ANTICANCER RESEARCH 24: (2004)

Transcription

1 Ursodeoycholic Acid Treatment in IBD-patients with Colorectal Dysplasia and/or DNA-aneuploidy: a Prospective, Double-blind, Randomized Controlled Pilot Study URBAN SJÖQVIST 1, BERNHARD TRIBUKAIT 2, ÅKE ÖST 3, CURT EINARSSON 4, LENA OXELMARK 5 and ROBERT LÖFBERG 5 1 Department of Medicine, GE-unit, Stockholm South Hospital, 2 Department of Oncology and Pathology, Karolinska Hospital, 3 Department of Pathology, Medilab AB, 4 Department of Gastroenterology, Huddinge University Hospital and 5IBD-Unit, Sophiahemmet, All at Karolinska Institutet, Stockholm, Sweden Abstract. Background & Aims: There is an increased risk of colorectal carcinoma (CRC) in patients with longstanding, etensive colonic inflammatory bowel disease (IBD). Primary sclerosing cholangitis, family history of CRC, mucosal dysplasia and DNA-aneuploidy are other risk factors. Recently, results from animal studies have shown that the bile acid ursodeoycholic acid (UDCA) has a favourable impact on eperimentally-induced CRC/neoplasia in rats. The aim of this proof of the concept study was to eplore the possible preventive/reverting effects of UDCA in patients with colorectal IBD with eisting findings of low grade dysplasia and/or DNAaneuploidy. Patients and Methods: Nineteen patients (13 UC, 6 CD, median age 43 years) with long-standing, etensive IBD (median duration 21 years), with previous findings of low-grade dysplasia and/or DNA-aneuploidy, were randomized to receive either UDCA (500 mg b.i.d) (n=10) or placebo (n=9) in a controlled, double-blind, two-year study. Colonoscopy with multiple biopsies for histopathology and for DNA-flow cytometry was performed at the start and at si-month intervals during the study period. The primary outcome was the need for colectomy due to progression of dysplasia. Changes in dysplasia and DNAaneuploidy scores were also assessed. Results: There were no Abbreviations: Colorectal carcinoma, CRC; Crohn s disease, CD; DNA-flow cytometry, DNA-FCM; inflammatory bowel disease, IBD; primary sclerosing cholangitis, PSC; ulcerative colitis, UC; ursodeoycholic acid, UDCA. Correspondence to: Dr Urban Sjöqvist, MD, PhD, Department of Medicine, Karolinska Institutet at Stockholm Söder Hospital, SE Stockholm, Sweden. Tel: , Fa: , urban.sjoqvist@sos.sll.se Key Words: Ursodeoycholic acid, inflammatory bowel diseases, ulcerative colitis, Crohn s disease, dysplasia, DNA-aneuploidy. significant differences in the overall composed score between the two groups, either at study start or during the study period. In the placebo group one patient had a progression of dysplasia into high-grade and one patient developed DALM with lowgrade dysplasia; both had a colectomy. In contrast, no UDCAtreated patient had progression of dysplasia. Conclusion: UDCA may prevent further progression of manifest low-grade dysplasia in colorectal IBD. Prolonged treatment or an increased dose may be needed to fully eploit the chemopreventive properties of this compound. There is an increased incidence of colorectal carcinoma (CRC) in patients with long-standing, etensive ulcerative colitis (UC). The cumulative risk for CRC is 10-15% after 25 years duration of etensive disease (1-3) and seems to be the single most important cause for increased risk of death in patients with etensive colonic disease (4). The risk for CRC in colorectal Crohn s disease (CD) appears to be of the same magnitude (5, 6). Patients with concomitant primary sclerosing cholangitis (PSC) have a higher CRC-risk, approaching 40% at 25 years from onset (7-9). Other risk factors, which confer an increased risk for CRC, include a positive family history of CRC (10, 11). Neoplastic mucosal changes such as definite dysplasia low-(lgd) or high-grade (HGD) and DNAaneuploidy, i.e. gross chromosomal abnormalities as detected by flow cytometry analyses (FCM), are both well established markers indicating impending malignant transformation. The above-mentioned groups of patients at risk of developing colorectal carcinoma are today usually followed in colonoscopic surveillance programs in order to detect dysplasia preceding the development of invasive carcinoma, or at least to detect early cancer at a curable stage (12, 13). Also, patients with chronic, etensive Crohn s colitis may warrant colonoscopic surveillance due to the high prevalence of dysplasia (14) /2004 $

