Readers could be forgiven for doing a double-take at the title of a recent paper linking amyloid precursor protein (APP)—best known for its role in Alzheimer’s disease—to amyotrophic lateral sclerosis (ALS). The authors, led by first author Petra Steinacker and senior author Johannes Brettschneider of the University of Ulm, Germany, report their results online in the August 15 PLoS One. They found that quantities of soluble fragments of APP released by α- (sAPPα) and β-secretases (sAPPβ) were lower than normal in the cerebrospinal fluid (CSF) of some people with ALS—those whose disease progresses rapidly. Thus, they suggest sAPP concentrations could help distinguish between people with fast-moving ALS and those whose disease will take a slower course.

“I think that sAPPα and sAPPβ are among the most interesting new biomarkers for a range of neurodegenerative diseases,” wrote Robert Perneczky of the Technical University in Munich, Germany, in an e-mail to ARF (see full comment below). Perneczky, who was not involved in the current work, recently reported on sAPP’s potential as an AD biomarker (see ARF related news story on Perneczky et al., 2011). ”What makes the present report so interesting is that the results suggest that sAPP might serve as accurate predictors of disease course,” he added. “This is a highly relevant topic, since the velocity of disease progression, and therefore the survival rate, varies from a few months to several years.”

Previously, Steinacker published a pilot study (Steinacker et al., 2009) that showed sAPP amounts helped distinguish between ALS and frontotemporal lobar degeneration, two sets of symptoms resulting from overlapping pathology (Geser et al., 2010). In the current cross-sectional study, the team recruited 68 people with ALS. They compared CSF from these subjects to that from 40 control participants who had tension headaches, but not neurodegenerative disease. In addition, the researchers included CSF from 20 people with Parkinson’s disease, to help them tease out ALS-specific effects from those common to neurodegeneration in general.

At first glance, the results are disappointing—the scientists found no correlation between sAPP levels and ALS in general. However, sAPP did help them distinguish between subgroups of the ALS participants. Concentrations of both sAPPα and sAPPβ were unusually low in people with quickly progressing ALS. In people with rapid ALS, CSF sAPPα levels had a median of 21 ng/ml, compared to 27-28 ng/ml in the other subjects. Similarly, sAPPβ levels in people with fast-moving disease averaged 29 ng/ml, compared to 37-41 ng/ml in other participants.

The authors also observed that sAPP levels were slightly lower the longer a person had had ALS. In people who had been sick for only a few months, sAPPα levels ranged from approximately 15-50 ng/ml; by five years in, the values were between approximately five and 25. There was a similar change for sAPPβ. Scientists hope such markers might be useful to track progression during clinical trials.

The reduction of sAPP in ALS is probably due to the death of neurons that make the protein, Perneczky suggested. If that is correct, then sAPP loss is a surrogate marker for degenerating neurons, not a key part of the pathology. The change might only be apparent in ALS, not other neurodegenerative diseases, because motor neurons manufacture particularly high levels of APP, speculated Hui Zheng of the Baylor College of Medicine in Houston, Texas. Zheng was not part of the PLoS study team.

In addition, the study authors pointed out that sAPP is believed to be neuroprotective (see ARF related news story on Li et al., 2010; Kögel et al., 2003). Thus, declining sAPP production could mean that remaining neurons are more vulnerable to protein aggregation or other stresses.

Steinacker and colleagues also examined neurofilament heavy chain, a standard marker for axon damage (reviewed in Petzold, 2005) that is already under consideration as an ALS biomarker (Boylan et al., 2009). CSF neurofilament levels had a median of 15-18 pg/ml in control and PD subjects; 52 pg/ml in those with slow-moving ALS; and 108 pg/ml in people with rapidly progressing ALS.

How well do these markers separate swift disease from sluggish? Using neurofilament alone, the researchers calculated a test would have a sensitivity of 0.8 and specificity of 0.57. That is, the test would correctly identify 80 percent of rapidly progressing ALS cases, and 57 percent of more gradual cases. Adding sAPPα levels to the neurofilament analysis, the team raised the sensitivity and specificity slightly, to 0.84 and 0.6, respectively. Combining neurofilament with sAPPβ achieved similar rates.

