Early View
White Paper
Open Access

Non-Labeled, Stable Labeled, or Radiolabelled Approaches for Provision of Intravenous Pharmacokinetics in Humans: A Discussion Piece

Graeme C. Young

Corresponding Author

Graeme C. Young

GSK Research &Development Ltd., Stevenage, UK

Correspondence: Graeme C. Young ([email protected])

Search for more papers by this author
Douglas K. Spracklin

Douglas K. Spracklin

Pfizer Inc., Groton, Connecticut, USA

Search for more papers by this author
Alexander D. James

Alexander D. James

Novartis, Basel, Switzerland

Search for more papers by this author
Mette G. Hvenegaard

Mette G. Hvenegaard

H. Lundbeck A/S, Copenhagen, Denmark

Search for more papers by this author
Mette L. Pedersen

Mette L. Pedersen

Drug Metabolism and Pharmacokinetics, Early Research and Development, Cardiovascular, Renal and Metabolism, Biopharmaceuticals R&D, AstraZeneca, Gothenburg, Sweden

Search for more papers by this author
David S. Wagner

David S. Wagner

AbbVie, North Chicago, Illinois, USA

Search for more papers by this author
Katrin Georgi

Katrin Georgi

The Healthcare Business of Merck KGaA, Darmstadt, Germany

Search for more papers by this author
Hanno Schieferstein

Hanno Schieferstein

The Healthcare Business of Merck KGaA, Darmstadt, Germany

Search for more papers by this author
Inga Bjornsdottir

Inga Bjornsdottir

Novo Nordisk, Maaloev, Denmark

Search for more papers by this author
Andrea A. Romeo

Andrea A. Romeo

Roche Pharma Research and Early Development, Basel, Switzerland

Search for more papers by this author
Kenneth C. Cassidy

Kenneth C. Cassidy

Eli Lilly and Company, Indianapolis, Indiana, USA

Search for more papers by this author
Georges Da-violante

Georges Da-violante

Technologie Servier, Orleans, France

Search for more papers by this author
Stefan Blech

Stefan Blech

Boehringer-Ingelheim Pharma GmbH & Co. KG, Biberach, Germany

Search for more papers by this author
Simone I. Schulz

Simone I. Schulz

Bayer AG, Wuppertal, Germany

Search for more papers by this author
Filip Cuyckens

Filip Cuyckens

Janssen R&D, Beerse, Belgium

Search for more papers by this author
Mai Anh Nguyen

Mai Anh Nguyen

Sanofi, R&D, Frankfurt, Germany

Search for more papers by this author
Graeme Scarfe

Graeme Scarfe

Sosei Heptares, Great Abington, Cambridge, UK

Search for more papers by this author
First published: 29 November 2023

Abstract

A review of the use of microdoses and isotopic microtracers for clinical intravenous pharmacokinetic (i.v. PK) data provision is presented. The extent of application of the varied approaches available and the relative merits of each are highlighted with the aim of assisting practitioners in making informed decisions on the most scientifically appropriate design to adopt for any given new drug in development. It is envisaged that significant efficiencies will be realized as i.v. PK data in humans becomes more routinely available for suitable assets in early development, than has been the case prior to the last decade.

