COMMENTARIES
`
`Translation of Drug Interaction
`Knowledge to Actionable
`Labeling
`Mongthuong T. Tran1 and Joseph A. Grillo1,*
`
`This paper focuses on the effective communication of drug
`interaction information in US prescription drug labeling. There are
`important implications and unique challenges related to drug
`interaction information, including the translation of scientific data
`into clinical recommendations and presentation of its breadth and
`complexity. This paper highlights strategies to enhance
`communication in labeling of essential drug interaction– related
`information to healthcare providers regarding the safe and
`effective use of a drug.
`
`BACKGROUND
`Unanticipated, unrecognized, or mismanaged
`drug interactions (DIs) are major contribu-
`tors to preventable morbidity and mortality.
`DIs are estimated to represent 3–5% of pre-
`ventable in- hospital adverse reactions1 and
`are recognized as an important contributor
`to emergency department visits and hospital
`admissions.2 A retrospective study reported
`26% of total hospital admissions directly due
`to adverse drug reactions involved a DI.3
`DIs may be complicated, not well eluci-
`dated, and lead to suboptimal therapy with
`reduced efficacy or increased drug- related
`toxicities. Understanding how to safely
`avoid, mitigate, or manage DIs is critical
`to patient care. However, in a recent cross-
`sectional study of 895 final- year European
`medical students, Brinkman et al.4 observed
`that the students had the lowest knowl-
`edge scores for DIs and contraindications.
`
`Furthermore, a 2016 newspaper investiga-
`tion reported that when their investigators
`attempted to fill prescriptions that had po-
`tentially dangerous interactions at 255 com-
`munity pharmacies in the United States,
`52% dispensed the drugs without warning
`the patient of the potential risk.5 These
`findings highlight the importance of effec-
`tive communication of DI- related informa-
`tion to the healthcare provider.
`
`DI- RELATED INFORMATION IN US
`PRESCRIPTION DRUG LABELING
`A key source for information regarding
`the interaction profile of a drug in the
`United States is the prescribing infor-
`mation (PI) portion of prescription drug
`labeling, which is also referred to as the
`package insert. The PI is the primary
`mechanism by which the US Food and
`Drug Administration (FDA) and drug
`
`manufacturers communicate essential in-
`formation to healthcare providers regard-
`ing the safe and effective use of a drug per
`the Code of Federal Regulations ((CFR)
`21 CFR 201.56 (a)(1)). In addition to the
`specific requirements on content and for-
`mat of labeling for human prescription
`drug and biological products that are cod-
`ified in the CFR, the FDA also publishes
`Guidances for Industry, which represent
`its current thinking on a topic but are not
`binding on the FDA or the public.
`Comprised of 17 major sections, the PI
`is based on the totality of evidence derived
`from a comprehensive review and analysis
`by the FDA of the new drug application or
`biologics license application submitted by
`an applicant. Following regulatory review
`and approval, the PI is available directly
`to providers and informs tertiary drug in-
`formation sources and clinical decision
`support tools, which may be more readily
`accessible.
`The majority of DI information in the
`United States PI is found in the DRUG
`INTERACTIONS
`and CLINICAL
`PHARMACOLOGY sections. In particu-
`lar, the DRUG INTERACTION section
`is a highly utilized section by healthcare pro-
`viders.6 The DRUG INTERACTIONS
`section presents clinically relevant regula-
`tory conclusions regarding the existence
`and mitigation of DIs, and the CLINICAL
`PHARMACOLOGY section describes de-
`tailed scientific information to support these
`conclusions. Additional pertinent information
`may also be found in other PI sections, such
`as DOSAGE AND ADMINISTRATION,
`CONTRAINDICATIONS, OR WARNINGS
`AND PRECAUTIONS.
`DI information in labeling should in-
`form prescribing decisions regarding phar-
`macokinetic (PK) or pharmacodynamic
`interactions by including clinically relevant
`findings and regulatory conclusions, if
`
`1Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, US Food and Drug Administration, Silver
`Spring, Maryland, USA. *Correspondence: Joseph A. Grillo (joseph.grillo@fda.hhs.gov)
`
`Received February 22, 2019; accepted March 6, 2019. doi:10.1002/cpt.1427
`
`1292
`
`CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 105 NUMBER 6 | June 2019
`
`PERSPECTIVES
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2043, Page 1
`
`

`

`appropriate, for such information as meta-
`bolic and transport pathways, clinically im-
`portant metabolites, clinical implications
`of clinically significant DIs and genetic
`polymorphisms, and recommended risk
`mitigation strategies.
