Blood Tests May Aid Alzheimer's Diagnosis (CME/CE)
Several novel markers detectable in blood could prove to be useful adjuncts in diagnosis and management of patients with Alzheimer's disease and other dementias, researchers suggested here.
In separate studies, researchers found significant associations between fragments of tau protein, metabolites of thiamine, and apolipoprotein J (also known as clusterin) in blood products such as serum with clinical diagnoses and/or results of neurocognitive tests, according to presentations at the Alzheimer's Association International Conference.
None of the researchers involved in the studies suggested that such tests would be the sole basis for diagnosing Alzheimer's disease. But, if the current results are confirmed in larger studies, they could help physicians with differential diagnoses without resorting to lumbar punctures or expensive, high-tech imaging scans.
Apolipoprotein J/Clusterin
Veer Bala Gupta, PhD, of Edith Cowan University in Joodalup, Australia, reported that levels of Apo J/clusterin were elevated in patients with either mild cognitive impairment or clear Alzheimer's disease, in a study of participants in the Australian Imaging, Biomarkers, and Lifestyle Flagship Study of Aging.
Apo J/clusterin had previously been linked to Alzheimer's disease, with a British study published 3 years ago showing that levels were significantly higher in patients diagnosed with the disease than in those with mild cognitive impairment or healthy volunteers. Protein levels also were higher in patients with rapidly progressing dementia compared with those with more stable impairments.
Another study reported in 2011 by Dutch researchers indicated that the protein was elevated primarily in patients with established impairments, not in those currently without impairments who subsequently developed cognitive deficits.
Nevertheless, Gupta and colleagues thought it was worthwhile to examine Apo J/clusterin in their research cohort, which is following nearly 1,000 people as they age. She reported data on 590 healthy controls, 93 with mild cognitive impairment, and 150 with Alzheimer's disease at baseline who had follow-up data 18 months later.
The protein is an attractive potential biomarker because it appears to play a role in clearance of beta-amyloid proteins, which most (but not all) Alzheimer's disease specialists believe is a critical pathological agent in the illness.
Mean plasma Apo J/clusterin levels at baseline were significantly higher in the Alzheimer's disease patients (about 360 pg/mL) than in either of the other two groups (320 pg/mL with mild cognitive impairment, 280 pg/mL in healthy controls, both P
<0 data-blogger-escaped-.001="">
Age, independent of cognitive status, also appeared to be associated with levels of the protein. Compared with participants younger than 65, those who were 75 to 85 or older than 85 had significantly higher levels, Gupta reported.
However, Apo J/clusterin levels appeared to be independent of APOE
genotype status. Levels in each of the three diagnostic groups were highly similar between those with the epsilon-4 genotype, which raises the risk of Alzheimer's disease dramatically, and those with other genotypes.
The protein also correlated significantly with scores on the Mini-Mental State Examination (MMSE) and an oral word-association test, with r
coefficients of 0.344 and 0.258, respectively, for baseline values (both P<0 data-blogger-escaped-.0001="">
Finally, protein levels correlated significantly with PET imaging results using a tracer for beta-amyloid protein plaques and with hippocampal volumes estimated from brain scans.
Results were similar for most of these analyses when data from the 18-month follow-up were used.
Gupta said that Apo J/clusterin levels could be helpful in screening patients for Alzheimer's disease risk. When protein levels were added to a model based on age, gender, and the presence of the APOE4
allele, the area under the receiver-operating characteristic curve (AUC) increased to 0.82 from 0.77.
Thiamine Metabolites
Another blood-based biomarker that may indicate Alzheimer's disease is metabolites of the vitamin thiamine, said Chunjiu Zhong, MD, PhD, of Fudan University in Shanghai.
Thiamine handling is related to glucose metabolism in the brain, which is attracting increasing attention in the Alzheimer's disease field as a possible mediator of neuronal dysfunction and death. "Decreased cerebral glucose metabolism is an invariable feature" in the disease, Zhong said, typically preceding onset of overt symptoms.
To examine whether two thiamine metabolites -- thiamine monophosphate (TMP) and thiamine diphosphate (TDP) -- were associated with Alzheimer's disease, he and colleagues determined blood levels of these molecules in 63 patients with clinically diagnosed Alzheimer's disease and 401 controls.
These data formed the basis of a statistical model that was then validated in a different cohort of 31 Alzheimer's patients, 65 with vascular dementia, 10 with frontotemporal dementia, and 200 controls. Revalidation was then performed with data from both cohorts combined.
