z-logo
Premium
Comparison of two analytical platforms for blood‐based surrogate biomarkers of amyloid pathology
Author(s) -
De Meyer Steffi,
Schaeverbeke Jolien,
Gille Benjamin,
Verberk Inge M.W.,
Luckett Emma Susanne,
Thijssen Elisabeth H.,
Gabel Silvy,
Mauroo Kimberley,
Bruffaerts Rose,
Stoops Erik,
Vanderstichele Hugo Marcel,
Teunissen Charlotte E.,
Vandenberghe Rik,
Poesen Koen
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.045110
Subject(s) - amyloid (mycology) , pathological , biomarker , surrogate endpoint , medicine , pathology , amyloid beta , amyloid β , oncology , disease , biology , biochemistry
Abstract Background Formation of amyloid plaques and fibrillary tau tangles in brain are key pathological hallmarks in Alzheimer’s disease (AD). Surrogate biomarkers of these hallmarks have been established in CSF. However, a less invasive sample source would simplify the recruitment and the follow‐up of response in clinical trials targeting the pathological hallmarks in AD. We assessed the performances of two analytical platforms for plasma beta‐amyloid (Aβ) species and total tau (t‐tau) as surrogates of amyloid‐PET status. Methods Plasma Aβ1‐42 and Aβ1‐40 were quantified in cognitively healthy controls (n=160, age[mean]=69, 46% female) and amnestic MCI (aMCI) patients (n=41, age[mean]=71, 43% female) by i) the Simoa Amyblood platform (Amsterdam UMC and ADx NeuroSciences) and ii) an ELISA platform (Euroimmun, Lubeck, Germany) employing identical antibody pairs. Plasma t‐tau was quantified with a prototype ELISA (ADx NeuroSciences). The same biomarkers were also measured in CSF (ELISA [Euroimmun]). Controls underwent [ 18 F]‐flutemetamol PET and aMCI patients received [ 18 F]‐florbetaben PET (SUVR comp cut‐off for amyloid positivity=1.38 and 1.29, respectively). Spearman correlations between CSF and plasma biomarkers were calculated and the performance of plasma biomarkers to identify amyloid PET positivity was determined through ROC analysis. Bonferroni correction adjusted for multiple comparisons. Results For ELISA, the strongest correlations were found among amyloid biomarker ratios between plasma and CSF within the aMCI cohort (r=0.6561‐0.8123, p<0.0023); most of these correlations were also present in controls, albeit weaker (r=0.4308‐0.5498, p<0.0078). Using Amyblood, correlations were weaker than for ELISA, most not surviving Bonferroni correction (Table 1). All plasma biomarker ratios could differentiate between amyloid‐PET positive and amyloid‐PET negative subjects in both cohorts (AUC[range]=0.719‐0.876, p<0.0001). Plasma t‐tau identified amyloid positivity only in the aMCI cohort (AUC[range]=0.818, p<0.0001), while plasma Aβ1‐42 could do so only in controls (AUC[range]=0.676‐0.679, p<0.0027). Within the aMCI cohort, Aβ1‐42/Aβ1‐40 performed better than Aβ1‐42 when measured with Amyblood (p=0.0064). No differences in performances were observed between platforms for any biomarker (p>0.543) (Table2). Logistic regression analysis including age, gender and APOEε4 as covariates did not significantly improve performances. Conclusion Measuring amyloid species with blood‐based immunoassays renders good performance to assess amyloid PET status, especially in aMCI. To this end, Amyblood did not outperform ELISA.

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here