Skip to main content
Premium Trial:

Request an Annual Quote

Danish Team Developing miRNA Signatures to Diagnose Skin Diseases


By Molika Ashford

Researchers from the University of Copenhagen, Exiqon, and Leo Pharma have published data showing that microRNAs from skin samples can be used to diagnose different skin diseases with similar clinical presentation.

In a study appearing online last week in the journal Experimental Dermatology, the group showed that miRNAs measured in formalin-fixed, paraffin-embedded skin tissue samples from patients with psoriasis accurately reflect those measured in fresh-frozen or other sample types — an important step toward establishing the validity of skin miRNA biomarker research using stored FFPE samples.

In an earlier study published last year in Blood, the same Danish group demonstrated its first major diagnostic milestone, showing that a signature of five miRNAs in the skin can predict with more than 90 percent accuracy whether a rash is due to cutaneous T cell lymphoma or to a chronic skin disease.

Overall, the team is working with the larger goal to explore the use of miRNAs to diagnose a number of skin diseases using Exiqon's locked nucleic acid technology. The project has been funded by the Danish National Advanced Technology Foundation.

Marianne Lovendorf, a researcher affiliated with Leo pharma and the University of Copenhagen and a co-author of both studies, told Gene Silencing News in an e-mail this week that amid growing evidence that miRNA molecules play a role in the human immune system and the repair function of the skin, the group became interested in finding miRNA signatures that could reliably distinguish different skin diseases.

"The researchers involved in the project have found … a difference in the expression of miRNA molecules in the skin of healthy subjects and patients with skin diseases, and in involved and non-involved skin in patients. Thus, the project aims to evolve this knowledge to develop new forms of miRNA diagnostics tailored to the individual patient needs," she wrote.

The first disease for which the group has shown a promising diagnostic miRNA signature is cutaneous T cell lymphoma. As described in their Blood paper, the researchers examined 200 patients with lymphoma, other skin diseases, and healthy controls. They successfully demonstrated that the level of five miRNAs in the skin could predict with more than 90 percent accuracy whether a patient had lymphoma or another chronic skin disease.

To confirm the expression measurements, the group also measured expression levels of its five-miRNA signature using qRT-PCR, finding that in 103 samples, three of the five miRNAs — miR-155, miR-203, and miR-205 — could distinguished patients with CTCL from benign skin diseases with 95 percent accuracy.

According to the group, cutaneous lymphoma is particularly hard to diagnose by current means. "It can easily be confused with eczema, psoriasis, or a fungal infection. With the existing methods, it may take several years to diagnose the disease, which may therefore in some cases reach a life-threatening stage,” Lovendorf wrote.

She called the team's discovery a "breakthrough that will make it possible to diagnose lymphoma of the skin much sooner and with greater accuracy than today."

In its more recent study, the group compared its ability to detect miRNAs in different skin sample types, establishing the validity of using FFPE tissue for diagnostic biomarker discovery.

Formalin fixing of tissue is known to induce RNA degradation and Lovendorf explained that extracting high-quality RNA from human skin is a further challenge because of high levels of RNases in skin tissue.

MiRNAs are thought to be less prone to RNA degradation than mRNA because of their short length and lack of a poly-A tail. "We thought that it would be interesting to investigate whether microRNA expression in FFPE skin samples reliably reflected miRNA expression levels in matched samples preserved as fresh-frozen and Tissue-Tek-embedding, which are thought to be less prone to RNA degradation," Lovendorf said.

In the study, the team analyzed global miRNA expression levels using Exiqon miRCURY LNA arrays in 25 psoriatic lesions obtained through the three different preservation methods. The researchers also validated the expression levels in three patients using qRT-PCR.

The group found a strong correlation between the three different preservation methods with correlation coefficients ranging from 0.91 to 0.95, the authors wrote.

The team also found a pattern of clusters of miRNA with higher correlation coefficients for each preservation method, suggesting that preservation methods may "to some degree" affect miRNA expression levels. Because of this, the group recommended that future investigations use only a single preservation method.

Lovendorf wrote that the results demonstrate that "miRNA detection in human skin is robust irrespective of preservation method. Thus, we believe that miRNAs offer an appropriate and flexible approach in clinical practices and may hold great promise for biomarker and novel target discovery for skin diseases in the future."

The project involved collaboration between academic and industry players. Lovendorf holds a position she described as an "industrial PhD," in partnership between Leo Pharma and the University of Copenhagen. The study also included researchers from Exiqon.

Lovendorf said the different parties contributed different things to the work. "Leo Pharma was, among other things, responsible for the data analysis. The departments at the hospitals provided samples, and Exiqon provided the LNA technology," she wrote.

Lovendorf could not provide details on the group's plans for developing a commercial test for either cutaneous T cell lymphoma or the other skin diseases the group is interested in. Leo Pharma also declined to comment on its plans regarding the team's research.

The company has several drugs in its pipeline for skin diseases, including psoriasis and non-melanoma skin cancer.

Have topics you'd like to see covered in Gene Silencing News? Contact the editor at mashford [at] genomeweb [.] com.