2 Table I. UDCA-pilot study outline. Visit # No. of months in study informed consent demographics medical history medication log randomization clinical chemistry and hematology colonoscopy histopathology DNA-FCM adverse events * dispense of study drug drug accountability drug compliance * Clinical activity; remission vs activity Much interest has been focused on the role of bile acids in colonic neoplasia, ever since Earnest et al. reported that dietary ursodeoycholic acid (UDCA) decreased the incidence of eperimental colonic carcinogenesis in a rat azoymethane model (15). Since then, UDCA, the 7 ß- epimer of chenodeoycholic acid, has been demonstrated to harbour several anticarcinogenic properties: a) it reduces the colonic concentration of the secondary bile acid, deoycholic acid, which is a strong cancer promoter (16, 17), b) it has been shown to stabilize the cell membranes of cultured hepatocytes and erythrocytes against other bile acids (18-20), c) it increases the epression of class I and II major histocompatibility antigens in premalignant colonocytes from rats treated with azoymethane (21), d) it blocks alteration in protein C kinase epression and distribution induced by azoymethane in eperimental colon carcinogenesis (22) and e) it increases the intracolonic level of alkaline sphingomyelinase, an enzyme which is believed to inhibit cell proliferation and induce apoptosis (23). It is tempting to speculate that UDCA might also be a potential chemoprotective agent in humans at high-risk of developing CRC. This aspect was addressed recently in two studies (24, 25) showing that UDCA use in patients with UC and PSC was associated with a lower risk of colonic neoplasia compared to patients not taking the compound. Our primary objective in this double-blind, prospective controlled pilot trial was to study patients with long-standing colorectal IBD with eisting premalignant findings of a certain magnitude in order to assess whether preventing/reverting actions of UDCA could be demonstrated in such a selected, high CRCrisk group. Our primary aim was to study the impact on the need for colectomy due to progression of dysplasia. Patients and Methods Overall study design. The study was performed as a single center trial at the IBD-unit at Huddinge University Hospital, Karolinska Institutet, Stockholm, Sweden and was of randomized, controlled, double-blind design. The patients were randomly allocated to therapy with either UDCA (Ursofalk, Meda AB, Sweden, 500 mg b.i.d) or placebo and the duration of treatment was two years. Table I outlines the study in general terms. The inclusion criteria into the study were as follows: 1) Definite diagnosis of UC or CD in the colon and rectum; etensive UC (mucosal involvement proimal to the splenic fleure), etensive CD in the colon and rectum (>2/3 of colon involved). 2) Colonoscopic surveillance should already have been initiated, either due to long duration (>8 years), etensive disease, or due to other risk factors (e.g. PSC). 3) Findings of DNA-aneuploidy and/or low-grade dysplasia (LGD), should already have been confirmed during the most recent preceding surveillance colonoscopy. 4) Patients not willing to undergo surgery in spite of certain predetermined histological findings eceeding the minimal criteria for colectomy (i.e. confirmed low-grade dysplasia in connection with a Dysplasia Associated Lesion or Mass (DALM), or high-grade dysplasia (HGD) either in flat mucosa, or associated with a DALM. 5) Age>18 years. 6) Informed consent to participate in the study. The eclusion criteria were: 1) Pregnancy, lactation or childbearing potential without adequate contraception. 2) Allergy to UDCA. 3) History of alcohol or other abuse, or history of psychosis, emotional or intellectual problems likely to limit the validity of the consent to participate in the study. Study events. Out-patients visits, including full colonoscopy with multiple biopsies, were performed every si months. Blood samples for hematology (B-Hemoglobin (g/l), B-Leukocytes, particle concentration (10 9 /L) and B-Platelets, particle concentration ( 10 9 /L)) and for clinical chemistry (B-Glucose (mmol/l), S-Sodium (mmol/l), S-Potassium (mmol/l), S-Creatinine (Ìmol/L), S-Bilirubin, total (Ìmol/L), S-Alkaline phosphatase (Ìkat/L), S-Alanine aminotransferase (Ìkat/L), S-Aspartate aminotransferase (Ìkat/L), S-GT, Á- Glutamyltransferase, (Ìkat/L), S-Albumin (g/l), S-C-reactive protein (mg/l), S-Orosmucoid (g/l), B-Erythrocyte sedimentation rate (B- ESR) 1 hour (mm/h), S-Cholesterol (mol/l), and S-Triglycerides (Ìmol/L)) were obtained at each visit as well as urinalysis (dipslides). Drug accountability and compliance were recorded at each visit as were any adverse events. Maintenance drugs such as 5-ASA, 3122

