Assay type | Study | Clinical study | Cohort size (average age) | Distribution of CN/MCI/AD | Amyloid positive rate | AUC (95% CI) of plasma Aβ42/40 in classifying amyloid PET status | PET tracer (amyloid status cut-off) | AUC (95% CI) of plasma Aβ42/40 in classifying amyloid CSF status | CSF amyloid status cut-off | Correlation | Sensitivity | Specificity | Plasma status cut-off as plasma Aβ42/40 ratio |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
IP-MS | Nakamura et al. 2018 [17] (Discovery cohort) | National Center for Geriatrics and Gerontology (NCGG) | n=121 (74.0 years) | 62/30/29 | 41.3% by PET | 0.967 (0.942–0.992) a | PiB (1.4) | - | - | For PET, Pearson’s r a= 0.767 | 0.960 | 0.873 | 3.93% a |
(Validation cohort) | Australian Imaging, Biomarkers and Lifestyle Flagship Study of Aging (AIBL) | n=252 (74.2 years) | 156/67/29 | 54.3% by PET | PiB: 0.889 (0.825–0.952) a Overall: 0.837 (0.787–0.887) a | PiB (1.40), FMM (0.55), Florbetapir (1.05) | - | - | For PET, Pearson’s r a= PiB: 0.601 FMM: 0.540 Florbetapir: 0.466 | PiB: 0.733 Overall: 0.657 | PiB: 0.922 Overall: 0.896 | 3.62% a | |
ECL | Yamashita et al. 2022 [38] (Discovery cohort) | MissionAD | n=197 (71.1 years) | MCI due to AD/mild AD 157/40 | 50.8% by PET | 0.941 (0.910–0.973) | FMM, Florbetapir, Florbetaben (based on visual read) | - | - | For PET, Spearman’s r b= −0.75 | 0.96 (0.90–0.99) | 0.84 (0.75–0.90) | 10.2% |
(Validation cohort) | MissionAD | n=200 (70.8 years) | MCI due to AD/mild AD 160/40 | 50.0% by PET | 0.868 (0.816–0.920) | FMM, Florbetapir, Florbetaben (based on visual read) | - | - | For PET, Spearman’s r b= −0.73 | 0.88 (0.80–0.94) | 0.72 (0.62–0.81) | 10.2% | |
IP-MS | Ovod et al. 2017 [7] | Knight Alzheimer’s Disease Research Center (ADRC) | n=41 (76.2 years) | 27/14 c | 43.9% by PET/CSF c | 0.887 | PiB | - | - | For CSF, r=0.699 | - | - | 12.43% |
ELISA | Pérez-Grijalba et al. 2019 [25] | AB255 Study | n=59 (72.7 years) | 39/20/0 | 30.5% by PET | 0.881 (0.779–0.982) | PiB (1.4) | - | - | For PET, Spearman’s r= −0.464 | 0.778 | 0.875 | 10.49% |
IP-MS | Schindler et al. 2019 [18] | Clinical study from Washington University in St. Louis | n=158 (63.7 years) | CDR 0/0.5/1 d: 148/9/1 | 27.2% by PET | 0.88 (0.82–0.93) | PiB (1.42), Florbetapir (1.22) | - | - | For PET, Spearman’s r= −0.55 (−0.65 to −0.43) | 0.88 (0.75–0.96) | 0.76 (0.67–0.83) | 12.18% |
ECL | Palmqvist et al. 2022 [40] | Panel A+ | n=227 (66.5 years) | 32/106/89 e | 48.5% by CSF | - | - | 0.87 (0.82–0.91) | 0.047 f | For CSF, Spearman’s r= 0.64 | - | - | - |
BioFINDER-1 | n=693 (72.4 years)g | 461/232/0 (174 of CU participants had SCD) | 41.8% by CSF | 0.85 (0.79–0.90) | FMM (1.42) | 0.83 (0.80–0.86) | 0.066 f | For CSF, Spearman’s r= 0.40 | - | - | - | ||
IP-MS | Hu et al. 2022 [33] | MissionAD | n=437 (72.4 years) | 0/411/17 e (9 with no criteria met) | 49.7% by PET | 0.86 (0.82–0.89) | Florbetapir, Florbetaben (Centiloid>25) | - | - | - | 0.90 (0.86–0.94) | 0.71 (0.65–0.77) | 9.2% |
