Though Alzheimer’s is not typically considered an immune disease, serum autoantibodies just may hold promise as a blood-based AD diagnostic, suggests a study published August 3 in PLoS ONE. Using protein microarrays to probe human blood samples, Robert Nagele of the University of Medicine and Dentistry of New Jersey in Stratford and colleagues identified 10 autoantibody markers that distinguished AD from control patients with 96 percent sensitivity and 92.5 percent specificity. These numbers are “really quite impressive—by far the best anybody has reported to date,” Tom Kodadek of the Scripps Research Institute, Jupiter, Florida, told ARF. However, he and others noted that, while the data point to the presence of AD biomarkers in human sera, a reliable AD blood test is still quite a way off.

Prior work by Nagele’s group indicates that the vast majority of people—young and old, diseased and healthy—have brain-reactive autoantibodies in their serum (Levin et al., 2010). The researchers had their first hunch that autoantibodies might play a role in AD when they noticed that many of the surviving neurons in their immunohistochemical studies on AD brains were loaded with immunoglobulins. “This was a consistent feature in regions with AD pathology,” Nagele said. He figured the antibodies came from the blood and reached brain neurons through a disrupted blood-brain barrier—an idea suggested by earlier studies on neuronal death in AD brain (see D’Andrea, 2003). And “when antibodies bind to cell surfaces, it’s not a good thing. They tend to clump things together and immobilize them,” Nagele said. If this applied to signaling receptors, neuronal function would be compromised, prompting Nagele to wonder whether autoantibodies might, in fact, trigger neurodegenerative disease. His lab showed recently that serum autoantibodies can increase intraneuronal Aβ deposition in adult mouse neurons in vitro (Nagele et al., 2011).

To get an idea what the immunoglobulins recognize in AD brains, first author Eric Nagele and colleagues used microarrays to screen human serum samples for autoantibodies against some 9,500 proteins—roughly a third of the human proteome. Eric Nagele, son of Robert, is a medical student at the University of Medicine and Dentistry of New Jersey, and a scientist at Durin Technologies, Inc., which his father founded last year. The team analyzed commercially supplied sera from 50 AD (early- and late-stage) and 40 non-demented controls (young and old).

Half of the AD and control samples served as the training set, from which 451 autoantibodies came up more frequently in the AD group. The researchers chose the 10 showing the greatest difference between AD and controls, and verified these markers in an independent analysis with the remaining 25 AD and 20 control samples. AD samples were also compared against sera from Parkinson's disease and breast cancer patients, with accuracy rates of the 10 markers exceeding 90 percent in all cases. Dot blot analysis confirmed that the two strongest AD-specific antigens were present at lower titers in the control sera.

“The data are very promising for a specific biomarker for AD in peripheral blood,” noted Kaj Blennow of the University of Gothenburg in Sweden. “But much work is needed before we have such a test in our hands.” For starters, the findings must be confirmed in independent AD and control populations, which is challenging in a biomarker field plagued by methodological variability (see ARF related news series) and “notoriously difficult” for plasma markers, Blennow wrote. Other groups have already jumped into the fray (see ARF related news story on Ray et al., 2007). Power3 Medical Products in The Woodlands, Texas, announced validation testing of its NuroPro Blood Test more than two years ago (ARF related news story), but did not respond when contacted for this story.

Using microarrays that run $2,000 a pop, the current study was expensive, Robert Nagele said—hence, the limited sample sizes. In addition, the samples could have been better characterized, noted Anne Fagan of Washington University School of Medicine, St. Louis, Missouri. Because the sera came from commercial suppliers rather than controlled research cohorts, no information was given on how the samples were collected or processed, or whether the patients took medications or had brain amyloid, for example. Given that up to a third of non-demented elderly show Aβ deposition (Aizenstein et al., 2008), some controls in the current study would be expected to have AD pathology, possibly even mild cognitive impairment, leading to questions about whether the identified autoantibodies are markers of AD dementia or AD pathology, Fagan said.

