Interstitial lung diseases represent both a diagnostic and therapeutic challenge because the clinical features of the disease are often nonspecific. “The classifications of the disease changes every few years so it’s very hard to diagnose,” says Naftali Kaminski, director of the Dorothy P. and Richard P. Simmons Center for Interstitial Lung Disease (ILD) and associate professor of medicine, pathology, and human genetics at the University of Pittsburgh Medical Center.
Kaminski and Moises Selman from Mexico City set out to ascertain whether gene expression patterns could help determine differences between various interstitial lung disease types, including idiopathic pulmonary fibrosis (IPF), hypersensitivity pneumonitis (HP), and nonspecific interstitial pneumonia (NSIP). IPF is the most common – a highly lethal, interstitial lung disease with an incidence of approximately 80,000 people in the U.S. alone. Survival rates for people to live more than three years is only 50 percent, with the only treatment being transplantation.
Using custom oligonucleotide arrays designed by Eos Biotechnology and Affymetrix, Kaminski and Selman and their collaborators found a functional genetic signature to distinguish between IPF and HP. “We found that the genes that distinguish HP from IPF were genes related to inflammation,” says Kaminski.
What makes this part of the story interesting is that, according to Kaminski, it has been previously thought that IPF lacks an inflammatory component. “The functional profile of genes upregulated in IPF are related to repair and modeling and injury as opposed to inflammatory genes,” says Kaminski. “The neat thing is that theoretically you wouldn’t need to know the patient’s diagnosis to decide on a treatment because if he has this enrichment with inflammatory genes, it would make sense that he would respond to immunosuppressive therapies. And if not, he would need an experimental drug.
“Most patients are prescribed a course of corticosteroids as first-line treatment, such as prednisone, which we know they will not respond to; so theoretically, we now have justification to perform studies which use microarray data to guide therapy,” explains Kaminski.
Kaminski’s team is currently doing another study at the Simmons Center and trying to integrate different phases of disease to see if a genetic signature of early disease versus late disease could also help to classify the different lung diseases.
When asked about comparing data across different platforms, Kaminski said they recently did an analysis using GE Healthcare’s CodeLink Bioarrays and globally found a lot of agreement between the Eos and Affymetrix data in terms of defining an IPF signature.
Moving forward, based on this microarray data, the research team is going to test panels of peripheral blood biomarkers. “The clinical premise here is that if the genes are upregulated in the lung, then they should also be found in other compartments, and then people ideally would not have to undergo biopsies,” explains Kaminski.
The study is published in the January 15 issue of the American Journal of Respiratory and Critical Care Medicine.
This article first appeared in the Bio-IT World newsletter Microarray Technology Review. For a free subscription, sign up here.