3 Sjöqvist et al: Ursodeoycholic Acid Therapy in IBD-patients with Colorectal Neoplasia Table II. Patient characteristics; se, age (median), IBD-type, disease duration (median), concomitant PSC, treatment (SASP=sulphasalazine, AZA=azathioprine), previous colonic surgery, eistence of aneuploidy or LGD at study start (% of total). a) Treatment group Pat.no M/F Age IBD-type Disease- PSC IBD-treatment Previous Aneuploidy LGD duration (dose g/day) colonic (% of total (% of total surgery biopsies) biopsies) 1 M 54 UC 46 - SASP (2) - An (100) LGD (100) 3 F 27 UC 18 + SASP (2) - An (20) LGD (44) 5 M 33 UC 26 + SASP (2) +1 An (28) LGD (28) 6 F 33 UC 14 - SASP (2) - - LGD (11) 9 M 43 UC 20 - SASP (2) - An (11) - 11 M 62 UC 47 - SASP (2) +2 An (67) - 14 F 27 UC An (33) - 16 F 46 CD An (11) LGD (11) 18 M 73 CD 14 - SASP (2) +3 An (67) - 19 M 62 CD 9 + SASP (2) - An (56) - Total 106M/4F M 44 M 19 90% (An) 50% (LGD) of the patients of the patients 1 Operated with resection of the sigmoid colon due to DALM+ LGD 1995 (refused pancolectomy) 2 Operated with IRA(ileorectal anastomosis) 1989 due to DALM with Dukes A in left colon. Only had the rectum left for surveillance 3 Operated with IRA due to Dukes B adenocarcinoma in the transverse colon in Only had the rectum left for surveillance. b) Placebo group Pat.no M/F Age IBD-type Disease- PSC IBD-treatment Previous Aneuploidy LGD duration (dose g/day) colonic (% of total (% of total surgery biopsies) biopsies) 2 M 39 UC 8 - SASP (2) - An (56) LGD (78) 4 M 32 CD 18 + Metronidazole (0.8) - An (56) LGD (33) 7 F 45 UC 21 - SASP (2) - An (33) LGD (56) 8 M 42 UC LGD 33) 10 F 42 UC 25 - SASP (2) - An (33) - 12 F 54 UC 24 - Olsalazine (1) - - LGD (22) 13 M 47 UC 41 - SASP (2) - An (44) - 15 M 43 CD 24 - Olsalazine (1) - - LGD (22) 17 F 43 CD 17 - Metronidazole - An (44) LGD (22) +AZA (0.8 and 0.75) Total 9 5M/4F M 43 M 24 67% (An) 78% (LGD) of the patients of the patients SASP, olsalazine, metronidazole or azathioprine were allowed provided the dose was kept constant. Other intercurrent drug therapy necessary for the patient s well being was recorded. Treatment with antacids containing aluminum hydroide, cholestyramine or colestipol was not allowed during the course of the study because of risk of interaction with UDCA. Patients. Nineteen patients with etensive colitis undergoing colonoscopic surveillance were selected from an ongoing surveillance program at Huddinge University Hospital, Stockholm, Sweden previously described in detail (26). LGD and/or DNA-aneuploidy had been present in all of these patients at the preceding colonoscopies. The patients enrolled were randomly allocated to receive capsules containing UDCA (Ursofalk, Meda AB) 500 mg b.i.d or an identical placebo. In all, 10 patients received active treatment and nine placebo. The demographics and medical history of the studied patients are displayed in Table II. Placebo group: In the placebo group the median age at onset of IBD was 20 years (range 6-31) and at study start the median age was 43 years (range 32-54). The median disease duration at study start was 3123

