Rapid Infliximab Infusions are Safe and Well-Tolerated in Patients with Inflammatory Bowel Disease: A Systematic Review and Meta-Analysis

EBV STATUS AND IMMUNOSUPPRESSANT USE IN PATIENTS WITH IBD WHO SUBSEQUENTLY DEVELOP LYMPHOMA: A RETROSPECTIVE AND PROSPECTIVE STUDY Grace Lam, Lindsy Ambrosio, Brendan P. Halloran, Anthea Peters, Richard N. Fedorak, Dan Sadowski, Robert Bailey, Jutta Preiksaitis Background: Immunosuppressive agents, such as thiopurines, methotrexate, and biologics, have revolutionized the treatment of inflammatory bowel disease (IBD). However, a number of studies over the last decade have identified a concerning link between immunosuppression and lymphoproliferative disorders (LPDs), the oncological phenomenon whereby lymphocytes divide uncontrollably. These LPDs have been associated with Epstein-Barr virus (EBV) infection in which the virus provides the impetus for malignant transformation while immunosuppression hampers the immune system's ability to detect and clear these malignant cells. While little is known about the LPD risk in IBD patients, one population study found an increased incidence of LPD in young (18-30 year old) males while on a thiopurine agent. Here, we conduct a retrospective study to determine the LPD incidence in our local IBD population and a prospective study to examine if adoption of an EBV serologic surveillance protocol, similar to the one used in the post-transplantation population, can minimize LPD risk in the IBD population. Method: The retrospective study was conducted by keyword search of the Zeidler Ledcor IBD Center's electronic medical records from 2014-2016 using the qualifiers "IBD" and "lymphoma". A manual chart review of patients in the Alberta Cancer Registry from 2014-2016 to identify those with IBD was also conducted. Identified charts were reviewed for age, gender, IBD status, type of immunosuppressants used, type of lymphoma and year of IBD and lymphoma diagnoses. Concurrently, a prospective study was initiated. IBD patients older than 18 years of age were recruited for an initial test of baseline EBV serology and monitored over time with regards to their EBV status and viral load as they initiate IBD treatment. Results: We identified 12 IBD patients who subsequently developed lymphoma after initiation of treatment out of the 8,607 unique IBD patients in the Edmonton Zone. There were 2 cases of Classical Hodgkin Lymphoma, 5 cases of indolent Non-Hodgkin Lymphoma and 5 cases of aggressive Non-Hodgkin Lymphoma. 3 were EBV associated LPD. Imuran was prescribed in all 3 cases with an average of 5.33 years from initiation of immunosuppressive therapy to LPD development. In the prospective arm, 121 patients have been recruited to date. Of those, 80 have been previously exposed to EBV, 8 have never been exposed to EBV, 4 have active EBV infections and the rest are pending serologies. None have been diagnosed with LPD since the start of the study one year ago. Conclusion: Our data indicates that LPD occurs locally at an incidence rate of 0.14% similar to the rates reported elsewhere in the world. As 21% of LPD were EBV associated, we await the results of the prospective study to determine if EBV monitoring in IBD patients on immunosuppressants can reduce the risk of LPD development.

