肠道微生物组的改变与阿尔茨海默病(AD)的发病机制有关,可用作诊断措施。然而,肠道微生物组的纵向数据及其对AD发生和进展的预后意义的认识有限.本研究的目的是开发基于肠道微生物组数据的AD发展的可靠预测模型。在这项纵向研究中,我们调查了49名轻度认知障碍(MCI)患者的肠道微生物组,平均(SD)随访3.7(0.6)年,使用猎枪宏基因组学。在4年随访(4yFU)结束时,27名MCI患者转化为AD痴呆,22名MCI患者保持稳定。从稳定的MCI患者中区分AD痴呆转化者的最佳分类模型包括24属,在BL处产生0.87的接受者工作特征曲线下面积(AUROC),1yFU为0.92,4yFU为0.95。通过分析25个GO(基因本体论)特征获得了具有功能数据的最佳模型,在BL时AUROC为0.87,1yFU时为0.85,4yFU时为0.81,33KO[京都基因和基因组百科全书(KEGG)直系同源]特征,BL时AUROC为0.79,1yFU为0.88,4yFU为0.82。对这三个模型使用集成学习,包括具有四个年龄参数的临床模型,性别,体重指数(BMI)和载脂蛋白E(ApoE)基因型,在BL时产生0.96的AUROC,1yFU为0.96,4yFU为0.97。总之,我们确定了新颖且及时稳定的肠道微生物组算法,该算法可准确预测MCI患者在4yFU期间进展为AD痴呆.
Alterations in the gut microbiome are associated with the pathogenesis of Alzheimer\'s disease (AD) and can be used as a diagnostic measure. However, longitudinal data of the gut microbiome and knowledge about its prognostic significance for the development and progression of AD are limited. The aim of the present study was to develop a reliable predictive model based on gut microbiome data for AD development. In this longitudinal study, we investigated the intestinal microbiome in 49 mild cognitive impairment (MCI) patients over a mean (SD) follow-up of 3.7 (0.6) years, using shotgun metagenomics. At the end of the 4-year follow-up (4yFU), 27 MCI patients converted to AD dementia and 22 MCI patients remained stable. The best taxonomic model for the discrimination of AD dementia converters from stable MCI patients included 24 genera, yielding an area under the receiver operating characteristic curve (AUROC) of 0.87 at BL, 0.92 at 1yFU and 0.95 at 4yFU. The best models with functional data were obtained via analyzing 25 GO (Gene Ontology) features with an AUROC of 0.87 at BL, 0.85 at 1yFU and 0.81 at 4yFU and 33 KO [Kyoto Encyclopedia of Genes and Genomes (KEGG) ortholog] features with an AUROC of 0.79 at BL, 0.88 at 1yFU and 0.82 at 4yFU. Using ensemble learning for these three models, including a clinical model with the four parameters of age, gender, body mass index (BMI) and Apolipoprotein E (ApoE) genotype, yielded an AUROC of 0.96 at BL, 0.96 at 1yFU and 0.97 at 4yFU. In conclusion, we identified novel and timely stable gut microbiome algorithms that accurately predict progression to AD dementia in individuals with MCI over a 4yFU period.