Rosetta Genomics uses miRNA to identify cancers of unknown origin.
By Kevin Davies
July 14, 2008 | During her 15 years at Novartis Institute for Biomedical Research, Dalia Cohen, the pharma’s former head of functional genomics, was fortunate to be allowed to pursue some of her own research interests, which included transcription (see “Genomics Provides the Kick Inside,” Bio-IT World, November 2003). She helped launch the company’s histone deacetylase program, identifying one gene family member.
Two years ago, Cohen joined Israeli biotech Rosetta Genomics as chief science officer, and her former interests are proving more relevant than she dreamed possible. “Rosetta is one of the first companies to identify micro RNAs (miRNAs) using bioinformatics tools,” Cohen says from Israel, where she spends about 1/3 of her time. “While others identified microRNAs using biology, one of the first microRNAs was discovered from a library of [non-coding] junk DNA.”
Rosetta uses a variety of public and proprietary bioinformatics programs to identify this increasingly interesting family of RNA molecules. miRNAs are naturally occurring RNAs that act as master regulators with the potential to form the basis for a new class of diagnostics (based on their differential expression in diseases) and therapeutics.
Pinpointing the Primary
Earlier this year, Cohen and colleagues published an interesting paper in Nature Biotechnology on the use of miRNAs to identify the origin of primary tumors that give rise to metastases in cancer patients. She says the problem of “cancers of unknown primary” (CUP) is an urgent unmet medical need, as conventional tools do not always allow pathologists and oncologists to identify the source of many metastases. About 70,000 patients in the U.S. are classified as CUP annually. “As a result, the patient gets many types of chemotherapy treatments while the right treatment isn’t often given,” says Cohen. “It’s clear that what’s available today is not sufficient or accurate enough for physicians to know how to treat the patient.”
About when Cohen joined the company, Rosetta scientists identified miRNAs and began asking whether they could profile miRNAs in primary to identify the tumor origins. “We looked at the microRNAs expressed in 22 different tumor types,” Cohen explains. “We identified the expression profile of the microRNAs in a variety of primary tumors. Then we asked whether we can see these microRNAs in metastases. The answer is yes—[we see] the same pattern of expression of micro RNA in primary tumors [and metastases], for example, a breast tumor and metastasis into the lung.”
In the report, Cohen and colleagues studied miRNA expression in 400 samples from 22 different tumor tissues and metastases. Using a subset of 48 miRNAs and training decision algorithms, they were able to classify two thirds of the samples with high accuracy (>90%). Unresolved for now is whether the specific miRNA in the primary tumor is mechanistically involved in the metastatic process. Other groups, Cohen says, have found some evidence that miRNA might be involved in metastasis.
“This paper will be translated to an assay used by physicians,” Cohen predicts. A diagnostic test to identify the CUP site should be ready and submitted for regulatory approval later this year.
Meanwhile, Rosetta has submitted a test, developed in conjunction with a CLIA-certified lab at Columbia University Medical Center, to distinguish two classes of the most common form of lung cancer (affecting 185,000 Americans each year), to New York State. If approved, the test will be the first to use Rosetta’s miRNA technology. “With advancements toward more targeted therapies for cancer, there is a growing need for better diagnostics,” said Amir Avniel, president and CEO of Rosetta Genomics. Another test—to differentiate (asbestos-associated) mesothelioma from adenocarcinoma—is also being readied for regulatory approval.
FURTHER READING:
Rosenfeld, N. et al. MicroRNAs accurately identify cancer tissue origin. Nat Biotechnology 2008; 26: 462–469.
___________________________________________________
This article appeared in Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply Today.