KANG-CHIH FANI-WENG YENCHIA-HUNG LINYang, Chung-YiChung-YiYangKuo, Chun-HengChun-HengKuoChen, Szu-ChiSzu-ChiChenLyu, Ya-PinYa-PinLyuJuan, Hsien-ChiaHsien-ChiaJuanMAO-SHIN LINSHU-HUEI WANGHUNG-YUAN LI2025-09-042025-09-042025-08-26https://scholars.lib.ntu.edu.tw/handle/123456789/731772Objectives: Type 2 diabetes (T2D) is a growing global health challenge, necessitating novel biomarkers for early risk assessment. This study investigates the associations between plasma sphingosine-1-phosphate (S1P), apolipoprotein M (ApoM), obesity measures, and the risk of incident T2D. Methods: We utilized data from the Taiwan Lifestyle Study, a prospective cohort initiated in 2006. A total of 1207 individuals without diabetes at baseline were included. Abdominal fat distribution was assessed using computed tomography, while plasma S1P and ApoM levels were measured using ELISA. Results: During an average follow-up of 5.93 years, 152 participants developed diabetes. Plasma S1P correlated with visceral, peritoneal, and retroperitoneal fat (all p < 0.001), while ApoM correlated with total, subcutaneous, and visceral fat (all p < 0.001). In adjusted models, plasma S1P significantly predicted diabetes (HR 0.52, 95% CI 0.35–0.77, p = 0.001), whereas plasma ApoM did not (HR 1.16, 95% CI 0.94–1.43, p = 0.170). The predictive model incorporating S1P showed an AUC of 0.7698 and a concordance statistic of 0.7637, outperforming traditional risk factors. Conclusions: Plasma S1P and ApoM are both associated with obesity. However, only plasma S1P is a good predictive biomarker for incident T2D, underscoring its potential for early risk assessment and intervention.en[SDGs]SDG3journal article10.1038/s41366-025-01890-740858792