Gut microbiota and SCFA biomarkers for early diagnosis of PD patients and differentiation of its motor subtypes

Early diagnosis of Parkinson's disease (PD) and distinguishing it from essential tremor (ET) remains a significant challenge. We analyzed fecal samples (gut microbiota via 16S rRNA gene sequencing with DADA2-denoising/OTU-clustering) and short-chain fatty acids (SCFAs) in 104 drug-naïve early PD patients (73 test/31 validation), 69 ET patients (48/21), and 61 healthy controls (HC; 43/18). Differential taxa were identified using ANCOM, ANCOM-BC, ALDEx2, MaAsLin2, and LEfSe; top diagnostic potential was selected by random forest and assessed by ROC curve analysis. Compared to HC, PD showed reduced Citrobacter/Haemophilus, increased Eggerthella, and elevated isovaleric/isobutyric acids (validation AUC = 0.864). PD vs. ET exhibited decreased Bilophila/Bacteroides/Haemophilus with elevated isovaleric/isobutyric/valeric acids (validation AUC = 0.825). Tremor-dominant PD (TD-PD) was distinguished from ET by lower Lachnoclostridium/Haemophilus/Bilophila and higher isovaleric/isobutyric acids (validation AUC = 0.780), while non-TD-PD differed from TD-PD only in decreased Dialister (validation AUC = 0.802). Gut microbiota and SCFAs might serve as specific, non-invasive candidate biomarkers for early PD diagnosis.