Abstract

Background: Systemic Lupus Erythematosus (SLE) is an autoimmune/inflammatory disease characterized by extremely variable clinical presentations. While across age groups, genetic determinants play a key pathomechanistic role, they are likely to have a more pronounced role in juvenile-onset (jSLE) when compared to adult-onset SLE patients, contributing to earlier disease-onset, more severe phenotypes and less favorable outcomes. Objectives: This study investigated the relationship between alternative allele counts and age at disease-onset, sex, ethnicity, organ involvement and clinical severity in a multiethnic jSLE cohort. It furthermore explored associations between gene level alternative allele scores, organ involvement and disease severity. Methods: A total of 315 jSLE patients enrolled in the UK JSLE Cohort Study were genotyped using a panel sequencing approach covering 61 genes/genomic regions associated with SLE. Demographic features, disease activity (pediatric British Isles Lupus Assessment Grade/pBILAG-2004) and damage (Systemic Lupus International Collaborating Clinics standardized damage index/SLICC-SDI) scores, and clinical features included in the America College of Rheumatology (ACR) criteria (1997) were recorded. DNA sequencing was performed using the Illumina MiSeq platform, sequences were aligned to the human reference genome (HG38) and genetic variants were identified using the Genome Analysis Toolkit. Alternative allele counts were generated by i) counting alternative alleles, and ii) weighing alleles considered the in silico predicted functional impact of each SNP (low or modifier=1, moderate=2, high=3). Alternative allele scores were calculated for N=269/315 (85.4%) jSLE patients with genotype information for all SNPs (N=4100) identified across the cohort. For N=289/315 (91.7%) patients with complete datasets, gene level allele scores were calculated, considering the frequency of alternative alleles, the impact of each SNP within individual genes and an allele score (homozygous reference=0, heterozygous=1, homozygous alternative=2). A generalized linear mixed model, adjusted for ethnicity, sex and family history was used to assess the relationship between gene level allele scores and clinical variables. Results were corrected for false discovery rate. Results: A (weak) inverse correlation between age at diagnosis and alternative allele counts was observed (R=-0.15, p=0.01) (Figure 1A). Notably, the inverse correlation between age at disease-onset and allele counts across ethnicities was primarily caused by Black African/Caribbean and South-Asian patients (Figure 1B). Allele counts were higher in jSLE patients of Black African/Caribbean descent when compared to White Europeans, East and South Asians (p<0.001) (Figure 1C). When considering organ involvement, disease activity in the constitutional (p<0.001), renal (p=0.001), hematological (p=0.001) and neuropsychiatric (p=0.03) pBILAG-2004 domains, and the presence of non-erosive arthritis (p=0.004) and malar rash (p=0.03) were significantly associated with alternative allele counts. No statistically significant correlation between the weighted alternative allele count and SLICC-SDI was observed (R=-0.07, p=0.2). Investigating possible relationships between gene level allele scores and clinical variables, associations were identified for 13 genes, including neuropsychiatric severity and ACP5 (p<0.001), RASGRP3 (p=0.04), RNASEH2A (p<0.001) and TYK2 (p<0.001), and renal severity and ACP5 (p<0.001), ITGAM (p<0.001), LYN (p<0.001), TNFAIP3 (p=0.007) gene score. Conclusion: Genetic variability contributes to early disease expression, especially in Black African/Caribbean and South-Asian jSLE patients. Gene level allele scores associate with organ involvement and severity across ethnicities. Observations from this study promise potential for future genetic risk assessment, to inform patient stratification towards individualized treatment and care. REFERENCES: NIL. Acknowledgements: We acknowledge the UK JSLE Study Group which provided patient samples and information. This work has been supported by LUPUS UK, Versus Arthritis UK for the EATC4Children and the NIHR. Disclosure of Interests: Valentina Natoli: None declared, Amandine Charras: None declared, Megan Hasoon: None declared, Eva Caamano Gutierrez: None declared, Eve MD Smith: None declared, Michael W Beresford: None declared, Christian M Hedrich Unrestricted research funding from Merck to study lupus nephritis.