Table 3.
Prognostic factors and their association with disease progression in eAD studies with low or moderate risk of bias
| Prognostic factor | Study | Population (N) | Measure of progression | Follow-up | Type of analysis | Statistical results | Effect on progression |
| Symptoms (apathy) | |||||||
| Using the GDS-15 assessment the presence of symptoms of apathy were dichotomized into any symptom versus no symptoms | Richard 2012 [23] •• | MCI N = 397 | MCI ⟶ AD | Average: 2.7 y (SD 1.0) | Cox proportional hazards models (adjusted for age, gender, education, and baseline MMSE score) | HR: 1.85; 95% CI: 1.09, 3.15; p = NR | Increase |
| Symptoms (depression) | |||||||
| Using the GDS-15 assessment the presence of symptoms of depressive affect were dichotomized into any symptom versus no symptoms | Richard 2012 [23] •• | MCI N = 397 | MCI ⟶ AD | Average: 2.7 y (SD 1.0) | Cox proportional hazards models (adjusted for age, gender, education, and baseline MMSE score) | HR: 1.15; 95% CI: 0.72, 1.83; p = NR | No evidence of effect |
| Risk factor (age) | |||||||
| Definition NR | Lee 2012 [25] •• | MCI N = 504 | MCI ⟶ AD | 1.47 y (Range: 5.5 mo to 5 y) | Multivariable Cox proportional hazards model analysis (adjusted model for age, education and KMMSE score) | OR: 1.03; 95% CI: 1.00, 1.06; p = 0.041 | Increase |
| Definition NR | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 0.996; 95% CI: 0.970, 1.023; p = NR | No evidence of effect |
| Risk factor (sex) | |||||||
| Female | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 1.152; 95% CI: 0.797, 1.665; p = NR | No evidence of effect |
| Risk factor (education) | |||||||
| Definition NR | Lee 2012 [25] •• | MCI N = 504 | MCI ⟶ AD | 1.47 y (Range: 5.5 mo to 5 y) | Multivariable Cox proportional hazards model analysis (adjusted model for age, education and KMMSE score) | OR: 1.08; 95% CI: 1.04, 1.13; p < 0.001 | Increase |
| Education duration | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 0.954; 95% CI: 0.900, 1.011; p = NR | No evidence of effect |
| Risk factor (amnesic MCI) | |||||||
| Amnesic MCI | Lee 2012 [25] •• | MCI N = 504 | MCI ⟶ AD | 1.47 y (Range: 5.5 mo to 5 y) | Multivariable Cox proportional hazards model analysis (adjusted model for age, education and KMMSE score) | OR: 2.47; 95% CI: 0.94, 6.52; p = 0.068 | No evidence of effect |
| Risk factor (traumatic brain injury) | |||||||
| Subjects with a history of traumatic brain injury with loss of consciousness | LoBue 2018 [27] •• | MCI N = 870 | Amnestic MCI ⟶ AD | Median: 4 y (IQR 2 to 5 y) | Cox proportional hazards models (adjusted for age of MCI diagnosis, race, presence of APOE4 alleles, and family history of dementia) | HR: 0.90; 95% CI: 0.70, 1.15; p = 0.39 | No evidence of effect |
| ADAS-cog 11 (impaired cognition indicated by higher score) | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 1.101; 95% CI: 1.061, 1.144; p < 0.001 | Increase |
| CDR SB (impaired cognition indicated by higher score) | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 1.526; 95% CI: 1.276, 1.824; p < 0.001 | Increase |
