In vivo, check; in vitro, check; in silico?
Have you ever heard of this third term, “in silico,” before? It means “performed on a computer or via computer simulation.” I’m not an expert on in silico, but I am pretty good at spotting emerging trends, and I sense one here. Let’s start with some terminology, recognizing that what follows is not a definitive treatise of the field, but rather an introduction to aspects of an emerging vocabulary and a powerful tool for the future.
Our readers are well aware of the expensive process of drug development, whereby one creates a new molecule, gets it through all of the appropriate clinical trials, and the drug is (one hopes) finally approved by the key regulatory agencies in our nation. This process can take anywhere from 6 to 13 years, at a price tag that can easily exceed billions of dollars per molecule. As a result, teams of interdisciplinary scientists are searching for a more cost-effective approach beyond in vivo and in vitro studies.
Building on the growth of computational biology, the new field often called “big data,” and the widespread availability of truly futuristic computational power, Allerheiligen and colleagues have described the emergence of a new field of science they call “pharmacometrics.”1 Pharmacometrics, “which builds on physiological, pharmacological, and biostatistical principles, provides additional powerful approaches for supporting important drug development and regulatory decisions. Numerous successful case studies published by academic, industry, and FDA scientists attest to the value of its contribution to decision making.”1
What exactly is pharmacometrics? According to Allerheiligen and colleagues, pharmacometric analyses are “quantitative analyses of data pertaining to pharmacokinetics, biomarkers, clinical outcomes, disease characteristics, and trial characteristics,”1 essentially, the entire waterfront of activity necessary to go from bench to bedside. However, pharmacometrics also includes mathematical modeling and simulation, statistical analyses, and the need for a multidisciplinary team consisting of quantitative clinical pharmacologists, statisticians, engineers, data management experts, and clinicians.1
Visser and colleagues offer a more nuanced definition of “modeling and simulation” (M&S).2 “M&S in preclinical drug development focuses on the translation of preclinical data into quantitative predictions of the pharmacokinetics, pharmacology (proof of mechanism and concept) and safety in man. This enables selection of drug candidates with the best efficacy—safety profile for clinical development and optimization of first-in-human clinical trial designs.”2
Another emerging term that is growing in popularity is “quantitative systems pharmacology” (QSP). According to Visser and colleagues, “QSP is an emerging discipline which is based on the integration of mechanism-based PKPD [pharmacokinetic/pharmacodynamic] modeling concepts and systems pharmacology concepts. A unique feature of QSP is that it significantly improves our ability to quantitatively understand and characterize pathways of disease and to predict the efficacy of compounds with novel mechanisms of action.”2
Whether we call it pharmacometrics, modeling and simulation, or quantitative systems pharmacology, I believe these terms are all about in silico pharmacology.
The question then is, does in silico work? Allerheiligen and colleagues note that there are some key advantages to using pharmacometric analysis.1 It may estimate the effect size and, thereby, in silico pharmacology may allow investigators to propose the best doses and estimate the clinical trial effect size, all by using the computer. Pharmacometric analysis may also enable us to target patient selection and maximize the value of previous data, and it could speed up drug approval and streamline the labeling process.
In fact, in another article by Allerheiligen, she contends that “Merck has measured cost avoidance for studies not conducted (not full development costs) and tracks critical governance-level and regulatory decisions—go and no-go decisions, phase II/III trial designs, approvals, dose selections, and development strategy—enabled by M&S each year.
In the past three years, the M&S group at Merck delivered more than half a billion dollars in cost avoidance, and it continues to enable approximately 10 critical decisions each year.”3 It would appear, then, at least at Merck, that in silico pharmacology is working.
How will large organizations adapt to in silico pharmacology, given its core interdisciplinary nature? Where in the hierarchy will this activity most readily flourish? Currently, we are unable to answer these important questions. The American Society for Clinical Pharmacology and Therapeutics has set up a pharmacometrics task force to build momentum for this new field and to attempt to accelerate its adoption.
I’m smitten with in silico pharmacology, and I view it as a critical evolutionary step that will streamline drug development and will reduce the economic burden of bringing new molecules to the marketplace.
1. Allerheiligen S, Gobburu J, Goldberger MJ, et al. Applying pharmacometrics in drug development. In: Evolution or Revolution? McKinsey Perspectives on Drug and Device R&D 2012. McKinsey & Co; 2012:48-51. www.mckinsey.com/~/media/mckinsey/dotcom/client_service/pharma%20and%20medical%20products/
pmp%20new/pdfs/evolution_or_revolution_compendium_2012.ashx. Accessed April 1, 2016.
2. Visser SAG, Manolis E, Danhof M, Kerbusch T. Modeling and simulation at the interface of nonclinical and early clinical drug development. CPT Pharmacometrics Syst Pharmacol. 2013;2:e30.
3. Allerheiligen SR. Impact of modeling and simulation: myth or fact? Clin Pharmacol Ther. 2014;96:413-415.