Information-based medicine maps a model for targeted drug therapies.
Blockbuster drugs—those with peak annual global sales of $1 billion and prescribed for general population use—can cost $800 million to $2.0 billion to develop and take more than a decade to advance from the lab to the pharmacy.1 These drugs are typically effective in only 40% to 60% of the patient population.2
Pharma is well aware of the blockbuster model’s limitations. The industry’s soaring spending on research and development (R&D) has not translated into higher success rates in reaching the market; in 2005, the US Food and Drug Administration (FDA) approved fewer new molecular entities (NMEs) than at any time in the last 20 years.3 Need for an alternative approach to R&D is obvious.
While the blockbuster model has served Pharma well, it is unable to build in a higher level of predictability in developing safe and effective treatments for patients with specific disease subtypes.4 By augmenting the blockbuster model with an information-based medicine approach, the pharmaceutical industry can shift its focus to R&D, supporting targeted treatments and personalized medicine.
Biomarkers and pharmacogenomics play a pivotal role in understanding patient differences and help transition the industry toward a business model that focuses on more targeted treatments. This approach uses diverse clinical and biomedical data—including genomic, proteomic, and imaging data—to determine if certain patients are more likely to benefit from certain treatments and therefore target the treatment accordingly.
Using more reliable information can change how patients are selected for clinical trials. Consider the confidence that comes with using cohorts of patients with similar genetic and phenotypic characteristics who are more likely to benefit from a treatment and less likely to experience toxicity. Approving treatments should take less time as knowledge advances with each phase of development. It also has great implications for changing the treatments that physicians select for their patients.
Biomarkers, adaptive trials, and modeling and simulation (M&S) are the cornerstones of information-based medicine. The model is also supported by innovations to accelerate how compounds are selected for clinical trials, new approaches to registering a treatment, and improving the manufacturing process. Technology helps harness this knowledge. This article focuses on the first set of drivers: biomarkers, adaptive trials, and modeling and simulation and the insights they bring to a more patient-focused development process.
Drug companies need to weave together traditional clinical information with genomics, proteomics, and medical imaging5 to gain insight into disease. However, scientific and technological advances can be inhibited by organizational obstacles such as the cultural roadblock of resistance to data-sharing with internal and external partners. An information-based model relies on new competencies for understanding data complexities that drive innovation and requires new information technology (IT) strategies to support R&D.
FDA has pressured the industry to shift toward personalized medicine that targets diagnostics and treatment based on patient-specific differences. FDA has partnered with the health care and the bio-pharmaceutical industries to modernize the regulatory process.6, 7
Biomarkers on target
Thanks to the mapping of the human genome, more information is available to optimize patients selected for studies. These data augment traditional clinical testing with greater precision to determine dosing, treatment response, and safety. Biomarkers contribute to developing new therapies in significant ways by identifying groups at risk for toxicity as well as groups who are more likely to respond; thereby addressing expensive, slow-progressing diseases.
While some organizations are concerned that a personalized medicine approach that targets treatment for limited patient groups decreases profits and increases research costs, others recognize that this direction has already created value beyond costs. They are building these capabilities into new R&D models and tactics. For example, Genetech’s Herceptin (Trastuzumab) is considered an effective, targeted treatment for breast cancer. Such treatments could actually increase both the medical and economic success of a therapeutic.8
Biomarkers can help diagnose disease, track a biological effect, and assess risk (such as cholesterol and blood pressure). They also are used for patient selection based on safety factors, patient stratification based on response factors, and monitoring treatment. Biomarkers also have a prognostic role.
By incorporating biomarkers into clinical trials, drug companies can generate more scientific precision and create knowledge feedback loops into its development strategy. These steps spawn innovation, which in turn can address unmet medical needs or give a company a competitive advantage.
Figure 1 shows how biomarkers, to connect bench to bedside and create a continuous feedback loop for innovation. When the right data are available, researchers can test hypotheses, help understand the mechanism of disease and its treatment, and simulate different clinical situations before testing—all without involving patients and the associated time or cost.
click to enlarge
Figure 2: Adaptive trials allow researchers to modify protocols, omit ineffective study treatments, or add more study participants who may respond to treatment. (Source: IBM)
Adapting to data
An adaptive trial is a clinical trial that is modified based on data analysis conducted while the study is ongoing. This approach can alter the course of the same trial without compromising its statistical validity. Figure 2 shows how adaptive trials allow researchers to redirect participants to modify protocols, omit study treatments that appear ineffective, or add more study participants who may respond to study treatment. Robust data management, rapid data-capture, and data-cleaning are prerequisites.