`Although the PI contains DI informa-
`tion required by regulation, inconsisten-
`cies with translating scientific data into
`clinical recommendations and present-
`ing the breadth and complexity of DI-
`related information may still exist. For
`example, conveying clinical implications
`of exposure changes driving interactions,
`method for providing representative ex-
`amples of interacting drugs, and inform-
`ing additional pathway implications or
`variability of an individual drug when
`representing DI as a group. As conveyed
`at the 2013 advisory committee meeting
`hosted by the FDA Office of Clinical
`Pharmacology, the desire is to communi-
`cate clinical pharmacology information,
`including DIs, in a manner that is more
`clinically intuitive and free of unnecessary
`information.7 Other unique challenges
`that
`impact effective communication
`may be attributed to the ever- changing
`nature of DI information that is difficult
`to capture in a timely way, the diverse un-
`derstanding of underlying pharmacology
`of metabolic- based and transporter- based
`DIs, inconsistent labeling development,
`and differences between PI and tertiary
`DI information sources.
`
`STRATEGIES TO ENHANCE
`COMMUNICATION OF DI- RELATED
`INFORMATION IN US PRESCRIPTION
`DRUG LABELING
`DI information should be communi-
`cated in an actionable and informative
`manner to healthcare providers who may
`not have specific expertise in clinical
`pharmacology.
`
`• DI information should be clear and in-
`clude essential details for safe and effec-
`tive prescribing of a drug.
`• Consistent and deliberate use of termi-
`nology in labeling is important to min-
`imize ambiguity or confusion within
`and across PIs.
`• DI information should be devoid of any
`technical jargon and distilled into what
`is clinically relevant.
`
`• In general, DI information should be
`presented in the format that enhances
`readability and best accommodates its
`breadth and complexity to ensure clar-
`ity and understanding (e.g., font and
`text attributes, bullets, headings, white
`space, and tables). For example, Figure 1
`displays a representative example that
`effectively conveys extensive informa-
`tion in the DRUG INTERACTIONS
`section.
`
`DRUG INTERACTIONS Section of US PI
`Regulation 21 CFR 201.57(c)(8)(i) and
`FDA guidance8 state that the DRUG
`INTERACTIONS
`section
`describe
`clinically significant interactions, either
`observed or predicted, with other prescrip-
`tion or over- the- counter drugs, classes of
`drugs, or foods (e.g., dietary supplements
`and grapefruit
`juice); include specific
`practical instructions for preventing or
`managing the DI; the mechanism of the
`DI (if known); and the clinical implica-
`tion of the DI. An interaction is clinically
`significant if concomitant use of drugs
`leads to safety, efficacy, or tolerability
`
`concerns greater than those present when
`the drugs are administered alone. In gen-
`eral, DIs should be listed in the order of
`clinical importance.
`The clinical implication of a DI is es-
`sential to convey as it links the mechanism
`of the interaction to a safety, efficacy, or
`tolerability concern that impacts the pro-
`vider’s decision making on therapeutic
`individualization and optimization for
`patients. Changes in relative drug con-
`centrations or other PK parameters alone
`do not sufficiently establish a clinical im-
`plication, and therefore, the PK changes
`should be linked with a clinical concern
`(e.g., reduced efficacy or increased bleed-
`ing risk). Information regarding the ab-
`sence of a DI should generally not appear
`in this section, unless this information
`is clinically relevant for the healthcare
`provider (e.g., if a drug does not have the
`same interaction as other drugs in the
`same class).
`Strategies to prevent or manage these
`DIs should be actionable, specific, and
`practical to the healthcare provider, such
`as contraindicate concomitant use, avoid
`concomitant use, temporarily discontinue
`
`Figure 1 Representative example of drug interaction information in a tabular format in the
`DRUG INTERACTIONS section of the prescribing information. The following table is meant to
`be a representative example of a possible format and should not be considered a template,
`limit other possible formats, or constrain the use of other information fields that may be
`required for a particular drug. Drug X is the proprietary name and drugoxide is the generic
`name. AUC, area under the curve; CYP, cytochrome P450.
`
`CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 105 NUMBER 6 | June 2019
`
`1293
`
`PERSPECTIVES
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2043, Page 2
`
`

`

`interacting drug, modify dosage, and mon-
`itor specific safety or efficacy parameters.
`Specific statements using active voice (e.g.,
`monitor serum creatinine or increase dose)
`are more informative than ambiguous or
`vague statements (e.g., use caution or ad-
`just dose). Standardization of “avoidance”
`terms (e.g., avoid, do not recommend, use
`if benefits outweigh the risk) will provide
`consistency and clarity for the healthcare
`provider.
