NOTE: It appears that the control population for many of the studies consists of normal individuals. It appears that comparing Alzheimer patients vs patients with other types of dementia has been somewhat limited.
A person with Alzheimer's disease should show both of the following:
(1) elevated CSF tau protein AND
(2) decreased CSF beta-amyloid(1-42)
Assays used by Sunderland et al:
(1) beta-amyloid (1-42): sandwich ELISA by IGEN International Inc, normal reference range not given
(2) tau protein: ELISA by Innotest Inc, reference range not given
|
Test
|
Cutoff
|
|
beta-amyloid (1-42)
|
<= 444 pg/mL
|
|
tau
|
>= 195 pg/mL
|
from Sunderland et al (Figure 2, page 2098)
Performance in the study of Sunderland et al:
• The sensitivity was 92% and specificity 89% (vs a normal control population).
• A tau protein level > 4 times the cutoff (> 800 pg/mL) is seen only in AD patients.
• A beta amyloid level > 2 times the cutoff is seen only in normals.
• The combination of beta-amyloid above the cutoff and tau protein below the cutoff identified a subpopulation almost entirely composed of normal individuals.
• As indicated by Andreasen et al (2001), prevalence of Alzheimer's disease in the study impacts the predictive values.
Limitations:
• The assays may not be completely standardized so variation can be expected between different manufacturers.
