Preclinical biomarker associations with both incident rheumatoid arthritis and its subsequent mortality: sex effects in a 41-year, community-based, case-control cohort study.

ABSTRACT

OBJECTIVES: To identify sex effects and preclinical serum biomarker associations with both incident rheumatoid arthritis (RA) and its subsequent mortality, using a 41-year, community-based, case-control cohort. METHODS: After cohort entry in 1974, incident RA cases (n=54) had clinical onsets between 1977 and 1994. Cohort control (CN) subjects were individually matched on entry to cases (4 CN1 RA, n=216). All subjects were followed for survival from 1995 through 2015. Ranks (1-5) of preclinical z-scores within each set of 1 RA and 4 matched CN were analysed for associations with incident RA and mortality. Survival was evaluated using Cox proportional hazards models. RESULTS: Preclinical serum IgG RF z-score ranks associated with incident RA in 90 males (18 RA, 72 CN). Cigarette smoking, androstenedione, pregnenolone, and sIL-2Rα ranks associated with incident RA in 180 females (36 RA, 144 CN). Total percentile mortality was greater (p=0.003) in RA (70.4) vs. CN (49.9) and equivalently increased in female RA (69.4) vs. CN (49.3) and in male RA (72.2) vs. CN (43.1) subjects. Percentile respiratory-related CODs were greater (p=0.009) only in the female RA cases (16.7) vs. CN (3.5). Ranks of preclinical hsCRP (p=0.028) and sIL-2Rα (p=0.030) independently associated with 140 total deaths, as did sTNF-R1 (p=0.003) and hsCRP (p=0.005) with 50 CVD deaths. Latter biomarker association were significant in females. Therapy responses in 1995 significantly associated with subsequent mortality. CONCLUSIONS: Sex effects were important in preclinical biomarker associations with incident RA, total and CVD mortality as well as occurrence of respiratory deaths.