`
`CLINICAL PHARMACOLOGY Section of
`US PI
`Regulation 21 CFR 201.57(c)(13)(i) and
`FDA guidance9 state the CLINICAL
`PHARMACOLOGY section should in-
`clude the results of clinical and in vitro
`studies (e.g., of metabolism or interaction)
`and other pertinent analyses that estab-
`lish the presence or absence of a clinically
`significant DI. Clinical information can
`include data and results from prospec-
`tive clinical DI studies, population PK
`analyses, modeling and simulations (e.g.,
`physiologically-based PK modeling), post-
`marketing reports, or data extrapolated
`from other information. Once deemed
`sufficient to inform a regulatory decision
`in place of a dedicated clinical study, popu-
`lation analyses or modeling and simulation
`
`approaches generally do not need to be ref-
`erenced as the source of data (e.g., “Based
`on physiologically-based pharmacokinetic
`modeling…”).
`study results
`significant
`Clinically
`can be presented in the CLINICAL
`PHARMACOLOGY section as text
`or in a table or figure depending on the
`number of studies and the level of detail
`needed for clarity and understanding.
`Figure 2 displays representative examples
`of tabular and figure presentations in this
`PI section. DI information should include
`only those study features that are essential
`for the safe and effective use of a drug.
`The relative change in exposure can be
`presented as a percentage or a fold change
`with a clinically meaningful measure of
`variability/dispersion, such as range. For
`example, although 90% confidence inter-
`vals are useful for regulatory review, the
`minimum and maximum potential expo-
`sure change may be more informative to
`the healthcare provider. Tables and fig-
`ures should clearly state the reference arm
`and display the relative change in key PK
`exposure measures.
`In vitro information should establish
`the absence of a DI effect or characterize
`protein binding, metabolic, and trans-
`porter pathways in the absence of clinical
`
`information. In vitro information alone
`does not establish a significant DI and,
`therefore, is rarely presented in the DRUG
`INTERACTIONS
`section. Generally,
`in vitro information that is superseded
`by clinical data should not be reported.
`However, in rare cases, in vitro studies may
`be included to provide additional context
`for related clinical studies.
`Studied drugs with no clinically signifi-
`cant interaction potential should be listed
`in a summary sentence without any study
`details. To minimize redundancy of infor-
`mation appropriately presented elsewhere
`in labeling, specific regulatory conclusions
`and actionable instructions regarding a DI
`should not be repeated in the CLINICAL
`PHARMACOLOGY section.
`
`CONCLUSION
`DIs contribute to preventable morbidity
`and mortality and may not be fully un-
`derstood by healthcare providers. How
`well information regarding the safe and
`effective use of a drug is ultimately com-
`municated to the prescribing provider in
`labeling is a key element of a drug devel-
`opment program. Appropriate presenta-
`tion of DI information in the PI is critical
`to enable interpretation and translation
`of this information for individualized
`
`Figure 2 Representative example of drug interaction information as either a tabular (a) and figure (b) format in the CLINICAL PHARMACOLOGY
`section of the prescribing information. The following table and figure are meant to be a representative example of a possible format and
`should not be considered a template, limit other possible formats, or constrain the use of other information fields that may be required for a
`particular drug. Drug X is the proprietary name and drugoxide is the generic name. AUC, area under the curve; CI, confidence interval; Cmax,
`peak plasma concentration; PK, pharmacokinetic.
`
`1294
`
`VOLUME 105 NUMBER 6 | June 2019 | www.cpt-journal.com
`
`PERSPECTIVES
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2043, Page 3
`
`

`

`patient care. This presentation can be
`optimized by communicating essential
`information in a clear, concise, nontech-
`nical manner and leveraging the use of
`text attributes and creative formatting
`techniques.
`
`FUNDING
`No funding was received for this work.
`
`CONFLICT OF INTEREST
`The authors declared no competing interests
`for this work.
`
`DISCLAIMER
`The contents of this article reflect the views
`of the authors and should not be construed to
`represent the US Food and Drug Administration
`(FDA)’s views or policies. No official support
`or endorsement by the FDA is intended or
`should be inferred. The mention of commercial
`products, their sources, or  their use in
`connection with material reported herein is
`not to be construed as either an actual or
`implied endorsement of such products by the
`FDA.
`
`Published 2019. This article is a U.S. Government
`work and is in the public domain in the USA
`
`1. Leape, L.L. et al. Systems analysis of ad-
`verse drug events. ADE Prevention Study
`Group. JAMA 274, 35–43 (1995).
`2. Raschetti, R. et al. Suspected ad-
`verse drug events requiring emergency
` department visits or hospital admissions.
`Eur. J. Clin. Pharmacol. 54, 959–963
`(1999).
`3. McDonnell, P.J., Jacobs, M.R., Monsanto,
`H.A. & Kaiser, J.M. Hospital admissions
`resulting from preventable adverse
`drug reactions. Ann. Pharmacother. 36,
`1331–1336 (2002).
