February 1, 2008 (Vol. 28, No. 3)
Technique Will Play an Important Role in Diagnostics, Arrays, and Therapeutics
Globally, the epigenomics market is estimated to reach $4.1 billion by 2012, up from $263.2 million in 2006, according to a March 2007 report by BCC Research. The key market segments include research tools, cancer diagnostics, and cancer therapeutics. Many companies are focused on developing specific molecules or assays that push the field of epigenomics forward.
Scientists have discovered that there exist many ways to control DNA, and changes in these control mechanisms may lead to abnormalities (disease). “Genetic and epigenetic changes is the field we’ve been focusing on in the clinic,” says Andrew Allen, M.D., Ph.D., executive vp and CMO at Pharmion (www.pharmion.com). “Some of the changes in epigenetics are reversible. If a cancer cell methylates DNA, we can demethylate that DNA and reverse that change, which may be therapeutically useful.”
Researchers at MethylGene (www.methylgene.com) developed a chemistry platform that enables them to make isoform-selective inhibitors. “Our medicinal chemistry group does rational drug design using mechanistic enzymology and the use of structural and computational information,” explains Donald Corcoran, president and CEO.
“Our competitors had pan-inhibitory compounds that were problematic in terms of toxicity—they weren’t discriminating isoforms. It has become clear through knock-outs that only a minority of these HDACs (histone deacetylases) regulate genes important in cancer,” adds Jeffrey Besterman, Ph.D., executive vp, R&D, and CSO.
The company’s lead compound, MGCD0103, is an oral isotype-selective HDAC inhibitor, currently in seven clinical trials. Several Phase I/II trials involve single use of this molecule; the other trials involve combining MGCD0103 with various clinically approved cancer therapeutics.
Recent study data results, conducted in collaboration with Pharmion, with Vidaza and MGCD0103 were reported at the American Society of Hematology meeting. There was a 53% response rate demonstrated at a 90 mg dose with a median response time of less than two months in AML and MDS patients. There was also a 36% response rate for patients evaluated at all doses. “The biology showed the tumor cells change not only methylate DNA, but also change the histone acetylation pattern around DNA encoding the p15 gene,” adds Dr. Allen. These results support the basis of further testing the combination therapy in a larger, randomized, three-arm study.
Additional ongoing trials include a newly initiated study using Vidaza in combination with MGCD0103 in lymphoma and one using MGCD0103 alone for Hodgkins disease. “We have a significant interest in the data because the biology of DNA demethylation is not completely understood in humans. The question is whether we can improve the efficacy of the data if we give Vidaza more continuously,” explains Dr. Allen.
Since Vidaza is an injection, it is only given seven days out of 28. After the three-week drug holiday, the DNA is almost fully remethylated. The company has developed an oral form, now in Phase I trials. “We’re pushing that forward as fast as we can to really test the hypothesis that continuous DNA demethylation activity may translate to a better clinical outcome.”
Methylation-Based Diagnostics
DNA methylation biomarkers provide several advantages. These include the ability to detect tumor DNA shed into body fluids (blood plasma/urine); analysis in fixed, paraffin-embedded tissue; and quantification of unmethylated DNA of the same gene in a sample to serve as an internal reference. Epigenomics (www.epigenomics.com) is focused on developing techniques to identify methylation-based biomarkers for early cancer detection.
“A DNA methylation pattern closely reflects the gene-expression pattern of a cell,” states Achim Plum, Ph.D., vp, corporate communications. “You’ll find DNA methylation changes occur early in disease and remain throughout cancer development. It’s the biomarker you’d like to have.”
In order to discover novel biomarkers, the company codeveloped a differential methylation hybridization (DMH) microarray with Affymetrix for efficient methylation profiling of human samples. It covers 50,000 genomic fragments and requires only 1 mg of DNA, according to Epigenomics. “The microarray is an efficient discovery tool used as a starting point in all of our biomarker projects,” says Cathy Lofton-Day, Ph.D., vp, molecular biology, diagnostics.
