At the “BIO International Convention” in Chicago, there were ten breakout sessions on biomarkers; intellectual property, commercialization, reimbursement, drug development, and diagnostics were among the topics discussed. This explosive new technology niche, which requires the co-development of diagnostics and drugs, has tremendous potential to exploit breakthroughs in the molecular basis of disease. Genomics has transformed biology, and as biomarker research advances we will likely see a similar impact on drug development.
Cancer Markers on a Fast Track
Cancer and Alzheimer disease are two disease areas that can be significantly advanced by biomarker research. Cancer is a priority because targeted drugs like Herceptin (trastuzumab), Gleevec (imatinib), and Iressa (gefitinib) are already on the market and can benefit from a more targeted therapy utilizing biomarkers.
In a session chaired by Bernward Garthoff, Ph.D., chair of cluster biotechnology of the Federal State of North Rhine-Westphalia, Germany, the use of biomarkers in the diagnosis and treatment of breast cancer and other tumors was explored. Breast cancer was highlighted because it is believed that biomarkers could help reduce the use of chemotherapy after surgery. As an example, it was noted that in the treatment of node-negative breast cancer, research has shown that only 30% of patients benefit from chemotherapy treatment.
Three diagnostic tests for breast cancer that are currently available to predict tumor recurrence were discussed. American Diagnostica’s PAI-1/uPA tumor tissue biomarker test (Femtelle®) costs $250, is performed on fresh frozen tissue, and can indicate a low risk of recurrence.
Mammaprint from Agendia is a gene signature test from 70 related genes in a breast tumor. It is an FDA-approved oligonucleotide array that can also be performed on fresh frozen tissue. It costs $3,000 and can reportedly identify which early-stage breast cancer patients are at risk of distant recurrence following surgery.
The Oncotype DX breast cancer test from Genomic Health predicts chemotherapy response from ER+ tumors from FFPE tissue and recommends a treatment regimen. The test costs about $3,200.
A number of promising biomarkers were also discussed, including the ErbB2 receptor for Her2, which could be used to rule out chemotherapy for those individuals who do not respond to trastuzumab. Another noteworthy biomarker is P13k; it controls important cellular pathways and several drugs are currently in development. In addition, the KRAS mutation predicts response to EGFR receptor drugs such as cetuximab and is a good predictive biomarker.
A commercial perspective was given by Stephen Little, Ph.D., vp of personalized healthcare at Qiagen. Dr. Little asserted his belief that the timing is right for biomarkers for a variety of reasons but mostly because they can aid in diagnosis and more accurate diagnostics can result in improved treatment.
Although the concept of companion diagnostics (CDx) is compelling, progress has been slow as they require extensive collaboration between diagnostic and drug companies. There are a myriad of political, scientific, and regulatory issues stymieing progress, including whether drug firms can expend the time and money to develop biomarkers in parallel with drugs; developing a robust, reliable assay; dealing with the fact that many proposed markers are run in CLIA labs and are not approved diagnostic kits; and onerous regulatory pathways.
A major challenge for biomarker discovery and evolution to a diagnostic kit is proving clinical utility and the major investment in clinical trials. Dr. Little estimated that a Phase III drug trial can cost $100 million and a companion diagnostic for the same indication could cost as much as $20 million more. The available market for an FDA-approved companion diagnostic is likely to be between $20 and $50 million a year.
Biomarkers have the potential to be transformative, but the business model has not yet been proven. According to Dr. Little, three possible business models include: a biomarker gets accepted as part of the drug development process; payors drive the evolution of a product when it becomes clear that a subpopulation can be targeted for improved drug efficacy and cost effectiveness; and patients pursue tests where they can obtain test data for their own treatment.
Alzheimer disease (AD) could benefit immensely from a biomarker that could predict the course of the disease. In a session entitled “Development of translational biomarkers to accelerate clinical treatment for Alzheimer disease,” chaired by Jeffrey Cummings, M.D., professor at UCLA, the latest ideas and key hurdles from industry and academia were discussed.
Although amyloid plaques are associated with AD, their presence does not predict how the disease will progress. The most likely R&D focus in the near term is amyloid protein precursors (APP) such as A4. The enzymes alpha- and gamma-secretase can cleave APPs and are potential drug targets as they disrupt the pathway and prevent accumulation of beta amyloid peptides in the brain. Eli Lilly has a gamma-secretase inhibitor in advanced clinical trials.
Randall Bateman, M.D., assistant professor at Washington University, presented a neural network approach using fcMRI to track whole-brain connectivity that may be correlated with clinical outcomes. A biomarker for AD would most likely come out of studies involving genetic markers such as APOEe4, the AB precursors, and imaging with PET and MRI. PIB-PET is a new imaging biomarker that recently confirmed that a humanized anti-AB antibody, bapineuzamab, can shrink plaques.
Kristian Hveem, M.D., Ph.D., of Hunt Biosciences, chaired a session on biobanks. He talked about his work showing the potential of population biobanks for tracking samples collected over time with a project involving urine and serum samples from 210,000 individuals in Norway.
Michael Hehenberger, Ph.D., business develop executive at IBM Research, confirmed the need for biobanks for long-term global studies to track genomic data with epigenetic changes.
Early- to Late-Stage Drug Development
A session chaired by Lewis Bender, CEO of Interleukin Genetics, focused on the need for biomarker and genetic-profiling tools at an earlier stage in development. All inflammatory diseases can benefit from biomarkers, he said, and he gave examples such as TNF, interleukins, and fLQT, a cardiac marker.
Obstacles to development were provided by John Schultz, senior vp of licensing and strategic development at Clinical Data. Discovery and validation of biomarkers can be expensive, he said, and he then posed the question of “who owns the IP?” He also asked, “which assay platform should be used?” as that affects pricing and marketing. Finally, he said it is important to determine how the diagnostic can be marketed.
Linda Pullen, Ph.D., an independent consultant, asked why no significant biomarker deals had been announced recently. It was speculated that the value proposition is unknown and the predictive power of the biomarker (or CDx) is also unclear.
Hans Bostrom, Ph.D., director, scientific liaison licensing and external research Europe at Merck, reiterated that big pharma is not interested in more targets but instead validation of existing targets through translational medicine.
The field is mind-boggling and moving fast as Chris Roberts, director of molecular profiles and research informatics at Merck, aptly stated. The discussions at BIO were just the tip of the iceberg. The market for biomarker products will develop slowly but this is clearly a long-term phenomenon that will transform drug and diagnostic R&D and make personalized medicine a reality in 15–20 years.