Furthermore, the current study does not say “how early the biomarkers come up, how sensitively we can measure these in early AD,” said Kodadek, who used a library of synthetic antigens to capture disease-specific antibodies (ARF related news story on Reddy et al., 2011). Early detection is important in light of new AD diagnostic criteria that reflect growing evidence that disease begins years before the appearance of obvious symptoms (ARF related news story on McKhann et al., 2011). Robert Nagele said his team would love to tap into existing longitudinal cohorts and track people before and after they show signs of cognitive decline. Their microarray studies thus far, some yet to be published, suggest that autoantibodies may not only serve diagnostic markers for AD, but also PD and non-neurodegenerative diseases. “If I were a betting man, I’d say there’s a common phenomenon here. These markers are detecting the presence of disease,” Robert Nagele told ARF.—Esther Landhuis

Comments

  1. Genomics and proteomics have become highly active research areas to search for novel genes and proteins involved in disease pathogenesis, or to identify novel biomarkers. Research has also expanded to include other “omics” such as metabolomics and lipidomics. In this PLOS-One paper, Nagele and coworkers now take this one step further, to autoantibodies, which could be called “autoantibody-omics,” if you like the “-omics” ending. They screen human serum samples for autoantibodies against a very large number (>9,000) of proteins and show that serum contains a very high number (>1,000) of such antibodies against proteins. Interestingly, the prevalence of autoantibodies was much higher in sera from AD patients. In the next step, they selected the 10 markers that showed the largest difference between AD and controls, ending up with a close-to-perfect diagnostic accuracy. These autoantibodies were directed against a wide range of proteins, not specifically (at least not as known today) linked to AD pathogenesis.

    The data are very promising for a specific biomarker for AD in peripheral blood. But much work is needed before we have such a test in our hands. As pointed out by the authors, a diagnostic test based on multiple indicators is complicated by the fact that many different combinations of biomarkers can be used to distinguish patients from controls with varying diagnostic accuracy. Further, findings from proteomic studies reporting protein panels for use as diagnostic markers have been notoriously difficult to replicate. For these reasons, it will be important to verify the diagnostic utility of these 10 autoantibodies in several independent AD patient and control populations, and also to further study their performance in different clinical stages of AD, as well as in other dementias such as frontotemporal and vascular dementia. Nevertheless, this study thus brings hope for the first blood-based diagnostic test for AD.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Worldwide Quality Control Set to Tame Biomarker Variation
  2. A Blood Test for AD?
  3. A New Test for Alzheimer Disease?
  4. New Strategy Nets Biomarkers for AD, and More

Paper Citations

  1. . Brain-reactive autoantibodies are nearly ubiquitous in human sera and may be linked to pathology in the context of blood-brain barrier breakdown. Brain Res. 2010 Jul 23;1345:221-32. PubMed.
  2. . Evidence linking neuronal cell death to autoimmunity in Alzheimer's disease. Brain Res. 2003 Aug 22;982(1):19-30. PubMed.
  3. . Brain-reactive autoantibodies prevalent in human sera increase intraneuronal amyloid-β(1-42) deposition. J Alzheimers Dis. 2011;25(4):605-22. PubMed.
  4. . Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007 Nov;13(11):1359-62. PubMed.
  5. . Frequent amyloid deposition without significant cognitive impairment among the elderly. Arch Neurol. 2008 Nov;65(11):1509-17. PubMed.
  6. . Identification of candidate IgG biomarkers for Alzheimer's disease via combinatorial library screening. Cell. 2011 Jan 7;144(1):132-42. PubMed.
  7. . The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement. 2011 May;7(3):263-9. PubMed.

Other Citations

  1. ARF related news story

Further Reading

Papers

  1. . Brain-reactive autoantibodies prevalent in human sera increase intraneuronal amyloid-β(1-42) deposition. J Alzheimers Dis. 2011;25(4):605-22. PubMed.
  2. . Brain-reactive autoantibodies are nearly ubiquitous in human sera and may be linked to pathology in the context of blood-brain barrier breakdown. Brain Res. 2010 Jul 23;1345:221-32. PubMed.
  3. . Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007 Nov;13(11):1359-62. PubMed.

Primary Papers

  1. . Diagnosis of Alzheimer's disease based on disease-specific autoantibody profiles in human sera. PLoS One. 2011;6(8):e23112. PubMed.