Identification of host-pathogen-disease relationships using a scalable multiplex serology platform in UK Biobank

Abstract

Certain infectious agents are recognised causes of cancer and other chronic diseases. To understand the pathological mechanisms underlying such relationships, here we design a Multiplex Serology platform to measure quantitative antibody responses against 45 antigens from 20 infectious agents including human herpes, hepatitis, polyoma, papilloma, and retroviruses, as well as Chlamydia trachomatis, Helicobacter pylori and Toxoplasma gondii, then assayed a random subset of 9695 UK Biobank participants. We find seroprevalence estimates consistent with those expected from prior literature and confirm multiple associations of antibody responses with sociodemographic characteristics (e.g., lifetime sexual partners with C. trachomatis), HLA genetic variants (rs6927022 with Epstein-Barr virus (EBV) EBNA1 antibodies) and disease outcomes (human papillomavirus-16 seropositivity with cervical intraepithelial neoplasia, and EBV responses with multiple sclerosis). Our accessible dataset is one of the largest incorporating diverse infectious agents in a prospective UK cohort offering opportunities to improve our understanding of host-pathogen-disease relationships with significant clinical and public health implications.

Fig. 1. Crude seroprevalence estimates and 95% confidence intervals for multiple infectious agents in 9695 biologically independent UK Biobank participants stratified by sex and age (A for males and B for females), self-reported ethnicity (C for males and D for females) and LSP (E for males and F for females).

Precise numbers of individuals in each category are provided in Supplementary Tables 9–11. The infectious agents represented are those with significant (P < 0.01) evidence of differences based on crude estimates when testing for association with either males or females alone, or combined with each of age, self-reported ethnicity and LSP where sufficient numbers of individuals were available in each group for models to converge. Evidence of interaction with sex was only observed for VZV (P = 0.01) and HPV-16 (P = 0.04) with age, and HSV-2 (P = 7.5 × 10⁻⁵) and HPV-18 (P = 0.01) with LSP. All P-values reported were from the likelihood ratio test with no adjustment for multiple testing.

Fig. 2. Association between HPV-16 L1 antigen seropositivity and risk of cervical cancer and cervical intraepithelial neoplasia (CIN) in 9695 UK Biobank participants.

A The diamonds represent unadjusted odds ratios of cervical cancer and cervical intraepithelial neoplasia by HPV-16 L1 seropositivity, shown with 95% confidence interval error bars (P-values calculated with two-sided Fisher’s exact test). B Odds ratio and 95% confidence interval of CIN risk after adjustment for age, sex, ethnicity, LSP and TDI (P-value calculated using multivariable regression analysis).

Fig. 3. EBV EBNA1 and VCAp18 antibody responses and associations with multiple sclerosis.

Log₁₀ transformed antibody levels against EBNA1 (A) and VCAp18 (B) antigens were compared between groups in the UKB Multiplex Serology subset with or without self-reported diagnoses of multiple sclerosis (MS; 34 cases). Box plot centre line, median; box limits, upper and lower quartiles; whiskers, 1.5x interquartile range. **P = 6.1×10⁻³ using linear regression. C The beta coefficients and 95% confidence intervals from the GWAS analyses of quantitative antibody responses against EBNA1 (blue) and VCAp18 (red) in the 9611 biologically independent individuals from the UKB subset (y-axis) were compared to the coefficients and 95% confidence intervals from the largest available case-control GWAS of MS (17,610 cases and 30,129 controls; x-axis) for HLA alleles recognised to be associated with MS risk. The points representing beta estimates are shaped by imputed HLA allele.