Abstracts of the 3rd ECCO Congress, Lyon, France, February 28–March 1, 2008 7 assessed the ability of adalimumab to maintain remission and response in infliximab-failure patients who completed the 4-week GAIN trial through 1 year of therapy in an ongoing open-label extension study. Materials and methods: As previously reported, 1 325 patients with Crohn' s disease who failed infliximab therapy enrolled in GAIN and were randomized to receive induction adalimumab 160/80 mg at Weeks 0 and 2 or placebo. At Week 4, all patients were eligible to enter an open-label extension dur- ing which they received open-label adalimumab 40 mg every other week. Patients could switch to weekly dosing for flares (increase in Crohn's Dis- ease Activity Index [CDAI] ≥70 points compared with Week 4 and CDAI>220) or non-response (CDAI decrease <70 points from baseline). Post-hoc analy- ses of maintenance of remission (CDAI<150) and response (drop in CDAI ≥70 [CR-70] or 100 [CR-100] points) were performed on both the intention-to- treat (all patients entering the open-label extension) and responder (CR-70 at Week 4 of GAIN) populations. Results: A total of 310 patients enrolled in the open-label extension, and 126 patients were responders. Remission/response results are shown for all patients who entered the open-label extension (Table). Table 1. remission/response with adalimumab in infliximab-failure patients through 1 year Endpoint Months since GAIN Intention-to-treat (N=310) Responder (N=126) baseline* n (%) n (%) Remission 6 107 (35) 72 (57) 12 89 (29) 50 (40) CR-100 6 157 (51) 94 (75) 12 135 (44) 83 (66) CR-70 6 187 (60) 116 (92) 12 156 (50) 82 (65) *Months 6 and 12 represent Weeks 24 and 48 in the open-label extension, in addition to the original 4 weeks in GAIN. Conclusions: Adalimumab shows sustained efficacy in maintaining clinical re- mission and response through 1 year of therapy in patients who failed prior infliximab therapy, with two-thirds of responding patients maintaining re- sponse and 40% achieving remission. Reference: 1. Sandborn WJ, et al. Ann Intern Med. 2007;146:829-838. P007 COLONOSCOPIC SURVEILLANCE FOR DYSPLASIA AND COLORECTAL CANCER IN LONG-STANDING ULCERATIVE COLITIS: A COHORT ITALIAN STUDY S. Ardizzone 1 , M. Bosani 1 , A. Cassinotti 1 , P. Fociani 2 , A. Massari 1 , G.M. Sampietro 3 , L. Ferrari 2 , G. Maconi 1 , G. Bianchi Porro 1 . 1 Department of Clinical Science, Gastroenterology Unit; 2 Pathology Unit; 3 Surgery Division, “L. Sacco” University Hospital, Milan, Italy Background: Colorectal cancer (CRC) can be a complication of long-standing ulcerative colitis (UC). Many risk factors has been reported. Periodic colono- scopic surveillance with multiple colonic biopsies is the mainstay to check for the presence of CRC or any preneoplastic lesion, such as low-grade (LGD) and high-grade (HGD) dysplasia, although it has not been proved to reduce mortality from CRC. Moreover, a few/some studies do not agree about the natural history of dysplasia, risk factors for its progression, and clinical man- agement of dysplasia, in particular LGD. Aim: To determine 1) the incidence of dysplasia and CRC in our cohort of long-standing UC; 2) any risk factors for dysplasia or CRC; 3) any risk factor for colectomy; 4) the clinical outcome and survival rate after the diagnosis of dysplasia/cancer or colectomy. Patients and methods: A cohort of 400 patients (M/F 250/150, mean age 38 years) with UC lasting more than 8 years (mean 18.5, range 8-30 years) was prospectively studied from 1995 to 2007 and underwent periodic surveillance colonoscopy for cancer risk. Biopsies were performed at 10 cm intervals and from strictures or any polipoid masses. Pathology was classified as normal, dysplasia (indefinite, LGD, HGD) or carcinoma, and was confirmed by a sec- ond independent pathologist. Results: No death for CRC was observed after a mean follow-up of 6 (1.5- 12) years. A total of 986 examinations (mean 2.46, range 1-14 for patient) was performed. The surveillance program detected dysplasia in 51 [indef- inite 0 (0%), LGD 49 (12,2%), HGD 2 (0.5%) and CRC in 9 (2.2%)] patients. Definite dysplasia was associated with disease duration (p=0.00) and steroid use (p=0.03); no significant factors had been found for CRC risk. Fifty-nine colectomy were performed, of which 18 (4.5%) for dysplasia or cancer. All cases of CRC were confirmed after colectomy, while no previously described LGD was confirmed. Among LGD cases where colectomy was not performed, no patients developed CRC on successive follow-up, while the diagnosis of LGD was confirmed only in 1 patient during follow-up. Conclusions: Colonoscopic surveillance is safe and effective in early diagno- sis of CRC. The cancer incidence was lower than in the other studies. The role of LGD as a risk factor of CRC was not confirmed. P008 6-THIOGUANINE NUCLEOTIDE (6-TGN) CONCENTRATIONS AND