Plasma autoantibodies as novel risk markers for stroke: A nested case-control study of the JPHC study

Background and aims: Recent reports have shown the development of autoantibodies in patients with atherosclerosis. The purpose of this study is to examine whether these autoantibodies are associated with the risk of incident stroke in community-based population. Methods: We performed a nested case-control study within the Japan Public Health Center-based Prospective Study involving 32,527 Japanese individuals aged 40-69 years at baseline (1990-1993). Incident stroke was followed up until the end of 2007. We identified 38 novel autoantibody biomarkers for atherosclerosis, and the plasma levels of these autoantibodies were measured by use of amplified luminescent proximity homogeneous assay-linked immunosorbent assay methods in 371 stroke cases (203 ischemic and 168 hemorrhagic) and 371 age-, sex-, and community-matched controls. Conditional odds ratios of total, ischemic, and hemorrhagic strokes according to quartiles of each autoantibody were calculated after adjustment for major cardiovascular risk factors. Results: Of the 38 the novel autoantibodies, the levels of anti-DIDO1 peptide, anti-LIMA1 peptide, anti-SH3BP5 peptide, anti-FOXJ2 peptide, anti-KIF20B protein, anti-DIDO1 protein, anti-TPM3 protein, anti-COPE protein, anti-MMP1 protein and anti-IFI16 peptide autoantibodies were significantly associated with the risk of total stroke (odds ratios of the highest versus lowest quartile >= 1.63) after adjustment for major cardiovascular risk factors. The association was generally similar but weakly observed for ischemic stroke but not for hemorrhagic stroke. Conclusions: These novel autoantibodies have the potential to predict future stroke incidence.