| K-MMSE score (normal at baseline) | Lee 2012 [25] •• | MCI N = 504 | MCI ⟶ AD | 1.47 y (Range: 5.5 mo to 5 y) | Multivariable Cox proportional hazards model analysis (adjusted model for age, education and KMMSE score) | OR: 0.90; 95% CI: 0.85, 0.95; p < 0.001 | Decrease |
| Genetic factors | |||||||
| APOE ɛ4 allele carrier | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 0.996; 95% CI: 0.674, 1.470; p = NR | No evidence of effect |
| Biomarkers (Aβ) | |||||||
| CSF Aβ | Spencer 2019 [31] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 3.5; 95% CI: 2.0, 6.1; p = 1.63x10-5 | Increase |
| CSF Aβ42 (abnormally low:<600 pg/ml) | Wolfsgruber 2017 [33] •• | MCI due to AD N = 134 | MCI due to AD ⟶ AD | 27.0 (0.95) mo | Cox-Proportional Hazard regression analyses (adjusted for age, gender) | HR: 6.4; 95% CI: 2.9, 14.2; p < 0.001 | Increase |
| CSF Aβ | Xue 2020 [32] •• | MCI N = 193 | MCI ⟶ AD | NR | Cox proportional hazard regression model (adjusted for age, sex, educational level, APOE ɛ4 genotype) | HR: 0.55; 95% CI: 0.41, 0.75; p < 0.001 | Decrease |
| Biomarkers (Aβ42/Aβ40) | |||||||
| Plasma Aβ42/Aβ40 (Abnormal) | Cullen 2021 [28] •• | MCI BioFINDER: n = 148 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education, and baseline MMSE | HR: 0.79; 95% CI:NR, NR; p = 0.5205 | No evidence of effect |
| Plasma Aβ42/Aβ40 (Abnormal) | Cullen 2021 [28] •• | MCI ADNI: n = 87 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education and baseline MMSE | HR: 0.8; 95% CI:NR, NR; p = 0.0835 | No evidence of effect |
| Biomarkers (p-tau) | |||||||
| CSF p-tau | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 1.006; 95% CI: 1.000, 1.013; p = NR | No evidence of effect |
| CSF P-tau | Spencer 2019 [31] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 2.9; 95% CI: 1.7, 4.7; p = 3.08x10-5 | Increase |
| CSF p-tau | Xue 2020 [32] •• | MCI N = 193 | MCI ⟶ AD | NR | Cox proportional hazard regression model (adjusted for age, sex, educational level, APOE e4 genotype) | HR: 2.31; 95% CI: 1.34, 3.93; p = 0.002 | Increase |
| Plasma p-tau 181 Abnormal | Cullen 2021 [28] •• | MCI BioFINDER: n = 148 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education and baseline MMSE | HR: 2.44; 95% CI:NR, NR; p = 0.0047 | Increase |
| Plasma p-tau 181 Abnormal | Cullen 2021 [28] •• | MCI ADNI: n = 87 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education and baseline MMSE | HR: 2.500.8; 95% CI:NR, NR; p < 0.0001 | Increase |
| Plasma p-tau 181 Abnormal | Palmqvist 2021 [29] • | MCI ADNI: n = 437 | MCI ⟶ AD | 4 y | Logistic regression models | OR: 2.84; 95% CI:2.06, 4.04; p = NR | Increase |
| Plasma p-tau181 > 17.71 (high) | Therriault 2021 [30] •• | MCI N = 604 | MCI ⟶ AD | 5 y | Cox proportional hazards models (adjusted for APOE ɛ4, age, sex and years of education) | HR: 2.06; 95% CI: 1.55, 2.74; p < 0.001 | Increase |
| CSF pTau 181 (abnormally high:>60 pg/ml) | Wolfsgruber 2017 [33] •• | MCI due to AD N = 134 | MCI due to AD ⟶ AD | 27.0 (0.95) mo | Cox-Proportional Hazard regression analyses (adjusted for age, gender) | HR: 6.3; 95% CI: 2.1, 16.3; p < 0.001 | Increase |