4 24 years (range 8-41). The patients had been under colonoscopic surveillance for a median of 33 months (range 7-172). Si patients had UC and three had Crohn s colitis of which one in the latter group also had a confirmed diagnosis of concomitant PSC. The maintenance treatment at study start included SASP, 5-ASA, metronidazole and azathioprine as shown in Table II. None of the patients had had previous bowel surgery. At the time of inclusion, si patients had findings of DNA-aneuploidy and seven patients had confirmed findings of LGD in at least one of the biopsies. Treatment group: In the treatment group the median age at onset of the disease was 20 years (range16-31) and at study start the median age was 44 years (range 27-73). The median disease duration at study start was 19 years (range 9-47). The patients had been under colonoscopic surveillance for a median of 62 months (range 7-137). Seven patients had UC and three colorectal CD. Three of the patients had PSC. In this group patient # 5 had been operated 18 months prior to study start with an unconventional (patient s preference) resection of the sigmoid colon due to a DALM with both HGD and DNA-aneuploidy. Patient # 11 was operated 106 months before study start by colectomy with ileorectal anastomosis (IRA) due to CRC (Dukes A) in the left colon. Patient #18 was operated with colectomy and an IRA 41 months before study start due to a CRC in the transverse colon (Dukes B). The two latter patients thus only had rectum left in situ for surveillance. The maintenance treatment consisted of SASP and 5-ASA in all patients ecept patient # 14 (no treatment). All patients ecept one in the treatment group had displayed DNAaneuploidy in at least one of the biopsies at prior colonoscopy and five patients had findings of LGD at the time of inclusion. Colonoscopy and biopsy sampling. Colonoscopy was always carried out by either US or RL using Olympus (CF 100) equipment. No or only very limited sedation was used (diazepam 5-10 mg i.v.). All colonoscopies performed within the frame of the study were complete. At each colonoscopy two biopsies for histopathological evaluation and one biopsy for FCM were sampled from each of nine predetermined segments of the colon and rectum. The biopsies were taken immediately adjacent to each other in all locations (< 1-2 mm:s apart). Histopathology. Biopsies for histopathology were fied in formalin and later cut in sections. The hematoylin-eosin sections were consecutively evaluated in a blinded manner with regard to dysplasia by one of the authors (ÅÖ). Dysplasia was classified using the criteria established by Riddell et al. (27) and was recorded only when the mucosa showed no active inflammation (defined as no more than infiltration of lymphocytes and plasma cells in the lamina propria (28)) to avoid misinterpretation of reactive, inflammatory changes. The histopathological grading of dysplasia was classified as not present (NP), indefinite; probably negative (probably reactive (IPR)), indefinite; unknown (IU), indefinite; probably positive (probably dysplastic (IPP)), or definite low-grade dysplasia (LGD), high-grade dysplasia (HGD) with or without a dysplasia associated lesion or mass (DALM). DNA-flow cytometry. Biopsies were fied in buffered formalin to optimise quality and later analysed as previously described (29). Preparation of cell nuclei was done from fresh tissues for highquality DNA flow cytometry. Briefly, the biopsies were fied in Table III. Composed score for dysplasia and aneuploidy. The mean sum from the histopathologic report was added to the mean sum from the flow cytometry analyses score and used for statistical analyses. Histopathological report Score No dysplasia 0 Indefinite, probably negative (probably reactive) 0 Indefinite, unknown 0 Indefinite, probably positive (probably dysplastic) 1 Low grade dysplasia 2 High grade dysplasia 3 DALM + 1 Flow cytometry analyses Score Diploid 0 Aneuploid 1 buffered formalin, the cytoplasm was digested by protease (Sigma protease XXIV) and the bare nuclei were stained directly by DAPI (Sigma D9542). This preparation, without any centrifugation steps, resulted in minimal damage and low frequencies of aggregates; thus, the background levels in the DNA-histograms were very low. The cell nuclei were analyzed with a PAS II flow cytometer (Partec, Münster, Germany). For cell cycle analysis, the sliced nuclei option for background subtraction of the Multicycle program (Version 3.0, Phoeni Flow System, San Diego, CA, USA) was used. DNA histograms with a single peak were classified as diploid and those with an additional peak and a corresponding G2+M peak as aneuploid. Composed neoplastic score. In order to evaluate whether progression or regression of the neoplastic lesions occurred during the observation period