EFFICACY OF AZATHIOPRINE (AZA) AND MERCAPTOPURINE (MP) IN INFLAMMATORY BOWEL DISEASE: THE METAZA STUDY J.P. Gisbert , on behalf of GETECCU (Grupo Español de Trabajo en Enfermedad de Crohn y Colitis Ulcerosa). La Princesa Hospital, Madrid, Spain Objective: Some authors have suggested that inflammatory bowel disease patients achieving remission with AZA/MP treatment have higher mean 6- TGN concentrations than those not responding to these drugs. Our aim was to evaluate the accuracy of the determination of 6-TGN concentrations as marker of AZA/MP treatment efficacy. Methods: In this multicenter national study, patients with inflammatory bowel disease starting treatment with AZA (2.5 mg/kg/day) or MP (1.5 mg/kg/day) for steroid-resistant or steroid-dependent disease were consec- utively included. TPMT activity was assessed before starting AZA/MP treat- ment, and patients with values < 5 U/mL were excluded. After starting AZA/MP treatment, 6-TGN concentrations were determined at 2 weeks, 1 and 2 months, and then every 2 months up to 1 year. Truelove modified in- dex and CDAI were used to assess clinical response. The follow-up finished once the steroid-resistance or steroid-dependence was overcame (remission for at least 6 months after stopping steroid treatment), when response was not achieved after 6 months of AZA treatment, or when clinical relapse was demonstrated. Results: One-hundred and fifty-three patients were included (mean age 36 years, 51% males, 73% with Crohn's disease and 27% with ulcerative colitis). Ninety-eight of these patients completed the study (e.g. were classified as responders or non-responders). Mean 6-TGN concentrations at 2 weeks, and at 1, 2, 4, 6, 8, 10 and 12 months after staring AZA treatment were: 262 ± 142, 310 ± 194, 309 ± 148, 293 ± 154, 307 ± 213, 329 ± 245, 306 ± 181 and 391 ± 269 pmol/8 x 10 8 red blood cells. The detection at 2 weeks of 6-TGN concentrations > 230 had a sensitivity and a specificity of only 48% and 52% for the diagnosis of clinical remission at the end of follow-up. In- creasing the cut-off point up to 250 did not change the sensitivity, and only slightly increased the specificity (62%). The area under the ROC curve (AUC) which assesses the global accuracy of the 6-TGN determination at 2 weeks to estimate the response to AZA/MP was of only 0.5. 6-TGN determination at 1 and 2 months did not improve previous results (AUC of 0.54 and 0.52, respectively). Finally, 6-TGN determination at 4 months was associated with a higher diagnostic accuracy, although it was still deficient (AUC of 0.65; sensitivity and specificity of the cut-off point set at 230 was 69% and 60% respectively). Conclusion: Determination of 6-TGN concentrations is not useful in estimat- ing the response to AZA/MP treatment. Therefore, systematic quantification of 6-TGN concentrations in patients receiving AZA/MP with the aim of pre- dicting treatment response cannot be recommended. P009 MUCOSAL HEALING PREDICTS SUSTAINED CLINICAL REMISSION IN EARLY CROHN' S DISEASE F. Baert 1 , L. Moortgat 1 , G. Van Assche 2 , P. Caenepeel 3 , P. Vergauwe 4 , M. De Vos 5 , P. Stokkers 6 , D. Hommes 7 , P. Rutgeerts 2 , S. Vermeire 2 , G. D' Haens 8 . 1 H-Hartziekenhuis, Roeselare, Belgium; 2 University Hospital Gasthuisberg, Leuven, Belgium; 3 ZOL, Genk, Belgium; 4 Groeninge ziekenhuis, Kortrijk, Belgium; 5 University Hospital, Gent, Belgium; 6 AMC, Amsterdam, The Netherlands; 7 LUMC, Leiden, The Netherlands; 8 Imelda GI Clinical Research Center, Bonheiden, Belgium Aims: At present no clinical data are available supporting the relevance of mucosal healing in Crohn' s disease (CD). We already reported that early com- bined immunosuppression (CIS) with infliximab (IFX) and azathioprine (AZA) led to significantly higher rates of complete ulcer healing after 2 years (yrs) compared to conventional management (CM) including corticosteroids. This study focused on the value of endoscopy performed after 2 yrs of treatment to predict clinical outcome in the following 2 yrs. Methods: 133 newly diagnosed CD pts at 21 centers were randomized to treatment with CIS (3 infusions of IFX with AZA) or to CM with corticos- teroids, to be repeated as clinically necessary. In the CIS group, patients with a relapse were given repeated IFX. In the CM group, AZA was added in case of corticosteroid dependency and IFX was only given after failure of AZA. Nineteen pts dropped out of the study before 1 yr. A subset of 44 pts (24 CIS, 20 CM) underwent an ileocolonoscopy at 2 yrs. Endoscopic CD activity was scored with the Simple Endoscopic Score for CD (SES-CD). All centers re- consented their pts for an additional 2 yrs of clinical follow-up. Fishers exact two sided test was used for comparison. Results: At the time of analysis year 4 data were available in 42/44 pts (23 CIS, 19 CM). A SES-CD score = 0 at yr 2 predicted a stable clinical remission