医学
生物信息学
重症监护医学
再生医学
疾病
计算生物学
糖尿病
精密医学
肠道菌群
盐皮质激素受体
干细胞
转化研究
2型糖尿病
概化理论
二肽基肽酶-4
梅德林
串扰
转化医学
生物标志物发现
个性化医疗
间充质干细胞
2型糖尿病
药物发现
糖尿病管理
作者
Lijun Zhao,Jiamin Yuan,Qiqi Yang,Jing Ma,Fenghao Yang,Yutong Zou,Ke Liu,Fang Liu
标识
DOI:10.1038/s41392-025-02401-w
摘要
Diabetic complications represent a formidable clinical challenge characterized by hyperglycemia-induced multiorgan dysfunction and dysregulated intercellular signaling networks. Advances in spatial multiomics and single-cell transcriptomic techniques, along with insights into aberrant signaling via myokines, cytokines, hormones, the gut microbiota, and exosomes, have revealed the molecular heterogeneity and dynamic inter-organ crosstalk underlying diabetes. Digital diabetes prevention programs have demonstrated effectiveness in high-risk populations through the use of remote tools to support lifestyle changes, reduce hemoglobin A1c, and delay the onset of type 2 diabetes. The therapeutic landscape for diabetic complications has been reshaped by agents with proven cardiorenal benefits, including sodium‒glucose cotransporter 2 inhibitors, glucagon‒like peptide-1 receptor agonists, and nonsteroidal mineralocorticoid receptor antagonists, with combination therapies offering potential additive or synergistic effects. However, their optimal application requires careful benefit-risk assessment across diverse patient populations. Novel therapeutic strategies involving mesenchymal stem cells and their derived exosomes, gut microbiota modulation, bioactive compounds from traditional Chinese medicine, and AI-assisted disease management systems offer promising approaches to correct molecular dysfunctions. This review summarizes recent advances in the mechanisms, prevention, and treatment of diabetic complications, alongside a critical examination of current bottlenecks in translational applications. The remaining challenges include establishing long-term safe regenerative therapies and effectively integrating AI into clinical workflows. Although AI shows promise, issues such as limited data diversity and low model interpretability hinder its generalizability and clinical trust. Addressing these challenges will be essential for transitioning toward a proactive, personalized, and patient-centered model of care.
科研通智能强力驱动
Strongly Powered by AbleSci AI