A paradox has plagued Alzheimer disease (AD) researchers for quite some time: People with AD have less Aβ42 in their cerebrospinal fluid (CSF) than do healthy individuals. This seems to fly in the face of the amyloid cascade hypothesis, which suggests that AD begins with accumulation of Aβ42, the small peptide that forms amyloid plaques. The face-saving explanation for the discrepancy is that those very same plaques act as a sink, mopping up peptide that would otherwise diffuse or get carried into the CSF. Now, with the aid of Pittsburgh compound B (PIB), a chemical that allows doctors to peer into the living brain and “see” plaques (see ARF related news story), researchers may have confirmed this hypothesis.

Anne Fagan (no relation to the author of this report) and colleagues at the Washington University School of Medicine, St. Louis, Missouri, and at the University of Pittsburgh, Pennsylvania, used positron emission tomography (PET) to quantify uptake of PIB in the brains of 24 volunteers. They compared those measurements, which directly relate to plaque burden (see ARF related news story), with levels of Aβ42 in lumbar puncture samples. Their report in an advanced online publication at the Annals of Neurology contains the first published data correlating plaque load and CSF Aβ42 in living patients.

Their findings confirmed the inverse relationship between plaque burden and CSF levels of Aβ42 that was previously seen in postmortem samples. By itself, this would be insufficient to support the sink hypothesis, because an alternative explanation for the missing peptide is simply that less of it gets produced as the disease progresses and neurons die. But the authors’ findings argue against this latter view. Remarkably, Fagan and colleagues found that in addition to the four AD patients tested, three of the 18 control volunteers also had low CSF Aβ and high uptake of PIB. Because cognitively normal people do not have significant neuron loss, the sink hypothesis seems more likely to explain the ratio in these subjects.

But perhaps more importantly, the authors write that “regardless of the underlying mechanism(s), the present finding suggests that in vivo amyloid imaging, together with CSF Aβ42 measures, may have utility as antemortem AD biomarkers.” This adds to growing optimism that PIB may deliver on its promise as a diagnostic agent (see ARF related news story).

Another promising diagnostic tool is magnetic resonance imaging (MRI). Because this technique reveals that the volume of specific brain regions—the hippocampus and the amygdala—is smaller in those suffering from Alzheimer disease (see Jack et al., 1992 and Cuenod et al., 1993), scientists wonder if brain MRI scans might predict who will get AD. The answer is yes, according to Monique Breteler and colleagues from Erasmus Medical Center, Rotterdam, and University Medical Center, Utrecht, both in the Netherlands.

Writing in the January Archives of General Psychiatry, Breteler, first author Tom den Heijer, and colleagues report the results of a longitudinal cohort study called the Rotterdam Scan Study (see ARF related news story), which is an offshoot of the much larger Rotterdam study that examines the causes, effects, and incidence of disease in elderly volunteers.

Between 1995 and 1996, den Heijer and colleagues used MRI to measure hippocampus and amygdala volumes in 511 volunteers. The people were free of dementia at that time, but over the next six years, 35 of them developed the disease, 26 confirmed as having AD. Now, den Heijer and colleagues report that those original volume measurements are strongly associated with risk for dementia. They reveal that those who developed dementia within 2-3 years of the MRI scans had baseline brain volumes that were about 17 percent smaller than normal. Even those who developed dementia after six years had baseline volumes that were about 5 percent reduced. After adjusting for age, sex, and education, the authors calculate that a single standard deviation below mean hippocampal volume translates into a threefold higher risk for dementia. The same loss in amygdala volume confers a slightly lower risk—about twofold above normal.

“Our study suggests that structural brain imaging can help identify people at high risk for developing dementia, even before they have any memory complaints or measurable cognitive impairment,” write the authors. However, they note that this measurement alone is insufficient to predict dementia, as many people with small volumes at baseline did not go on to develop AD. Moreover, the reported brain shrinkage, in the hippocampus and overall, of AD patients who responded with antibodies and with a hint of visual memory improvement to immunotherapy in Elan’s halted AN-1792 trial has, at least temporarily, complicated the picture on the utility of volumetric measurements in AD (Fox et al., 2005), at least where their usefulness for monitoring treatment effects is concerned. The ability to make these kinds of predictions about diagnosis and progression in a robust way will take on a new level of urgency when drugs are developed that slow or halt the progression of the disease.—Tom Fagan


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Amyloid Ligand Looks Suited for Future Diagnostic Test
  2. Pittsburgh Compound-B Zooms into View
  3. One Thumb Up on Dietary Fat, One Down on Vitamins, Two Down on Estrogen

Paper Citations

  1. . MR-based hippocampal volumetry in the diagnosis of Alzheimer's disease. Neurology. 1992 Jan;42(1):183-8. PubMed.
  2. . Amygdala atrophy in Alzheimer's disease. An in vivo magnetic resonance imaging study. Arch Neurol. 1993 Sep;50(9):941-5. PubMed.
  3. . Effects of Abeta immunization (AN1792) on MRI measures of cerebral volume in Alzheimer disease. Neurology. 2005 May 10;64(9):1563-72. PubMed.

Further Reading

Primary Papers

  1. . Use of hippocampal and amygdalar volumes on magnetic resonance imaging to predict dementia in cognitively intact elderly people. Arch Gen Psychiatry. 2006 Jan;63(1):57-62. PubMed.
  2. . Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Abeta42 in humans. Ann Neurol. 2006 Mar;59(3):512-9. PubMed.