Abstract
Sales of botanical dietary supplements and other purported medicinal natural products (NPs) have escalated over the past ∼25 years, increasing the potential for NPs to precipitate clinically significant pharmacokinetic interactions with U.S. Food and Drug Administration–approved medications [NP-drug interactions (NPDIs)]. However, published NPDI studies to date often lack consistency in design, implementation, and documentation, which present difficulties in assessing the clinical significance of the results. Common hurdles include large variability in the admixture composition of phytoconstituents between and within batches of a given NP, limited knowledge on the pharmacokinetics of precipitant NP constituents, and use of animal and/or in vitro models which, in some cases, are not mechanistically appropriate for extrapolation to humans. The National Center for Complementary and Integrative Health created a Center of Excellence for Natural Product-Drug Interaction Research (NaPDI Center) to address these unmet research needs. The NaPDI Center has two overarching goals: 1) develop Recommended Approaches to guide researchers in the proper conduct of NPDI studies, which will evolve over time concurrent with emerging technologies and new research data, and 2) apply the Recommended Approaches in evaluating four model NPs as precipitants of NPDIs with clinically relevant object drugs. The major objectives of this commentary are to 1) explain the rationale for creating the NaPDI Center; 2) describe the decision trees developed by the NaPDI Center to enhance the planning, rigor, and consistency of NPDI studies; and 3) provide a framework for communicating results to the multidisciplinary scientists interested in the NaPDI Center’s interaction projects.
History and Unique Challenges of Pharmacokinetic Natural Product-Drug Interactions (NPDIs)
Sales of botanical dietary supplements and other purported medicinal natural products (NPs) in the United States have more than tripled over the past ∼25 years since passage of the Dietary Supplement Health and Education Act in 1994, nearing $7.5 billion as of 2016 (Smith et al., 2017). Based on the general perception that ‘natural’ means ‘safe’, the lay public often turns to NPs as a means to alleviate self-diagnosed illnesses, allay health care costs, and/or supplement prescribed therapeutic regimens, often without discussing with their health care practitioners (Blendon et al., 2001; Gufford et al., 2014). Consequently, patients frequently take NPs in conjunction with Food and Drug Administration–approved medications, potentially leading to NPDIs that may alter the safety or effectiveness of conventional medications. Despite considerable investigation, pharmacokinetic elucidation and clinical relevance of many of these complex interactions remain inconclusive (Gufford et al., 2014).
Assessing the risk of NPDIs is more challenging than that of drug-drug interactions. Common difficulties include the large variability in the admixture composition of phytoconstituents among marketed products of supposedly the same NP and the relatively scant pharmacokinetic knowledge of precipitant NP constituents (Paine and Oberlies, 2007; Brantley et al., 2014; Gufford et al., 2014). Many purported NPDIs are based on studies involving animal and/or in vitro models; it is difficult, and often not mechanistically appropriate, to extrapolate results from these preclinical models to humans and assess probable clinical relevance (Gufford et al., 2014). Even when human studies are available, the data often are equivocal or conflicting due in part to: 1) variable study design, which ranges from case reports to uncontrolled observational studies to controlled human studies, and/or 2) faulty study design or inadequate methods. Another difficulty is reports of mild to moderate NPDIs, particularly with widely used drugs, which have unclear clinical significance (National Center for Complementary and Alternative Medicine, 2012) and often lead to the question of whether vulnerable subpopulations (e.g., low prevalence genotype carriers, renally and hepatically impaired patients, pregnant women, and elderly or pediatric patients) exist that would be uniquely susceptible to a NPDI. Clear standards for publication of NP-related research, including the Recommended Approaches of the Natural Product Drug Interaction Research Center of Excellence (NaPDI Center), can mitigate some of these challenges. Such standards should spur other pharmacology journals to consider adopting practices similar to those of the American Society for Pharmacology and Experimental Therapeutics journals.
Addressing NPDI Research Challenges
Feasible solutions to the aforementioned challenges of NPDI research are available; however, organized efforts to rally the research community to develop evidence-based, standardized approaches were previously lacking. The National Center for Complementary and Alternative Medicine, Office of Dietary Supplements, and National Cancer Institute jointly sponsored a roundtable meeting of experts and stakeholders on “Dietary Supplement-Drug Interactions” in March 2012 (National Center for Complementary and Alternative Medicine, 2012). The National Center for Complementary and Alternative Medicine (now the National Center for Complementary and Integrative Health) subsequently put forth a request for application in June 2014 to create a center of excellence focusing on NPDI research methods; our investigative team was awarded this U54 cooperative agreement in September 2015. The overarching goals of the NaPDI Center are twofold:
Develop a set of Recommended Approaches to guide investigators conducting NPDI research. These approaches are expected to be updated at regular intervals to reflect new research findings based on emerging technologies, including initial lessons learned from the NaPDI Center’s interaction projects (described subsequently).