of two years, we recorded the eistence of aneuploidy or dysplasia at the entry colonoscopy and at each of the four scheduled colonoscopies performed at 6, 12, 18 and 24 months during the course of the study. In order to compare and include all of the biopsies taken at each colonoscopy, we developed a scoring system presented in Table III. For the histopathological report numerical values were obtained by giving biopsies with no dysplasia, indefinite, probably negative and indefinite, unknown a zero point. Biopsies with indefinite, probably positive (probably dysplastic) findings scored one point. LGD rendered two points and HGD three points. If DALM was observed an additional point was added. The sum was then divided by the total number of biopsies at each colonoscopy resulting in a numerical mean value. DNA-FCM diploid samples were rated zero. Aneuploidy rendered one point and the sum was then divided by the total number of biopsies at each colonoscopy resulting in a numerical mean value. The mean values from the histopathology report were then added to the results from DNA-FCM resulting in a total composed score. This system was used throughout the entire study at each colonoscopy. Statistical analysis. Mean values ± SD or median values with range were used in the tet for the descriptive tables. The primary analysis was based on the all-patients-treated and the last value-etended 3124

5 Sjöqvist et al: Ursodeoycholic Acid Therapy in IBD-patients with Colorectal Neoplasia Figure 1. a. Composed neoplastic score for the treatment group at study start (0.87±0.23) and study end (0.71±0.23), p=0.63. Boplot with means and SD. b. Composed neoplastic score for the placebo group at study start (1.12±0.48) and study end (0.87±0.95), p=0.51. Bo plot with means and SD. principles. For analysis of composed score, two-way analysis (Anova) was applied to the changes from entry to study end. This pilot study was not powered to ensure statistical significance in differences concerning the primary endpoint between the two groups. However, from our point of view and others (12, 30) eperience, it is known that definite dysplasia and/or DNA-aneuploidy rarely reverts spontaneously, making it possible to detect trends in this proof of concept trial. Ethical considerations. The study was performed in accordance with the principles stated in the Declaration of Helsinki and the study was approved by the local ethics committee at the Karolinska Institutet as well as the Swedish Medical Product Agency (Läkemedelsverket) before the trial started. Results The groups were comparable with respect to age at onset of disease and disease duration at study start. By chance, the treatment group included three patients operated on for DALM (n=1) and colorectal cancer (n=2) 18 and 106 months before study start. On the other hand, the patients randomized to UDCA had a shorter duration of disease (median 19 years vs 24, p=0.97) There were no differences in composed score between the two groups at study start. For the placebo group the mean neoplastic composed score was 1.12±0.48 and for the treatment group 0.87±0.23 (p=0.47). During the study, two of the patients in the placebo group had a colectomy due to increased degree and severity of mucosal changes. Patient # 7 developed HGD in one location in the descending colon at the planned one-year assessment colonoscopy. At this time the patient also had LGD in two locations and IPP in five locations. DNA-FCM showed five samples with aneuploidy including high S-phase fractions in two of those. This patient was subsequently operated on with colectomy and an ilealpouch anal anastomosis (IPAA) at 17 months. No colorectal cancer was detected in the colectomy specimen. Patient # 17 was operated on due to sessile dysplastic polyps interpreted as DALM:s with LGD in the ceacum after 12 months in the study. This patient displayed aneuploidy in the same area at the previous colonoscopy together with LGD. The patient received an ileorectal anastomosis because of CD. No cancer was found in the surgical specimen. Besides these two patients, no significant differences could be seen in any of the groups during the study. Figure 1 shows the composed score for the treatment group (a) and the placebo group (b) at the