Some experts believe these trials can reduce the duration, cost, and size of clinical trials. Interest in such techniques is growing rapidly among regulators and contract research organizations (CRO). In July, 2006, FDA reported that each of its review divisions had “received at least one adaptive trial submission in the past year.” The agency plans to release guidance on adaptive trials in FY2008.9 Similarly, Quintiles Transnational, a CRO based in Research Triangle Park, N.C., has devised at least 10 adaptive designs for reassessing sample sizes, combining dose selection with confirmatory Phase 3 testing, and using Bayesian principles to incorporate data from earlier trials or earlier stages of the same trial.10
Modeling the trial
M&S, which can help reduce drug development costs by up to 50%,11 creates a feedback loop into research that produces more relevant compounds to feed into the development cycle.
Some pharmaceutical companies have already received regulatory approval for drugs using M&S. Hoffmann-La Roche AG, Basel, Switzerland received approval for PEGASYS (Peginterferon alfa-2a), a combination drug for the treatment of Hepatitis C. M&S was used to determine the proper dose for a subpopulation of patients that took into account a variety of factors, including the genotype of the virus and the weight of the patient. Pfizer, New York, used M&S to augment their traditional studies for Neurontin (gabapentin), which was approved for a variety of neuropathic pain conditions, including post-herpetic neuralgia.11
Many companies struggle with M&S tools that rely on data that resides in different databases and therefore, do not communicate to each other. As these organizations adopt more flexible integration approaches, like Service Oriented Architecture (SOA) and embrace standard data formats and metadata to enhance its ability to manage structured (e.g., data in databases) and unstructured data (e.g., documents), they will bring more efficiency into clinical development processes.
Augmented with pharmacogenomic data, M&S can uncover important safety trends. FDA recognizes M&S in their Critical Path initiatives12 as an important part of the drug approval process. In addition, an information-based medicine model can help incorporate basic research from publicly-funded academic medical research into early identification of safe and effective compounds. It can integrate knowledge and feedback from clinical care of patients into the development of new medical treatments and diagnostics.
While the blockbuster model will continue, growing evidence demonstrates the need for another model, which draws from biomarkers, modeling and simulation, and adaptive trial designs, to emerge and coexist. New insights bring the confidence that the treatments on the market are based on an understanding of disease subtypes as well as patient cohorts for whom the treatment is intended.
About the Authors
Kathleen Martin is experienced in both business and clinical consulting and has also worked in the pharmaceutical and healthcare industries. She has an MPH in Epidemiology and has published in the areas of medical malpractice, corporate learning, and clinical development.
Michael Hehenberger is responsible for the development and implementation of IT solutions that support health care and life sciences. Recently, he developed solutions for clinical genomics and biobanking, pharmacogenomics, clinical decision intelligence, bio-medical and molecular imaging.
This article was published in Drug Discovery & Development magazine: Vol. 10, No. 12, December, 2007, pp. 34-37.
- Adams & Brantner 2006 Mar-Apr; 25(2):420-8; DiMasi & Grabowski J Clin Oncol. 2007 Jan 10; 25(2):209-16
- “Pharmacogenomics to replace pharma’s business model” (PwC)
- US Food and Drug Administration. “FDA Facts.” March 29, 2006.
- Martin, Kathleen, Mark Hammond and Stuart Henderson. The eClinical Equation: Part 2 – Bridging Connections for Innovation.” IBM Institute for Business Value. October 2006.
- McCormick, Terrence, Kathleen Martin and Michael Hehenberger. The Evolving Role of Biomarkers: Focusing on Patients from Research to Clinical Practice”. IBM Institute for Business Value, July 2007.
- von Eschenbach, Dr. Andrew C. Commissioner of Food and Drugs, U.S. FDA. Keynote address. Imaging Biomarker Summit II. June 2006.
- McCormick, Terrence, Kathleen Martin and M. Hehenberger. Advancing the Utility of Imaging Biomarkers: Insights from the Second Imaging Biomarker Summit. IBM Institute for Business Value, Dec 2006.
- Trusheim, M. R., E. R. Berndt, and F. L. Douglas. “Stratified Medicine: Strategic and Economic Implications of Combining Drugs and Clinical Biomarkers.” Nature Reviews/Drug Discovery, Vol. 6, April 2007, pp. 287-293.
- Kaisernetwork.org, FDA to Develop Regulatory Guidelines for Trials That Change Midcourse to Accommodate Early Results. July 10, 2006. (accessed August 24, 2006).
- Quintiles Transnational press release. Quintiles Creates Strategic Biostatistics Unit to Help Sponsors Reduce Time, Costs of Clinical Drug Development. July 24, 2006.
- McGee, Patrick. Modeling Success with In Silico Tools. Drug Discovery and Development. April 1, 2005.
- US Food and Drug Administration, “Challenge and Opportunity on the Critical Path to New Medical Products.”
Filed Under: Drug Discovery