In the raging war against disease, now more than ever before, the leading minds in drug discovery and development recognize that joining together in battle means more effective outcomes. In fact the wonder, and complexity, of the human body demands it.
Increasing demands for more and better disease diagnostics and therapeutics, combined with new technologies and growing regulatory requirements, have spurred the growth of biomarker applications in drug discovery and development, according to market researcher Global Industry Analysts (GIA). Biomarkers: A Global Strategic Business Report projects that the global biomarkers market will reach $34 billion by 2017. While oncology is a key area for biomarker development, cardiology is considered a major segment of growth. The report goes on to say that 90 percent of candidate drugs fail in clinical trials. Thus, biomarkers are important tools used for honing in on a specific population that can benefit from a given drug and for detecting the possibility of drug failure at an early stage.
There are thousands of potential biomarkers for particular cancers and other diseases, but each one needs to be validated before it can be turned into a useful diagnostic test. According to Food and Drug Administration (FDA) requirements issued in 2011, such tests must be used in the process of proving that a drug works for a particular disease on a subset of the population. Thus, pharmaceutical companies have to identify the subpopulations where a given test works, and researchers have to develop a diagnostic test as they figure out what a drug is doing.
"This is a new paradigm in business," explains Steve Hamilton, Ph.D., SLAS director of education. "The regulatory structure in diagnostics was well defined for a long time, but people in pharma did not have to think of it before. In the past, a researcher could develop an assay that would not live on beyond the research. These research-quality tools may not be good enough anymore. For SLAS members, it's a matter of being more upstream in the drug discovery process but knowing what happens downstream."
Representatives of instrumentation companies, drug companies and research institutions cite challenges of throughput, specificity, validation and translation as keys to making biomarkers more effective predictors of drug action. Various collaborations are in the works to make that effectiveness a reality.
Biomarkers, such as temperature to indicate infection and cholesterol count to indicate predisposition to heart disease, have been used in clinical settings for many years. According to John Robinson, Ph.D., senior director of collaborative research management at Abbott Diagnostics in Abbott Park, IL, "What's changed is the science. Today's drug development is more targeted because of an increased understanding of disease pathways and mechanisms that have come from breakthroughs such as the sequencing of the human genome and the analysis of the impact of specific genomic changes in various patient populations. Now companion biomarkers can be used to predict which targeted patient group will respond to a specific therapy."
Robinson defines a biomarker as "any biomolecule that identifies something that is abnormal in disease and has demonstrated value in a well-controlled clinical study that provides a physician with actionable information ultimately used at the point of diagnosis or treatment."
Kevin Halling, M.D., Ph.D., a molecular pathologist at the Mayo Clinic in Rochester, MN, says, "Biomarkers can be DNA, RNA, proteins or metabolites. These markers can provide useful clinical information such as a patient's prognosis or whether a patient is likely to respond to a particular drug. Our improved understanding of the genetic basis of cancer has resulted in finding logical targets for therapies and the development of ‘targeted therapies.'"
SLAS member Kevin Hrusovsky, president of the new Life Sciences and Technology Division at PerkinElmer in Waltham, MA, explains that biomarkers can be classified as diagnostic, prognostic, predictive and pharmacodynamic. "By targeting patients on the basis of distinct biological characteristics that can be objectively measured, enriched subpopulations that respond differently to treatment are identified, thereby generating more favorable risk-benefit profiles for the therapeutics than with a traditional empirical all-comers approach," he explains.
Hrusovsky adds that biomarkers may be used throughout the drug discovery and development process – early in development to elucidate the method of action of a drug and provide preliminary evidence of effectiveness, and later as indicators of safety and efficacy.
Richard Ellson, chief technical officer of Labcyte in Sunnyvale, CA, also an SLAS member, says maintaining sample integrity along the chain from patient to measurement is critical for effective application. "The handling process should ensure that biomarkers are not lost or damaged on the way to the detector."
According to Halling, biomarkers need to be validated for large numbers of samples and be done on the right population. People doing the studies have to understand clinical sensitivity and specificity.
Doug Ward, vice president of translational diagnostics at Ventana Medical Systems, a unit of the Roche Group in Tucson, AZ, believes that there is a need to detect and identify biomarkers using multiple technologies. Some of the technologies involved are mass spectrometry, microarrays and microfluidics, fluorescent in situ hybridization (FISH) and immunohistochemistry (IHC). There is also a need to use algorithms to "take scanning data and convert it into meaningful information."
According to SLAS member Mark Stolowitz, Ph.D., director of the proteomic core facility of the Canary Center at Stanford for Cancer Early Detection in Palo Alto, CA, "To qualify as a biomarker, there must be enough studies completed to statistically validate a biomolecule. We have to address diversity across the human genome. Historically, diagnostics were based on a single molecule, but now we're looking at multiple markers."
Hrusovsky adds, "Given that a single biomarker is rarely sufficiently informative and focusing on a single biomarker during clinical development is very risky since it may subsequently prove to be inadequate or misleading, it is critical to efficiently and economically gather as much biomarker information as possible throughout the course of drug development. Multiplexing is an ideal way of maximizing efficiency, cost-effectiveness and information gathering, and consequently is rapidly gaining momentum in both research and diagnostics."
