001 Risk of serious infection associated with biologic therapies in psoriasis: A prospective cohort study from the British Association of Dermatologists Biologic Interventions Register (BADBIR) ZZ Yiu, CH Smith, A Ormerod, DM Ashcroft, M Lunt, S Walton, R Murphy, N Reynolds, CE Griffiths and RB Warren British Association of Dermatologists Biologic Interventions Register (BADBIR), Manchester, United Kingdom In a prospective cohort study from BADBIR (09/2007-10/2016), we compared the risk of serious infections (SI) associated with the use of biologic therapies in patients with psoriasis against a cohort on non-biologic systemic therapies (non-biologics). Etanercept, adalimumab, and ustekinumab were compared to non-biologics, inclusive of any exposure to metho- trexate, ciclosporin, acitretin, fumaric acid esters, psoralen-ultraviolet A, hydroxycarbamide. SI were defined by association with hospitalisation; use of intravenous antimicrobial therapy; and/or led to death. Incidence rates to first SI were measured. Propensity-score weighted Cox proportional hazards models were used to compare the risk of SI and hazard ratios were derived. 1352, 3271 and 994 biologic-naı ¨ve participants were included in the etanercept, adalimumab and ustekinumab cohorts respectively and 3421 biologic-naı ¨ve participants were included in the non-biologic cohort. In total, 283 patients had a SI; the incidence rates were: non-biologic 14.2/1000 person-years (95% confidence interval [CI] 11.5 to 17.4); etanercept 15.3/1000 person-years (11.6 to 20.1); adalimumab 14.8/1000 person-years (11.4 to 16.6); ustekinumab 15.1/1000 person-years (10.8 to 21.1). No statistically significant in- creases in the risk of SI were observed for etanercept (hazard ratio [HR] 1.10; 95% CI 0.75 to 1.60); adalimumab (HR 0.93, 95% CI 0.69 to 1.26) or ustekinumab (HR 0.92, 95% CI 0.60 to 1.41) compared to non-biologics. Our results suggest that the risk of SI should not be a key discriminator when choosing between non-biologics, etanercept, adalimumab and usteki- numab for the treatment of psoriasis. 002 Switching treatments of etanercept biosimilar GP2015 with originator product does not impact efficacy, safety and immunogenicity in patients with chronic plaque-type psoriasis CE Griffiths 1 , K Reich 2 , D Thac ¸i 3 , S Gerdes 3 , P Arenberger 4 , K Kingo 5 , J Weglowska 6 and H Woehling 7 1 University of Manchester, United Kingdom, Manchester, United Kingdom, 2 Dermatologikum Hamburg, SCIderm GmbH, Hamburg, Germany, 3 University Hospital Schleswig-Holstein, Lu ¨beck and Kiel, Germany, 4 Charles University, Prague, Czech Republic, 5 Tartu University Hospital, Tartu, Estonia, 6 Szpital Specjalistyczny we Wroclawiu, Wroclaw, Poland and 7 HEXAL AG/Sandoz, Holzkirchen, Germany The objective of the phase III EGALITY study was to show equivalence in efficacy, safety, and immunogenicity of the etanercept biosimilar GP2015 and the originator product (ETN) in patients with chronic plaque psoriasis. We present the 30-week data comparing repeated switching with continued treatment. Patients (N¼531) with moderate to severe plaque pso- riasis were randomized to receive 50 mg GP2015 or ETN twice weekly s.c. up to week 12. Patients with 50% improvement in psoriasis area and severity index (PASI 50) at week 12 were re-randomized to continue with the same treatment or to undergo 3 consecutive switches until week 30, each at 50-mg once-weekly. The primary endpoint of equivalence in PASI 75 response at week 12 was met. At week 30, the pooled continued and pooled switched treatment groups (n¼446, per-protocol set) showed similar PASI 75 response rates of 86.6% and 85.9%. The proportion of patients with 1 treatment-emergent adverse event (AE) and the types of AEs reported were similar between groups (continued 34.9%; switched 