IP-MS | Schindler et al. 2022 [30] | Knight ADRC | n=152 (68.4 years) h | CDR 0/0.5/1 d: 138/9/5 | 32.5% by CSF | - | - | 0.86 (0.79–0.92) | 0.0673 f | For CSF, Spearman’s r= 0.61 (0.50–0.70) | 0.86 | 0.73 | 9.95% |
Knight ADRC | n=103 (68.4 years) h | CDR 0/0.5/1 d: 99/3/1 | 25.2% by PET | 0.86 (0.77–0.95) | PiB (1.42), Florbetapir (1.19) | - | - | For PET, Spearman’s r= −0.44 (−0.58 to −0.27) | 0.85 | 0.71 | 9.85% | ||
Multiple | Janelidze et al. 2021 [22] (Validation cohort) | Alzheimer’s Disease Neuroimaging Initiative (ADNI) | n=122 (72.4 years) | 51/51/20 | 48.3% by PET | IP-MS: 0.845 (0.772–0.917) ECL: 0.740 (0.651–0.829) SIMOA: 0.685 (0.590–0.781) IP-MS: 0.662 (0.565–0.758) SIMOA: 0.634 (0.534–0.734) | Florbetapir (1.11) | - | - | - | - | - | - |
(Development cohort) | Biomarkers For Identifying Neurodegenerative Disorders Early and Reliably (BioFINDER) | n=286 (71.6 years) | 182/104/0 | 41.2% by CSF 38.4% by PET | IP-MS: 0.833 (0.787–0.879) IP-free LC-MS: 0.753 (0.696–0.811) ECL: 0.727 (0.669–0.784) ELISA: 0.672 (0.609–0.735) SIMOA: 0.655 (0.591–0.719) | FMM (1.42) | IP-MS: 0.855 (0.810–0.899) ECL: 0.778 (0.725–0.832) IP-free LC-MS: 0.776 (0.721–0.830) ELISA: 0.697 (0.635–0.758) SIMOA: 0.687 (0.626-0.748) | 0.059 i | For CSF, Spearman’s r b= IP-MS: 0.66 ECL: 0.48 ELISA: 0.36 SIMOA: 0.31 IP-free LC-MS: 0.46 | - | - | - | |
IP-MS | Li et al. 2022 [31] | AIBL | n=183 (74.2 years) | 71/66/46 | 57.4% by PET | 0.84 (0.78–0.90) | PiB, FMM, Florbetapir (Centiloid>25) | - | - | - | 0.90 (0.82–0.95) | 0.69 (0.58–0.79) | 12.3% |
BioFINDER | n=100 (71.1 years) | 28/51/0 (SMC or SCD=21) | 50.0% by PET | 0.83 (0.75–0.91) | FMM (1.42) | 0.81 (0.73–0.89) | 0.066 i | For CSF, Spearman’s r= 0.59 (0.45–0.71) For PET, Spearman’s r = −0.52 (−0.65 to −0.35) | 0.74 (0.60–0.85) | 0.78 (0.64–0.88) | 12.3% | ||
ADNI | n=182 (72.5 years) | 71/86/0 (SMC or SCD=25) | 48.9% by PET | 0.82 (0.76–0.89) | Florbetapir (1.11) | 0.92 (0.84–1) | 0.064 i | For CSF, Spearman’s r = 0.60 (0.31–0.79) For PET, Spearman’s r= −0.57 (−0.66 to −0.46) | 0.73 (0.63–0.82) | 0.84 (0.75–0.91) | 12.5% | ||
SIMOA | Tanaka et al. 2021 [37] | Recruited from the memory clinic at the National University Hospital, Singapore | n=68 (74.5 years) | 14/0/15 (CIND=30, VaD=9) | 33.8% by PET | 0.816 (0.704–0.900) | PiB (visually interpreted) | - | - | For PET, Pearson’s r= 0.123 | 0.696 | 0.889 | - |
Multiple | Keshavan et al. 2021 [24] | Insight 46 | n=441 (70.7 years) | 410/7/0 (24 with prior neurological condition) | 18.6% by PET | IP-MS: 0.817 (0.770–0.864) SIMOA: 0.62 (0.548–0.691) | Florbetapir (0.61) | - | - | - | IP-MS: 0.866 SIMOA: 0.451 | IP-MS: 0.719 SIMOA: 0.780 | IP-MS: 9.5% SIMOA: 5.8% |
Multiple | Zicha et al. 2022 [28] | ADNI | n=121 (77.9 years) | 49/54/18 | 49.6% by PET | IP-MS: 0.814 (0.736–0.892) 0.715 (0.625–0.805) ECL: 0.710 (0.617–0.803) SIMOA: 0.661 (0.563–0.760) 0.645 (0.545–0.745) IP-MS: 0.643 (0.542–0.743) | Florbetapir (1.11) | - | - | For PET, Spearman’s r j IP-MS: −0.533 −0.369 ECL: −0.390 SIMOA: −0.293 −0.329 IP-MS: −0.242 | - | - | - |
