September 15, 2007 (Vol. 27, No. 16)

miRNA, Imaging Agents, and the Burgeoning Field of Epigenomics Lead the Charge

Later this month, researchers will converge at the “AACR Molecular Diagnostics in Cancer Therapeutic Development” conference in Atlanta to review the newest advances in cancer biomarkers, including developments in the emerging field of epigenomics, which is expected to burgeon within the next several years as genomic data accumulates.

New assays for probing stored tissue samples, a novel agent for detecting angiogenesis in situ and in vivo, analysis of DNA-methylation patterns, miRNA characterization, and mitochondrial mutation patterns are also on the agenda for discussion.

Genetic profiling of malignant tissue is already lending more power to clinical care, as in the case of the recently characterized 2+Edel marker for prostate cancer that does PSA one better by determining the level of tumor aggression and hence allowing a more intelligent response to disease management.

Zvi Bentwich, M.D., chief scientist at Rosetta Genomics ( and professor of medicine at the Hebrew University Hadassah Medical School, will be offering a first look at a new technique that will hopefully give rise to a diagnostic assay based on a small sample of miRNA, which will be able to detect specific cancer signatures.

Dr. Bentwich explains that “50 percent or more of miRNA sequences are located in the region of cancer-promoting genes, and the remarkable tissue specificity of miRNA has led Rosetta to explore and develop assays that would enable one to diagnose the tissue of origin, particularly as a number of cancers present as cancer of unknown primary (CUP).”

Rosetta says it has developed a highly sensitive and powerful assay to identify miRNA patterns as a diagnostic of particular tumors. The technique is currently based on paraffin-embedded fixed tissues, but, Dr. Bentwich says, “we know by now that miRNA can be found in serum and other bodily fluids; looking into the future, we are indeed looking at developing similar diagnostic tests that are miRNA-based but can be derived from urine and blood, for example. The two immediate candidates for our miRNA-based assays are CUP and distinguishing between adeno- and squamous-cell carcinomas inside of the lung.”

“In the broader context, we would like to diagnose not just the presence of cancer but also to offer prognostic markers,” he adds. “And in terms of miRNA function, inhibiting the expression of miRNA may be an approach to treating cancer. It certainly opens the door to new modes of therapy.”

Investigational PET Agent

Pamela Cohen, M.D., global oncology therapeutic area head of GE Healthcare (, will discuss a new investigational PET agent, [18F]-AH111585, which binds to the alpha-V-beta-3 integrin receptor. This new entity is designed to be an imaging biomarker for tumor-related angiogenesis and is currently in early Phase I/II development.

The alpha-V-beta-3 integrin receptor is preferentially expressed in endothelial cells unique to tumor angiogenesis. According to Dr. Cohen, imaging angiogenesis using PET with [18F]-AH111585 may have distinct advantages over dynamic contrast-enhanced MRI (DCE-MRI) in that it can be quantitative and could be used in certain regions of the body not easily assessable by DCE-MRI.

“Angiogenesis is a sine qua non during early tumorigenesis. Given the high costs of antiangiogenic treatments, there is great demand for a biomarker that could be used to personalize treatment choices by identifying those patients who will benefit most from antiangiogenic cancer therapy.”

[18F]-AH111585 will be tested both for its ability to be used diagnostically, for pre-selection of patients, and for monitoring of early therapy response after diagnosis and initial treatment. The hope is that PET imaging biomarkers such as [18F]-AH111585 will be useful not just in management of patients receiving currently approved antiangiogenic therapy but also may be helpful in speeding development of the newer antiangiogenic drugs currently in development.

Dr. Cohen has great optimism for the utility of [18F]-AH111585 PET imaging and notes that, in the first Phase I trial, it identified an asymptomatic brain tumor in an undiagnosed volunteer participant of the study.

Resequencing Array-based Assays

Joseph A. Califano Jr., M.D., otolaryngologist, surgeon, and associate professor at Johns Hopkins University School of Medicine, has had nearly a decade of experience in mutation detection and has recently devised a resequencing array-based assay for detecting tumor-specific mitochondrial mutations in head-and-neck cancers.

