Table 2 CSF biomarkers for predicting the development of AD in patients with MCI
Study | Number of AD patients | Number of MCI/ non-AD patients | Number of control patients | CSF biomarkers | Optimal cut-off* | Sensitivity % (95% CI)** | Specificity % (95% CI)** | AUC (95% CI) | SL (p value)# |
|---|---|---|---|---|---|---|---|---|---|
Hansson et al. [31] | 0 | 137 (MCI) | 0 | Aβ42 | 0.64 ng/mL | 93 (82–98) | 53 (41–64) | 0.77 (0.69–0.84) | – |
Aβ42/40 ratio | 0.95 | 87 (76–95) | 78 (67–86) | 0.87 (0.80–0.92) | < 0.05 | ||||
Hertze et al. [32] | 94 | 166 (MCI) 29 | 38 | MCI-AD vs MCI | |||||
(DD) | Aβ42 MSD | < 523 | 67 | 71 | 0.73 (0.66–0.80) | – | |||
Aβ42 MSD/40 ratio | < 0.069 | 85 | 71 | 0.86 (0.79–0.91) | NP | ||||
Aβ42 MSD/38 ratio | < 0.37 | 88 | 71 | 0.85 (0.79–0.91) | NP | ||||
Parnetti et al. [33] | 28 (AD) 32 | 58 (MCI-MCI) | 0 | MCI-AD vs MCI-MCI | |||||
(MCI-AD) | 28 (OND) | Aβ42 | 420.5 pg/mL | 56 (38–74) | 96 (88–99) | 0.85 | – | ||
Aβ42/40 ratio | 5.3 | 71 (48–89) | 92 (79–98) | 0.82 | NP | ||||
AD vs non-AD (OND) | |||||||||
Aβ42 | 500.0 pg/mL | 63 (42–81) | 79 (59–92) | 0.70 | – | ||||
Aβ42/40 ratio | 5.0 | 74 (54–89) | 79 (59–92) | 0.78 | NP | ||||
Lewczuk et al. [21] | 75 (AD-MCI) | 0 | 45 | Aβ42 | 691 pg/mL | 69.3 | 88.9 | 0.895 (0.819–0.946) | – |
Aβ42/40 ratio | 0.06 | 93.3 | 100 | 0.970 (0.916–0.993) | < 0.0001 | ||||
Baldeiras et al. [27] | 168 | 197 (MCI) | 66 | Aβ42 | 585 pg/mL | 82 | 83 | 0.882 (0.837–0.927) | – |
Aβ42/40 ratio | 0.068 | 79 | 86 | 0.874 (0.827–0.921) | n.s. | ||||
Frölich et al. [34] | 0 | 115 (MCI) | 0 | Aβ42 | < 600 pg/mL | 74 | 64 | 0.68 | – |
Aβ42/40 ratio | – | 59 | 75 | 0.66 | NP | ||||