Send to printer »

Feature Articles : Apr 1, 2009 (Vol. 29, No. 7)

MDx Will Drive Personalized Medicine

Prediction Expected to Hold Only if the Novel Suite of Technologies Can Save Healthcare $$$
  • Vicki Glaser

Global trends in molecular diagnostics was a key topic of discussion for an expert business panel at the recent “CHI Molecular Medicine Tri-Conference” in San Francisco.

The panel predicted that some of the approximately $50 billion of President Obama’s stimulus package will be allocated to electronic medical record (EMR) opportunities and will directly impact the diagnostics market. By expanding access to a patient’s records and making them portable, EMRs will foster the use of point-of-care diagnostic tests.

Harry Glorikian, a partner at Scientia Advisors, spoke of the growing recognition that diagnostics will be a significant enabler of reduced healthcare costs. He identified several major global drivers of an expanding market for in vitro and molecular diagnostics: an aging population; an increase in chronic diseases (as therapeutic regimens improve); the availability of high-value diagnostics; and growth in decentralized healthcare (e.g., China plans to establish about 60,000 healthcare clinics by 2012).

Glorikian described increasing scientific innovation in Eastern countries, where products can be made for a fraction of the cost in the U.S., with similar quality, and projected a shift in the flow of products from Asia to the West in the near future.

Diagnostics offers the potential for shorter timelines to market and profitability, noted a venture capital panel during a discussion entitled, “How Is Personalized Medicine Moving Forward?”

Although, historically, diagnostics were less heavily regulated than drugs, that equation appears to be changing for molecular diagnostics, at least within the U.S. FDA. The panelists advised that when a company develops a companion diagnostic it has to consider whether the product can reap a profit, but it also has to assess the cost/benefit for all concerned, including the patients (Will insurance cover it?), the physicians (Will they be reimbursed?), and the healthcare system (Will the test reduce the overall cost of care by enabling the use of targeted, more effective treatments or identifying early on therapies not likely to be effective?).

New Twists in Point-of-Care

To bring a new biomarker-based diagnostic test to market takes an estimated three years and $45–50 million to complete outcome studies. From first investment to revenue production, this estimate rises to $70–100 million.

A medical expert panel that discussed the “Future of Point-of-Care” pointed out  that with the emergence of “quick clinics” in grocery stores and shopping centers, for example, it is clear that people are willing to pay out-of-pocket to seek advice for managing their healthcare outside traditional treatment settings.

Quantitative diagnostic tests are facilitating a shift in trust, from relying on a physician to deliver and interpret test results, to a point-of-care model, noted the panelists. This trend will not likely impact test volumes at central laboratories, which are enjoying increasing revenues at a rate of about 5% per year, largely due to the aging of the population.

Technology that enables the development of robust, easy-to-use test kits will drive growth in the home diagnostic testing market, as patients will begin to manage their own diagnoses and treatments, bypassing the physician’s office.

What did the participants identify as the most significant current barrier to the development of point-of-care diagnostic tests? The issue of reimbursement, which will largely determine physician adoption. In-office testing would save physicians time and enable a test-and-treat model of care in a single visit. These products will give physicians faster and better answers, but they must be affordable.

Setting an Example

Genomic Health’s Oncotype DX® breast cancer assay is a prime example of a clinically relevant biomarker-based molecular diagnostic test. The 21-gene quantitative real-time PCR (RT-qPCR) assay produces a Recurrence Score® (RS) that guides patients and clinicians in the decision of whether to pursue treatment with adjuvant chemotherapy.

While many women with breast cancer will be offered chemotherapy, only a select group of estrogen receptor-positive (ER+), node-negative patients are likely to benefit from this toxic and debilitating therapy. As a result, some women who might benefit are undertreated, while many are overtreated.