study start and at the end. The mean values of the sum for the treated were 0.87±0.23 at study start and 0.71±0.23 at study end (p=0.63). Corresponding values for the placebo group were 1.12±0.48 and 0.87±0.95 (p=0.51). There were no differences in mean S-phase fraction for diploid and aneuploid samples at study start compared to study end. For diploid biopses, in the treatment group the mean S- phase fraction was 2.7±2.4% at study start vs 2.94±1.01 at study end (p=0.47). In the placebo group the values were 3.17±1.97 and 3.01±1.87, respectively (p=0.86). For aneuploid samples, in the treatment group the mean S- phase fraction was 2.95±1.9% at study start compared to 3.44±1.8% at study end (p=0.3). Corresponding values in the placebo group were 3.18±1.2 and 3.97±1.21% (p=0.28). The mean percentage of aneuploid cells from each aneuploid cell population analysed seperately, as well as the mean percentage of aneuploid cells per total number of biopsies, were also compared but no differences could be seen in either group throughout the study period (data not shown). 3125

6 Compliance and side-effects. All of the patients in the treatment group were followed as per protocol, ecept patient #14 who became non-compliant after 12 months due to diarrhea and was thereafter withdrawn. Patient #18 could not be investigated by colonoscopy at 24 months for personal reasons. In the placebo group, patient #8 did not attend colonoscopy at 6, 18 and 24 months but was compliant in taking medicine. One of the patients in the treatment group with known PSC had an attack of cholangitis (#19) and was hospitalized and treated with antibiotics for a week. Pat #12 in the placebo group had a flare-up of colitis which was treated with oral steroids (Prednisone). Both of these patients recovered well after treatment. Laboratory tests throughout the study did not show any abnormalities in either group, ecept for patient #19 who eperienced cholangitis and patient #12 when eperiencing an acute attack of colitis (high inflammatory parameters). Discussion We investigated for the first time, in a prospective, doubleblinded study the possible chemoprotective effects of UDCA in patients with longstanding etensive IBD. For this proof of the concept study only high-risk patients already with findings of premalignant lesions (aneuploidy or LGD) were selected. The reason was to evaluate any possible impact on manifest dysplasia and/or DNAaneuploidy in the colorectal mucosa over time. The main finding in this study was that two of the patients in the placebo group had to be referred for colectomy due to progression of dysplasia (HGD and DALM with LGD, respectivelly). In contrast, no patient in the UDCA-group eperienced dysplasia progression, and no patient in this group was deemed to be in immediate need for surgery. This difference was also related to the fact that three patients in the treatment group had previously had either HGD+DALM (patient #5) or manifest CRC (patient #11 and 18). This indicates that the active UDCA-treatment group was at least not less prone to developing advanced dysplasia in comparison to the control group. It is known, from previous colonoscopic surveillance studies (12, 30), that during a five-year follow-up period up to 50% of patients with LGD progress into HGD. However, overall in our study, no systematic changes could be found between the groups regarding the presence of dysplasia, aneuploidy or percentage of S-phase fraction. There has been so far two studies adressing the role of UDCA as a potential chemoprotective agent in patients with UC and PSC. The first one from Tung and collaborators (25) showed a strong association between UDCA use and a lower prevalence of colonic dysplasia (odds ratio 0.18, p=0.005). This, retrospective, study comprised 59 patients of which 41 were treated and 18 not. In that study, the treated patients were older and had had shorter duration of disease than the untreated patients, putting them at lower risk of developing dysplasia. Ambiguites regarding the etent of the disease as well of the dosage of UDCA used make that particular study difficult to evaluate. Recently, Pardi et al. (24), in a prospective study, analyzed the outcome of colonic neoplasia in 52 UC/PSC patients who were