IP-MS | West et al. 2021 [32] | Clinical studies from University of Wisconsin, the Banner Alzheimer’s Institute, University of Florida, Washington University School of Medicine, and Precision for Medicine | n=414 (70.0 years) k | - | 39% by PET/CSF l | 0.81 (0.77–0.85) | PiB, Florbetapir, Florbetaben k | - | - | - | - | - | 9.75% |
IP-MS | Tosun et al. 2021 [29] | ADNI | n=173 (72.5 years) | 87/86/0 | 48.0% by PET | 0.80 (0.65–0.94) m 0.87 (0.75–0.99) n | Florbetapir (1.11) | - | - | - | 0.64 m 0.79 n | 0.77 m 0.75 n | - |
IP-MS | Kaneko et al. 2014 [21] | NCGG | n=62 (74.1 years) | 33/12/17 | 64.5% by PET | 0.798 | PiB (Scored by nuclear medicine physicians) | - | - | For PET, Pearson’s r= −0.316 | 0.750 | 0.773 | |
ECL | Palmqvist et al. 2019 [39] (Development cohort) | BioFINDER | n=842 (72 years) | 513/265/64 | 43.7% by CSF | - | - | 0.77 (0.74–0.81) | 0.059 i | For CSF, Spearman’s r= 0.476 | 0.75 (0.68–0.80) | 0.72 (0.65–0.77) | 6.5% |
SIMOA | Vergallo et al. 2019 [35] | INSIGHT-preAD | n=276 (76.8 years) | 276 with SMC | 26.4% by PET | 0.794 a | Florbetapir (0.7918) | - | - | - | 0.781 | 0.749 | 5.61% a |
IP-MS | Hu et al. 2022 [33] | Plasma Test for Amyloidosis Risk Screening (PARIS) | n=249 (74.6 years) | 0/172/77 | 64.7% by PET | 0.79 (0.73–0.85) | Florbetapir, Florbetaben, FMM (analyzed by radiologists) | - | - | - | 0.85 | 0.63 | 8.9% |
Multiple | De Meyer et al. 2020 [36] | Flemish Prevent AD Cohort KU Leuven (F-PACK) & Biomarker-based adaptive development in Alzheimer’s disease (BioAdaptAD) | n=199 (70 years) | 161/38/0 | 19.1% by PET | SIMOA: 0.79 (0.73–0.85) ELISA: 0.78 (0.72–0.84) | FMM (1.38), Florbetaben (1.29) | - | - | For PET, Spearman’s r= SIMOA: −0.32 ELISA: −0.32 For CSF, Spearman’s r b,o= SIMOA: 0.29 ELISA: 0.41 | SIMOA: 0.74 (0.57–0.87) ELISA: 0.78 (0.62–0.90) | SIMOA: 0.80 (0.72–0.86) ELISA: 0.75 (0.68–0.82) | SIMOA: 23.0% ELISA: 15.9% |
IP-free LC-MS | Janelidze et al. 2022 [34] | BioFINDER-1 | n=381 (71.6 years) h, p | 241/140/0 | 43.1% by CSF 42.5% by PET | 0.788 (0.729–0.847) m 0.771 (0.687–0.856) n | FMM (0.74) | 0.790 (0.730–0.851) m 0.714 (0.628–0.801) n | 0.091 q | - | - | - | - |
BioFINDER-2 | n=514 (66.2 years) h, r | 350/164/0 | 33.1% by CSF 31.7% by PET | 0.752 (0.691–0.814) m 0.670 (0.585–0.755) n | FMM (0.69) | 0.786 (0.732–0.841) m 0.703 (0.621–0.784) n | 0.0752 s | - | - | - | - | ||
SIMOA | Verberk et al. 2018 [20] | Subjective Cognitive Impairment Cohort (SCIENCe) Project | n=248 (61.0 years) | 248 with SCD | 23% by CSF | - | - | 0.77 (0.69–0.84) | CSF Aβ42 used t: 813pg/mL | For CSF, Pearson’s r= 0.38 | 0.76 | 0.75 | 4.5% |
SCIENCe Project | n=69 | 69 with SCD | 33% by PET | 0.68 (0.55–0.82) | Florbetaben, Florbetapir, FMM, PiB (Scored by a nuclear medicine physician) | - | - | - | 0.70 | 0.78 | 4.4% |