Background: Antinuclear antibodies (ANA) to intranuclear particles are found in the blood of people with and without autoimmune diseases. To our knowledge, only 1 past genome-wide association study (GWAS) has sought to identify genetic factors related to ANA positivity, reporting association with HLA-DR in a Japanese population (Terao C, Arthritis Rheum , 2014). Dissecting the genetic basis for ANA positivity would allow new insights into autoimmunity susceptibility.

Objectives: We conducted a genome-wide association study for ANA positivity.

Methods: We used data from the Mass General Brigham Biobank, containing linked electronic health records and genotyping (Illumina’s Multi-Ethnic Genotyping Array or Global Screening Array chips) for >65,000 consented patients. We identified subjects with genotyping and results for either hep2 or ML immunofluorescence ANA assays (1989-2022). ANA+, those with ≥1 result of ≥ 1:40 titer, were compared to ANA-, those with no positive ANA results. After quality control, we imputed using the Haplotype Reference Consortium, excluding variants with imputation score <0.5. Classical HLA alleles and amino acids for HLA genes were imputed with SNP2HLA using TopMed HLA data as reference. We restricted analyses to those with predicted European ancestry (1000 Genomes EUR group; 82% of biobank subjects). Diagnostic codes identified 8 autoimmune diseases: systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), scleroderma (SSc), Sjögren’s syndrome, mixed connective tissue disease (MCTD), multiple sclerosis (MS), primary biliary cirrhosis (PBC), and autoimmune thyroid disease (ATD). GWAS for ANA + vs – was done in PLINK2, with logistic regression adjusted for age, sex and 4 genetic principal components. Linkage disequilibrium (LD) block was defined as r2>0.5 using EUR information. We also tested for associations with ANA ≥ 1:80, excluding anti-nucleolar or anti-centromere patterns, excluding those with autoimmune diseases vs. ANA-, and HLA alleles and amino acids.

Results: We studied 12,875 subjects with ANA and genotyping results: 7,035 ANA + (≥ 1:40) vs. 5,840 ANA -. The female/male ratio was similar in both groups: 62.9% female in ANA+ vs. 62.1% in ANA- (p 0.39). Patients who were ANA+ were older than the ANA- (56.1 ± 15.8 vs. 52.1 ± 15.2 years, p<0.0001). Of ANA+ patients, 33% had ≥1 of the 8 autoimmune diseases, as did 19% of ANA- patients. Overall, SLE, RA, SSc, Sjogren’s syndrome, MCTD, MS, PBC and ATD were present in 538, 1425, 159, 370, 65, 354, 63 and 408 subjects (with ANA+ proportions 71%, 55%, 78%, 74%, 82%, 55%, 63%, and 62% respectively). 8,073,314 SNPs with MAF>=0.01 were included. The strongest GWAS association with ANA+ ≥1:40 (vs. ANA -) was for rs3134951 on chromosome 6 in the HLA region (OR 1.2, 95% CI 1.1-1.3, p 1.6×10 ^-9 ) (Figure 1). Within the HLA region, HLA-DQB1:0201 allele had the strongest association (OR 1.3, 95% CI 1.2-1.3, p 9.2×10 ^-11 ). We identified two suggestive novel loci for ANA +, one in LOC105373419 (rs60343674) on chromosome 2 and the other in ABHD5 (rs12489159) on chromosome 3. (Figure 1) HLA-DRB1:03 and HLA-B:08 alleles, not in strong LD, each had significant associations (Figure 2). Repeated GWAS for ≥1:80 titer and ≥ 1:80 excluding anti-centromere and anti-nucleolar patterns vs. ANA - showed similar, but stronger signals for rs3134951 on chromosome 6 (p=5.6×10 ^-13 and p=2.3×10 ^-14 , respectively). GWAS excluding those with known autoimmune diseases had weaker signal for rs3134951 (p=2.6×10 ^-7 ). This SNP, rs3134951 at 6p.21.32, is intergenic and associated with 2 genes, FKBPL and PRRT1 , previously associated with SLE (Yin X, Annals Rheum Dis , 2020).

Conclusion: HLA-DRB, HLA-DQB1 and HLA-B, and newly identified rs3134951 at 6p.21.32 are genetic factors predisposing to ANA positivity in this European ancestry population. We are pursuing stratified analyses by ANA pattern and titer, sex, and autoimmune disease.

REFERENCES: NIL.

Acknowledgements: NIL.

Disclosure of Interests: Jing Cui: None declared, Jeong Yee: None declared, Emily G. Oakes: None declared, Hongshu Guan: None declared, Liming Liang: None declared, Karen Costenbader Bristol Myers Squibb, Glaxo Smith Kline, Cabaletta Bio, Merck, Gilead,