The science of clinical drug development is a complex business and there is a myriad of options available to assist with understanding the pharmacokinetics (PKs) of a New Chemical Entity (NCE) by the intravenous (i.v.) route in particular.1 With the recent advancements in the emergence of sensitive analytical technologies, such as accelerator mass spectrometry (AMS)2, 3 and further enhancements in that regard to liquid-chromatography tandem mass spectrometry (LC–MS/MS), Clinical Pharmacokineticists, Bioanalysts, and Drug Metabolism scientists have a broad spectrum of options available to them in choosing a preferred approach (e.g., microdose vs. microtracer) and study design to provide the desired information (e.g., 2 period crossover vs. single period with or without concomitant oral/i.v. dosing). The decision as to which approach is taken as the preferred one is influenced by many aspects from sound scientific knowledge and judgment to historical bias and potentially flawed financial analyses; such as those relying on the use of “free” in-house LC–MS/MS resources rather than on external expenditure via the provision of more specialized analytical support by a Contract Research Organization offering technology not available in-house (e.g., AMS).4 These aspects and more will be discussed in this article – prepared by the members of the European Federation of Pharmaceutical Industries and Associations (EFPIA) drug metabolism and pharmacokinetic (DMPK) Network sponsored human absorption, distribution, metabolism, and excretion (ADME) paradigm shift(s) working group – in the hope that the decisions that are made around such choices are done so in an informed and pragmatic way and to improve the overall efficiency of drug development as it pertains to the generation of intravenous pharmacokinetic (i.v. PK) data in humans. The benefits to clinical development through provision of i.v. PK data in humans have been known for years and are many-fold5 and for brevity are not re-stated here. Historically (likely more than 25 years ago) i.v. data were generated in phase I clinical studies and in fact entry into human subjects often started with an i.v. dose escalation phase, but this practice stopped across the pharmaceutical industry as it was deemed that oral dose escalation designs were a more efficient use of resources for drugs that were not intended to be parenterally administered medicines. Suffice to say that through mention of the generation of i.v. PK data (for absolute bioavailability assessment within the human mass balance study) in the recent draft guidance from the US Food and Drug Administration (FDA)6 there is recognition that these data can now be generated via efficient microdose/microtracer designs. In this paper, the term microdose is as per the International Conference on Harmonization (ICH) guidance,7 that is, for NCEs, it is ≤ 100 μg and ≤ 1/100th of the pharmacologically active dose; whichever is the lower dose. The term microtracer pertains to concomitant administration of an isotopically labeled substance (radioactive or stable isotope) administered intravenously along with a considerably higher extravascular dose(s).

A variety of strategies are outlined in the following, with reflections on the relative merits of each approach with consideration of the overall premise toward more routine generation of i.v. PK data in humans in support of the development of medicines for extravascular administration.

SINGLE PERIOD MICRODOSE/MICROTRACER DESIGN VS. CONVENTIONAL TWO PERIOD CROSSOVER

As previously outlined,8 traditionally a two-period crossover design was conducted to investigate i.v. compared with extravascular dosing, with a washout period between. For example, the first period where drug is administered by the i.v. route and the second period where the drug is administered by the extravascular route (often oral). The study would often involve between 12 and 16 participants and the doses used would be at or near therapeutic levels for both study periods. To enable such a study design infers a significant commitment of resources, including the need to generate nonclinical and clinical safety data and extensive formulation development activities to facilitate dosing for both routes of administration at a therapeutic dose level to humans. The more contemporary approach is to use concomitant administration with the isotopically labeled drug, as a so-called microtracer, administered by the i.v. route in the same dosing period as the non-labeled extravascular dose at a therapeutic level.

The various isotopic microtracer approaches have previously been thoroughly reviewed1 and are becoming an industry standard for the generation of i.v. PK data. Lappin et al., highlighted 10 such studies to 2016, but the number of studies reported in the literature has increased markedly since then (see Table 1 below) and a search of the clinicaltrials.gov website (using search terms “absolute bioavailability” and “microtracer“) produced more than 50 studies. However, there is often a predisposition within pharmaceutical companies toward one favored approach, for example, use of stable labelled (usually 13C) vs. of radiolabeled (14C) drug for the i.v. formulation, or visa versa depending on experience, application of resources, capabilities, etc. In each case, the “microdose” i.v. administration within the concomitant two dose route design,4 is commonly referred to as a microtracer. Equally, it should not be overlooked that unlabeled drug can be administered as the microdose, indeed, as will be outlined later, this approach provides significant opportunity to provide i.v. PK data early, in a single dosing route period embedded as part of the first-in-human (FIH) dose escalation study. As discussed later, this type of approach can often be the “easy sell” to drug development project teams as there are relatively fewer resources (and less time) involved to conduct enabling activities as compared with the isotope label incorporation designs.