Ward agrees, saying, "There are multiple ways that a tumor is created and multiple ways that a tumor grows. There might be more than one biomarker. Tissue is valuable, and you need technology to use it sparingly while looking at as many biomarkers as possible to get to the right therapy."
Fully optimized biomarker assays on the appropriate testing platform, along with the identification of well-characterized case and control clinical specimens, are critical to the validation of these assays, according to Robinson.
Continual advances in laboratory automation are a factor enabling these diagnostic processes. "Reducing reagent costs, maintaining specimen integrity and high throughput are all desirable and will contribute to successful and economical biomarker diagnostics. Miniaturization and automation are means to these ends," notes Ellson.
Hrusovsky believes that one of the biggest hurdles in drug development is the issue of translational medicine – the translation of early promising in vitro or in vivo results to clinically useful medicines. Biomarkers that are mechanistically related to the disease and/or drug increase the success of translational medicine, he says.
"The biggest impact on the diagnostics community is the fact that pharma companies need to partner to develop commercially available biomarker assays," explains Hamilton. "The FDA says that if a pharmaceutical company wants to get a drug approved for a subpopulation, it has to get a diagnostic test approved at the same time so the population can be identified. Whether it leads to more approvals remains to be seen."
According to Marta Hamilton, a consultant to pharmaceutical companies, to select patients for the clinical trial, a pharmaceutical company needs to have an approved companion biomarker or work with someone to do that while developing the drug. She explains, "To secure approval for the biomarker at the same time as the drug, you're adding a whole other parallel development track. Now you can't say on the drug package that it works on a specific group of patients without saying which test was used, so you have to develop biomarkers that work and assays that multiple people can run."
Ideally, a drug and its corresponding biomarker test should be co-developed and launched concurrently, Hrusovsky says. There are multiple reasons to support co-development: it is recommended by the FDA, as described in the recent draft guidelines; it maximizes the financial value of both the drug and the diagnostic tool; and it maximizes the clinical benefit to patients from the outset.
Steve Hamilton adds, "People are forced to talk to each other and interact with people they've never talked to before. It's a way to educate pharma about what goes into a diagnostic test and vice versa."
He thinks the need to collaborate will change the mindset of some researchers. "Researchers in early stages of drug discovery develop assays as tools for research. They may have to make the transition into diagnostics. Thus, they may get asked more questions. They could be under pressure to do a better job of documenting research and passing it on to new owners. Instead of discovering and developing tools solely for their own research, they would have to make the tools robust, documented and understood enough to hand off to someone else, especially if they are going to be commercialized. It could involve a whole change in perspective."
Some companies, such as Roche and Abbott, have both diagnostic and therapeutic arms, according to Marta Hamilton. Pharmaceutical companies need to decide whether to invest in building their own therapeutic departments or to partner with other companies.
Many biomarker collaborations have been in the news recently. Last summer, there was simultaneous FDA approval of Abbott's Vysis Anaplastic Lymphoma Kinase (ALK) Break Apart FISH Probe, which identifies ALK-positive non-small-cell lung cancer (NSCLC) patients who are candidates for Pfizer's approved NSCLC therapy, Xalkori (crizotinib), an oral ALK inhibitor for non-small-cell lung cancer.
In January, Ventana entered into a collaboration agreement with Pfizer and a license agreement with Cell Signaling Technology of Danvers, MA, to develop a fully automated and standardized immunohistochemistry (IHC) companion diagnostic test for ALK gene rearrangements. Ventana has also partnered with Syndax to develop a companion assay to select non small cell lung cancer patients who might benefit from treatment with erlotinib and entinostat.
Researchers can partner with instrumentation companies to develop assays that are efficient and effective. One example, says Ellson, is the partnership of Labcyte and the Canary Center at Stanford to combine "novel physics with novel biology" to develop protein-based diagnostic tools such as blood tests and molecular imaging approaches to detect and localize cancer at early stages, as well as engage in the translation of early cancer detection research into clinical trials and practice. Thanks to a $196,000 award from the National Cancer Institute (NCI) to fund a collaborative cancer biomarker validation program, Labcyte acoustic droplet technology is part of The Canary Center process.
Drugs directed at specific targets pinpointed by biomarkers can reduce development time for some drugs, according to Robinson. Ellson adds that something uniquely expressed eliminates false positives and makes early detection a reality, especially with certain kinds of cancer. Marta Hamilton cites good efficacy in small market segments as a result of biomarker technology.
Increased interest in specialized assays will eventually "bring diagnostics closer to the patient than further away," according to Steve Hamilton. "Now most newly developed biomarkers are only being tried at large, sophisticated labs. There's been a decades-long push to make assays more local. If it's not a point of care situation, at least it might be a ‘region of care.'" He believes that microfluidic technology is well suited to being closer to the patient when people who are doing the assays are not experienced laboratory researchers. "It's always a trend to simplify test procedures," possibly with "assays in a box."
"By lowering clinical risk with the chance of breakthrough efficacy results, increasing the speed to market and reducing overall development cost, biomarker-based strategies can present a value-creating, science-driven approach to drug development," Hrusovsky concludes.
April 3, 2012