Factors related to blood-based biomarkers for neurodegenerative diseases and their intergenerational associations in the Young Finns Study: a cohort study

Abstract

Background: Blood-based biomarkers (BBM) of neurodegenerative diseases are emerging as cost-effective tools in the differential diagnostics of Alzheimer's disease and other dementias. Scarce data exist about factors explaining BBM variation in population-based cohorts, and their intergenerational associations are unknown. This study aimed to characterise BBM distributions among a population-based cohort, investigate the association of a wide array of factors with BBM both in midlife and old age, and investigate intergenerational associations of BBM.

Methods: We measured BBM detecting amyloid β and tau pathologies, including amyloid β42, amyloid β40, and phosphorylated Tau (pTau)-217, as well as glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL) in the multigenerational Young Finns Study participants (n=1237, age 41-56 years) and their parents (n=814, age 59-90 years) using the Quanterix Simoa HD-X analyser. Standard statistical methods were used to examine the associations between BBM and age, sex, genetic factors, a plethora of cardiometabolic markers, liver and kidney function, and lifestyle factors, as well as their intergenerational associations.

Findings: Increased age was associated with adverse BBM concentrations. Of the various investigated factors, the most robust associations towards adverse BBM concentration were found for APOE ε4 carrier status among parents (amyloid β42:40 ratio, pTau-217, and GFAP) and high serum creatinine concentration in both generations (pTau-217, GFAP, and NfL). Several factors related to glucose metabolism and dyslipidaemia were negatively associated with all BBM, but adjusting for BMI diluted many of these associations. Statistically significant intergenerational correlations ranged from 0·20 to 0·33 and were mostly observed between mothers and offspring in pTau-217, GFAP, and NfL. No intergenerational correlations existed in amyloid β42:40 ratio.

Interpretation: We identified several factors that might influence BBM concentrations, parental transmission being one of them. For reliable use of BBM in clinical practice, it is important to identify which factors directly link to amyloid β and tau pathology and which factors influence BBM concentrations due to other physiological processes.

Funding: Research Council of Finland, Social Insurance Institution of Finland, Competitive State Research Financing of the Expert Responsibility area of the Kuopio, Tampere and Turku University Hospitals, Juho Vainio Foundation, Paavo Nurmi Foundation, Finnish Foundation for Cardiovascular Research, Finnish Cultural Foundation, The Sigrid Juselius Foundation, Tampere Tuberculosis Foundation, Emil Aaltonen Foundation, Yrjö Jahnsson Foundation, Signe and Ane Gyllenberg Foundation, Jenny and Antti Wihuri Foundation, Diabetes Research Foundation of the Finnish Diabetes Association, EU Horizon 2020, European Research Council, Tampere University Hospital Supporting Foundation, Finnish Society of Clinical Chemistry, the Jane and Aatos Erkko Foundation, and the Finnish Brain Foundation.

Figure 1

Distribution of BBM within generations G0 and G1 Boxplots are based on untransformed data, but the difference between generations is tested using logarithmic transformations for skewed BBMs (pTau217, GFAP, and NfL). ∗p<0·001 for the difference between the generations. BBM=blood-based biomarkers. GFAP=glial fibrillary acidic protein. NfL=neurofilament light chain. pTau217=phosphorylated Tau 217.

Figure 2

Associations of different variables with BBM concentrations adjusted for age and sex (A; model 1) and adjusted for age, sex, and BMI (B; model 2) A separate linear model was created for each variable. BBMs are Z score standardised. All variables are dichotomised into value 0 (indicating the reference, no diseases, laboratory values within reference, and a healthy lifestyle) and value 1 (disease, laboratory values outside reference values, and an unhealthy lifestyle). Therefore, the effect sizes express how many SD units a given BBM changes when having an adverse condition compared with a healthier condition. For the amyloid β42:40 ratio, negative effect sizes indicate adverse associations, and for pTau-217, GFAP, and NfL, positive effect sizes indicate adverse associations. Filled circles represent statistically significant effect sizes and open circles non-significant effect sizes. BBM=blood-based biomarkers. GFAP=glial fibrillary acidic protein. HDL=high-density lipoprotein. LDL=low-density lipoprotein. NfL=neurofilament light chain. pTau-217= phosphorylated tau 217. PRS=polygenic risk score.

Figure 3

Distribution of BBM within generations G0 and G1 for different APOEε4 carrier status (A; zero, one, or two alleles) and serum creatinine concentration within or outside reference value (B) Boxplots are based on untransformed data, but the difference between the generations are tested using logarithmic transformations for skewed BBM (pTau217, GFAP, and NfL). Statistically significant differences between the groups within generations are marked with an asterisk. ∗p<0·05 for the difference between the groups. †p<0·01 for the difference between the groups. ‡p<0·001 for the difference between the groups. BBM=blood-based biomarkers. GFAP=glial fibrillary acidic protein. NfL=neurofilament light chain. pTau-217= phosphorylated tau 217.