Value of CSF Aß/tau Profiles in Subjective and Mild Cognitive Impairment

Abstract & Commentary

By Michael Lin, MD, Assistant Professor of Neurology and Neuroscience, Weill Medical College of Cornell University. Dr. Lin reports no financial relationships relevant to this field of study.

Synopsis: Biomarkers in the cerebrospinal fluid may help in the early diagnosis of Alzheimer's disease.

Source: Visser PJ, Verhey F, Knol DL, et al. Prevalence and prognostic value of CSF markers of Alzheimer's disease pathology in patients with subjective cognitive impairment or mild cognitive impairment in the DESCRIPA study: A prospective cohort study. Lancet Neurol 2009;8:619-627.

Using current criteria, Alzheimer's Disease (AD) is not diagnosed until cognitive impairment becomes severe enough to cause a functional decline in cognitive function and behavior. However, given its slowly progressive course, AD must pass through earlier stages: perhaps an asymptomatic stage, a stage of subjective cognitive impairment (SCI), and a stage of mild cognitive impairment (MCI) with objective deficits but no functional decline. Since potential disease-modifying therapies are most likely to be effective when started early, identifying which subjects in these pre-dementia states actually have AD becomes important.

In a recent Lancet Neurology article, Visser and colleagues suggest that cerebrospinal fluid (CSF) Aß42 and tau levels may assist with this identification. CSF Aß42 levels tend to be decreased and tau levels increased in AD subjects. The prevalence and prognostic value of this profile was therefore examined in neurologically healthy controls and subjects with SCI, non-amnestic MCI (naMCI), and amnestic MCI (aMCI). Subjects were drawn from the DESCRIPA study, a multi-center prospective study by the European Alzheimer's Disease Consortium to Develop Screening Guidelines and Criteria for Pre-dementia AD.

Out of 881 subjects enrolled at 20 centers, 168 had CSF analysis at baseline and up to three years of follow-up. An abnormal CSF Aß42/tau ratio was increasingly common in controls (28 of 89 [31%]), SCI (31 of 60 [52%]), naMCI (25 of 37 [68%]), and aMCI (56 of 71 [79%]). A normal CSF Aß42/tau profile was associated with stable or improving cognitive scores in SCI, naMCI, and aMCI. In contrast, a CSF AD profile was associated with declining cognitive scores in naMCI and aMCI. A subsequent clinical diagnosis of AD was made in none of 58 SCI subjects, eight of 34 naMCI subjects, and 27 of 70 aMCI subjects. All of the clinically diagnosed AD cases occurred in subjects with a CSF AD profile. The increase in AD risk associated with a CSF AD profile was statistically significant for aMCI (OR 26.8, 95% CI 1.6-456.4, p=0.02), but not quite for naMCI (OR 10.2, 0.55-188.10, p=0.12).


This was a large study involving multiple centers, with statistically significant results despite the use of different clinical scales at different centers. This suggests generalizability. On the other hand, the study needs replication outside of a memory clinic setting.

Several other caveats should be noted. A positive CSF AD profile was not specific for AD, since it was associated with eventual clinical diagnosis of a non-AD type dementia in one SCI case and one naMCI case. Longer follow-up will be of interest to determine predictive value with greater certainty, particularly in the controls and SCI cases. Serial CSF evaluations and correlation with symptom progression would also be of interest. Additionally, CSF Aß42/tau should be compared with other biomarkers, including amyloid imaging and volumetric MRI studies.

This study provides further support for recent proposals to incorporate disease biomarkers into the diagnostic criteria for AD, which should eventually permit identification at earlier stages. Future studies should consider examination of such biomarkers, including CSF Aß42 and tau measurements.