血浆代谢组、长期体重变化与美国人群2型糖尿病
研究在7,499名美国健康专业人员中检测260种血浆代谢物,用弹性网回归建立预测BMI变化的代谢组评分,并分析其与2型糖尿病发病的关联。预测BMI变化的代谢物因基线体重状态而异,瘦人组有46种代谢物显著预测BMI变化斜率,超重组20种、肥胖组13种,三组间无重叠代谢物。瘦人组代谢组评分与BMI斜率相关(Spearman r=0.31),且与更高2型糖尿病风险相关(HR 1.42;95%CI 1.26-1.60),超重组亦相关(HR 1.29;95%CI 1.17-1.42),肥胖组无显著关联。
要点
- 预测体重变化的血浆代谢物谱因基线体重状态不同而异,三组间无重叠。
- 瘦人组代谢组评分与BMI变化斜率相关(r=0.31),超重组r=0.16,肥胖组r=0.10。
- 瘦人组评分每增1SD,2型糖尿病风险HR 1.42(95%CI 1.26-1.60)。
核心结论
- 预测未来体重变化的血浆代谢组谱因体重状态而异,瘦人组与超重组评分与2型糖尿病风险相关,肥胖组无显著关联。
- 特定代谢特征可能有助于识别未来体重增加和2型糖尿病风险升高的瘦人个体,为早期针对性预防提供机会。
研究设计与样本
- 在7,499名美国健康专业人员中检测260种血浆代谢物。
- 采用弹性网回归与训练/测试方法建立预测BMI变化的代谢组评分。
- 采用多变量Cox模型分析代谢组评分与2型糖尿病发病的关联。
主要发现
- 瘦人组有46种代谢物显著预测BMI变化斜率,超重组20种,肥胖组13种,三组间无重叠代谢物。
- 瘦人组代谢组评分与BMI斜率相关(Spearman r=0.31),超重组r=0.16,肥胖组r=0.10。
- 瘦人组评分每增1SD,2型糖尿病风险HR 1.42(95%CI 1.26-1.60)。
- 超重组评分每增1SD,2型糖尿病风险HR 1.29(95%CI 1.17-1.42),肥胖组无显著关联。
临床怎么用
- 对瘦人个体可考虑利用代谢组评分识别未来体重增加和2型糖尿病风险升高者,以实施早期针对性预防。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVE: Body weight influences the blood metabolome, but whether specific plasma metabolomic profiles predict long-term weight change remains unclear. We aimed to investigate associations among plasma metabolites, current weight, weight change trajectories, and incident type 2 diabetes (T2D) in adults. RESEARCH DESIGN AND METHODS: We profiled 260 plasma metabolites in 7,499 U.S. health professionals. A metabolomic score predicting BMI change over time was derived using elastic net regression with a training/testing approach. Associations with incident T2D were examined using multivariable Cox models. RESULTS: Metabolomic predictors of BMI change varied by baseline weight status. In lean individuals, 46 metabolites significantly predicted BMI change slope compared with 20 individuals in the group with overweight and 13 in the group with obesity, with no overlapping metabolites across groups. A metabolomic score of BMI slope derived in lean participants correlated with BMI slope in the lean group (Spearman r = 0.31), but the correlation was weaker in the groups with overweight (r = 0.16) and obesity (r = 0.10). The score was associated with higher T2D risk in the lean group (hazard ratio [HR] per 1 SD 1.42; 95% CI 1.26-1.60) and group with overweight (HR 1.29; 95% CI 1.17-1.42) but not in the group with obesity. CONCLUSIONS: Plasma metabolomic profiles predictive of future weight change differ by body weight status. Distinct metabolic signatures may help identify certain lean individuals at an elevated risk of future weight gain and T2D, offering opportunities for earlier, more targeted prevention.
引用
- Wu Z, Liu B, Li J, et al. Plasma Metabolome, Long-term Body Weight Change, and Type 2 Diabetes in U.S. Individuals. Diabetes Care. 2025. PMID: 40802109. doi:10.2337/dc25-0902
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。