The company’s lead project is a biomarker for colorectal cancer that measures DNA methylation of a region of the Septin-9 gene in blood plasma. A 2006 clinical study showed that measurement of the biomarker was able to detect 70% of all colorectal cancers at a specificity of 90%. “The technologies we have are sensitive enough to detect the DNA equivalent of two tumor cells in the bloodstream,” states Dr. Plum.
A clinical trial, planned for next year, will target adults ages 50 and over. The goal is to detect cancer in stage I or stage II. Abbott Molecular Diagnostics recently licensed the biomarker to develop and commercialize as a screening test on its m2000 system.
Researchers at the NCI are working to develop an improved DNA methylase inhibitor called zebularine. Designed in 1977 by Victor Marquez, Ph.D., chief of the laboratory of medicinal chemistry in the Center for Cancer Research, it was shown to inhibit cytidine deaminase. Then in 2003 it was shown to have additional activity. “Zebularine inhibits DNA methylation and reactivates some important tumor suppressors. If you discontinue dosage, however, you lose the expression of the gene,” explains Dr. Marquez.
His group is focused on enhancing the activity of zebularine. New toxicity studies will be conducted in dogs, since recent studies showed that monkeys have a liver enzyme (aldehyde oxidase) that inactivates the drug. Dogs lack this enzyme. Another study will combine zebularine with raloxifene (Evista®, Eli Lilly), used in postmenopausal women for prevention of bone loss and breast cancer. If the dog study shows acceptable toxicity, the researchers will administer both agents.
“It is hoped that either zebularine alone, or in combination with raloxifene, will prove to be safe,” Dr. Marquez states. In addition, the group is using prodrugs (pronucleotides of 2´-deoxyzebularine monophosphate) to chemically mask zebularine so the liver won’t metabolize it and to expedite its conversion to 2´-deoxy-triphosphate required for incorporation into DNA.
Oligodeoxynucleotides with zebularine are being incorporated into the recognition sequence of the methylase. “These smaller pieces of DNA that contain zebularine have shown potent inhibition in vitro of both bacterial and mammalian DNA methylase,” explains Dr. Marquez.
Zebularine is exciting because of its potential for preventing tumor progression, he adds. A recent study showed that zebularine is stable in water and nontoxic in mice. After drinking zebularin-treated water for seven weeks intestinal polyps were reduced in mice with a gene mutation that caused polyps. “This proves that the use of DNA methylation inhibitors offers a great deal of promise in cancer prevention.”
An Assay
After the launch of Illumina’s (www.illumina.com) GoldenGate Assay last year, there was high customer demand to scale-up the number of methylation sites for analysis, says Carsten Rosenow, Ph.D., senior marketing manager. “It was a natural progression from GoldenGate to really allow, for the first time, CpG site-specific methylation patterns on a whole-genome scale, which currently doesn’t exist on the market.”
The Infinium DNA Methylation Panel will include a 12-sample format (chip) with 60,000 oligonucleotides and will accommodate analysis of around 27,000 CpG sites per sample. This is a big increase from the GoldenGate, which allows analysis of 1,536 sites. The panel also requires only 1 mg of sample for automated processing and provides single CpG resolution, according to Dr. Rosenow.
Additional advantages to the assay include high reproducibility (as reported by beta test sites), and the methylation sites are “hand-selected based on their relevance and importance in the methylation field. We don’t have any restraints on the assay side, so we can pick whatever we feel is valuable for the researcher,” explains Dr. Rosenow.
The company anticipates availability early next year. “There are 30 million methylation sites, compared to genotyping with 10 million SNPs. We need to know we have a product that’s viable. The data and results of the research will define what can be done in the future.”
There’s no doubt regarding the current intense focus on methylation and cancer. There is growing interest, however, in other enzymes related to gene expression, such as histone-methyl transferases and sirtuins, not only in cancer but other diseases. “This will be a fascinating area of exploration for us over the next few years, as we bring histone-methyl transferase inhibitors into the clinic and combine it with other enzymes,” says Dr. Allen. This is just the beginning of the epigenomics story.