36.7%). Similar proportion of patients experienced injection site reactions with continued (4.3%) or switched treatments (4.6%). The incidence of antidrug antibodies (ADAs) over 30 weeks was low (1.9%, n¼5, all in ETN group). At week 36, 12 weeks after the last of multiple switches, another patient switching from ETN to GP2015 found ADA positive. All ADAs were transient and non-neutralizing. Equivalence in efficacy of GP2015 and ETN was observed in moderate to severe psoriasis patients. Multiple switching between treatments did not impact efficacy, safety, or immunogenicity. AEs were consistent with previous ETN trials in psoriasis. 003 Crisaborole ointment improves global atopic dermatitis severity: Pooled results from two phase 3 clinical trials J Fowler 1 , R Sidbury 2 , AL Zaenglein 3 , DM Pariser 4 and FE Cook-Bolden 5 1 Dermatology Specialists Research, Louisville, KY, 2 Seattle Children’s Hospital, Seattle, WA, 3 Penn State Milton S. Hershey Medical Center, Hershey, PA, 4 Eastern Virginia Medical School & Vir- ginia Clinical Research, Inc., Norfolk, VA and 5 Skin Specialty Dermatology, New York, NY Atopic dermatitis (AD), a chronic inflammatory skin disease, affects children and adults. Pooled efficacy and safety results are presented from 2 identically designed, multicenter, vehicle-controlled Phase 3 studies in children and adults given crisaborole ointment, a nonsteroidal, phosphodiesterase 4 inhibitor. Patients 2 years with mild to moderate AD were randomly assigned 2:1 to receive crisaborole or vehicle twice daily for 28 days. The primary endpoint defined success in Investigator’s Static Global Assessment (ISGA, 5-point scale graded from clear [0] to severe [4]) as clear (0) or almost clear (1), with a 2-grade improvement from baseline (BL) at day 29. Secondary endpoints included proportion of patients who achieved ISGA score of clear (0) or almost clear (1) and success in ISGA at weekly in-clinic evaluations (days 8, 15, and 22). Of 1522 enrolled participants, 1016 were randomly assigned to receive crisaborole and 506 to receive vehicle. Significantly more crisaborole-treated than vehicle-treated patients achieved success in ISGA at day 29 (day 29: 32.1% vs 21.8%; P<0.001). More crisaborole-treated patients than vehicle-treated patients achieved success in ISGA at every weekly in-clinic evaluation (day 8: 14.7% vs 5.4%; day 15: 24.4% vs 11.0%; day 22: 29.8% vs 15.9%). More crisaborole-treated than vehicle-treated patients achieved an ISGA score of clear (0) or almost clear (1) at day 29 (50% vs 35.2%, P<0.001). Mean percentage change from BL in ISGA scores at every in-clinic evaluation was greater in crisaborole-treated patients than in vehicle-treated patients. Treatment-emergent adverse events occurred at a low frequency, and most were mild or moderate. Pooled analysis from 2 large Phase 3 trials showed that crisaborole ointment was well tolerated and improved global disease severity. 004 Patients with hidradenitis suppurativa have a high psychiatric disease burden: A Finnish nationwide registry study L Huilaja 1 , H Tiri 1 , J Jokelainen 2 , M Timonen 2 and K Tasanen 1 1 Department of Dermatology, University of Oulu, Oulu, Finland and 2 Center for Life Course Epidemiology and Systems Medicine, University of Oulu, Oulu, Finland Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease of hair follicles which is associated with various comorbidities. The aim of our study was to clarify the associations between HS and psychiatric disorders. We conducted a nationwide retrospective study that included 4381 patients with HS and 39554 psoriasis and 43248 melanocytic nevi (MN) patients as controls. Patient data were obtained from the statutory Finnish Care Register for Health Care. Statistical analyseswere performed using STATA and the SAS software