initially enrolled in an ursodiol, placebo-controlled trial, investigating UDCAs roll in improving liver function in these patients (31). Those originally assigned to receive UDCA had a relative risk of 0.26 for developing colorectal dysplasia or cancer (95% confidence interval, ; p=0.034). In those two studies only patients with PSC who have the highest risk of developing dysplasia were studied, but at the same time they constitute the smallest part of the population of patients with etensive UC. In our study the patients were well defined and represented a group with a fairly high-risk of developing CRC. They were followed in a predetermined period, including repeated colonoscopies. Histopathological evaluation was done by the same pathologist with long-term (i.e.>27 years) of dysplasia assessment in IBD (13, 26, 28, 32, 33). Interpretation of dysplasia was made in a blinded fashion without knowledge of the patient treatment or phase in study. Both patients who were selected for surgery also had, apart from HGD and DALM with LGD, widespread findings of aneuploidy as an objective marker for gross chromosomal changes highlighting the neoplastic transformation in their colorectal mucosa. We chose a study time of two years which, in retrospect, may have been too short. A longer treatment period than two years may be needed in order to fully eploit the possible reverting impact on the neoplastic lesions already present. The dose of UDCA given was based on eperience from the recommendation in primary biliary cirrhosis (10-12mg/kg body weight). However, it may have been too low for the purpose we evaluated UDCA for in this particular study. Thus, prolonged treatment and/or increased dose of UDCA may be worthwhile aspects to study in future trials. This small proof of concept study encourages the further clinical approach of a chemoprotective role for UDCA in patients with long-standing IBD. A larger study in IBDpatients seems warranted. Such a study should most probably focus on the high-risk patient group not yet having developed dysplasia and/or aneuploidy and be of sufficient length (i.e. more than 2 years). Acknowledgements The study drug and placebo were provided by Dr Falk Pharma, Germany via Meda AB, Sweden. Research grants were provided by Pharmacia and Astrazeneca, Sweden. 3126

7 Sjöqvist et al: Ursodeoycholic Acid Therapy in IBD-patients with Colorectal Neoplasia References 1 Gyde SN, Prior P, Allan RN, Stevens A, Jewell DP, Truelove SC et al: Colorectal cancer in ulcerative colitis: a cohort study of primary referrals from three centres. Gut 29(2): , Ekbom A, Helmick C, Zack M and Adami HO: Ulcerative colitis and colorectal cancer. A population-based study. N Engl J Med 323(18): , Gilat T, Fireman Z, Grossman A, Hacohen D, Kadish U, Ron E et al: Colorectal cancer in patients with ulcerative colitis. A population study in central Israel. Gastroenterology 94(4): 870-7, Ekbom A, Helmick CG, Zack M, Holmberg L and Adami HO: Survival and causes of death in patients with inflammatory bowel disease: a population-based study. Gastroenterology 103(3): , Ekbom A, Helmick C, Zack M and Adami HO: Increased risk of large-bowel cancer in Crohn's disease with colonic involvement. Lancet 336(8711): 357-9, Gillen CD, Andrews HA, Prior P and Allan RN: Crohn's disease and colorectal cancer. Gut 35(5): 651-5, Brentnall TA, Haggitt RC, Rabinovitch PS, Kimmey MB, Bronner MP, Levine DS et al: Risk and natural history of colonic neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis. Gastroenterology 110(2): 331-8, Broome U, Lofberg R, Veress B and Eriksson LS: Primary sclerosing cholangitis and ulcerative colitis: evidence for increased neoplastic potential. Hepatology 22(5): , Shetty K, Rybicki L, Brzezinski A, Carey WD and Lashner BA: The risk for cancer or dysplasia in ulcerative colitis patients with primary sclerosing cholangitis. Am J Gastroenterol 94(6): , Askling J, Dickman PW, Karlen P, Brostrom O, Lapidus A, Lofberg R et al: Family history as a risk factor for colorectal cancer in inflammatory bowel disease. Gastroenterology 120(6): , Askling J, Dickman PW, Karlen P, Brostrom O, Lapidus A, Lofberg R et al: Colorectal cancer rates among first-degree relatives of patients with inflammatory bowel disease: a