OR WAIT null SECS
A look at the different testing methods and how the results impact drug development.
Over the past 30 years the cost of taking a drug from conception to FDA approval has increased from $138 million to $1 billion. R&D spending in the United States rose from $4 billion in 1985 to $55 billion in 2006.1 This increase has been caused by many factors, including increasing regulatory requirements, more technically demanding science and manufacturing procedures, and the introduction of innovative new drugs and molecular entities for complex disease targets.
(CHAD BAKER, GETTY IMAGES)
The number of successful product approvals for new drugs has decreased in the last three years, with a high rate of attrition. Current estimates suggest that only one in 10 Investigational Medicinal Products (IMPs) entering Phase I clinical trials actually makes it through to the end of the drug development process and obtains marketing approval.2 Since the cost to develop and market is so high and constantly increasing, it would be advantageous to reduce the number of unsuccessful drugs at an early stage of development before embarking on expensive trials.
A potential solution to the high number of failures in clinical trials could be the use of biomarkers. The U.S. National Institutes of Health defines a biomarker as "a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes or pharmacologic responses to a therapeutic agent."
Identifying, studying, and quantifying biomarkers can increase our understanding of drug action, its metabolism in the body, efficacy, and safety. All of which can collectively act as key decision making tools when evaluating the performance of an IMP in the clinical environment.
The scope of what constitutes a biomarker is in itself wide ranging. The nomenclature of biomarker can be applied both at the gene level or post-gene expression as well as analytes, which are indicative of a physiological change post drug administration. Therefore, with such a wide range it is easy to understand the impact that biomarker discovery, identification, and subsequent monitoring could play in the life cycle of a potential therapeutic agent.
Novel biomarkers can be used as research tools during early drug development to select candidates with therapeutic potential. Patients can be screened at both the genetic and/or protein level to indicate or predict their response (therapeutic or not) to administration of an IMP and to monitor any potential toxicity caused by a drug, which could lead to an adverse drug reaction.
Qualifying a biomarker—that is linking it to a clinical endpoint—is essential if the biomarker is to be used to measure a clinical outcome. And by linking biomarkers at the different levels, an increasing and ultimately enhanced understanding of the receptiveness of the patient to respond either well or adversely can more rapidly be qualified. Genetic and/or metabolic expression biomarkers can classify patient responses to treatments, but these must be correlated with phenotypic endpoints or surrogate endpoints.
Biomarkers have become more important in clinical practice because the tools and technologies available to researchers have facilitated better understanding of:
It is the continual advancement of new technologies that can measure these markers that is resulting in the ability to make go/no-go decisions earlier in the development process. This article will describe some of these technologies and the methods used to measure biomarkers.
With the completion of the Human Genome Project, the ability to generate genome-wide association data, next generation sequencing, and gene expression profiles has helped researchers in the study, understanding, and explanation of disease mechanisms and epidemiology. By looking at the data available through these technologies, biomarkers in monogenic and polygenic diseases have been identified.
In the study of human genetics and pharmacogenomics, it was historically thought that there would be one "wild-type" and one "defective allele." With the passage of time, this has changed to a position whereby most researchers think that there will be multiple Single Nucleotide Polymorphisms (SNPs) in one or a few genes that will be associated with any complex disease or drug response.
Technology has advanced to such a level that these SNPs can now be identified, and the downstream consequences of them in relation to both disease epidemiology and reactions to drugs can be assessed and measured.
Within the clinical trial arena the extraction and purification of DNA from patients is currently being undertaken. This provides clinical researchers the potential for subsequent genotyping experiments on multiplexed arrays such as the Illumina IScan. These genome-wide association studies form the initial point of biomarker discovery and identification, providing a high-throughput solution for screening up to over 1 million potential markers across the trial population. Complex bioinformatics can then be used to narrow down specific SNPs of interest right down to a singleplex approach, looking specifically at SNPs within candidate genes.
The launch of the DETPlus panel by Affymetrix provides a high-throughput screen that captures more than 90% of the current ADME Core markers as defined by the PharmaADME group. This panel provides coverage of a wide range of genetic variations, including common and rare SNPs, which enables the identification of significant, new biomarker associations.
As an example, warfarin, a well-known drug prescribed to prevent heart attacks and strokes, is also known to be one of the top drugs for causing severe adverse events. In a recent warfarin study using the DMET Panel, researchers discovered a new variant in CYP4F2 (a drug metabolizing enzyme) that explained 8% of dosing variability in select patient populations.3 This new biomarker has subsequently gone on to being tested in a Phase III prospective trial.
Using this information in conjunction with the raft of other clinical data available provides a level of information that could determine the efficacy and safety of a drug during a trial or that could be used as a screening diagnostic to prevent the administration of products to recipients who will have an adverse effect or demonstrate no therapeutic value.
