MicroRNA Biomarkers Show Potential for Alzheimer's Diagnosis

Daniel M. Keller, PhD

June 13, 2014

ISTANBUL, Turkey — From a small study looking at the 84 microRNAs (miRNAs) most commonly reported to be differentially deregulated in serum, Italian researchers have found 4 that were significantly decreased in patients with Alzheimer's disease (AD) compared with controls.

Three of these were also decreased in cerebrospinal fluid (CSF), suggesting that circulating miRNAs, combined with other biomarkers, have the potential to be used as noninvasive indicators of AD.

Many biomarkers are useful in indicating AD, including structural imaging (MRI), functional imaging (positron emission tomography/single-photon emission computed tomography), CSF components, and detection of genetic autosomal dominant mutations. However, they are cumbersome, invasive, or costly.

A blood test would be easier to apply in the clinic and could overcome some or all of these negatives, Daniela Galimberti, PhD, from the Department of Pathophysiology and Transplantation, "Dino Ferrari" Center at the University of Milan, Italy, told delegates here are the 24th Meeting of the European Neurological Society (ENS).

"In this scenario, serum circulating microRNAs would be ideal due to the high stability," she said.

Aberrant Expression

These short, single-stranded RNAs modulate gene expression by binding to the 3'-untranslated region of target messenger RNA and thus repress translation.

Several miRNAs have been identified that have aberrant expression in various central nervous system diseases and can be found in several biological fluids, including serum, plasma, and CSF. They are protected from degradation by complexing with microvesicles, high-density lipoprotein, or various proteins.

To detect specific circulating miRNAs in serum and CSF from patients with AD, researchers performed a "discovery analysis" study and a validation study comparing materials from patients with AD to materials from control patients.

Comparisons were made with control patients with noninflammatory neurologic disease (NIND), control patients with inflammatory neurologic disease (IND), and patients with behavioral variant frontotemporal dementia (FTD; also called Pick's disease).

In the discovery phase, miRNA was extracted from serum, amplified, and analyzed on plates that could assess the expression of the 84 miRNAs most commonly reported to be differentially deregulated in serum.

The researchers found 4 miRNAs whose expression was downregulated in AD serum compared with that of NIND controls.

Table. Fold-Regulation of Serum miRNAs in AD vs NIND

miRNA Fold-Change P Value
miR-125b –5.5 .013
miR-223 –4.5 .004
miR-23a –5.0 .012
miR-26b –6.3 .038


Validating the results in a larger population (22 patients with AD, 18 with NIND, 8 with IND, and 10 with FTD), researchers extracted miRNAs from serum and CSF. They validated 3 of the miRNAs (miR-23a, miR-26b, and miR-125b) but not miR-223 in AD serum when compared with serum from NIND controls. Downregulation was also reflected in CSF for miR-26b and miR-125b. Failing validation in CSF was miR-23a.

Receiver-operating characteristic curves showed that AD could be distinguished from NIND by using serum miR-125b (P = .005) or serum miR-26b (P = .019). On the basis of these findings, Dr. Galimberti concluded that miR-125b could be useful as a peripheral biomarker of AD.

Although miR-223 failed validation, she said there was a trend toward decreased levels in both the AD and the FTD groups. It is also decreased in the primary progressive form of multiple sclerosis and correlates with disability. She said miR-223 provides neuroprotection by targeting glutamate receptors. Decreased levels are linked with neurodegeneration.

For the future, she said she plans validation of miR-125b specificity for AD by comparison with other neurodegenerative dementias, as well as a complete profiling of all regulatory noncoding RNAs in AD.

Constant Search

Murat Emre, MD, professor of neurology at Istanbul University Medical School in Turkey, told Medscape Medical News that for the past 20 years there has been a constant search to find a urine or blood marker for AD that would be much simpler than using CSF or brain imaging. So, he said, the current study is important, interesting, and exciting, "but we always have to take it with a grain of salt because we have been excited several times before. You have to be careful that it's validated."

Previous suggestions for protein combinations in blood or in urine have not been replicated, and besides the need for replication in further studies, any candidate biomarkers must show up early in the disease process.

"If it is in a full-blown Alzheimer's patient, it's not very helpful... and I think last year we had another paper from German researchers on microRNAs [that] may be helpful in diagnosis, and we have yet to see replication in larger populations, and the same is true for this new suggestion that 2 specific microRNAs may be diagnostic for Alzheimer's," he said.

Additionally, Professor Emre noted that miRNAs are "shared by hundreds of genes," so one cannot say any single miRNA is a specific regulator for amyloid or tau, for example.

There have been suggestions that a constellation of several miRNAs or proteins in blood or urine may be a better marker for AD, he said. "If the Italians can get along with 2, that would be even better. So this is the first time I'm hearing that 1 or 2 microRNAs may make it." he said, adding that validation for practical use will require hundreds of patients and not just a few dozen.

To discover whether a compound can be an early marker for AD, mining of existing blood and CSF banks and the related databases that exist in many laboratories may reveal markers for progression from good health or mild cognitive impairment to AD. "We have to get it out of the freezer," Professor Emre advised.

The study did not receive any commercial funding. Dr. Galimberti and Dr. Emre have disclosed no relevant financial relationships.

24th Meeting of the European Neurological Society (ENS). Abstract OS1204. Presented May 31, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.