Steve Shak, M.D., CMO at Genomic Health, reported that utilization of Oncotype DX is increasing, with more than 85,000 tests delivered since its launch in 2004, and the test is now reimbursed by Medicare and all major insurance payers. Dr. Shak emphasized the need for standardization in the molecular diagnostics arena, particularly in technology implementation, and the importance of quality control to ensure consistent and reliable results with biomarker-based diagnostics.

In a presentation entitled, “The BCR-ABL qRT-PCR Assay: Status of a Molecular Diagnostic that Is a Current Standard of Care,” J. Milburn Jessup, M.D., chief of the diagnostics evaluation branch, cancer diagnosis program at the National Cancer Institute, described how a qPCR assay to detect the BCR-ABL fusion gene—the cause of chronic myelogenous leukemia (CML), which accounts for 20% of leukemias—became the basis for a personalized medicine treatment paradigm with the drug imatinib (Gleevec), a targeted therapy that can reduce BCR-ABL by 3 logs from baseline and yield durable remissions.

Issues Remain

The problem with this model, according to Dr. Jessup, is that, while more than 150 hospitals in the U.S. currently perform the qRT-PCR assay for BCR-ABL, and it has proven quite sensitive for molecular monitoring of CML. “None of these hospital laboratories can directly compare their results to any other.” He proposes that daily use of a calibrator/control for the assay could “harmonize results between labs, but a control is not readily available now.”

Dr. Jessup said that, “each lab becomes a silo of efficiency,” and “reproducibility within labs is excellent”; however, studies have shown considerable differences in the copy number of BCR-ABL between laboratories in samples taken at the time of diagnosis and after treatment. “The CLIA process reinforces precision,” but not as much accuracy, added Dr. Jessup.

He described global harmonization efforts under way to standardize the multitude of home-brew qRT-PCR assays, including an International Standardization Project in Australia and efforts to develop reference standards for use across Europe. In the U.S., the NCI will facilitate development of standards and reagents for the BCR-ABL assay.

Leveraging MicroRNA

Rosetta Genomics is applying its miRNA discovery process to the identification of diagnostic biomarkers for cancer, quantifying miRNA levels in a range of tissues and body fluids and creating disease-linked miRNA signatures.

Late last year the company introduced its first three miRNA-based cancer diagnostic tests. miRview™ squamous can differentiate between samples of squamous versus non-squamous non-small cell lung cancer (NSCLC) to stratify patients for treatment.

The company’s miRview mets product can identify the tissue-of-origin for 25 metastatic tumor types with a specificity of 97%. Knowledge of the tissue of origin of metastatic disease can guide therapeutic decision-making. For differentiating malignant mesothelioma or metastatic cancer affecting lung and pleural tissue from adenocarcinoma the company developed miRview meso.

In 2010, Rosetta plans to commercialize miRscreen™ Colon, a serum-based miRNA biomarker test that can differentiate patients with and without colorectal cancer. The test, with 91% sensitivity and 72% specificity, can be used to screen for colon cancer in blood samples to identify high-risk patients who should undergo colonoscopy.

In Silico Approach

Megan Laurance, Ph.D., senior scientist at Ingenuity Systems, described the company’s computational workflow for linking overexpression of miRNAs to cancer pathways to identify candidate pools of cancer-related genes. By layering and integrating molecular profiles, this in silico approach searches for profiles that are coregulated to affect cell function and disease phenotype.

A study of gene profiles in ovarian cancer to search for inter-related pathways found an association between differential miRNA expression in neoplastic versus non-neoplastic samples and deregulation of the DNA damage repair response and of pro-angiogenesis genes.

Yuriy Gusev, Ph.D., assistant professor at the University of Oklahoma, described computational methods his group is developing to identify genes that are collectively targeted by aberrantly expressed miRNAs in human cancer. In colon cancer, for example, Dr. Gusev presented a molecular signature of 20 overexpressed miRNAs.

Computational analyses point to miRNA targets that establish tissue-specific biomarkers of cancer and identify cancer-specific pathways enriched for these miRNA targets.

