Ulcerative colitis and colorectal cancer: a population-based study N Engl J Med 1990;323:1228-1233. Rutter M, Saunders B, Wilkinson K et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology 2004;126:451-459. Heuschen UA, Hinz U, Allemeyer EH et al. Backwash ileitis is strongly associated with colorectal carcinoma in ulcerative colitis. Gastroenterology 2001;120:841-847. Soetikno RM, Lin OS, Heidenreich PA et al. Increased risk of colorectal neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis: a meta-analysis. Gastrointest Endosc 2002;56:48-54. Askling J, Dickman PW, Karlen P et al. Family history as a risk factor for colorectal cancer in inflammatory bowel disease. Gastroenterology 2001;120:1356-1362. Chen R, Rabinovitch PS, Crispin DA, Emond MJ, Bronner MP, Brentnall TA. The initiation of colon cancer in a chronic inflammatory setting. Carcinogenesis 2005;26:1513-1519. Staib F, Robles AI, Varticovski L et al. The p53 tumor suppressor network is a key responder to microenvironmental components of chronic inflammatory stress. Cancer Res 2005;65:10255-10264. Maia L, Dinis J, Cravo M et al. Who takes the lead in the development of ulcerative colotis-associated colotrectal cancers: mutador, supresor or methylator pathway? Cancer Genet and Cytogenet 2005;162:68-73. Schulmann K, Mori Y, Croog V et al. Molecular phenotype of inflammatory bowel disease-associated neoplasms with microsatellite instability. Gastroenterology 2005;129:74-85. Tahara T, Inoue N, Hisamatsu T et al. Clínical significance of microsatellite instability in the inflamed mucosa for the prediction of colonic neoplasms in patients with ulcerative colitis. J Gastroenterol and Hepatol 2005;20:710-715. Issa J-PJ, Ahuja N, Toyota M, Bronner MP, Brentnall TA. Acceleratied age-related CpG island methylation in ulcerative colitis. Cancer Res 2001;61:3573-3577. Fujii S, Tominaga K, Kitajima K et al. Methylation of the oestrogen receptor gene in non-neoplastic epithelium as a marker of colorectal neoplasia risk in longstanding end extensive ulcerative colitis. Gut 2005;54:1287-1292. Sato F, Shibata D, Harpaz N et al. Aberrant methylation of the HPP1 gene in ulcerative colitis-associated colorectal carcinoma. Cancer Res 2002;62:6820-6822. Azarschab P, Porschen R, Gregor M et al. Epigenetic control of the E-cadherin gene (CDH1) by CpG methylation in colectomy samples of patients with ulcerative colitis. Genes Chromosomes Cancer 2002;35:121-126. Aust DA, Terdiman JP, Willenbucher RF et al. Aletered distribution of beta-catenin and its binding proteins E-cadherin and APC in ulcerative colitis-related colorectal cancers. Mod Pathol 2001;14:29-39. Ying L, Marino J, Hussain SP et al. Chronic inflammation promotes retinoblastoma protein hyperphosphorylation and E2F1 activation. Cancer Res 2005;65:9132-9136.