The earliest pathological underpinnings of Alzheimer disease often precede symptoms by untold numbers of years, presenting researchers with a two-edged sword. On the one hand, early intervention seems difficult when rogue proteins and pathways have ravaged the brain long before a person can tell a doctor something is amiss. On the other hand, optimists see in this long “latent” phase tremendous potential for new drugs to stop the disease before it gains a foothold. At this year’s Human Amyloid Imaging meeting, held 24 April in Seattle, the latter sentiment percolated through the day’s presentations and conversations. With continued advances in brain imaging techniques that measure atrophy, functional changes, and amyloid deposition in vivo, recognition is growing that the field by now indeed has worthy biomarkers in its arsenal, and may be poised to identify individuals at high risk for AD. This should help streamline clinical trials by providing quicker and more reliably measured endpoints, and refining selection of “normal” participants to those on the verge of cognitive decline. This part of our HAI series will summarize several biomarker presentations.

A question that has come to the fore in recent years is whether cognitively normal seniors with high levels of brain amyloid are in fact in greater danger of succumbing to AD than fellow seniors without amyloid. “Various biomarkers inform the issue…but they are not equal,” said Keith Johnson of Massachusetts General Hospital in Boston, one of the conference organizers. “Figuring out how these biomarkers relate to one another, and with cognitive change, will be important for disentangling the normal controls issue. Another thing that will be important is longitudinal data—how normals change over time,” Johnson said. This is critical to make the best use of biomarkers in future trials. “If you’re testing a drug in a big sample of people, you’re going to have a limited budget. You can’t run every marker on everybody. You have to understand how these biomarkers relate to each other to make an informed choice.”

In Seattle, John Morris of Washington University in St. Louis, Missouri, described a study designed to do just that. Preliminary analysis of its data suggests that the earliest biochemical hints of AD are not captured by amyloid imaging but by cerebrospinal fluid (CSF) biomarkers. This provocative conclusion emerged from a study of 241 cognitively normal seniors that used these two modalities to put into better perspective how age and ApoE genotype influence preclinical AD. By plotting amyloid burden (measured by the positron emission tomography [PET] tracer Pittsburgh Compound-B [PIB]) against age for subpopulations stratified by absence or presence of ApoE4 or ApoE2 alleles, Morris showed that ApoE genotype has a clear effect on PIB binding. This is consistent with a recent publication (Reiman et al., 2009 and ARF related news story). In Morris’s study, people who carried at least one E4 allele deposited amyloid with age more quickly than did those who lacked E4. For E2, the effect was protective: in people with an E2 allele PIB did not increase with age, Morris said. Similar trends appeared with CSF Aβ42, a biomarker that is believed to correlate inversely with PIB. Having an E4 allele accelerated the rate of decrease in CSF Aβ42 levels with age, while E2 carriers did not have this age-related decrease; if anything, their CSF Aβ42 levels edged upward as they got older.

When put through the same statistical gauntlet, CSF tau, unlike Aβ, did not seem to associate with ApoE4 genotype in normal elderly. To Morris, the study confirmed that the earliest pathological hints detected thus far in AD involve Aβ42 and not tau. “Tau abnormalities are important, of course, but occur later in the AD process, typically after dementia is fully expressed,” he noted in a post-meeting e-mail to ARF.

A case study from within this cohort of 241 provided further clues about the order of pathological events revealed, thanks to AD biomarkers. Two years before this man was diagnosed with very mild AD (Clinical Dementia Rating of 0.5) at age 90, his performance on episodic memory measures had begun a steady decline. However, the PIB-PET did not show elevated binding levels, and postmortem neuropathological analysis after his death at age 91 showed only small numbers of neurofibrillary tangles (Braak Stage II) and very few neuritic plaques or fibrillar plaques, the type detectable by PIB. The researchers did find numerous neocortical diffuse plaques, though, and CSF assays done at age 88.5 years showed markedly reduced levels of Aβ42, as well as minimal elevations of tau and phospho-tau.

Noting that firm conclusions cannot be drawn from a single case, Morris proposed nonetheless that this particular one “is consistent with a biomarker sequence for AD where the initial abnormality is represented by reduced levels of CSF Aβ42, corresponding to the deposition of diffuse amyloid plaques. Later, as the diffuse plaques become fibrillar, they also are detected by PIB and, as cognitive decline and dementia symptoms occur, by elevations of CSF tau and phospho-tau,” he explained.

In a paper published 11 May in the Archives of Neurology, work by a Washington University team that included Morris, David Holtzman, and first author Barbara Snider, among others, underscores the prognostic value of CSF markers. Studying 49 people with very mild AD, the researchers found that those with lower baseline CSF levels of Aβ42, higher tau or phospho-tau levels, or high tau:Aβ42 ratios, worsened more quickly in a follow-up assessment an average of 3.5 years later (Snider et al., 2009). Another recent paper by an independent group in Munich, Germany (Grimmer et al., 2009), confirmed that reduced CSF Aβ42 levels tracked with high PIB load by and large, but showed some regional differences.