`4. Brinkman, D.J. et al. Hospital admissions
`resulting from preventable adverse drug
`reactions. Clin. Pharmacol. Ther. 101,
`281–289 (2017).
`5. Roe, S., Long, R. & King, K. Pharmacies
`miss half of dangerous drug combinations.
`Chicago Tribune <http://www.chicagotri-
`bune.com/> (2016). Accessed February
`22, 2019.
`6. Sullivan, H.W., O’Donoghue, A.C. & Aikin,
`K.J. Primary care physicians’ use of
`
`FDA- approved prescription drug labels. J.
`Am. Board Fam. Med. 27, 694–698 (2014).
`7. US Food and Drug Administration (FDA).
`FDA Advisory Committee Meeting
`Transcript. Advisory committee for phar-
`maceutical science and clinical pharma-
`cology (ACPS-CP); 210–294, <https://bit.
`ly/2QK9z4s> (2013). Accessed February
`22, 2019.
`8. U.S. Department of Health and Human
`Services, Food and Drug Administration,
`Center for Drug Evaluation and Research
`(CDER). Clinical Drug Interaction
`Studies — Study Design, Data Analysis,
`and Clinical Implications: Guidance
`for Industry. Draft Guidance. October
`2017. <https://bit.ly/2BRNbl7> (2017).
`Accessed February 22, 2019. When
`finalized, the guidance may be found at
`<https://bit.ly/2SYdmRF>.
`9. U.S. Department of Health and Human
`Services, Food and Drug Administration,
`Center for Drug Evaluation and Research
`(CDER). Clinical Pharmacology Section
`of Labeling for Human Prescription Drug
`and Biological Products — Content and
`Format: Guidance for Industry. December
`2016. <https://bit.ly/2VcXesb> (2016).
`Accessed February 22, 2019.
`
`Clopidogrel as a Perpetrator
`of Drug– Drug Interactions:
`A Challenge for Quantitative
`Predictions?
`Manthena V.S. Varma1,*
`, Yi-an Bi1, Sarah Lazzaro1 and Mark West1
`
`Clopidogrel perpetrates pharmacokinetic interactions primarily
`due to time- dependent inhibition (TDI) of cytochrome P450
`(CYP)2C8 by a circulating metabolite, clopidogrel acyl- β- d-
`glucuronide (Clop- Gluc).1 Additionally, Clop- Gluc is a reversible
`inhibitor of organic anion transporting polypeptide (OATP)1B1
`in vitro . Given many CYP2C8 substrates show hepatic uptake
`via OATP1B1, clopidogrel interaction mechanisms and
`inhibition potential at clinical doses have been debated—
`particularly with the quantitative predictions of dasabuvir–
`clopidogrel interactions.2,3 Here, we summarize clinical data
`and evaluate mechanistic models to further our understanding of
`clopidogrel interactions.
`
`Gemfibrozil and clopidogrel are recom-
`mended inhibitors to probe CYP2C8 ac-
`tivity in vivo .4 Gemfibrozil is metabolized
`by uridine 5′- diphosphate glucuronosyl-
`transferase 2B7 to form gemfibrozil acyl-
`β- d- glucuronide
`(Gem- Gluc), whereas
`clopidogrel is an ester prodrug that con-
`verts to an inactive carboxylic acid (~ 85%
`of dose), which is further glucuronidated
`to form Clop- Gluc. Both Gem- Gluc
`and Clop- Gluc show CYP2C8 TDI,
`and parent- metabolite pairs also inhibit
`OATP1B1 in vitro .1,5
`
`EFFECT OF CLOPIDOGREL ON CYPs—
`CLINICAL EVIDENCE
`to a
`Clopidogrel
`inhibits CYP2C8
`moderate- to- strong degree (area under
`the curve (AUC) ratio 2–5) and weakly
`(AUC ratio 1.25–2) inhibits CYP2B6
`and CYP2C19 at a loading dose of 300 mg
`(Figure 1). Relatively lower effects are
`noted with 75 mg maintenance dose.
`On the other hand, gemfibrozil 600 mg
`b.i.d.
`increased AUC of daprodustat
`and dasabuvir by ~ 19- fold and 11- fold,
`
`1Pharmacokinetics, Dynamics, and Metabolism, Medicine Design, Worldwide R&D, Pfizer, Inc., Groton, Connecticut, USA. *Correspondence:
`Manthena V.S. Varma (manthena.v.varma@pfizer.com)
`Received December 13, 2018; accepted February 8, 2019. doi:10.1002/cpt.1398
`
`CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 105 NUMBER 6 | June 2019
`
`1295
`
`PERSPECTIVES
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2043, Page 4
`
`