Biomarker Panel Separates Alzheimer's From Other Dementias

Pauline Anderson

September 27, 2012

September 27, 2012 — Using a panel of 5 biomarkers may help provide a differential diagnosis of patients with dementia or parkinsonian disorders, new research suggests.

The study showed that 4 of these biomarkers — T-tau; P-tau; Aβ1-42; and a relatively new biomarker, α-synuclein — can differentiate patients with Alzheimer's disease (AD) from those who have dementia with Lewy bodies (DLB) or Parkinson's disease with dementia (PDD).

Another biomarker, neurofilament light chain (NF-L), used alone can separate patients with PD from those with atypical parkinsonism, such as progressive supranuclear palsy, multiple-system atrophy, and corticobasal degeneration, the researchers also report.

The study is the first to assess the added value of combining several cerebrospinal fluid (CSF) biomarkers simultaneously in the differential diagnosis of all these disorders, said study author Annika Öhrfelt, PhD, a neurochemist researcher at the Sahlgrenska Academy, University of Gothenburg, Sweden.

Although this study is exciting and represents a major move forward in the field, it's still a research project, she said, and more work is needed before the biomarker panel is ready for prime time in clinical practice.

"We need more longitudinal studies of those 4 biomarkers together," although all but α-synuclein are already available and in use to differentiate AD, she said.

The study was published online August 27 in the Archives of Neurology.

Clear Value

For the study, researchers analyzed 453 CSF samples from patients with PD (n = 90), AD (n = 48), PDD (n = 33), DLB (n = 70), progressive supranuclear palsy (n = 45), multiple-system atrophy (n = 48), and corticobasal degeneration (n = 12). They also collected 107 samples from healthy individuals serving as controls.

Researchers at Lund University, Skåne University Hospital, University of Gothenburg, and Sahlgrenska University Hospital in Sweden used a newly developed multiplex assay (Luminex Corp.) to simultaneously quantify α-synuclein, Aβ1-42, T-tau, and P-tau.

Although Aβ1-42, T-tau, and P-tau are commonly used biomarkers, the "new" element in this study is the addition of α-synuclein to differentiate AD from DLB or PDD, said Dr. Öhrfelt.

The study confirmed that CSF T-tau and P-tau levels are increased in patients with AD compared with patients with PDD and DLB. It also showed that although Aβ1-42 is reduced in PDD and DLB, the levels are further reduced in AD, a finding consistent with neuropathologic data showing that A&beta-related pathologic abnormality is more pronounced in AD than in DLB and PDD.

Multivariate analysis found that the 5 biomarkers together could differentiate AD from DLB and PDD, with an area under the curve (AUC) of 0.90 (95% confidence interval, 0.85 - 0.96). The diagnostic accuracy of this panel of biomarkers that represent tau-, Aβ- and α-synuclein–related pathologic abnormality was "high enough to be of clear value" in differentiating patients with AD from those with PDD and DLB, commented Dr. Öhrfelt.

The diagnostic accuracy of the biomarkers for differentiating AD from DLB is at least as good as that obtained with use of dopamine transporter imaging (the imaging of the protein that pumps dopamine out of the synapse and that provides a marker for presynaptic neuronal degeneration), and it's less expensive, she noted.

The study showed "a clear difference" in the levels of α-synuclein between patients with AD and those with PDD or DLB. However, although α-synuclein decreased in PD, PDD, DLB and multiple-system atrophy compared with controls, these changes were only modest.

"Therefore, CSF measurements of α-synuclein alone would be of little use for early diagnosis of PD," the authors write.

Disappointing Marker

"We can't use it as a single marker yet," added Dr. Öhrfelt. "There is a clear difference between AD and the movement disorders — the PD and atypical Parkinson's disorders — but we can't use α-synuclein to differentiate between PD and controls. To be honest, that's disappointing because we always like to have differences between healthy controls and patients."

As well, she said, including this biomarker in the assay can be "very tricky" because there is potential for it to leak from red blood cells in blood-contaminated CSF samples, which can raise levels of α-synuclein. This casts some doubt on its future as a reliable biomarker, she said.

The study found that in patients with predominant parkinsonism, all 5 CSF biomarkers together could differentiate patients with PD from those with atypical parkinsonism disorders (progressive supranuclear palsy, corticobasal degeneration, and multiple-system atrophy) with an AUC of 0.93 (95% confidence interval, 0.89 - 0.97).

As for NF-L, levels of this biomarker correlated with more severe parkinsonian symptoms in patients with PD and progressive supranuclear palsy, and with worse cognitive performance in patients with AD. However, the NF-L level did not correlate with the duration of disease in these conditions, indicating that NF-L is associated with disease severity but not disease duration.

"NF-L might be a disease severity biomarker that could be used in clinical trials when evaluating the effects of new disease-modifying therapies," said Dr. Öhrfelt.

Although previous studies also showed that NF-L can differentiate PD from atypical parkinsonian disorders, this new study was much larger and it included patients with corticobasal degeneration, which the others had not, said Dr. Öhrfelt. She stressed that assessing NF-L levels is relatively simple because it involves only 1 biomarker instead of a panel of 4.

Making an early determination of pathologic brain abnormality that underlies symptoms is important because future disease-modifying therapies are likely to be more efficient if initiated when the patient first exhibits subtle symptoms. An accurate diagnosis would also be important in clinical practice because early and accurate diagnosis speeds initiation of relevant symptomatic therapies, according to Dr. Öhrfelt.

Expands Field

Asked to comment, Maria C. Carrillo, PhD, senior director, Medical & Scientific Relations, Alzheimer's Association, said the study is important because it takes the field "beyond the typical 3 markers and moves it to an additional 2 markers — the NF-L and α-synuclein."

Another positive point about the study, she said, was that it included a significant number of patients in each of the different dementia groups. "There are multiple pathologies that exist within any 1 dementia diagnostic, so trying to dig deeper into that is what this paper is focused on," she said.

She commented that the only pathologic abnormality underlying the dementia process that was not represented in the paper was microhemorrhages, or vascular changes that are related to vascular dementia.

The study "is directly speaking to" the issue of how to use CSF to define the population with AD who don't yet have symptoms of memory problems, said Dr. Carrillo. She noted that various laboratories around the world, including the one involved in this present research, have been following patients for up to 10 years and have already demonstrated a very high degree of prediction for people who have "this biosignature of CSF" to go on to develop AD.

"So we're gaining more and more confidence in that, but how we then standardize it so that you could actually get it at your doctor's office if you need it is the next step."

The Alzheimer's Association is already making moves in that direction, she added. It hosts an international global consortium that works to develop such standardization, with the goal of eventually having biomarkers available for use in the clinic, said Dr. Carrillo.

This study was supported by the Swedish Research Council, Swedish Alzheimer Foundation, Torsten and Ragnar Soderberg Foundation, Swedish Brain Power consortium, and Regional Agreement on Medical Training and Clinical Research between the Skåne County Council and Lund University and the Sahlgrenska University Hospital and the Sahlgrenska Academy. The authors have disclosed no relevant financial relationships.

Arch Neurol. Published online August 27, 2012. Abstract