Su1159 Risk of Infections With Natalizumab Therapy Among Patients With Crohn's Disease: An Analysis of the Food and Drug Administration Adverse Event Reporting System Parakkal Deepak, Derrick J. Stobaugh, Eli D. Ehrenpreis Background: Natalizumab is a humanized monoclonal antibody to alpha-4 integrin that is approved by the Food and Drug Administration (FDA) for the induction of remission and maintenance of moderate to severe Crohn's disease (CD). Progressive Multifocal Leu- koencephalopathy (PML) secondary to an opportunistic infection with John Cunningham virus has been reported with natalizumab usage in CD. We sought to clarify the risk of other opportunistic infections in CD patients treated with natalizumab by analyzing reports to Adverse Event Reporting System (AERS) since its approval. Methods: The AERS is a publicly available database of voluntary reports for post marketing surveillance of all FDA approved drugs. 2,053,106 files between January 2008 and June 2011 were downloaded and analyzed using SPSS 20 (IBM Co. Armonk, NY, USA). They were queried for Primary Suspect (PS) reports of adverse events where natalizumab (or its trade names) was used for the treatment of Crohn's disease. Control drugs (5-aminosalicylic acid and sulphasalazine, including trade names) served as a basis of comparison. Further queries for outcomes of viral, bacterial, fungal and parasitic infection as well as control reactions (syncope, hernia, deafness and vertigo- predefined to have no association with natalizumab or control drugs) were performed using reactions terms from the Medical Dictionary for Regulatory Activities. Each report also had concomitant medications analyzed. Any patients reported who used a concomitant biologic medication (including a tumor necrosis factor - α inhibitors) were excluded from analysis. Use of immunomodulators (methotrexate, azathioprine, and mercap- topurine) or systemic corticosteroids was noted. Odds ratios for the risk of opportunistic infections with natalizumab were calculated using the Fischer's exact test. Results: 141 PS reports of infections were identified with natalizumab among CD patients, with a majority having a bacterial origin (115 reports, 82%). The reports demonstrated a female predomi- nance (64.54%) with a mean age of 41 ±14 years. Similar usage of corticosteroids was seen among natalizumab cases and controls (14.18% versus 25.93%, p = 0.27) while a higher usage of immunomodulators was seen among control cases (0.71% versus 14.82%, p = 0.004). Significant odds of developing infections with natalizumab were seen only in the subcategory of bacterial infections (table 1). The majority of bacterial infections (figure 1) occurred in the respiratory system (33 reports, 29%) followed by 22 reports (19%) of disseminated infections. Conclusion: Risk of infections in patients receiving natalizumab for CD appears to be restricted to bacterial infections, with a female predominance and mainly affecting the respiratory system. Using this method, significant risk of other opportunistic infections was not found. Table 1: Reported risk of various infections in Crohn's disease patients treated with Natali- zumab Figure 1: Distribution of bacterial infections reported with natalizumab in Crohn's disease patients (legend - arranged in descending frequency of distribution) Su1160 Risk of Melanoma and Non-Melanoma Skin Cancers With Pharmacotherapy Among Inflammatory Bowel Disease Patients: An Analysis of the Food and Drug Administration Adverse Event Reporting System Derrick J. Stobaugh, Parakkal Deepak, Eli D. Ehrenpreis Background: An increased risk of melanoma skin cancers (MSC) with any exposure to tumor necrosis factor alpha (TNF- α) inhibitors and non-melanoma skin cancers (NMSC) with thiopurine exposure has been reported. We sought to examine the reported risk of skin cancers with drug therapy in IBD and further differentiate the risk with monotherapy with TNF-α or immunomodulators compared to combination therapy. Methods: The Food and S-415 AGA Abstracts Drug Administration Adverse Event Reporting System is a publicly available database based on voluntary reports for post marketing surveillance. 3,171,655 files between January 2003 and June 2012 were downloaded and analyzed using SPSS 20 (IBM Co. Armonk, NY, USA). They were queried for Primary Suspect (PS) reports of adverse events with TNF- α inhibitors (Adalimumab, Certolizumab, Infliximab), corticosteroids, and immunomodulators (thiopur- ines, methotrexate and calcineurin inhibitors), using both trade and generic names, for treatment of IBD along with control drugs (5-aminosalicylic acid and sulphasalazine, includ- ing trade names) as a basis of comparison. These were further queried for outcomes of skin cancer as well as control reactions (syncope, hernia, deafness and vertigo) predefined to have no association with study drugs or control drugs, utilizing reactions terms from the Medical Dictionary for Regulatory Activities. Any reports with comorbidities of human immunodeficiency virus infection, organ transplantation or rheumatologic diseases associated with a risk of skin cancer as well as concomitant exposure to a non- TNF- α inhibitor biologic medication were excluded from analysis. Odds ratios for the risk of skin cancers with TNF- α inhibitor and immunomodulator monotherapy as well as combination therapy were calculated with the Fischer's exact test. Reports with more than one type of skin cancer within the same report were counted as a single entry for calculating overall odds ratios for individual drug categories. Further analysis to determine the reported risk of the MSC and NMSC with the various drug combinations excluded reports of skin cancer not specifying MSC or NMSC. Results: 385 PS reports of skin cancers were identified, with a majority being NMSC (181, 47%). A female predominance (64.08%) was seen with a mean age of 49 ±16 years, reported most commonly in Crohn's disease (317, 82.34%). Significantly elevated odds of reports of skin cancers were seen with TNF- α inhibitors and immunomodu- lators both in monotherapy and in combination with immunomodulators and corticosteroids (table 1). Subgroup analysis (table 2) revealed significant odds of MSC and NMSC with both TNF-α or immunomodulators monotherapy and in combination with each other. Conclusion: Elevated odds of MSC as well as NMSC are reported in patients receiving TNF- α inhibitors in IBD, both as monotherapy and in combination with immunomodulators. Table 1: Reported odds of developing skin cancer with drug therapy in IBD patients 1Fischer's exact test (using approximation of Woolf) Table 2: Reported odds of melanoma and non-melanoma skin cancers with drug therapy in IBD patients 1Counts for control reactions similar to the counts for each drug category in table 1 regardless of type of skin cancer 2Fischer's exact test (using approximation of Woolf) AGA Abstracts