A Nomogram Based On Plasma FAM19A5 And Radiomics For Prediction Of Parkinson’s Disease And Parkinson’s Disease With Depression

Background and Objective:As the second-largest degenerative disease in the world,the currently known pathogenesis of Parkinson’s disease includes protein overexpression and aggregation,mitochondrial dysfunction,neuroinflammation,oxidative stress,etc.Among them,neuroinflammation mediated by glial cells is one of the pathogenic mechanisms of the disease.Family with sequence similarity 19(Family with sequence similarity 19,member A5,FAM19A5)is mainly secreted by astrocytes and glial cells in the brain,and is involved in the injury response of the central nervous system.This leads to neuronal death,which is structurally manifested by a reduction in gray matter volume,which is also reflected in structural MRI.Therefore,in this paper,by studying the expression differences of plasma FAM19A5 between healthy controls and Parkinson’s disease(PD)patients and different non-motor symptoms of Parkinson’s disease,the imaging group of plasma FAM19A5 levels combined with various nuclei was investigated.Implications of science for the diagnosis of Parkinson’s disease and its nonmotor symptoms.Neuroinflammation mediated by glial cells is one of the pathogenesis of Parkinson’s disease.Family with sequence similarity 19,member A5,FAM19A5,member 5(Family with sequence similarity 19,member A5,FAM19A5)is mainly secreted by astrocytes and glial cells in the brain,and participates in the damage response of the central nervous system(Central Nervous System,CNS)A prolonged inflammatory response can lead to neuronal death,which is structurally manifested as a decrease in the volume of gray matter,which can also be reflected in structural magnetic resonance.Therefore,this paper studies the expression difference of plasma FAM19A5 between Parkinson’s disease(Parkinson’s disease,PD)and the non-motor symptoms of Parkinson’s disease,and explores the level of plasma FAM19A5 combined with radiomics to diagnose Parkinson’s disease.The meaning of disease and its non-motor symptoms.Methods:This experiment screened 176 PD patients from May 2018 to September 2020,and included 181 healthy controls who matched the age and gender of PD patients.The clinical information of PD patients was collected: gender,age.Course of onset,age of onset,modified Hoehn-Yehr(the modified Hoehn-Yehr,H&Y)classification,left-line dopa equivalent dose(levodopa daily equivalent dose,LEDD),and collected their blood samples,using enzyme-linked immunosorbent assay(enzyme-linked immunosorbent assay,ELISA)technology to measure all The participant’s plasma FAM19A5 concentration.According to the standard of UPDRS Part III(motor symptoms),the PD group is further divided into stiffness group(akinetic-rigid,AR),tremor group(tremor-dominant,TD)and mixed subtype(MT);according to statistics,The three non-motor symptoms of the patients are divided into constipation group and non-constipation group,depression group and non-depression group,rapid eye movement(REM)sleep behavior disorder group(Rapid eye movement(REM)sleep behavior disorder,RBD)and non-RBD group.In the PD group and the control group,82 people with perfect magnetic resonance examination were selected,and the bilateral amygdala,the head of the caudate nucleus,putamen,black and red nuclei were marked in the 1.5T or 3.0T magnetic resonance image.Region of interest(ROI),the PD group and the control group are further divided into training group and validation group at random,using Lasso Logistical Regression and the weight method to screen and normalize the imaging omics features,and combine Plasma FAM19A5 concentration and other risk factors established a logistic regression model for Parkinson’s disease and Parkinson’s disease with depression(PDD),and then a five-fold cross-validation regression model was selected.The prediction results are visualized using a nomogram and a receiver operating curve(ROC)curve is drawn.Results:(1)The level of plasma FAM19A5(2.46±0.51)in the PD group was significantly higher than that of the control group(2.23±0.46)(t=4.433,P<0.001).The difference was statistically significant;The Pearson correlation showed that plasma FAM19A5 was correlated with age There is no significant correlation between the disease course and age of onset.Spearman correlation analysis shows that there is no correlation between plasma FAM19A5 and gender,H&Y grade,LEDD;(2)In the PD group,the age of onset was earlier in females(63.03±9.05)than in males(65.03±10.95),and the incidence of constipation was higher than that in males;(3)AR group,TD group and MT group are in gender,age,age,course of disease,There was no statistical difference between age of onset,H&Y,LEDD and plasma FAM19A5 levels;(4)The T-test between the two groups showed that plasma FAM19A5 levels were not significant between constipation and non-constipation groups(P=0.67,t=-0.472),as well as the RBD group and the non-RBD group(P=0.862,t=-0.174);There is a difference between the depression subgroup(2.576±0.408)and the non-depressive subgroup(2.406±0.549)(P=0.045,t=-2.012),and the age of onset of patients with depressive symptoms is lower than that of PD patients without depression;(5)The model of plasma FAM19A5 combined with radiomics can predict Parkinson’s disease with an accuracy rate of 85.6%(AUC)= 0.913,95% confidence interval is 0.861–0.955);(6)The accuracy of logistic regression model in distinguishing PD from PDD is 87.8%(AUC = 0.937,95% confidence interval is 0.845–0.970).Conclusion:Compared with the normal control group,the expression level of plasma FAM19A5 is higher in PD patients,and plasma FAM19A5 is related to PD with depression.This study also developed a nomogram model based on radiomic characteristics,plasma FAM19A5 and clinical risk factors,which may be used as a potential tool for predicting early PD and PDD in a clinical setting.