POS0007 HLA-DQ2 IS ASSOCIATED WITH ANTI-DRUG ANTIBODY FORMATION TO INFLIXIMAB ACROSS IMMUNE-MEDIATED INFLAMMATORY DISEASES

Background Immunogenicity is a leading cause of treatment failure to TNF inhibitors, and also affects drug safety. Variations in HLA class II genes have been suggested predispose anti-drug antibody formation (ADA), but characterisation biologically relevant haplotypes, based on high-resolution genotyping, lacking. Objectives To assess associations between loci ADA infliximab across different immune mediated inflammatory diseases. Methods Patients with diseases therapy (N=612; 181 spondyloarthritis, 120 rheumatoid arthritis, 72 psoriatic 114 ulcerative colitis, 80 Crohn’s disease 45 psoriasis) participating the Norwegian Drug Monitoring (NOR-DRUM) trials (1, 2) were included present analyses. Neutralising assessed an automated fluorescence assay at each infusion. Next generation sequencing-based typing was performed. Associations locus, allele, haplotype amino acid level. Peptide binding predictions for Results detected 147 patients (24%). Significant shown several loci, whereas conditional analyses indicated HLA-DQB1 (p=1.4x10-6) as primary risk locus. Highest seen carrying least one HLA-DQ2 haplotypes; DQB1*02:01~DQA1*05:01 DQB1*02:02~DQA1*02:01 (OR 3.18, 95% CI 2.15 4.69, p=5.9x10-9) (Figure 1). These findings consistent diagnoses (Table 1), remained significant when adjusting other possible predictors ADA. Computational that both these haplotypes could strongly bind two peptide motifs (INTVESEDI VYACEVTHQ) heavy light chain. Table 1. carrier frequencies according phenotypes all diagnosis combined Diagnosis carrier-frequency among without P-value RA (N=120) 0.316 0.134 0.02 PsA (N=72) 0.55 0.231 0.01 SpA (N=181) 0.364 0.182 UC (N=114) 0.556 0.264 0.006 CD (N=80) 0.429 0.303 0.33 Ps (N=45) 0.867 0.267 0.0004 All 0.469 0.217 5.9x10 -9 Conclusion The three-fold higher A biological role molecules encoded by further supported predictions. novel provide promise future incorporation testing facilitate personalised decisions. References [1]Syversen SW et al. Jama. 2021;326(23):2375-84. [2]Syversen 2021;325(17):1744-54. Disclosure Interests Marthe Kirkesæther Brun: None declared, Kristin Hammersbøen Bjørlykke: Marte K. Viken: Bitte Stenvik Employee of: former employee UCB Pharma, Rolf A. Klaasen: Johanna Gehin: David J Warren: Joe Sexton: Øystein Sandanger: Cato Mørk Speakers bureau: Novartis Norway, LEO ACO Hud Norge, Cellgene, Abbvie, Galderma Nordic AB., Consultant Tore Kvien Amgen, Celltrion, Evapharma, Gilead, Hikma, Mylan, Oktal, Pfizer, Sandoz, Sanofi, UCB, Grant/research support from: AbbVie, BMS, Novartis, Espen Haavardsholm Celgene, Janssen, Eli-Lilly, Jørgen Jahnsen Boerhinger Ingelheim, Giliad, Janssen Cilag, Orion Roche, Takeda, Guro Løvik Goll Boehringer pharma, Benedicte Lie: Kaasen Jørgensen Norgine., Nils Bolstad Roche Pharmaceuticals Silje Watterdal Syversen: declared