The end result, Zhong said, was that TDP as a solitary marker had an AUC o.808 for predicting the presence of Alzheimer's disease, using a cutoff of 103.26 nmol/L.
He and his colleagues also developed a model combining TDP and TMP values plus patient age which, in the pooled cohorts, yielded an AUC of 0.909 in APOE4
-negative participants. Specifically, its sensitivity was 85.1% and the specificity was 90.5%. In APOE-positive participants, the performance was slightly worse, with an AUC of 0.885.
Zhong also noted that TDP was not associated with scores on the MMSE or the Clinical Dementia Rating (CDR) scale, indicating that it wouldn't be a marker for disease severity.
Tau Fragments
Neurofibrillary tangles composed of abnormal tau proteins are the other major hallmark of Alzheimer's disease in addition to beta-amyloid plaques. That observation naturally has prompted efforts to find tau-based biomarkers that could serve as diagnostic tools.
However, said Kim Henriksen, PhD, of Nordic Biosience in Herlev, Denmark, those efforts have been frustrated because intact tau proteins, which are expressed only in the brain, are too large to penetrate the blood-brain barrier and therefore aren't amenable to a blood test.
But proteases in the brain do carve them up, producing fragments small enough to pass into circulation, he said. His firm has developed quantitative assays for specific fragments that would allow these tau fragments to be tested as possible blood-based biomarkers.
He and his colleagues obtained serum samples from 39 Alzheimer's disease patients who had received placebo in an Eli Lilly drug trial. Disease severity in these patients ranged from mild to moderate, and the samples were accompanied by data on the patients' cognition scores in the Alzheimer's Disease Assessment Scale (ADAS-Cog). The data of interest were change from baseline at weeks 28 and 52 in the trial.
The researchers were especially interested in a tau fragment cleaved by a caspase-3 enzyme, which they dubbed Tau-C. For comparison, they also analyzed the samples for another fragment called Tau-A produced by a different enzyme.
Henriksen reported that Tau-C was significantly correlated with change from baseline ADAS-Cog scores, but -- against expectations -- the correlation was negative. That is, higher levels of Tau-C were associated with less disease progression during the Lilly trial. In contrast, Tau-A showed a weak trend toward more progression with higher levels.
That led the researchers to explore use of both fragments in tandem. Henriksen said that the ratio of Tau-A to Tau-C proved to be a reasonably strong predictor of ADAS-Cog score changes during the trial.
Higher ratios were associated with greater changes in scores -- that is, faster disease progression -- with r
coefficients of 0.46 at 28 weeks and 0.50 at 52 weeks (both P=0.006).
The study did not examine whether the tau fragments could be used to discriminate between Alzheimer's disease and other types of cognitive impairment, Henriksen noted. But he said it would be relatively straightforward to develop a biomarker panel that included markers specific to other forms of dementia, such as Lewy body disease and frontotemporal dementia, along with Tau-A and Tau-C.
Henriksen acknowledged that his team had not confirmed that the enzymatic action producing the tau fragments had occurred in the brain. If it actually occurs in the blood, that could affect the strength of the correlations with cognitive function or other brain pathologies. He said it would be important to examine that issue in future studies.
The study of apolipoprotein J/clusterin was supported by Australian government agencies. Study investigators had no other disclosures.
The study of thiamine metabolites was funded by Chinese government agencies. Study investigators had no other disclosures.
The study of tau peptides was funded by Nordic Biosciences. Eli Lilly supplied the serum samples. All the investigators were employees of Nordic Biosciences.
Primary source: Alzheimer's Association International Conference
Source reference:
Gupta V, et al "Establishing apolipoprotein j as a potential candidate for Alzheimer's disease biomarker panel: Australian Imaging, Biomarker and Lifestyle (AIBL) Flagship Study of Aging" AAIC
2013; Abstract O1-02-03.
Additional source: Alzheimer's Association International Conference
Source reference:
Zhong C, et al "Altered blood thiamine metabolism in Alzheimer's disease" AAIC
2013; Abstract O1-02-02.
Additional source: Alzheimer's Association International Conference
Source reference:
Henriksen K, et al "A caspase-generated epitope of tau measured in serum predicts progression of disease in people with early Alzheimer's disease" AAIC
2013; Abstract O1-02-01.