Indeed, the best approach will likely be not one biomarker, but a set, Zheng said; researchers are pursuing this approach for several neurodegenerative conditions.—Amber Dance


  1. I think that sAPPα and sAPPβ are among the most interesting new biomarkers
    for a range of neurodegenerative disorders, including ALS and AD. In line
    with their previous report on this topic, Steinacker et al. clearly show
    that ALS has a significant impact on sAPP levels in cerebrospinal fluid (CSF). What makes the
    present report so interesting is that the results suggest that sAPPs might
    serve as accurate predictors of the disease course. This is a highly
    relevant topic, since the velocity of disease progression, and therefore the
    survival rate, varies from a few months to several years. The results will
    surely have to be replicated before they will have a significant impact on
    the diagnostic algorithms, but the findings are clearly relevant and

    Even though it has been shown that sAPP levels in CSF are also
    altered in AD, the reasons are probably different, and I do not think that
    the findings offer a link between ALS and AD. sAPP levels are increased in
    AD, which is a probably a direct result of faulty Aβ clearance or
    overproduction, whereas Steinacker et al. report decreased CSF sAPP levels, probably related to dying neurons, resulting in lower sAPP production. Therefore, I think that sAPP concentration alterations in AD are specific
    for the disorder, whereas they are an unspecific phenomenon in ALS. The
    specificity of sAPP for AD pathology is probably the most important point;
    in my opinion, sAPP will not only be very useful for the early, pre-dementia
    diagnosis of AD (and maybe as an unspecific marker of neuronal damage such as
    in ALS), but it will also significantly improve the differentiation between
    AD versus other non-amyloid dementias such as frontotemporal lobar
    degenerations. The feasibility of sAPP measurements in blood, which have
    also been shown in the present paper, are of particular interest since they
    may provide the means for the first accurate non-invasive biological
    diagnostic procedure in this field.

    View all comments by Robert Perneczky

Make a Comment

To make a comment you must login or register.


News Citations

  1. Soluble APP, Glycosylated Fragments Correlate With AD
  2. APP’s Better Half? Soluble Fragment Activates Protective Genes

Paper Citations

  1. . CSF soluble amyloid precursor proteins in the diagnosis of incipient Alzheimer disease. Neurology. 2011 Jul 5;77(1):35-8. Epub 2011 Jun 22 PubMed.
  2. . Concentrations of beta-amyloid precursor protein processing products in cerebrospinal fluid of patients with amyotrophic lateral sclerosis and frontotemporal lobar degeneration. J Neural Transm. 2009 Sep;116(9):1169-78. PubMed.
  3. . Amyotrophic lateral sclerosis and frontotemporal lobar degeneration: a spectrum of TDP-43 proteinopathies. Neuropathology. 2010 Apr;30(2):103-12. PubMed.
  4. . Soluble amyloid precursor protein (APP) regulates transthyretin and Klotho gene expression without rescuing the essential function of APP. Proc Natl Acad Sci U S A. 2010 Oct 5;107(40):17362-7. PubMed.
  5. . The amyloid precursor protein protects PC12 cells against endoplasmic reticulum stress-induced apoptosis. J Neurochem. 2003 Oct;87(1):248-56. PubMed.
  6. . Neurofilament phosphoforms: surrogate markers for axonal injury, degeneration and loss. J Neurol Sci. 2005 Jun 15;233(1-2):183-98. PubMed.
  7. . Immunoreactivity of the phosphorylated axonal neurofilament H subunit (pNF-H) in blood of ALS model rodents and ALS patients: evaluation of blood pNF-H as a potential ALS biomarker. J Neurochem. 2009 Dec;111(5):1182-91. PubMed.

Further Reading


  1. . Biomarker-based dissection of neurodegenerative diseases. Prog Neurobiol. 2011 Dec;95(4):520-34. PubMed.
  2. . Combination of neurofilament heavy chain and complement C3 as CSF biomarkers for ALS. J Neurochem. 2011 May;117(3):528-37. PubMed.
  3. . Cystatin C: a candidate biomarker for amyotrophic lateral sclerosis. PLoS One. 2010;5(12):e15133. PubMed.
  4. . Elevated CSF TDP-43 levels in amyotrophic lateral sclerosis: specificity, sensitivity, and a possible prognostic value. Amyotroph Lateral Scler. 2011 Mar;12(2):140-3. PubMed.
  5. . PATH45 Cytoskeletal and inflammatory protein biomarkers for amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2010 Nov;81(11):e20. PubMed.
  6. . CSF glial markers correlate with survival in amyotrophic lateral sclerosis. Neurology. 2010 Mar 23;74(12):982-7. PubMed.

Primary Papers

  1. . Soluble beta-amyloid precursor protein is related to disease progression in amyotrophic lateral sclerosis. PLoS One. 2011;6(8):e23600. PubMed.