In either case, the decision pathway required to get a drug to market is greatly enhanced as potentially toxic products could be quickly identified and dropped, or conversely, more precise clinical data could be accrued to support authorization of a product for a very specific genotype.
Proteomics is the scientific discipline that studies proteins and their association with diseases by means of their altered levels of expression and/or post-translational modification between control and disease states. It enables correlations to be drawn between the range of proteins produced by a cell or tissue and the initiation or progression of a disease state and the effect of therapy. Proteomics can be used for the discovery of novel drug targets and in the discovery and validation of diagnostic and prognostic disease biomarkers.
An example of a disease where proteomic research of biomarkers is being used is the diagnosis of ovarian cancer. Researchers collected serum from 50 ovarian cancer patients and 50 controls and used a computer algorithm to search for the protein patterns that distinguished cancer from noncancer. When they tested this pattern with a set of blinded serum samples, the test pattern correctly identified all 50 patients with cancer, and was able to discriminate them from 63 out of 66 patients who were unaffected or had benign disease.4
Some of the analytical methods used in Proteomic research are gel electrophoresis and mass-spectrometry.
Enzyme Linked Immunosorbent Assay (ELISA) is probably the most well-established quantitative and/or qualitative biomarker testing method. There are many test kits on the market designed to measure markers specifically associated with diseases (e.g., cardiovascular, oncology, inflammatory). Other biomarkers may be indicators of infection, such as skeletal, muscle and acute phase proteins, and toxicity (e.g., liver toxicity) or immunotoxicity. They usually have a "sandwich" format where biomarker-specific antibodies are coated on a microtitre plate, the sample is added, then after wash steps another specific labeled antibody is added and the bound complex detected by color/fluorescence development.
Multiplex platforms where several markers can be measured in one sample are becoming more popular due to the reduction in cost, time, and sample volume required. They often use fluorescent beads to which antibodies are attached. For example, several cytokines can be measured using the Luminex xMAP (multi-analyte profiling) platform by covalently linking each capture antibody to Luminex beads, enabling measurement of each cytokine level in plasma.
Flow cytometry is also used in biomarker detection by employing specific fluorescence-tagged antibodies that bind to antigens expressed on the target cells, which can then be detected in the cytometer. The fundamental mechanism of biomarker detection is the same as ELISA (antigen–antibody affinity), however, the flow cytometry platform can be used to examine the proteins that are expressed by the different populations of cells.
Metabolic profiling involves the characterization of the metabolites (usually small molecules) found in an organism, which are the end products of gene expression—it is therefore closely linked with genomics and proteomics. Methods of metabolite analysis usually involve:
Radioimmunoassay (RIA) is a well-established method that can be used to measure metabolites (e.g., hormones such as insulin in biological samples).
It is based on the antigen–antibody reaction in which tracer amounts of the radio-labeled antigen competes with the endogenous antigen for limited binding sites of the specific antibody against the same antigen. Another example of the application of RIA in biomarker research has been its use to quantitate biomarkers of breast cancer.5
The study of biomarkers, whether genomic, proteomic or metabolomic, has the potential to revolutionize clinical research. The complexity of the systems and advanced technologies mean that data are being collected at a fast pace.
Efforts to qualify/validate, interpret, and integrate biomarker data with clinical endpoints will ultimately provide a better understanding of biomarkers in disease processes and will benefit clinical diagnosis and prognosis.
Emma Waite is manager, biopharmaceutical laboratory, for Tepnel Research Products & Services, Appleton Place, Appleton Parkway, Livingston, West Lothian EH54 7EZ, United Kingdom, www.tepnel.com.
1. Business Insights Market Report, Drug Approval Trends at the FDA and EMEA: Process Improvements, Heightened Scrutiny and Industry Response (April 2008).
2. Tufts Center for the Study of Drug Development, News, "Despite More Cancer Drugs in R&D, Overall U.S. Approval Rate is 8%," http://csdd.tufts.edu/NewsEvents/NewsArticle.asp?newsid=83.
3. M.D. Caldwell, T. Awad, J.A. Johnson, B.F. Gage et al., "CYP4F2 Genetic Variant Alters Required Warfarin Dose," Blood, 111 (8), 4106-4112 (2008).
4. E.F. Petricoin, A.M. Ardekani, B.A. Hitt, P.F. Levine et al., "Use of Proteomic Patterns in Serum to Identify Ovarian Cancer," Lancet, 369, 572-577 (2002).
5. N. Fujino, Y. Haga, K. Sakamoto, H. Egami et al., "Clinical Evaluation of an Immunoradiometric Assay for CA 15-3 Antigen Associated with Human Mammary Carcinomas: Comparison with Carcinoembryonic Antigen," Japanese Journal of Clinical Oncology, 46, 335-334 (1986).