Review Article

Cerebrospinal Fluid Biomarkers for Dementia with Lewy Bodies

Table 3

CSF biomarkers for DLB.

StudyBiomarker(s)TechniqueNo. of subjectsNo. of controlsResults

Mollenhauer et al. [51] -synucleinSandwich ELISA38 DLB; 13 AD; 8 PD; 8 CJD13 neurological controlsIn PD and DLB, -synuclein levels significantly reduced ( ) compared to AD and control subjects.
Mukaetova-Ladinska et al. [45] -synuclein -synucleinIgGDot blot5 LBD; 9 AD; 3 VaD8Postmortem ventricular CSF analysis. Elevation of both - and -synucleins in AD, LBD, and VaD compared to controls. An increase in - and -synucleins seen from Braak stage III onwards. Results not influenced by age at death or postmortem delay.
Noguchi-Shinohara et al. [46] -synucleinELISA Assay16 DLB; 21 AD(A subgroup of 13 DLB patients matched for duration of disease and MMSE score to those of the AD subjects) -synuclein levels do not differ between DLB and AD patients. Lower levels of -synuclein in CSF correspond to DLB duration (P  .05).
Ohrfelt et al. [49] -synucleinELISA15 DLB; 66 AD; 15 PD55Similar levels of -synuclein in PD, DLB, and control subjects, whereas, in AD, α-synuclein levels significantly lower compared to controls (P   .001). AD subjects with MMSE 20 had significantly lower level of -synuclein than AD subjects with MMSE 20.
Spies et al. [48] -synucleinELISA40 DLB; 131 AD; 39 FTD; VaD 28Two groups: Group A 57 (aged  50); Group B 55 healthy volunteersNo significant difference in -synuclein levels between DLB, AD, FTD, VaD, or control subjects.
Ballard et al. [44] -synucleinWestern blot12 DLB9Significant lower levels of -synuclein in DLB compared to controls (P   .05). Mildly cognitively impaired DLB subjects (MMSE   24) also had lower levels than controls (P < .007).
Kasuga et al. [52] -synuclein/ t-tau/p-tau181/A 42ELISA34 DLB (including 2 with SNCA duplication); 31 AD; 21 other dementias (12 FTD; 2 PSP, 2 normal pressure hydrocephalus; 2 VaD; 3 unclassified)No control group -synuclein significantly lower in DLB than in AD ( ) or other dementias ( ). CSF -synuclein levels correlated with A 42 level in DLB only ( ; ). CSF t-tau and p-tau181 levels as well as A 40/A 42 ratio levels significantly lower in DLB in relation to AD (P   .01), but similar to other dementias.
Arai et al. [54]t-tauSandwich ELISA6 DLB; 8 FTD; 6 PSP; 3 CBD19 (data taken from previous study)Similar levels of tau in AD and DLB (P = .78), but higher than in controls.
Parnett et al. [55]t-tau/p-tau181/A 42HT7-AT270 Assay and ROC analysis43 DLB; 80 AD40Strong correlation between t-tau and p-tau independent of diagnostic group ( ). No differences between DLB and AD for A 42. The significant increase in p-tau181 in AD (P = .039) has 80% sensitivity at differentiating between AD and DLB.
Clark et al. [56]t-tau/A 40/A 42 (includes also postmortem correlation)ELISA3 DLB; 74 AD (including 4 genetic AD and 10 LBVAD); 10 FTD; 5 CJD; 3 GSS syndrome; 11 miscellaneous neurologic conditions73DLB subjects had 2-fold higher level of t-tau in relation to controls, but two-fold lower levels in relation to AD. No differences in t-tau between LBVAD and AD ( ). Aβ42 highly depleted in DLB in comparison to both control (8-fold) and AD (4.4-fold) subjects.
Hampel et al. [57]p-tau181/p-tau231/p-tau199ELISA22 DLB; 108 AD; 24 FTD; 7 VaD; 22 OND23Decrease in p-tau199, p-tau231, and p-tau181 (P < .001) in DLB compared to AD, with similar levels to other studied dementia groups.
Parnetti et al. [58]t-tau/p-tau181/ Ab 42ELISA19 DLB; 23 AD; 20 PD; 8 PDD20DLB mean CSF t-tau levels significantly lower than in AD patients ( ), but significantly higher in PD, PDD, or control subjects. p-tau181 elevated in AD, but similar between DLB, PD, and PDD groups.
Vanderstichel et al. [59]t-tau/ p-tau181/ A 42ELISA assay60 DLB; 94 AD Higher levels of p-tau181 in AD than in DLB and controls. p-tau181 was the most statistically significant single variable of the 3 biomarkers to discriminate between AD and DLB.
Simic et al. [60]t-tau/ p-tau181/ p-tau199ELISA assay2 DLB; 11 AD; 5 FTD; 8 VaD13p-tau181 differentiates AD and DLB with a sensitivity of 91% and a specificity of 95%.
Koopman et al. [61] (autopsy confirmed dementias)t-tau/p-tau181/A 42ELISA18 DLB; 95 AD; 10 FTD; 6 CJD; 16 VaDNo control groupDLB group had similar level of t-tau, p-tau181, and A 42 as the other dementia groups (FTD, CJD, and VaD); these dementia subjects had significantly lower t-tau ( ) and p-tau (P  ) and higher A 42 ( ) in comparison to AD.
Mattson et et al. [62] (autopsy confirmed dementias)t-tau/p-tau181/A 42ELISA750 MCI (420 stable MCI; 14 incipient DLB; 271 incipient AD; 28 incipient VaD; 7 incipient FTD; 10 other dementias); 529 AD304MCI subjects who developed DLB had significantly lower levels of t-tau and p-tau181 at baseline compared to AD and incipient AD ( ), significantly lower A 42 CSF levels in relation to control ( ), stable MCI, and AD subjects ( ).
Spies et al. [63]A 1-42/A 1-40/t-tau/p-tau181ELISA16 DLB; 69 AD; 26 VaD; 27 FTD47Significantly lower levels of A 40 in DLB and VaD in relation to AD ( ). AD had similar A 40 level to controls ( ). The A 42/A 40 ratio significantly lower in AD in comparison to other dementia groups (P  .001). A 42/A 40 ratio improves differentiating AD from VaD, DLB, and FTD than A 42 measures alone (P  .01). A 42/A 40 ratio performed equally well as the combination of A 42, p-tau181, and t-tau in differentiating AD from FTD and non-AD dementias.
Bibl et al. [64]Aβ peptidesA -SDS-PAGE/ immunoblot21 DLB; 23 AD; 21 PDD23The significant increase of a novel peptide with an A -like immunoreactivity (A 1- ) in DLB patients relative to PDD has a sensitivity of 81% and a specificity of 71% using a cut of point of 0.954% but failed to be classified as a sole biomarker.
Bibl et al. [65]A peptides/tauA -SDS-PAGE/ immunoblot and ELISAs for Aβ1-42 and tau25 probable DLB; 18 probable AD14The ratio of A 1-42/A 1-38 and A 1-42/A 1-37 when combined with absolute tau levels produced a diagnostic test with 100% sensitivity and 92% specificity. This ratio discriminated between AD and DLB with a high specificity (P = 6.6   10 ).
Wada-Isoe et al. [66, 67]Aβ42/p-tau 181ELISA assay22 DLB; 34 AD37No significant difference in p-tau levels in AD and DLB, but a significant increase in the p-tau/Aβ42 ratio in AD in comparison to DLB.
Vanderstichele et al. [59]Aβ42ELISA6 LBD; 39 AD; 10 other dementias, neurological and psychiatric disorder patients 12Significant decrease in CSF A 42 levels in both AD (P .000l) and LBD (P = 0.002), relative to the control group.
Maetzler et al. [18]A 42ELISA9 DLB; 12 PDD; 14 PD no dementiaNo control groupLower levels of A 42 in DLB and PDD compared to the nondemented PD subjects (P = .024). DLB-PIB-positive subjects had lower levels than the PIB-negative subjects (P = .044), but similar A 42 levels to the PIB-negative subjects who had dementia (P = .42).
Boström et al. [68]Mg/Ca/CuMass spectrometry29 DLB51Levels of Mg/Ca/Cu increased in CSF in DLB relative to controls, although increases in Cu not significant. The CSF-Mg concentration had a sensitivity of 93% and a specificity of 81% to detect DLB.
Molina et al. [69]Nitric-oxide metabolites (L-arginine to L-citrulline)Ionic -exchange chromatography22 DLB13Not statistically significant difference in NO metabolite concentration between DLB and controls.
Molina et al. [70]Neurotransmitter (NT) amino-acid (AA) concentrationsIon-exchange chromatography21 DLB No significant differences between control and DLB groups in relation to glutamate, aspartate, and GABA levels; however, higher concentrations of asparagine ( 25%) and glycine ( 21%) in DLB.
Schultz et al. [71]Cocaine- and Amphetamine-Regulated Transcript (CART)Radio-immunoassay12 DLB; 14 AD12Significant decrease (30%) in CART in DLB versus controls ( ), DLB, and AD ( ), but concentrations of CART did not indicate DLB progression. CART levels correlated with p-tau protein concentration.
Schultz et al. [72]Transthyretin (TTR)Radio-immunoassay13 DLB; 59 AD13No significant differences of TTR concentrations between AD and DLB.

