study we were unable to find a beneficial effect of starting anti-TNF early compared to starting anti-TNF late on disease complications and surgery. It seems that an inappropriate selection of pts leads to a suboptimal drug efficacy lacking the ability to prevent the develop- ment of complications. Future research should aim to elucidate the method for selecting pts. Su1381 Vagus Nerve Stimulation in Active Crohn's Disease Bruno L. Bonaz, Sonia Pellissier, Nicolas Mathieu, Dominique Hoffmann, Candice Trocme, Magalie Baudrant-Boga, Valérie Sinniger, Chloe Picq, Olivier David, Laurent Vercueil, Cecile Dantzer, Jean-Luc Cracowski, Didier Clarençon Aim: The vagus nerve (VN) has an anti-inflammatory role through the cholinergic anti- inflammatory pathway (Mol Med 2003;9:125-34). VN stimulation (VNS) improves colitis in rats (Auton Neurosci 2011;160:82-9) but has never been used in the treatment of Crohn's disease (CD). We performed a pilot study of VNS in patients with active CD (ClinicalTrials.gov NCT01569503). Methods: Consecutive patients with moderate to severe (220<CDAI<450) CD were included. Clinical evaluation (CDAI), CRP, fecal calprotectin (FC), ileo-colonoscopy (CDEIS), contrast-enhanced ultrasound (CEUS), heart rate variability (HRV, a marker of the sympatho-vagal balance) were performed before VNS and during the follow-up for one year. Under general anesthesia, an electrode (Cyberonics, Lyon, France) was wrapped around the left VN in the neck, tunnelized subcutaneously, and connected to a pulse generator (Cybero- nics) located subcutaneously in the left chest wall. The following parameters were used for VNS: 0.5 mA, 10 Hz, 30 s ON, 5 min OFF, continuous cycle. In case of aggravation, patients were removed from the study and treated with anti-TNF/immunosuppressants or surgery. Results: 4 patients have been included: 3 men/1 women; mean age: 42.5 years (32-50); Montreal classification (A3L1B2, A2L1B3, A2L2B1, A3L2B1); length of the disease (9.5 years; 0.5-26); smoking (2/4). Two patients were Naive of treatment on inclusion, one was under azathioprine (AZA) since 8 years and one had stopped AZA 3.5 years before. The mean CDAI at entrance was 335 (320-358), CRP: 78 (8-166), FC: 677 (20-1762), CEUS: active disease, CDEIS: 16 (8-30), vagal hypoactivity on HRV was observed in 3/4 patients and uninterpretable in one. Two patients are currently in clinical remission (CDAI<150) with mucosal healing (CDEIS: 3-0) with a respective follow-up of 20 and 8 months. The patient in deep remission under AZA+VNS stopped AZA after 14 months of VNS and is in remission after 6 months of follow-up. One patient presented an improvement of CDAI and FC but switched to surgery after 2.5 months of VNS because of a persistent CD activity with an entero-enteric fistula and abscess; the patient agreed to pursue VNS alone post- surgery and had endoscopic healing (i0) at 6 months post-surgery. The last patient switched to Infliximab (IFX) and AZA because of an uncontrolled disease after 3 months of VNS despite a transient improvement. Two patients significantly improved HRV vagal tone after 6 months of VNS but not the third one who switched to IFX+AZA. VNS was well tolerated and no patient withdrew from the study due to complications or side-effects. Conclusion: The preliminary data of this pilot study show the feasibility of VNS which was well tolerated and efficient in 2 patients currently in deep remission; one patient with a failure of VNS is in deep remission 6 months post-surgery. We are currently pursuing the study. Su1382 Adalimumab and Infliximab As Induction Therapy for Crohn's Disease - A Prospective Observational Study in Germany Bernd Bokemeyer, Ulf Helwig, Niels Teich, Carsten Schmidt, Thomas Krummenerl, Ann- Katrin Rupf, Heinz Hartmann, Michael Bläker, Annette Krummenerl, Marc Düffelmeyer, Regina Hinrichs, Petra Hartmann, Susanna Nikolaus, Dietrich Hüppe, Stefan Schreiber Background: The nationwide BioCrohn Registry (Biological Registry with Crohn ` s Disease Patients in Germany) of the German Competence-Network IBD is a five-year prospective registry of about 1.500 patients with Crohn ` s disease (CD) in Germany. This is a sub-study of the BioCrohn Registry reporting the induction therapy steroid-free remission rates in 393 anti-TNF-Naive CD-patients with adalimumab (ADA) or infliximab (IFX). Methods: Within the framework of this non-interventional prospective online documentation, data on the course of disease, on psychosocial burden of disease, on health economics and on the genetic profile were examined. End of 2012 the recruitment was stopped having 1.525 CD-patients included by 59 different gastroenterology practices and hospitals with IBD-experience. All patients have a 5 year follow-up period. The databank for baseline and 6-months data has been closed in 07/2013 and after databank cleansing now we have the finalized data including the 6-months visit. Results after 6 months are calculated only on the basis of an interim analysis. Results: 392 TNF-naive CD-patients (ADA: n=263; IFX: n=128) have been analysed (average age: 36 years; female: 52%; smokers 34%; disease duration: 9.4 years; bowel resection: 33%; prior immunosuppressive therapy: 75%). Baseline characteristics were in the two groups. The IBD-therapy followed an accelerated step-up management. The indication for TNF-therapy