At HAI, Cliff Jack of the Mayo Clinic in Rochester, Minnesota, offered his perspective on the sequence of AD pathological events as determined by serial PIB-PET and structural magnetic resonance imaging (MRI). Jack’s team assessed 21 healthy elderly, 32 people with amnestic mild cognitive impairment (aMCI), and eight AD patients in a longitudinal study involving clinical assessments, MRI, and PIB studies at two time points about a year apart. As previous longitudinal PIB-PET studies have shown, annual change in brain amyloid was small and about equal among all three clinical groups. In contrast, the rate of ventricular expansion, as determined by structural MRI, showed the expected trend (i.e., slowest for controls, intermediate for aMCI, and fastest for AD) and hence was able to distinguish the clinical groups. As for how these measures relate to cognition, PIB change did not correlate with performance in the CDR sum of boxes and only associated weakly with Mini-Mental State Examination (MMSE) scores. Here, too, ventricular expansion rates did correlate with cognitive decline as measured by these two tests. Jack proposed a model whereby amyloid accumulates at a slow, steady rate in late life—in contrast to neurodegeneration, which accelerates.

These data, published recently (Jack et al., 2009), suggest that amyloid imaging is useful for prediction while MRI is most useful for progression tracking during the clinical phase of AD, implying complementary roles for these two measures, Jack said. The same might be said for fluorodeoxyglucose (FDG)-PET and MRI. In a recent study using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), FDG-PET appeared more sensitive at predicting cognitive decline in normal seniors, whereas MRI seemed to more aptly predict further decline for those who were already memory-impaired (Walhovd et al., 2008 and ARF related news story).

As for how changes in FDG-PET and brain amyloid load relate to each other during disease progression, Agneta Nordberg of the Karolinska Institute, Stockholm, Sweden, offered new insight (for more, see Part 1 of this series). At their five-year follow-up, members of an initial cohort of mild AD patients who had received PIB- and FDG-PET scans since 2002-2003 (see Klunk et al., 2004 and ARF related news story) showed very little change in PIB retention from levels at baseline and a two-year follow-up, Nordberg reported. This contrasts with FDG-PET results in these patients; their five-year cerebral glucose metabolism dropped considerably relative to the previous two time points. An ADNI data presentations meeting held in Seattle on 26 April, two days after HAI, featured more on the relative utility of various AD biomarker approaches. Stay tuned for the skinny on that.—Esther Landhuis.

This story concludes our conference series. See Part 1, Part 2, and Part 3.

All attendees are invited to send additions and corrections to esther@alzforum.org.

Comments

Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments

References

News Citations

  1. More ApoE4 Means More Amyloid in Brains of Middle-Aged
  2. Sharper Image—Do More Studies Improve Picture of Cognitive Decline?
  3. HAI Seattle: Does Brain Amyloid Correlate With Sagging Metabolism?
  4. Pittsburgh Compound-B Zooms into View
  5. HAI Seattle: Aβ May Disrupt Brain Function in Normal Seniors
  6. HAI Seattle: Not Just Amyloid, Not Just PIB

Paper Citations

  1. . Fibrillar amyloid-beta burden in cognitively normal people at 3 levels of genetic risk for Alzheimer's disease. Proc Natl Acad Sci U S A. 2009 Apr 21;106(16):6820-5. PubMed.
  2. . Cerebrospinal fluid biomarkers and rate of cognitive decline in very mild dementia of the Alzheimer type. Arch Neurol. 2009 May;66(5):638-45. PubMed.
  3. . Beta amyloid in Alzheimer's disease: increased deposition in brain is reflected in reduced concentration in cerebrospinal fluid. Biol Psychiatry. 2009 Jun 1;65(11):927-34. PubMed.
  4. . Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: implications for sequence of pathological events in Alzheimer's disease. Brain. 2009 May;132(Pt 5):1355-65. PubMed.
  5. . Multi-modal imaging predicts memory performance in normal aging and cognitive decline. Neurobiol Aging. 2010 Jul;31(7):1107-21. PubMed.
  6. . Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004 Mar;55(3):306-19. PubMed.

Other Citations

  1. esther@alzforum.org

Further Reading

Papers

  1. . Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: implications for sequence of pathological events in Alzheimer's disease. Brain. 2009 May;132(Pt 5):1355-65. PubMed.
  2. . Cerebrospinal fluid biomarkers and rate of cognitive decline in very mild dementia of the Alzheimer type. Arch Neurol. 2009 May;66(5):638-45. PubMed.
  3. . Beta amyloid in Alzheimer's disease: increased deposition in brain is reflected in reduced concentration in cerebrospinal fluid. Biol Psychiatry. 2009 Jun 1;65(11):927-34. PubMed.
  4. . Preclinical evidence of Alzheimer changes: convergent cerebrospinal fluid biomarker and fluorodeoxyglucose positron emission tomography findings. Arch Neurol. 2009 May;66(5):632-7. PubMed.