NuGEN has developed a new RNA amplification and labeling system to enable RNA profiling from FFPE on Affymetrix microarrays. The system provides a key technology that completes the workflow going from total RNA to microarray analysis for FFPE samples.

“Based on everything that we’ve seen thus far, people are going to be very pleased with the kinds of results and the biology that they will be able to access from FFPE samples,” said Gianfranco de Feo, Ph.D., Senior Director for Customer Solutions, NuGEN. “We predict that Affymetrix arrays together with our system will work for a large percentage of the FFPE samples that are banked right now.”

The NuGEN FFPE amplification system is a whole transcriptome amplification assay optimized for degraded and limited amounts of input RNA. Researchers can start with as little as 50 nanograms of degraded total RNA and get enough product to process on Affymetrix gene expression arrays, like the GeneChip® Human Genome U133 microarray.

The company plans to launch the product in May 2007, but researchers have already been able to access the assay as a service offered through Expression Analysis since March.

 Gianfranco de Feo sat down with UserForum editor Tommy Broudy to discuss NuGEN’s recent developments in RNA profiling from FFPE. The two discussed:

• The types of research enabled by accessing FFPE samples
• The technology required and the challenges of getting quality RNA from FFPE samples
• The performance of microarray expression profiling from FFPE samples

ENABLING ACADEMIC AND PHARMACEUTICAL RESEARCH
Broudy: What will academics performing biomedical research be able to do, now that they can leverage FFPE banks for gene expression studies?

de Feo: Large clinical centers, like M.D. Anderson, the Cleveland clinics and a lot of university research hospitals, have a wealth of samples that they’ve already gathered from patients undergoing clinical trials, new therapies or standard treatment. The samples have been formalin-fixed and paraffin-embedded and much of the patient’s information—both demographics and behavioral, as well as therapeutic outcome or disease outcome—has already been established.

These groups want to leverage these samples
for biomarker discovery research to understand which markers correlate to disease or therapeutic phenotypes. Up till now, these groups have been limited as to the numbers of samples they could access for this work. You really need larger
studies to validate biomarker signatures and
you need the sample sets containing outcome information; you can’t always do that
prospectively.

Our customers are very interested in looking at whole-genome expression. You may have a little easier time looking at a small panel of genes, but most people feel that you’re going to miss something. Basically, why look at a subset of genes
when you can do exactly the same thing and
look at everything in the transcriptome? Few of our customers have been willing to look at a list of
genes that somebody else came up with. Especially in cases like FFPE, where the sample is the most valuable part of the experiment, not being able to look at the whole transcriptome limits the value of that experiment too much.

Broudy: What about drug discovery and development? How is FFPE RNA profiling being used there?

de Feo: There are two general areas within the pharma market where FFPE samples can be very relevant. One of them essentially overlaps with everything that we’ve discussed regarding clinical centers and biomarker validation; in those cases, they have their own sample banks from clinical trial patients that they want to go back to. In other cases, they have relationships with some of the research hospitals and clinical hospitals and get access to some of those clinical samples to do biomarker discovery and validation.

The aspect of the market that is unique to pharma is toxicology. So when we talk to our big pharma customers, one of the first areas they discuss with regard to FFPE is in toxicology. And in those cases you’re looking at a variety of different samples, including those where they have retrospective clinical samples of people undergoing therapy where they know their outcomes were negative. They want to go back and take a look at whether they can identify markers to stratify those patients that are actually showing the toxicological effect from a particular therapy.