Rubicon Genomics is developing a DNA methylation test for the diagnosis and prognosis of a variety of cancers that it expects will compete directly with microarray-based tests that are likely to hit the market over the coming years.
The new MethylPlex test is being developed to detect patterns of abnormal DNA methylation in the serum or urine of patients during tumor formation and progression. The Ann Arbor, Mich.-based firm expects the first commercial MethylPlex tests will be used in patients who have a high PSA level or other indications of prostate cancer to identify cases that need aggressive treatment.
“We are working right now on clinical samples from esophageal adenocarcinoma and from prostate adenocarcinoma,” said John Langmore, vice president of commercial development for Rubicon Genomics. He noted that the number of esophageal cancer cases in the US has increased by about 350 percent in the last 10 years.
The test is based on the firm’s OmniPlex technology, which creates in vitro libraries of nucleic acid fragments that are amplified. Langmore said the technology allows for up to a million-fold amplification, and can be used for genome-wide genetic, epigenetic, or expression analysis.
“It doesn’t work using methylation-specific PCR, and it doesn’t work using bisulfite conversion, or any of the other normal ways of looking at methylation,” Langmore told BioArray News. “It’s a type of amplification where we preferentially amplify the methylated sequences in the genome.”
The amplification process takes a few hours, “and at the end of the day we have all the methylated sequences in the genome and we can compare different genomes — a controlled to an experimental genome — and we can find out right away which promoters or other sequences are methylated preferentially in the experimental cells,” Langmore explained.
“If the methylation is in the promoters, then that tells you which genes have been down-regulated in the cancer. So, when the tumor-suppressor genes are turned off that’s one of the markers of a cancer cell.”
According to Rubicon, MethylPlex will be capable of producing high-resolution maps of methylation for more than 10,000 genes in a single day, at a price of less than $.10 per gene. The result would be a reduction in time and cost for discovery and validation of new markers by nearly 100-fold.
In addition to competing with current cancer tests, the MethylPlex will compete with microarray-based molecular diagnostic products that are likely to reach the market over the next few years. Competition could come from Nanogen, which will soon launch its NanoChip 400 platform for microarray-based diagnostics. The firm’s acquisition of Epoch Biosciences last week (see brief, p. 7) also means that it will add a line of analyte-specific reagents — which will eventually run on the NanoChip system — including reagents for detecting mutations associated with cancer.
Dutch firm Agendia also is in the early stages of developing microarray-based tests for prostate and esophageal cancer, Chief Operating Officer Laura Van ‘t Veer told BioArray News. The company will soon launch its MammaPrint prognostic breast cancer gene-expression microarray in the US (see BAN 12/15/2004).
Another potential competitor could be Affymetrix, which recently released its first tiling array containing data from the Encyclopedia of Complete DNA Elements (ENCODE) initiative (see BAN 10/27/2004). But beyond gene-expression analysis, the tiling arrays can be used for analyzing DNA methylation. Affy also recently filed for US Food and Drug Administration clearance of its GeneChip 3000Dx system, a version of its GeneChip gene-expression platform that was designed for diagnostic applications (see BAN 12/15/2004).
Langmore said that the amplified methylated sequences can be analyzed on a microarray, and the firm has customized arrays that can do that for all known genes. “We use those arrays to discover new genes associated with cancer,” he said.
“We are currently in the evaluation stage with two manufacturers of oligonucleotide arrays,” he noted. Though he declined to say who the firm was collaborating with, “these are very reputable array companies working in the research and diagnostics area.”
Langmore said the firm believes its technology will prove superior to other molecular diagnostic products because of its focus specifically on methylation, and because it is easier to use than other microarray-based technologies. “The existing technologies are very awkward to use, because [scientists] have to do a lot of chemical processing of the DNA before they use the arrays — and they also need a tremendous amount of starting DNA, whereas we can do our test from very small amounts of DNA,” said Langmore.
The firm claims its microarrays can determine methylation profiles of more than 50,000 sites using less than 10 nanograms of DNA. In addition, it believes the MethylPlex test will enable comparison of methylation profiles from hundreds of patients within several weeks, which will help determine which methylated DNA sites are the most effective diagnostic markers.
Rubicon will most likely market MethylPlex as an analyte-specific reagent first, Langmore said. But, “we are working toward it being a properly FDA-approved test and being distributed by a major pharmaceutical or diagnostics company.” He declined to name who those firms might be. He also noted that the firm was in discussions with “several large companies” regarding co-development of the final MethylPlex test, but declined to name them as well.
As for an estimated date for commercialization, he said, “With the FDA approval, it’s anyone’s guess. As an ASR, if we choose to do that in the beginning, I would hope that we could have something going by the end of 2005.”
Langmore also said the firm has yet to determine a cost for the test. “We would like to eventually replace the PSA [test],” for which there are several million tests run each year in the US. “Our test would be more expensive than the PSA test, but it should be much more specific,” he said.
“What our job is between now and releasing a test is to look at these thousand-or-so candidate markers and validate which ones give us the highest specificity and the highest sensitivity for a patient test.”