population-based cohort study. Lancet 357(9252): 262-6, Connell WR, Lennard-Jones JE, Williams CB, Talbot IC, Price AB and Wilkinson KH: Factors affecting the outcome of endoscopic surveillance for cancer in ulcerative colitis. Gastroenterology 107(4): , Lofberg R, Brostrom O, Karlen P, Tribukait B and Ost A: Colonoscopic surveillance in long-standing total ulcerative colitis- -a 15-year follow-up study. Gastroenterology 99(4): , Friedman S, Rubin PH, Bodian C, Goldstein E, Harpaz N and Present DH: Screening and surveillance colonoscopy in chronic Crohn's colitis. Gastroenterology 120(4): 820-6, Earnest DL, Holubec H, Wali RK, Jolley CS, Bissonette M, Bhattacharyya AK et al: Chemoprevention of azoymethaneinduced colonic carcinogenesis by supplemental dietary ursodeoycholic acid. Cancer Res 54(19): , Rodrigues CM, Kren BT, Steer CJ and Setchell KD: The sitespecific delivery of ursodeoycholic acid to the rat colon by sulfate conjugation. Gastroenterology 109(6): , Batta AK, Salen G, Holubec H, Brasitus TA, Alberts D and Earnest DL: Enrichment of the more hydrophilic bile acid ursodeoycholic acid in the fecal water-soluble fraction after feeding to rats with colon polyps. Cancer Res 58(8): , Galle PR, Theilmann L, Raedsch R, Otto G and Stiehl A. Ursodeoycholate reduces hepatotoicity of bile salts in primary human hepatocytes. Hepatology 12(3 Pt 1): , Heuman DM, Pandak WM, Hylemon PB and Vlahcevic ZR. Conjugates of ursodeoycholate protect against cytotoicity of more hydrophobic bile salts: in vitro studies in rat hepatocytes and human erythrocytes. Hepatology 14(5): 920-6, Guldutuna S, Zimmer G, Imhof M, Bhatti S, You T and Leuschner U: Molecular aspects of membrane stabilization by ursodeoycholate. Gastroenterology 104(6): , Rigas B, Tsioulias GJ, Allan C, Wali RK and Brasitus TA: The effect of bile acids and piroicam on MHC antigen epression in rat colonocytes during colon cancer development. Immunology 83(2): , Wali RK, Frawley BP Jr, Hartmann S, Roy HK, Khare S, Scaglione-Sewell BA et al: Mechanism of action of chemoprotective ursodeoycholate in the azoymethane model of rat colonic carcinogenesis: potential roles of protein kinase C-alpha, -beta II, and -zeta. Cancer Res 55(22): , Duan RD, Cheng Y, Tauschel HD and Nilsson A: Effects of ursodeoycholate and other bile salts on levels of rat intestinal alkaline sphingomyelinase: a potential implication in tumorigenesis. Dig Dis Sci 43(1): 26-32, Pardi DS, Loftus EV Jr, Kremers WK, Keach J and Lindor KD: Ursodeoycholic acid as a chemopreventive agent in patients with ulcerative colitis and primary sclerosing cholangitis. Gastroenterology 124(4): , Tung BY, Emond MJ, Haggitt RC, Bronner MP, Kimmey MB, Kowdley KV et al: Ursodiol use is associated with lower prevalence of colonic neoplasia in patients with ulcerative colitis and primary sclerosing cholangitis. Ann Intern Med 134(2): 89-95, Sjoqvist U, Ost A and Lofberg R: Increased epression of proliferative Ki-67 nuclear antigen is correlated with dysplastic colorectal epithelium in ulcerative colitis. Int J Colorectal Dis 14(2): , Riddell RH, Goldman H, Ransohoff DF, Appelman HD, Fenoglio CM, Haggitt RC et al: Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol 14(11): , Geboes K, Riddell R, Ost A, Jensfelt B, Persson T and Lofberg R: A reproducible grading scale for histological assessment of inflammation in ulcerative colitis. Gut 47(3): 404-9, Castro J, Heiden T, Wang N and Tribukait B: Preparation of cell nuclei from fresh tissues for high-quality DNA flow cytometry. Cytometry 14(7): , Ullman TA, Loftus EV Jr, Kakar S, Burgart LJ, Sandborn WJ and Tremaine WJ: The fate of low grade dysplasia in ulcerative colitis. Am J Gastroenterol 97(4): 922-7, Lindor KD: Ursodiol for primary sclerosing cholangitis. Mayo Primary Sclerosing Cholangitis-Ursodeoycholic Acid Study Group. N Engl J Med 336(10): 691-5, Brostrom O, Lofberg R, Ost A and Reichard H: Cancer surveillance of patients with longstanding ulcerative colitis: a clinical, endoscopical, and histological study. Gut 27(12): , Lofberg R, Brostrom O, Karlen P, Ost A and Tribukait B: DNA aneuploidy in ulcerative colitis: reproducibility, topographic distribution, and relation to dysplasia. Gastroenterology 102(4 Pt 1): , Received April 7, 2004 Accepted June 22,