has been steroid-dependent or steroid-refractory course of disease or patients haven ` t responded or were intolerant to an immunosuppressive therapy. Immunosuppressants were used in 19% at 6 months, in 23% after 12 months and 17% after 24 months. Accordingly to the TNF therapy, the use of systemic glucocorticoids dropped over time (baseline until 6 and 12 months) from 22.3% to 6.3% and 9.2%, respectively (p<0.001). The remission rate (PGA) at 6 months was 70.7% and 70.1% after 12 months. In spite of the TNF-induced clinical remission (> 70 %) the psychosocial impairments with anxiety/depression (EQ-5D) show only minor improvement and remain on a relatively high level (baseline: 37%, 6 months: 31%, 12 months: 29%). In the induction therapy with TNF we found a steroid- free remission (HBI< 5) in 66.7% at 6 months. Evaluating the efficacy of ADA vs. IFX as an induction therapy we do not find any difference in steroid-free remission rates at month 6 (ADA: 67.5%; IFX: 65.2%; p=n.s.). After databank cleansing in 03/2014 we will additionally show the finalized data including the 12-months visit. Conclusion: In this real life setting anti-TNF therapy could induce steroid-free remission in about 70% with the relatively early escalation of therapy in IBD-experienced centres. In comparison there is no difference in steroid-free remission between ADA vs. IFX. S-453 AGA Abstracts Su1383 Pilot Study of Oral Delivery of Monoclonal Anti-CD3 Antibody (OKT3) in Moderate to Severe Ulcerative Colitis Elisa K. Boden, James B. Canavan, Christopher J. Moran, Katelyn McCann, Francis A. Farraye, Ashwin N. Ananthakrishnan, Vijay Yajnik, Deanna D. Nguyen, Joshua R. Korzenik, Scott B. Snapper Background: Ulcerative colitis (UC) is a chronic disease characterized by infiltration of T cells in the colon. Antibodies against the T cell receptor (anti-CD3) are approved for intravenous use in the treatment of organ transplantation and are being evaluated for the treatment of autoimmune diseases. However, intravenous dosing of anti-CD3 has been limited by significant toxicity including cytokine storm. Preclinical data suggest that orally delivered anti-CD3 is safe and can prevent development of colitis in murine models. Methods: An open-label pilot study of orally delivered anti-CD3 monoclonal antibody (OKT3) in UC was performed at two referral centers. Inclusion criteria were a baseline Mayo score 6-12. Subjects taking concurrent biologic or immunomodulators were excluded. Subjects received 1 mg oral OKT3 with Omeprazole for 30 days. Primary endpoints were safety and change in immunologic parameters including lymphocyte count and proliferation of peripheral blood mononuclear cells (PBMCs) to ex vivo anti-CD3. Secondary endpoints were parameters of clinical efficacy including Simple Clinical Colitis Activity Index (SCCAI) and partial Mayo score. Response to therapy was defined as a reduction of 3 or more points in partial Mayo score with a decrease in rectal bleeding subscore of 1 point or an absolute rectal bleeding score of 0-1. Results: Six subjects were enrolled in the study and two subjects withdrew before completion. One subject withdrew due to worsening of UC and one subject withdrew for personal reasons. The study drug was well-tolerated. Two serious adverse events occurred in two subjects during the study period (worsening of bloody diarrhea and development of perianal abscess). Neither was deemed likely related to the study drug. Subjects did not experience symptoms of cytokine storm and there was no change in the absolute lymphocyte count during treatment. Subjects did not develop anti-drug antibodies up to 10 weeks after treatment initiation. As previously demonstrated in a study of oral anti-CD3 in healthy subjects, there was a significant increase in ex-vivo proliferation of PBMCs to anti-CD3 during treatment that returned to baseline after discontinuation of therapy (Fig 1). Three of six subjects had a response to therapy defined by reduction in partial Mayo score. These subjects also had significant decreases in SCCAI during treatment (Fig 2). Conclusion: In this study, orally delivered anti-CD3 appeared to be safe in subjects with moderate to severe UC and did not cause cytokine storm or depletion of lymphocytes. There was evidence of biologic activity, as PBMCs from subjects showed increased proliferation to anti-CD3 during therapy. These preliminary results are encouraging, however, a larger study of oral anti- CD3 is necessary to evaluate efficacy in UC. Figure 1: Proliferative response of PBMC to OKT3 in vitro. PBMC were obtained from subjects during the study period from baseline (week 0) to week 5. Oral anti-CD3 was administered from week 0 to 4. PBMC were stimulated in vitro with soluble anti-CD3 (5 mcg/mL). Proliferation was measured as counts per minute (CPM) by pulsing the cells for 12 h with 1 μCi [3H]thymidine after 60 h of stimulation. **p< .05 as measured by ANOVA. Figure 2: (A) Simple Clinical Colitis Activity Index (SCCAI) over study period. SCCAI was recorded for each individual subject from baseline (week 0) to week 5. (B) Partial Mayo Score over study period. Partial Mayo score was recorded for each individual subject at baseline (week 0), week 3 and week 5. Oral anti-CD3 was administered from week 0 to 4. Two subjects withdrew from the study after week 3. AGA Abstracts