Based on small quantities (less than 50 ng) of DNA from frozen tissue samples, the method employs the Affymetrix ( MitoChip v 2.0 mitochondrial genome resequencing array, “which sequences the entire 16-kb mitochondrial genome at least twice,” according to Dr. Califano.

“We developed a simple algorithm to look at the secondary signal, representing hybridization signals from minor populations of mutant DNA. We have also demonstrated proof-of-principle in mixed samples of head-and-neck (tumor) cell lines and have found we can reliably detect mutations in a 1:200 solution.” In the real world, this translates to the much-desired ability to detect cancers in salivary rinses in lieu of invasive biopsy.

Thus far, Dr. Califano’s technique has been shown to detect malignancies in 10 of 13 known cancer patients, “including tumor-specific mutations. The advantage of this technique is the detection of rare mutant signals. There are seven or eight assays out there that perform the same task, but they must be optimized for individual mutations, require research-grade conditions, and incorporate finicky fluorescence-based assays that can take weeks to complete. This technique takes advantage of standard off-the-shelf arrays, standardized array cores, standard software, and a simple algorithm.

“It is widely applicable, easy to use, requires no special expertise, and is objective and quantitative, in contrast to the subjective interpretation of gels or shifts. In addition, mitochondrial mutations, found in virtually all solid tumors, are often many, thus an individual tumor may yield multiple opportunities for detection.”

The technique also has a potential application for monitoring patients after cancer surgery and applicability for detecting, in bodily fluids, mutations specific to other tumor types, Dr. Califano notes.

Archival Tissue

Malek Faham, M.D., Ph.D., vp, oncology technology research, Affymetrix (, will discuss a novel method for determining DNA copy-number in formaldehyde-fixed, paraffin-embedded (FFPE) tumor samples, opening the door to examination of archival tissue.

The technique, which uses molecular inversion probes, is not new but uses “solution probes as distinguished from array probes,” says Dr. Faham. The product, not yet commercialized, “is flexible and capable of probing across the genome panel with between 100 and 50,000 probes,” he notes. “We have shown that the technique works well in a variety of FFPE samples where all we want is a 40 bp footprint for the two homology ends.”

This technique uses probes that can be circularized, does not require preamplification, uses less than 50 ng of DNA per assay, and will allow researchers to access the large historical archive of fixed-tissue samples available in tumor registries, thus acting as a survey tool for identifying biomarkers, Dr. Faham says, adding that the technique offers “good precision in copy number and genotype measurement, and good dynamic range (0–60 copies).”


Mathias Ehrich, M.D., Ph.D., group leader, cancer biology, Sequenom (, will be addressing epigenetic biomarker discovery and validation and will be describing Sequenom’s new research tool for high-throughput analysis of DNA-methylation patterns.

Dr. Ehrich allows that “while epigenomics is still a pretty young field, the techniques have come a long way. Two main projects are under way. The first is an analysis of a panel of 200 genes in about 250 patients with adult myelogenous leukemia where we have been able to show a pattern predictive of patient survival.”

In the second project, Sequenom has analyzed “a large panel of 500 genes in different cancer cell lines (lung, colon, ovarian, and kidney) and found that around 90 percent of differentially methylated genes are confirmed in clinical samples. Of 50 genes differentially methylated in colon cancer, for example, 42 genes were confirmed in clinical specimens.

“The next step would be to apply this to an earlier diagnostic setting, obtaining DNA from stool and blood, to see if there are detectable changes in more biologically accessible tissue and then move to an ultrasensitive detection method.” Dr. Ehrich says that the forecast for epigenomic drugs, the logical successor to elucidation of epigenetic patterns, is quite robust and is expected to increase to roughly $2.6 billion in 2012, a large proportion of the total projection of $4 billion for epigenomics.

Previous articleVaxGen Slashes More Jobs to Preserve Cash
Next articleSontra Medical and Echo Therapeutics Merge