Building on Biomarkers

PDZ domains are highly conserved protein globules that serve a structural function in cells, linking signaling cascades, for example. The human genome codes for about 300 PDZ proteins, each 80–90 amino acids long. Arbor Vita is developing PDZ-based diagnostics and therapeutics, applying its PDZ discovery platform to three target areas: neuroischemia, oncology (cervical cancer), and infectious disease (seasonal influenza).

The technology uses modeling to develop a library of PDZ interactions. In cervical cancer, for example, the cancer-causing human papilloma virus (HPV) targets a PDZ inside cervical cells, triggering cancerous changes. Notably, only high-risk HPV (the form most likely to cause cancer) specifically interacts with a PDZ.

Whereas a test to detect HPV can only tell you if a virus is present, Arbor Vita has developed a test that detects the PDZ ligand sequence that is present only in high-risk HPV-E6. “E6 oncoprotein activity depends on PDZ binding,” explained Linda McAllister, M.D., Ph.D., executive vp of diagnostics and CMO at Arbor Vita.

A positive test indicates increased cancer risk and can guide treatment decisions. The test “detects oncogenic activity of the virus” to help “triage women to surgical intervention,” said Dr. McAllister. The company plans to initiate a clinical trial in the third quarter of this year.

A single molecule counting approach to early disease detection developed by Singulex, was the focus of a talk by John Todd, Ph.D., vp of R&D. This technology is based on the Erenna™ paramagnetic microparticle (MP)-based sandwich immunoassay, targets low abundance proteins in the bloodstream and can detect protein biomarkers at pg/mL concentrations.

The company is evaluating the utility of cardiac troponin-I as a biomarker for acute myocardial infarction, and of plasma cytokines in inflammation. An Erenna assay against VEGF (a biomarker for angiogenesis) in the plasma of healthy women compared to women with breast cancer showed statistically significant differences in VEGF concentrations with <1 pg/mL sensitivity using 10 µL of plasma. Dr. Todd described the company’s programs to develop immunoassays capable of detecting small changes from a healthy baseline in plasma protein biomarkers for early detection of breast cancer and recurrent prostate cancer.

High Throughput Genomics (HTG) presented its qNPA™ technology based on a nuclease protection assay for predictive multiplexed gene expression diagnostics. The qNPA platform offers several advantages, according to Bruce Seligmann, Ph.D., SCO and vp R&D at HTG: whole assay CV’s <10%; detection of <1.2-fold changes; single copy gene sensitivity at 1,000 cells; multiplexed mRNA and miRNA measures; and suitability for use with fresh tissue samples, archived FFPE tissue, or fixed and sorted cells.

Researchers from the University of Illinois at Urbana-Champaign have developed a colorimetric “dipstick” test for molecular diagnosis of a range of targets based on functional DNA nanotechnology. Yi Lu, Ph.D., professor of chemistry, described the detection of metabolites—such as metal ions and organic molecules—as “a new approach for diagnostics.”

Dr. Lu’s group chose to utilize functional DNA because it is biocompatible and cost effective, stable under harsh conditions, generally nonimmunogenic, can be denatured and renatured repeatedly, and can be readily modified to generate a signal. Additionally, DNA can recognize a variety of molecular targets, both large and small.

Dr. Lu presented a general method to convert DNAzymes into fluorescent sensors using catalytic molecular beacons and described the design of a simple colorimetric biosensor that functions like pH paper. The method incorporates both positive selection for target detection and negative selection to reduce unwanted activity.

The ability to detect cancerous lesions at an earlier and smaller stage will drive changes in clinical care. James Mulshine, M.D., professor and director, Rush Translational Sciences Consortium at Rush University Medical Center in Chicago, discussed the impact of high resolution computed tomography (CT) on lung cancer management.

“CT resolution has been doubling every two years for more than a decade,” said Dr. Mulshine, and “image-processing capabilities are rapidly evolving.” Higher resolution CT imaging is allowing clinicians to measure tumor volume with greater accuracy and to monitor changes in tumor volume over time and in response to drug treatment.