Quantitative Clinical Pharmacology: Making Paradigm Shifts a Reality
Rajesh Krishna · The Journal of Clinical Pharmacology · 2006
Today, drug development is at a crossroads. A number of scientific and regulatory initiatives including the Critical Path Initiative of the United States1 and the Innovative Medicines Initiative of the European Union2 have acknowledged a pipeline problem, with a declining number of new molecular entities (NME) related to a suboptimal application of biomarkers, predictive models of human disease, computational disease model libraries, and adherence to model-based drug development. Although these initiatives attempt to address the stagnation versus innovation conundrum somewhat, we as a scientific community must try to address the problem at the core. For instance, it is not the number of NMEs reaching regulatory acceptance that is the issue; rather, it is our ability to quantitatively understand pathways of disease such that the therapies that best modulate these pathways without unacceptable risk are pursued. Let us try to ponder current drug-development hypotheses a bit. The current approach to drug development assumes that the target is univariate and that a single NME would be sufficient for drug effect on that target, that is, a “one target, one molecule” paradigm. Indeed, this is how most drug-development hypotheses have evolved in recent years. Our lessons learned in oncology and infectious diseases have proven this approach wrong, in that multiple pathways are involved, implicating multiplicity of targets and requiring NME combinations for optimal therapeutic success. Perhaps one of the best examples of a systems approach to medicine is the success using inhaled corticosteroids for asthma. Is it because asthma as a disease paradigm is multifactorial in nature and that steroids are able to modulate multiple pathways effectively? We know that the risk/benefit of inhaled corticosteroids is likely closely tied to their being able to modulate multiple pathways. In a target-centric drug-development approach, potential for toxicities due to off-target drug action are likely to be unpredictable. On the other hand, a systems-based approach to human disease pathophysiology and pharmacology has been recently proposed, which could potentially revolutionize drug development such that therapies that meet unmet medical needs as well as retain a favorable risk/benefit profile can be best pursued. For a systems-based approach to be successful, disease fingerprinting is essential, opening up opportunities for disease progression and biodynamic modeling using complex but credible computational tools. These approaches, when coupled with intelligent clinical trial development, can positively influence the quality, efficiency, and productivity of drug development. A second aspect of current drug development is relying on subjective clinical opinion on risk versus benefit. We have to move away from a study-centric philosophy to a seamless data mining and knowledge management philosophy that effectively integrates data across studies such that long-term hazards could be predicted with a more stochastic sense of uncertainty. Population modeling and simulations have allowed us to quantitatively integrate pharmacokinetic and pharmacodynamic data across individual studies. However, there is still a lot to be leveraged in the very significant amount of information generated during the life cycle of a new drug product, specifically data on safety, pharmacokinetics, pharmacodynamics, efficacy, and clinical outcomes. Effective data integration can form the foundation for a more quantitative outlook on risk versus benefit leveraging the predictive power of advances in computational science. The ultimate goal for clinical pharmacology and therapeutics is the development and delivery of safe and effective medicines. It is hoped that paradigm shifts in drug development will achieve that vision. In keeping with these changing times and to encourage a systems-based approach to model-based drug development, The Journal of Clinical Pharmacology announces the launch of a new section devoted exclusively to the emerging area of quantitative clinical pharmacology (QCP). The goal of the new section is to encourage research and discussion on the broader elements of hypothesis-based drug development and, specifically, on quantitative risk versus benefit. Using these approaches, it is hoped that a more quantitative framework on risk versus benefit will be fostered. Given the importance of systems biology in early drug discovery and experimental medicine, the scope of QCP ranges from translational discovery to the postmarketing phase for NMEs and approved drugs, as indicated in the following list: QCP Subject Areas Systems biology and biomarker modeling Disease progression modeling Population pharmacokinetic/pharmacodynamic modeling and simulations Simulation of clinical trial designs and quantifying uncertainty in drug response Novel clinical trial designs (eg, adaptive designs, N of 1 trials, seamless trials, etc) Modeling of missing data Clinical utility index assessments Modeling epidemiologic and outcomes data Confirmatory risk versus benefit I encourage you to submit manuscripts relevant to QCP subject areas directly to The Journal of Clinical Pharmacology via its online submission portal at http://www.rapidreview.com, with a notation in the cover letter that the manuscript should be considered for the QCP section. If you are engaged in QCP research, I would also encourage you to contact the editorial office to discuss opportunities on how you can participate in this new section activity as a reviewer.