Apply the Recommended Approaches toward assessing the potential of four carefully selected NPs to precipitate clinically significant pharmacokinetic interactions with relevant object drugs in model studies referred to as interaction projects.
Creation of the Recommended Approaches will ideally lead to consensus on an evidence-based, standardized approach to address the unique challenges of NPDI research. Addressing these challenges requires a multidisciplinary team of clinical pharmacologists, natural products and analytical chemists, and biomedical informaticists.
The NaPDI Center consists of three scientific cores (Pharmacology, Analytical, and Informatics), along with an Administrative Core that coordinates activities among the scientific cores and liaises between NaPDI Center investigators and National Center for Complementary and Integrative Health program officials. The Pharmacology Core will: 1) use a systematic evaluative approach to select four NPs as potential precipitants of NPDIs and 2) develop statements of work for a coordinated series of studies, which are guided by evidence-based decision trees, ranging from investigations with human-derived in vitro systems to static and dynamic mathematical modeling to well-designed clinical studies. The Analytical Core will: 1) develop rigorous guidelines for the NP chemistry investigations, focusing primarily on characterizing and sourcing NPs, and 2) analyze the biofluids (e.g., plasma, urine) obtained from clinical interaction studies for the precipitant NP constituents and the object drug and its metabolites. The Informatics Core will: 1) create a repository of the data generated from the interaction projects using the most up-to-date open-source technologies and cloud-based infrastructure and 2) develop a public portal to inform researchers of the NaPDI Center’s results. The Cores will work together to achieve the NaPDI Center’s two overarching goals. The main innovation offered by the NaPDI Center is the combined deployment of state-of-the-art in vitro cell-based assays, pharmacokinetic modeling and simulation, and clinical approaches to improve NPDI research methods.
Recommended Approaches
There is a range of challenges (Table 1) associated with NPDI research that can best be addressed with evidence-based decision trees (Figs. 1–3). These challenges need to be tackled in a variety of ways, spanning from prioritization of which NPDIs should be rigorously evaluated, to NP characterization that accounts for structural isomers and metabolites of phytoconstituents, to using only human-derived in vitro systems for screening altered activity of drug metabolizing enzymes (DMEs) or transporters, to necessitating follow-up NPDI studies with narrow therapeutic index object drugs if a clinical NPDI occurs with probe substrates. The list of unique research challenges (Table 1), anticipated Recommended Approaches (Table 2), and decision trees (Figs. 1–3) developed by the NaPDI Center to date serve as a roadmap and toolset for addressing key NPDI research needs. Because of the numerous challenges associated with NPDIs, the Tables and Figures present how to address these challenges in various formats to provide a comprehensive overview from the perspective of the challenges (Table 1), the Recommended Approaches (Table 2), and how an individual NP will progress through the Decision Trees (Figs. 1–3).
The first charge of the NaPDI Center was to select and prioritize the NPs to be studied as potential precipitants of NPDIs. A systematic method was developed to identify four priority NPs that will serve as exemplars for the Recommended Approaches (Johnson et al., 2018). Beginning with a list of 47 commonly used NPs, these four NPs were selected based on existing in vitro and clinical data suggesting the potential to precipitate a NPDI. Each identified NP will be entered into the decision trees to evaluate the requirements for the interaction project.
Upon prioritization and approval of the NPs, the Pharmacology Core will develop detailed statements of work for each of the four NPs. These statements of work will include a comprehensive literature review and preliminary data to identify the particular needs for NP characterization and sourcing, details about the target DME(s) and/or transporter(s) mediating the potential NPDIs, and according to the decision trees (Figs. 1–3), build the plan for in vitro and clinical studies to determine the likelihood of a NPDI.