Table 2. Predictors of LRTI AEs in GEMINI 1 and GEMINI 2

Su1878 RAPID INFLIXIMAB INFUSIONS ARE SAFE AND WELL-TOLERATED IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE: A SYSTEMATIC REVIEW AND META-ANALYSIS Ali Raza, Pramoda Koduru, Vineet Gudsoorkar, Wajeeha Yousaf, Bincy Abraham Background: Infliximab, a monoclonal antibody against tumor necrosis factor-alpha, is used in the treatment of inflammatory bowel disease (IBD). Food and drug administration (FDA) approved duration of infusion is at least two-hours. Rapid infliximab infusions (60-minutes or less) have been implemented recently in a few studies. We aimed to conduct a systematic review & meta-analysis to compare the safety profile of rapid infliximab infusions (60 minutes or less) to the standard two-hour infusions. Additionally, we aimed to assess the rate of infliximab discontinuation due to side effects. Methods: Two independent reviewers performed systematic literature search in PubMed, ISI Web of Science, grey literature and relevant references of included studies from inception of idea to November 5th, 2016 according to institute of medicine guidelines. Human studies in English with documentation of number of rapid and standard infliximab infusions and related adverse reactions were included. In case of overlapping data, most recent publication was included. A random effect model was used with assessment of heterogeneity by the I2 statistic. Results: After review of titles, abstracts, and removal of duplicates, 23 studies were fully reviewed. Seven studies were excluded, as those did not meet inclusion criteria. Sixteen studies (nine retrospective and seven prospective) were included in the final analysis. All studies used historical controls (two-hour standard infusions), and most controls crossed over to the rapid infusion arm (15 studies). A total of 20571 infusions were analyzed (10176, 9657 and 738 infusions in 2-hour, 60-minutes and 30-minutes groups respectively). Infusion reactions occurred in 170 (mild: 162; moderate to severe: 8) of 10176 infusions in 2-hour group (1.76%); 115 (mild: 95; moderate to severe: 20) of 9657 infusions in 60-minutes group (1.19%) and 23 of 738 infusions in 30-minutes group (3.1%). Overall rate of discontinuation was 0.25%, 0.11% and 0.27% in 2-hour, 60-minutes and 30-minutes groups respectively. Meta-analysis of 13 studies comparing 60-minutes to 2-hour group showed a trend towards decreased infusion reactions in 60-minutes group with an OR of 0.68 (95% confidence interval (CI) 0.37-1.24), but it was not significant (p=0.2), with moderate heterogeneity (I2: 66%; Figure 1). Meta-analysis of five studies comparing 30-minutes to 2-hour group showed a trend towards decreased infusion reactions in 30-minutes group with an OR of 0.67 (95% confidence interval (CI) 0.33-1.39), but it was not significant (p=0.28), without any significant heterogeneity (Figure 2). Conclusion: Rapid infliximab infusions (60-minutes or less) are not associated with increased risk of adverse reactions in IBD patients. By saving time, rapid infusions will be associated with conserving health care resources and may increase patient satisfaction.

Su1880 PERIOPERATIVE USE OF VEDOLIZUMAB IS NOT ASSOCIATED WITH SHORT-TERM POSTOPERATIVE INFECTIOUS COMPLICATIONS IN PATIENTS WITH ULCERATIVE COLITIS UNDERGOING (PROCTO)COLECTOMY WITH ILEAL POUCH-ANAL ANASTOMOSIS Marc Ferrante, Nikkie Schils, Anthony de Buck van Overstraeten, Severine Vermeire, Gert A. Van Assche, Albert Wolthuis, Andre D'Hoore Background: Vedolizumab, a bowel focused anti-adhesion molecule, is effectively used in patients with Crohn's disease (CD) and ulcerative colitis (UC). Preoperative use of vedolizumab has recently been associated with increased risk of short-term postoperative infectious complications. We assessed this risk in a single-center cohort of patients with UC undergoing (procto)colectomy with ileal pouch-anal anastomosis (IPAA). Methods: A chart review was performed in all patient undergoing (procto)colectomy with IPAA between September 2006 (initiation of vedolizumab in clinical trials) and September 2016. Patients receiving an investigational medical product besides vedolizumab within 14 weeks of (procto)colectomy or receiving a permanent ileostomy were excluded. Short-term postoperative infectious complications were evaluated within 30 days after (procto)colectomy and included pouch related complications, surgical site infections, and infections outside the surgical site. The comprehensive complication index (CCI) was calculated based on all events reported within 30 days of (procto)colectomy. Results: We identified 170 patients undergoing (procto)colectomy (46% female, median age 38 years, median disease duration 6 years). Thirty-four patients (20%) received vedolizumab within 14 weeks, 60 (35%) received anti-TNF within 8 weeks, 32 (19%) received a moderate-to-high dose (≥20 mg/day) of prednisone, and 71 (42%) received no therapy at time of (procto)colectomy (Figure 1). Surgery was laparoscopyassisted in 131 patients (77%). Pouch construction was performed at first stage in 47 patients (28%), more frequent in patients with dysplasia/cancer (85% vs. 13%, p<0.001), and less frequent in patients under vedolizumab (9% vs. 32%, p=0.005), anti-TNF (15% vs. 35%, p=0.006), or steroids (0% vs. 34%, p<0.001). In multivariate analysis, the only risk factor for short-term postoperative infectious and overall complications was the construction of the pouch at first stage [Odds ratio 2.40 (95%CI 1.18-4.90), p=0.016 and 3.11 (1.52-6.40), p=0.002, respectively]. As shown in Figure 2, no significant difference could be observed between different treatment categories and development of short-term postoperative complications. Also the CCI and postoperative hospitalization stay were comparable between each treatment category, and only elevated in patients undergoing pouch construction at first stage [20.9 (0.0-30.8) vs. 0.0 (0.0-20.9), p=0.001, and 11 (9-17) vs. 7 (5-10) days, p<0.001, respectively]. Conclusion: In this large single-center cohort of patient with UC undergoing

S-581

AGA Abstracts

AGA Abstracts

Su1879