Efficacy and Safety of Oral Tofacitinib as Maintenance Therapy in Patients with Moderate to Severe Ulcerative Colitis: Results from a Phase 3 Randomized Controlled Trial

steroid-free remission (remission at Wks 24 and 52; no steroid use ≥4 wks prior to each visit) among pts in remission at BL. Results: At Wk 52, tofacitinib 5 and 10 mg BID had significantly greater effect vs PBO for the primary endpoint of remission, and secondary endpoints of mucosal healing, clinical response as well as sustained remission, sustained mucosal healing and sustained clinical response (Table; p<0.001 all comparisons). Among pts in remission at BL, both tofacitinib groups had significantly higher proportions of pts with sustained steroid-free remission vs PBO (p<0.001 all comparisons). Adverse event (AE), serious AE and serious infection rates were similar among all groups. Despite more frequent infections with tofacitinib vs PBO, discontinuations due to AEs were numerically lower for tofacitinib vs PBO. A dose-dependent increase in herpes zoster (HZ) rate was observed. There were no deaths, malignancies (excluding non-melanoma skin cancer) or intestinal perforation AEs in either tofacitinib group. Changes in lipid and creatine kinase levels were consistent with results from tofacitinib studies in other populations. Conclusion: Tofacitinib 5 and 10 mg BID were significantly more effective vs PBO as maintenance therapy over 52 wks in pts with moderately to severely active UC. Despite a dose-dependent increase in HZ, overall, AE rates were similar among both tofacitinib groups. No new safety findings emerged from those previously reported in studies of rheumatoid arthritis. Reference:1. Sandborn WJ et al. J Crohns Colitis 2016;10:S15

1079 NEW TREATMENT OPTIONS FOR PROCTITIS. A PROSPECTIVE, DOUBLE BLIND, RANDOMIZED STUDY OF BUDESONIDE SUPPOSITORIES Wolfgang Kruis, Viacheslav Neshta, Marina Pesegova, Olga Alexeeva, Pavel Andreev, Oleksii Datsenko, Olena Levchenko, Sayar Abdulkhakov, Yuriy Lozynskyy, Yuriy Mostovoy, Konstatin Soloviev, Andrey Dorofeyev, Ralf Mohrbacher, Roland Greinwald, Britta Siegmund Though proctitis is the smallest form of ulcerative colitis it causes particularly awkward symptoms. Topical 5-ASA is the preferred therapy but not always successful. We studied alternative therapy with budesonide suppositories. Aims: To determine therapeutic efficacy and safety of two different dosages of a novel budesonide suppository vs a mesalamine suppository vs a combined budesonide/mesalamine suppository for proctitis. Methods: A prospective, double-blind, double-dummy, randomized, multicenter trial was performed in 337 patients with active proctitis (modified UC-DAI 4-10, with endoscopic subscore ≥ 1) comparing 4 different suppository treatment arms over 8 weeks: budesonide 2mg (BU2; n= 89 patients), budesonide 4mg (BU4; n=79), mesalamine 1000mg (ME; n=81); BU2 plus ME (BUME; n=88). Primary endpoint was the time to resolution of clinical symptoms defined as the first day of 3 consecutive days with a score of 0 for rectal bleeding and stool frequency. Results: Mean time to resolution of symptoms was similar in all groups ([days] 35.5 in BU2, 29.8 in BU4, 29.2 in ME and 29.3 in BUME). Also not different were outcomes for clinical remission (modified UC-DAI subscores "rectal bleeding" and "stool frequency" = 0): [%] 49.4 in BU2, 57.0 in BU4, 66.7 in ME and 58.0 in BUME; hallmark symptoms (mean time to 1st day with no rectal bleeding and normal stool frequency): [days] 24.7 in BU2, 20.2 in BU4, 17.5 in ME and 20.6 in BUME; mucosal healing (modified UC-DAI subscore "mucosal appearance" < 2): [%] 76.4 in BU2, 81.0 in BU4, 87.7 in ME and 84.1 in BUME; deep remission (total modified UC-DAI ≤ 1, with a score of 0 for rectal bleeding and stool frequency, no mucosal friability, and at least a 1-point reduction in the subscore for mucosal appearance): [%] 32.6 in BU2, 44.3 in BU4, 50.6 in ME and 44.3 in BUME; and histologic improvement (difference in histological index between baseline final visit ≥ 1): [%] 44.9 in BU2, 51.9 in BU4, 60.5 in ME and 56.8 in BUME. Patient's assessment of efficacy was very good/good in [%] BU2 71.9, in BU4 84.8, in ME 87.6 and in BUME 89.7. A number of factors influencing adherence to the treatment were analysed. Interference with daily routine was less with application of the suppositories before bedtime compared to morning application. Acceptance of treatment was very good: most patients rated morning (67 to 79% of patients) as well as evening application (69 to 85% of patients) with "easy"; no significant differences between groups were reported for easy retaining and administration of the suppositories. No signals for adverse or unexpected reactions and no serious adverse events were observed. Conclusions: BU4 was superior over BU2 which points to dose dependency. Efficacy and safety of BU4 was not different to ME. Novel budesonide suppositories offer an alternative therapy to mesalamine for topical treatment of proctitis.