| Plasma P-Tau 217 Abnormal | Palmqvist 2021 [29] • | MCI N = BioFINDER: n = 176 | MCI ⟶ AD | 4 y | Logistic regression models | OR: 3.88; 95% CI:2.42, 6.66; p = NR | Increase |
| t-tau | |||||||
| CSF t-tau | Spencer 2019 [31] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 1.9; 95% CI: 1.3, 2.8; p = 1.15x10-3 | Increase |
| CSF tau (abnormally high: > 300 pg/ml) | Wolfsgruber 2017 [33] •• | MCI due to AD N = 134 | MCI due to AD ⟶ AD | 27.0 (0.95) mo | Cox-Proportional Hazard regression analyses (adjusted for age, gender) | HR: 8.6; 95% CI: 2.0, 36.7; p < 0.001 | Increase |
| CSF total tau | Xue 2020 [32] •• | MCI N = 193 | MCI ⟶ AD | NR | Cox proportional hazard regression model (adjusted for age, sex, educational level, APOE ɛ4 genotype) | HR: 1.63; 95% CI: 1.09, 2.44; p = 0.016 | Increase |
| t-tau/Aβ ratio | |||||||
| CSF t-tau/Aβ ratio | Spencer 2019 [28] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 3.6; 95% CI: 2.2, 6.1; p = 9.83x10-7 | Increase |
| p-tau/Aβ ratio | |||||||
| CSF p-tau/Aβ ratio | Spencer 2019 [31] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 3.3; 95% CI: 1.9, 5.9; p = 4.03x10-5 | Increase |
| Ng | |||||||
| CSF Ng | Xue 2020 [32] •• | MCI N = 193 | MCI ⟶ AD | NR | Cox proportional hazard regression model (adjusted for age, sex, educational level, APOE ɛ4 genotype) | HR: 0.90; 95% CI: 0.72, 1.11; p = 309 | No evidence of effect |
| NFL | |||||||
| Plasma NFL (abnormal) | Cullen 2021 [28] •• | MCI BioFINDER: n = 148 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education, and baseline MMSE | HR: 2.56; 95% CI:NR, NR; p = 0.0177 | Increase |
| Plasma NFL (abnormal) | Cullen 2021 [28] •• | MCI ADNI: n = 87 | MCI ⟶ AD | 4 y | Cox regression modelling adjusted for age, sex, education, and baseline MMSE | HR: 1.75; 95% CI:NR, NR; p = 0.0001 | Increase |
| Imaging Biomarker | |||||||
| MTA: Evaluated using a five-point rating scale developed by Scheltens et al. | Pyun 2017 [26] •• | MCI N = 258 | MCI ⟶ AD | Up to 3 y; Median 24 mo | Multivariate Cox regression analysis (adjusted for MTA, PA, age, sex, education, APOE ɛ4 carrier, ADAS-cog 11, CDR SB, and CSF p-tau) | HR: 1.424; 95% CI: 0.997, 2.034; p < 0.05 | Increase |
| Hippocampal volume as a percent of intracranial volume (HC % ICV) | Spencer 2019 [31] •• | MCI N = 185 | MCI ⟶ AD | Mean 4.3 y (SD: 2.8) | Cox proportional hazards regressions controlling for age | HR: 2.4; 95% CI: 1.6, 3.6; p = 2.19x10-5 | Increase |
Aβ, amyloid β-protein; AD, Alzheimer’s disease; ADAS-cog, the Alzheimer’s Disease Assessment Scale-Cognitive Subscale; ADNI, Alzheimer’s Disease Neuroimaging Initiative; APOE, apolipoprotein E; CDR-SB, The Clinical Dementia Rating –sum of boxes; CSF, cerebrospinal fluid; GDS, Geriatric Depression Scale; HR, hazard ratio; K-MMSE, Korean Mini-Mental State Examination; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; mo, month; MSCI, Moderate/Severe Cognitive Impairment; MTA, medial temporal lobe atrophy; N, number; NFL, neurofilament light; NG, neurogranin; NR, not reported; OR: Odds Ratio; PA, posterior atrophy; P-Tau, phosphorylated tau; RR, Risk Ratio; t-tau, total tau; wk, week; y, year. •low risk of bias •• moderate risk of bias.