. Novel CSF biomarkers for Alzheimer's disease and mild cognitive impairment. Acta Neuropathol. 2010 Jun;119(6):669-78. PubMed.

Recommends

Please login to recommend the paper.

Comments

Make a Comment

To make a comment you must login or register.

Comments on this content

  1. In their recent paper in Acta Neuropatholgica, Hu and colleagues investigated whether novel biomarkers (including those associated with inflammation, apoptosis, growth factor signaling, etc.), on their own or in combination with the established AD biomarkers Aβ42, tau, and ptau could improve the specificity of CSF-based AD diagnosis. They also tested whether these markers could facilitate disease staging and/or predict the rates of cognitive decline, as evidenced by changes in the MMSE. The study has many strengths, including autopsy confirmation of disease state, inclusion of a group of non-AD dementia cases along with standard AD and MCI cases with longitudinal follow-up, a decent number of non-demented controls (n = 33), assessment of a large, standardized, quantitative panel of markers (n = 151 assayed by Rules Based Medicine, RBM), and model-fitting by three analytic methods (Logistic Regression, Random Forest, and PAM).

    Consistent with many previous findings, CSF Aβ42 and tau were found to do well on their own in discriminating AD from non-demented controls, with a sensitivity of 93 percent and a specificity of 82 percent. Impressively, adding to the logistic regression model, two molecules found to differ between the AD and control groups (NrCAM and PDGF) improved the sensitivity to 97 percent and the specificity to 94 percent. The same results were obtained when the data were analyzed by the other two models. CSF Aβ42, ptau, and the agouti-related peptide (AgRP) discriminated AD from non-AD cases in all models as well, whereas the further addition of tau, eotaxin-3, and hepatocyte growth factor (EGF) was found to be useful when analyzed by the RF and PAM models. Interestingly, among CSF biomarkers for AD that were identified by at least one approach, six (C3, CgA, IL-1α, I309, NrCAM, and VEGF) correlated with MMSE score in AD but not the other neurodegenerative disorders tested (including FTLD-TDP, FTLD-tau, FTLD-ALS, and DLB). Also of particular interest was the finding that CSF IL-1α and thymus-expressed chemokine (TECK) correlated with the rate of cognitive decline in a longitudinally followed cohort of MCI subjects (n = 38).

    Although several proteomics studies have identified a variety of markers related to inflammation and signaling cascades that differ in abundance in AD versus control CSF, many of which overlap with the present set, to our knowledge this is the first study to quantify these markers using a high-throughput multiplex platform and rigorously test their ability to discriminate clinical/neuropathologic groups and correlate with the rate of cognitive decline. Defining the generalizability and ultimate clinical significance of their findings will require replication by other groups in other cohorts. We predict assessment of the RBM marker panel in the ADNI cohort is underway. Although the present findings show that some markers correlate with MMSE in AD and were associated with subsequent cognitive decline in longitudinally followed MCI subjects, it will be especially important to determine whether these (or other) analytes are actually useful in predicting future cognitive decline in non-demented, clinically normal cohorts. For disease-modifying therapies to have the greatest chance of preserving normal brain function, it will be critical to be able to identify individuals while they are still cognitively normal. We and others have shown that the ratio of CSF tau/Aβ42 is very good at predicting future cognitive decline in non-demented controls (Fagan et al., 2007; Li et al., 2007). What isn’t known is whether this predictive power can be improved upon by the addition of novel biomarker candidates such as those identified in the present study.

    References:

    . Cerebrospinal fluid tau/beta-amyloid(42) ratio as a prediction of cognitive decline in nondemented older adults. Arch Neurol. 2007 Mar;64(3):343-9. PubMed.

    . CSF tau/Abeta42 ratio for increased risk of mild cognitive impairment: a follow-up study. Neurology. 2007 Aug 14;69(7):631-9. PubMed.