Table 2 Characteristics of the reviewed longitudinal studies
No. | Author | Study population (cohort) | Follow-up duration | AT(N) Biomarkers | Biomarker method / assay | NPS instrument | NPS with association | Significant findings |
|---|---|---|---|---|---|---|---|---|
1 | Binette et al. 2020 [45] | 115 CU at-risk for AD (PREVENT-AD) | 3 years | A T | [18F]NAV4694-PET, global SUVR [18F]AV1451-PET, SUVR in Braak stages I, III, IV | GDS-S, AES, Geriatric anxiety scale, Stress subscale, Big5 inventory | Apathy, anxiety | • ↓ A associated with ↓ anxiety and apathy at baseline • ↓ T associated with ↓ apathy at baseline • No longitudinal association between A, T and NPS |
2 | Babulal et al. 2016 [54] | 118 CU at baseline (9 with mood disorder) 66 CU on follow-up (Washington University) | 1 year | A T N (CSF total tau) | PiB-PET, MCBP cut-off used median split CSF Aβ42, total tau, ptau181 ELISA. Cut-off values used median split | POMS-SF, GDS-S, NPI-Q | Total mood disturbance, NPI-Q score, anxiety and depressive symptoms | • ↑ A longitudinally associated with ↑ total mood disturbance, anxiety, depression scores on POMS-SF, ↑ NPI-Q score and ↑ GDS-S score • T and N predicted longitudinal changes in NPI-Q score |
3 | Blasko et al. 2010 [55] | 331 CU (VITA cohort) | 2.5 years | A | Sandwich Aβ ELISA | DSM-IV criteria, GDS | Depression | • ↑ A at baseline predicted late-onset depression |
4 | Direk et al. 2013 [56] | 657 nondemented (Rotterdam study) | 11 years | A | Sandwich Aβ ELISA | CES-D, HADS | Depressive symptoms | • ↑ A associated with ↑ depressive symptoms at baseline and subsequent incident dementia |
5 | Donovan et al. 2018 [57] | 270 CU (HABS) | 5 years | A | PiB-PET, DVR as continuous variable | GDS | Anxious-depressive symptoms | • ↑ A longitudinally associated with ↑ GDS and anxiety-concentration scores |
6 | Gatchel et al. 2019 [58] | 276 CU - 22 with mild depression (HABS) | 1 year | A N (HV, glucose metabolism) | PiB-PET, DVR cut-off > 1.20 3 T MRI, FDG-PET | GDS | GDS score | • ↑ A moderated the longitudinal association between ↑ depressive symptoms and ↑ cognitive decline |
7 | Harrington et al. 2017 [59] | 359 CU (AIBL) | 54 months | A | PiB-PET, SUVR cut-off ≥1.5 | GDS-S | Incident depression | • ↑ A longitudinally associated with ↑ incident depression |
8 | Johansson et al. 2020 [60] | 104 CU 53 MCI (BioFINDER) | 4 years | A N (GM thickness, subcortical volumes) | [18F] flutemetamol PET, SUVR cut-off > 1.42 3 T MRI | AES, HADS | Apathy, anxiety | • ↑ A longitudinally associated with ↑ apathy, ↑ anxiety over time • ↑ N longitudinally associated with ↑ apathy over time (in CU and MCI subjects) |
9 | Perin et al. 2018 [61] | 585 CU (AIBL) | 72 months | A | PiB-PET, SUVR cut-off ≥1.5 | GDS-S | GDS-S score, apathy-anxiety symptoms | • ↑ A longitudinally associated with ↑ GDS-S score and apathy-anxiety symptoms |
10 | Qiu et al. 2015 [62] | 223 CU - 68 depressed (NAME cohort) | 6.2 years | A | Sandwich Aβ ELISA, Aβ + cut-off used median split of plasma Aβ40/Aβ42 ratio | CES-D | Nil | • No association between A and depression |
11 | Ng et al. 2017 [63] | 33 preclinical AD – A + T+ 60 at-risk for AD - A + T- or A-T+ 22 healthy control (ADNI) | 2 years | N (Glucose metabolism) | FDG-PET | NPI | NPI-Q score | • ↑ NPI-Q score predicted longitudinal ↓ glucose metabolism in preclinical AD |