More than 90% of study participants had detectable IgG against the EBV-encoded nuclear antigen-1 (EBNA-1), an antigen expressed in both latent and lytic modes of infection and that is essential for efficient EBV genome replication, persistence, and transcription in dividing cells. A large number of HLA SNPs were significantly associated with anti-EBNA-1 IgG levels (Figure S4A), consistent with the results of a previous study.25 Among amino acids, positions 11 and 26 of HLA-DRβ1 showed strong, independent associations (Table 2, Figure S4B), which together explained the top SNP association from the genome-wide association studies (GWASs) (conditional p = 0.04, Table S6). Among classical alleles, HLA-DRB1∗07:01 showed the strongest association result (Figure S4C), and there was an independent association of HLA-DRB1∗03:01. However, these two alleles could not completely explain the top GWAS hit (residual p = 2.6 × 10−11, Table S6), pointing to a possible relevance of additional classical HLA alleles that did not reach the significance threshold in our study. EBV is known or suspected to play a role in multiple sclerosis26 and systemic autoimmune diseases,27 and we observed an overlap between the genetic determinants of IgG response to EBV and the HLA class II variants that have been reported to associate with autoimmunity. Carriage of a glycine at HLA-DRβ1 position 11 was associated with decreased antibody levels (Figure S4B) and a lower risk of seropositive28 and seronegative29 rheumatoid arthritis (MIM: 180300). On the contrary, IgG levels were higher in individuals carrying a proline at position 11 (Figure S4B), which is in strong LD with alanine at position 71 (r2 = 0.83, Table S8) and carried by the HLA-DRB1∗15:01 classical allele. HLA-DRβ1 position 71 and HLA-DRB1∗15:01 have been reported as the strongest genetic risk factors for multiple sclerosis at the level of amino acids and classical HLA alleles, respectively.21 However, the directionality of immune response and autoimmune disease risk associations is not always consistent, and larger study cohorts will be required to determine the role and extent of the contribution of EBV immune response to autoimmune disease susceptibility.