Please note that ELISA assays for t-tau, p-tau, and A 42, unless otherwise specified, refer to commercially available sandwich ELISA assays.
Abbreviations: DLB: Dementia with Lewy bodies; LBD: Lewy Body disease; AD: Alzheimer’s Disease; LBVAD: Lewy Body variant of Alzheimer’s disease; PD: Parkinson’s Disease; PDD: Parkinson disease dementia; CJD: Creutzfeldt-Jakob disease; GSS: Gerstmann-Straussler-Scheinker syndrome; FTD: Frontotemporal dementia; VaD: Vascular Dementia; PSP: Progressive supranuclear palsy; OND: other neurologic disorders (e.g., mild psychiatric or neurologic symptoms); CSF: Cerebrospinal fluid; A : Amyloid-beta peptide; t-tau: total tau; p-tau: phosphorylated tau; ELISA: Enzyme-Linked Immunosorbent assay; A -SDS-PAGE: A -sodium dodecylsulphate-polyacrylamide gel electrophoresis; ROC: Receiver Operating Characteristic; MMSE: Mini Mental State Examination; PIB: C-labelled amyloid ligand Pittsburgh Compound B.
Specifically vascular dementia ( ); hypoxia during cardiac arrest ( ); cerebrovascular lesion ( ); unspecified dementia ( ); depression ( ).
Age-matched control.