1081 MICROBIOTA METABOLITE SHORT CHAIN FATTY ACID ACETATE PROMOTES INTESTINAL IGA RESPONSE TO MICROBIOTA WHICH IS MEDIATED BY GPR43 Wei Wu, Mingming Sun, Feidi Chen, Suxia Yao, Zhanju Liu, Yingzi Cong Background and Aims: Intestinal IgA, which is regulated by gut microbiota, plays a crucial role in maintenance of intestinal homeostasis and in protecting the intestines from inflammation. However, the means by which microbiota promotes intestinal IgA responses remain unclear. Emerging evidence suggests that the host can sense gut bacterial metabolites in addition to pathogen-associated molecular patterns and that recognition of these small molecules influences host immune response in the intestines and beyond. Methods: Fecal IgA in WT and GPR43-/- mice as well as in SCFA-treated mice was measured by ELISA, IgA+ B cells measured by FACS and IgA class switching events measured by qRT-PCR/nested PRC. Results: GPR43-/- mice demonstrated lower levels of intestinal IgA and IgA+ gut bacteria compared to those in WT mice. Feeding WT but not GPR43-/- mice acetate but not butyrate promoted intestinal IgA response. Feeding TCRbxd-/- mice, which lack T cells, acetate also promoted intestinal IgA response, indicating a T cell-independent manner in acetate-induction of intestinal IgA. Acetate did not directly act on B cells to promote IgA production in vitro, but promoted B cell IgA class switching and IgA production in the presence of WT but not GPR43-/- dendritic cells (DC). Mechanistically, acetate did not induce DC TGFb production and expression of BAFF/APRIL. Blockade of BAFF/APRIL pathway did not affect acetate-induction of B cell IgA production. Acetate, however, induced DC expression of Aldh1a2, which converts Vitamin A into its metabolite retinoic acid (RA). Conditioned media of acetate-treated DC promoted B cell IgA production through production of RA. Moreover, blockade of RA signaling inhibited the acetate induction of B cell IgA production. Conclusions: Collectively, our data demonstrated that microbiota metabolite short-chain fatty acid acetate promoted intestinal IgA responses, which was mediated by "metabolite-sensing" GPR43 and DC production of RA. Our studies thus identified a new pathway by which microbiota promotes intestinal IgA response through its metabolites.

1080 EFFICACY AND SAFETY OF ORAL TOFACITINIB AS MAINTENANCE THERAPY IN PATIENTS WITH MODERATE TO SEVERE ULCERATIVE COLITIS: RESULTS FROM A PHASE 3 RANDOMIZED CONTROLLED TRIAL William J. Sandborn, Bruce E. Sands, Silvio Danese, Geert R. D'Haens, Severine Vermeire, Stefan Schreiber, Brian G. Feagan, Walter Reinisch, Gary Friedman, Deborah A Woodworth, Haiying Zhang, Nervin Lawendy, Wojciech Niezychowski, Chinyu Su, Julian Panés Background: Tofacitinib is an oral, small molecule Janus kinase inhibitor that is being investigated for ulcerative colitis (UC). Efficacy and safety of tofacitinib 10 mg twice daily (BID) were reported in OCTAVE Induction 1 & 2. 1 Methods: OCTAVE Sustain (NCT01458574) was a Phase 3, randomized, double-blind, placebo (PBO) -controlled study that enrolled PBO or tofacitinib patients (pts) who completed OCTAVE Induction 1 or 2 with at least clinical response (≥3 points and ≥30% decrease from baseline (BL) Mayo score plus a decrease in rectal bleeding subscore of ≥1 or absolute rectal bleeding subscore ≤1). Pts were re-randomized (1:1:1) to maintenance treatment with PBO (N=198), tofacitinib 5 (N=198) or 10 mg BID (N=197) for 52 weeks (wks). The primary endpoint was remission (total Mayo score ≤2, no subscore >1, rectal bleeding subscore of 0) at Wk 52. Key secondary endpoints were mucosal healing (Mayo endoscopic subscore ≤1) at Wk 52, and sustained

S-199

AGA Abstracts

AGA Abstracts

Mayo sub-score ≥2) was performed. Anaerobically prepared donor stool (pooled from 3-4 donors) or autologous FMT (placebo) were stored frozen at -80°C, thawed and then administered via colonoscopy on day 0 followed by 2 enemas by day 7. The primary outcome was steroid-free remission of UC as defined by a total Mayo score of ≤ 2 with an endoscopic Mayo score of ≤ 1 at week 8. Secondary end points included clinical response (≥3 point reduction in Mayo score), clinical remission (Simple Clinical Colitis Activity Index ≤ 2), endoscopic remission (Mayo ≤ 1) and safety. A mandatory taper of oral corticosteroids was performed; those patients unable to cease oral corticosteroids were considered FMT nonresponders. Results 73 patients with UC were randomised; 38 received donor FMT and 35 received autologous FMT. 3 dropped out from the donor group and 1 from the autologous group during the 8 week observation period. In the intention to treat (ITT) analysis, 12/38 (32%) patients who received pooled donor FMT achieved the primary end point of steroidfree remission, as compared to 3/35 (9%) who received autologous FMT (p=0.02). In the per-protocol analysis, 12/35 (34%) vs 3/34 (9%) achieved the primary end point (P=0.02). In the ITT analysis, clinical response and clinical remission rates were 55% vs 20% (p<0.01) and 50% vs 17% (p<0.01) respectively. Steroid-free endoscopic remission occurred in 55% vs 17% (p<0.01). UC disease extent and disease duration were not significantly associated with achieving the primary endpoint in the donor FMT group. The frequency of serious adverse events (SAE) was not different between the donor and autologous FMT groups; 3 SAE's were recorded in the donor FMT group (1 worsening colitis, 1 Clostridium difficile colitis requiring colectomy, and 1 pneumonia) and 2 SAE's in the autologous FMT group (both worsening colitis). Conclusions In active UC, one week of induction therapy with anaerobically prepared pooled donor FMT is more effective than placebo (autologous FMT) in inducing both clinical and endoscopic remission at 8 weeks.