Evaluation of an interaction project begins with part I of the comprehensive decision tree (Fig. 1) by asking the following key question: Are the NP constituents commercially available? If such constituents are available, the Analytical Core will subsequently determine if the NP constituents meet their predetermined reference standard specifications. If these two criteria are met, the Pharmacology Core will evaluate the available in vitro metrics, specifically the inhibition or induction parameters described in the Food and Drug Administration draft guidance on drug-drug interactions (https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm292362.pdf), published in the literature, and other publicly accessible sources (e.g., new drug applications and medication labels). The Informatics Core will provide input, as necessary, regarding the availability of these data. If such constituents are not commercially available, the Pharmacology Core will work with the Analytical Core to isolate, purify, and scale-up sufficient quantities of the constituents for determination of the in vitro metric(s) using relevant human-derived systems. Once robust in vitro metrics are obtained, the interaction project will be advanced to part II (Fig. 2).
Part II of the decision tree begins with the following key question: Do in vivo human data exist about the concentrations of the NP constituent(s)? If the data do not exist, a phase 0 human pharmacokinetic study involving selected NP formulations and doses will be conducted. The choice of formulation(s) will be decided by the Analytical Core. Once in vivo data are obtained, they will be combined with in vitro data from part I and applied to a predictive static model based on an average steady-state concentration of the precipitant NP constituents; alternatively, a dynamic (e.g., physiologically based pharmacokinetic) model could be developed to simulate multiple NP and drug dosing regimens using the maximum tolerated dose of the NP. Modeling and simulations will be used to predict the likelihood and magnitude of a NPDI (https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm292362.pdf) under two circumstances: 1) if the maximum plasma concentration of one or more NP constituents is near or exceeds the in vitro metric (https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm292362.pdf) or 2) if the target DME/transporter is pharmacologically important in the gut (e.g., cytochrome P450 3A, UDP-glucuronosyltransferase 1A, P-glycoprotein, organic anion transporting polypeptide 2B1) (Won et al., 2010, 2012) and the estimated gut tissue/lumen concentration is near or exceeds the in vitro metric. If a NPDI is predicted, the interaction project will be advanced to part III (Fig. 3).
Part III of the decision tree begins with the following task: determine the NP dose needed to precipitate an interaction using physiologically based pharmacokinetic modeling and existing in vivo or phase 0 data from part II. Once the dose is determined, a clinical NPDI study will be conducted using a probe object drug or cocktail of object drugs. If the magnitude of the NPDI interaction [e.g., the ratio of the object drug area under the concentration-time curve (AUC) in the presence of the NP to the AUC in the absence of the NP, AUCNP/AUC] is significant (i.e., ≥30% (https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm292362.pdf), a follow-up NPDI study involving a different drug (e.g., one with a narrow therapeutic index) will be considered. If the magnitude of the NP-probe drug interaction is not significant, physiologically based pharmacokinetic modeling and simulation may be used to further assess the likelihood of a clinically significant NPDI in the following situations: a narrow therapeutic index object drug, human subjects carrying clinically important DME/transporter pharmacogenetic variants, or vulnerable subpopulations. If the modeling and simulation suggest a clinically significant NPDI is likely, follow-up clinical studies with a different object drug or within different patient populations will be considered. All data generated from the interaction projects will be organized and archived into a repository that is publicly accessible through a web-based public portal created by the Informatics Core.
Future Directions
The NaPDI Center’s Recommended Approaches for NPDI research are working documents that should be updated as new information comes forth from ongoing interaction projects and any ensuing studies. Priority and emphasis of NPDIs also will need reconsiderations with major shifts in the dietary supplement and herbal product markets and changes in standard guidelines for the prescribing and use of the involved medications. As sales of NPs continue to increase annually, the NaPDI Center will continually strive to provide evidence-based recommendations that balance the risks of drug interactions against the benefits of NPs.
Acknowledgments
M.F.P. dedicates this article to the memory of her father, Dr. David P. Paine. All authors acknowledge the participants and research staff of their previous NPDI studies. Both groups were essential to improving the scientific understanding of NPDIs, and thus the ideas expressed in this commentary.
Authorship Contributions
Wrote or contributed to the writing of the manuscript: Paine, Shen, McCune.
Footnotes
- Received February 28, 2018.
- Accepted May 3, 2018.
↵1 Current affiliation: Department of Population Sciences, City of Hope, Duarte, California.
This work was supported by the National Institutes of Health (NIH) National Center for Complimentary and Integrative Health [Grant U54 AT008909]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Abbreviations
- AUC
- area under the concentration-time curve
- DME
- drug metabolizing enzyme
- NaPDI Center
- Natural Product Drug Interaction Research Center of Excellence
- NP
- natural product
- NPDI
- natural product-drug interaction
- Copyright © 2018 by The American Society for Pharmacology and Experimental Therapeutics