package. The incidence of HS in Finland was 2.7/100,000 persons/year. At least one psychiatric diagnosis was found in 24.1% of the HS patients compared with 19.1% of those with psoriasis (odds ratio [OR] 1.34; 95% confidence interval [CI] 1.24e1.46) and 13.5% of those with MN (OR 2.04; CI 1.88-2.22). Every mental disorder examined was significantly more frequent in HS than in the two other groups. Major depression was more common in the HS group than in the psoriasis (15.3% vs. 12.1%, OR 1.31, 95% CI 1.19e1.44) and in the MN group (8.3%, OR 2.00, 95% CI 1.81-2.22). Anxiety disorders were diagnosed in 6.9% of HS patients compared with 5.0% of those with psoriasis (OR 1.41, 95% CI 1.23 - 1.62) and 3.8% of those with MN (OR 1.90, 95% CI 1.65-2.19). The total prevalence psychotic disorders was 4.7% in the HS group compared with 3.3% in the psoriasis group (OR 1.46, 95% CI 1.24e1.72) and 1.7% the MN group (2.74, 95% CI 2.29-3.28). Specifically, “schizophrenia or schizotypal disorder” was more frequent in the HS group than in patients with psoriasis (2.4% vs. 1.5%, OR 1.57, 95% CI 1.24e1.98) or in patients with MN (2.4% vs.0.7%, OR 3.38, 95% 2.59- 4.39). The main finding of our study is that HS patients have a higher risk for mental disorders than patients with psoriasis. For dermatologists treating HS patients it is important to take into account the high psychiatric comorbidity burden in HS. 005 WITHDRAWN 006 Abdominal obesity and insulin resistance are strongly associated with liver fibrosis in people with severe psoriasis C Maybury 1 , HF Porter 3 , C Lewis 3 , R Miquel 3 , T Wong 2 , CH Smith 1 and J Barker 1 1 St John’s Institute of dermatology, Kings College London, London, United Kingdom, 2 Guys and St Thomas Hospital, London, United Kingdom and 3 Kings College London, London, United Kingdom People with severe psoriasis are at risk of developing liver fibrosis due to obesity, alcohol and methotrexate. In Europe only people taking methotrexate are screened for liver fibrosis. Aims were:(1) To identify the proportion of patients attending our service with liver fibrosis(2) Identify the most important risk factors associated with fibrosis(3) Evaluate non-invasive tests for fibrosis.400 adults with severe psoriasis were recruited 2012-2015. Data collection to assess risk factors included a questionnaire, PASI, anthropometric indices and fasting blood draw. Abdominal ultrasound measured fatty liver disease. Diagnosis of fibrosis was made by transient elastography (Fibroscan). Diagnostic accuracy of blood tests and clinical measures were compared to Fibroscan results. Characteristics at enrolment were; age (years): 49.5 13, 27% female, body mass index: 29.2 7, waist (cm): 102 16, and PASI: 5 5. 48% had fatty liver disease, 20% liver fibrosis (Fibroscan 7 kPa);14% had advanced fibrosis (Fibroscan 8.7 kPa). Prognostic risk factors significantly associated with fibrosis on univariate analysis included age, BMI, waist, fatty liver disease and HOMA-IR (Homeostatic Model Assessment of Insulin Resistance calculated from fasting glucose and insulin). Sex, methotrexate duration, psoriasis duration and alcohol use were not significantly associated with fibrosis. Multivariate analysis identified the model of waist measurement, HOMA-IR and AST as the best per- forming model of fibrosis (R 2 ¼ 0.38), determined by stepwise variable selection. For both fibrosis and advanced fibrosis, HOMA-IR and waist measurements provided the strongest predictive measures with all AUCs > 0.78. One fifth of our patient population have liver fibrosis. Metabolic parameters were the most important factors associated with fibrosis: We should no longer restrict screening to people taking methotrexate but should screen ac- cording to waist measurement. www.jidonline.org S193 Clinical Outcomes | ABSTRACTS