Skip to main content
Premium Trial:

Request an Annual Quote

Defending Decision to Ride Arrays Into the Clinic, Roche Dx Lays Out IVD Agenda for Next 10 Years

Premium

Sticking up for a technology that his company has embraced, Roche Molecular Diagnostics' CEO Heiner Dreismann this week dismissed recent claims that arrays are not ready for the clinic, and backed up his defense with a timeline for when the company plans to begin selling its next three array-based in vitro diagnostics.

"This market is not an easy one, I agree with other people," Dreismann told BioArray News last week. "The burden of molecular diagnostics is to prove the clinical value of all the new information that genetics might bring [and] that has been completely underestimated."

Dreismann made his remarks one week after experts in both academia and industry told BioArray News that, while they think microarray technology is an excellent discovery platform, DNA chips are not "ready for prime time." Rather, older technologies, especially RT-PCR, are "more powerful" and could have a quicker route to the clinic (see BAN 2/28/2006).

In contrast, Dreismann said this week that microarray technology will eventually become more clinically useful than it is at the moment for molecular diagnostics applications.


"This market is not an easy one, I agree with other people. The burden of molecular diagnostics is to prove the clinical value of all the new information that genetics might bring [and] that has been completely underestimated."

"Diagnostics is an information business, and nothing else. It's just information," Dreismann said. "As soon as we have tools to process more information, we will use more information to gain more medical precision [and] to gain [better] evidence against disease," he continued. "It's a no-brainer for me, but this is my personal opinion, and even in my company we have debates on this issue," he said.

Roche helped put arrays on the molecular diagnostics map last year when it won regulatory approval in the US and Europe to sell its CYP450 AmpliChip as an IVD (see BAN 1/5/2005). That product, which is designed to help physicians assess the way in which a patient might metabolize certain drugs, is based on Affymetrix's GeneChip and accompanying GeneChip Scanner 3000Dx.

But if arrays truly are the platform of the future, why aren't more companies abandoning RT-PCR to develop diagnostic assays on DNA chips? And if more information is most desirable in the clinic, why have Affy and Roche been mostly mum about the performance of the CYP450 AmpliChip since it was released? As reported in BioArray News' sister publication Pharmacogenomics Reporter in November 2003, before the AmpliChip was cleared, Roche said the product would generate $100 million in annual revenues by 2008.

Dreismann said he believes that the real market opportunity for arrays does not exist now, and will probably not exist in two years. But the company is betting that the clinical need for tests built on the technology will be evident over the next decade.

"The really broad mainstream use of arrays will take another decade or so," Dreismann said. "It's starting now, and as a company you need to be part of that game now -- if you want to compete in 10 years, you want to be there now. For users, it's still some time to go," he said.

Leukemia, Lymphoma, and P53 in Queue

In order to "be there now," plans are currently underway to add three new diagnostics on the Affymetrix platform to Roche's roster of array-based tests. The first, and most widely disclosed, is a leukemia chip, similar in format to the CYP450 AmpliChip, that for the first time will use gene-expression profiling in a clinical setting to categorize leukemia patients according to disease subtype.

As recently as its third-quarter earnings presentation in October, Roche claimed that the leukemia chips would be available for research use by the end of 2005. Now, Dreismann said Roche will have the product ready for research use sometime in 2007. The launch of the CYP450 AmpliChip was also delayed, though at that time Roche blamed "market-preparation" setbacks, and the Swiss diagnostic giant was forced to scuttle its plans to sell the product as an analyte-specific reagent after the FDA took it to task for making clinically specific claims (see BAN 6/4/2003, BAN 11/5/2003).

Dreismann also said that the leukemia chip could contain between 150 and 400 genes, and that it is currently being evaluated at 12 undisclosed testing sites worldwide in a study that is likely to provide Roche with information on how to proceed by this fall.

"We have narrowed it down to a couple of hundred genes, and they are still in the run, and we will make a final selection only after this ... study will be finished," Dreismann said. Roche's objective will be to replace existing technologies, like RT-PCR assays, as well as its own LightCycler product, in the clinic.

"The [leukemia] chip that we are developing aims at eventually replacing these other technologies, and thus providing an economic advantage [to the user]," Dreismann said. "The chip diagnostic will be cheaper than running the sum of the seven or eight other technologies that must be run now for comprehensive genotyping," he said.

After Roche begins selling the product for research use sometime in 2007, the company will then seek European, US, and Japanese clearance to sell the test for clinical use.

"That might take up to three months in Europe and up to two years in the US and Japan. So the routine diagnostic product, fully approved, could only be available in 2009," Dreismann said. However, he said there is little value in his predictions. "I would have told you the same thing two years ago for the CYP450 chip," Dreismann added. "We thought it would take FDA two years at least, but they gave us expedited approval and we got approval in six months."

On the heels of the leukemia chip, Dreismann said that Roche also has two additional tests in the pipeline: one for lymphoma, and the other a P53 sequencing assay for assessing response to cancer treatment.

"The lymphoma chip will be used for a precise classification of [the] most important subtypes of lymphoma, based on the genotypes," Dreismann said. "The information the chip provides will enable the physician to tell accurately which subtype of the disease an individual patient has."

Both the leukemia chip and the lymphoma chip are expected to cost more than the CYP450 AmpliChip, according to Dreismann, because they "provide more information." BioArray News' sister publication Pharmacogenomics Reporter reported last December that LabCorp charges $1,360 for a typical AmpliChip test per patient sample.

Dreismann said that it was "textbook marketing" that a new technology like arrays would be more expensive than competitive, older technologies in the market. However, he said that in the long run, the price of its array-based products would come down from their current levels.

"The cost for any new technology always comes down. This happened in the computer industry and the cell phone industry, and it will happen to chip technology as well," he said.

He predicted that the costs for array-based tests would eventually settle in the "three digit space -- maybe around $100."

But as for the current market, Dreismann said it was too early to gauge the success of the CYP450 AmpliChip. "The product has started to grow, Roche focused on the installation of the instrument base in the first year," he said. "That is a pre-condition for the product's use. Now we are creating demand. We have not published any sales figures, though."

When the leukemia chip hits the market, Dreismann said he expects Roche will have an easier time placing the test than it has with CYP450. "The nice thing with leukemia is that there's a finite number of labs specialized on leukemia and a finite number of physicians -- so the universe for our marketing is quite limited," he said.

Dreismann said that Roche's "marketing and service approach to these labs will be pretty different from CYP450, where [it] basically had to go to a couple million doctors."

— Justin Petrone ([email protected])

The Scan

Shape of Them All

According to BBC News, researchers have developed a protein structure database that includes much of the human proteome.

For Flu and More

The Wall Street Journal reports that several vaccine developers are working on mRNA-based vaccines for influenza.

To Boost Women

China's Ministry of Science and Technology aims to boost the number of female researchers through a new policy, reports the South China Morning Post.

Science Papers Describe Approach to Predict Chemotherapeutic Response, Role of Transcriptional Noise

In Science this week: neural network to predict chemotherapeutic response in cancer patients, and more.