Context: The relationship between dehydroepiandrosterone sulfate (DHEAS) and mortality is of scientific and public health interest, yet it remains poorly understood.
Objective: We examined the association between DHEAS and mortality from cancer, cardiovascular disease, and all causes in middle-aged and older men and women.
Methods: DHEAS was measured in stored serum samples collected from 1994 to 1998 from a case-cohort nested within EPIC-Heidelberg, that included 7370 men (mean age = 55.0) and women (mean age = 52.4 years). Median follow-up for incident mortality events was 17.7 years. All deaths due to cancer (n = 1040), cardiovascular diseases (n = 598), and all causes (n = 2407) that occurred in EPIC-Heidelberg until end of 2014 were included.
Results: The association between DHEAS and mortality was nonlinear such that both participants in the lowest (Q1) and highest (Q5) sex- and 5-year age-group specific quintiles of DHEAS were at increased hazard ratios (HR) of mortality from cardiovascular [Q1: HR = 1.83 (95% CI: 1.33-2.51), Q5: 1.39 (1.00-1.94)], cancer [Q1: 1.27 (1.01-1.60), Q5: 1.27 (1.02-1.60)] and all causes [Q1: 1.51 (1.25-1.82), Q5: 1.31 (1.08-1.58)], compared with participants in Q3. In men and women with below-median DHEAS levels, doubling of DHEAS was associated with lower hazards of cardiovascular [0.87, (0.78-0.96)], cancer [0.90, (0.83-0.97)], and total mortality [0.89, (0.83-0.95)]. In contrast, a doubling in DHEAS among participants with above-median levels was associated with 1.20, (1.01-1.42), 1.28, (1.01-1.62), and 1.19 (1.03-1.37) higher hazards of mortality from cancer, cardiovascular, and all causes, respectively.
Conclusion: In this large population-based study, DHEAS showed a J-shaped association with mortality. Both participants with lowest and highest levels experienced higher hazards of mortality from cancer, cardiovascular disease, and all causes.
Keywords: DHEAS; all-cause; cancer; cardiovascular disease; mortality.
© The Author(s) 2022. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.