Table 1. A variety of outline study designs
N Design Outputs Comments Supporting references
Primary Secondary +
1 Non-labeled (12C) i.v. microdosea [separate dose period] i.v. PK; definitive Vd and CLp Absolute bioavailability (via crossover to oral dose period) Can be included as part of FIH with relative ease as separate cohort to, for example, oral SAD 9
2 Non-labeled therapeutic oral +14C microtracer (i.v.)c [concomitant dose] i.v. PK and absolute bioavailability Additional ADME end points, for example, renal clearance and systemic metabolic load following i.v. (parent vs. RDM) Gold standard for generic maximum sensitivity by AMS

10-18

3 Non-labeled therapeutic oral +13C microtracer (i.v.) [concomitant dose] i.v. PK and absolute bioavailability N/A Faster data turnaround than design #1

19-22

4 14C therapeutic oral (human ADME) + 13C microtracer (i.v.) [concomitant dose] Mass balance and metabolite characterization (human ADME) i.v. PK and absolute bioavailability Likely conducted late in development to try to fulfill human ADME and absolute bioavailability questions in one study periodb 23-27
  • ADME, absorption, distribution, metabolism, and excretion; AMS, accelerator mass spectrometry; CLp, clearance from plasma; FIH, first-in-human; N/A, not applicable; PK, pharmacokinetic; RDM, radioactive drug-related material (also commonly referred to as total radioactivity); SAD, single ascending dose; Vd, volume of distribution.
  • a Phase 0 microdose studies are not considered in this summary. See elsewhere for a comprehensive review of phase 0 studies.28
  • b Notwithstanding the option to consider bringing this study forward as outlined in the recent white paper by this human ADME paradigm shift(s) working group.29
  • c This design may be included as a period of the human ADME study (see section below describing the wording from the US Food and Drug Administration draft guidance).

In considering the experiences for the industry in the application of microtracer approaches to generate i.v. PK data in humans, a poll was carried out across the 16 companies represented by the authors of this article. The questions posed included “what approach to generation of i.v. microtracer data in humans has your company used over the last 5 years (i.e., use of 13C or 14C-labeled drug?).” The data gathered are shown in Figure 1, reflecting that, unsurprisingly, there were a variety of responses ranging from 100% application of the 14C approach (5 companies), a 20–95% application in favor of the 14C approach (6 companies), to a default toward use of 13C (3 companies). There were also two companies who had no experience of using the microtracer approach and/or were still using a traditional crossover design. Note that the question was confined to studies where isotopic labels of carbon were used (rather than to include alternatives such as deuterium and/or 15N-labeled drug) for simplicity.

Details are in the caption following the image
Representation of number of companies (out of polled total of 16) with experience of application of an isotope label design for provision of i.v. PK in humans for NCEs. Notes – Value within each circle represents the number of companies that responded with the preferred approach stated below each circle. Percentage figures indicate extent of application of the specific approach, for example, range of 20–95% indicates that across 6 companies the use of the 14C-label is as low as 20% or as high as 95% of i.v. PK studies conducted. NCE, New Chemical Entity; PK, pharmacokinetic.

It has been highlighted,1 that for regulatory agencies in Europe, the United States, and Japan, there are no specific requirements for the generation of PK parameters, such as clearance from plasma and volume of distribution (Vd) following i.v. administration for drugs intended for use by the non-parenteral route. However, the recent FDA draft of the mass balance guidance6 indicates that “Information on the absolute bioavailability of the investigational drug can help interpret mass balance data and understand the overall drug elimination pathways.” This statement supports a general consensus by the industry toward conducting such study designs29 and perhaps provides an indication that the agency may encourage i.v. PK data in humans to be available with increased frequency than in even the recent past, although scientific rationale should of course dictate the justification. The draft guidance goes on to outline a design in which the information can be efficiently derived: “When only the oral formulation is being developed, an absolute bioavailability study can be combined with the mass balance study in a single protocol in a two-part study. For example, Part A can be the human radiolabelled mass balance study for the orally administered investigational drug. Part B can determine the absolute bioavailability of the investigational drug administered as an oral non-radiolabelled dose….and an intravenous radiolabelled microdose (without the need for an intravenous toxicology program if the existing oral toxicity studies provide adequate exposure margins).” It should be noted that the characteristics of the drug under development will have an impact on the desire to generate i.v. PK data in humans (e.g., if there is ample nonclinical evidence to support assignment of the NCE as a Biopharmaceutics Classification System (BCS) class 1 compound),30 clinical assessment of absolute bioavailability is not likely to be prioritized during early development.

It is difficult to see why any company in modern times would revert to the use of a two-period crossover design using therapeutic level intravenous dose administration, purely to generate single dose absolute bioavailability data and the view of the majority of the authors of this article is that this historical design is largely consigned to history.

In addition to the now common combined human ADME and absolute bioavailability study design using either low, microtracer doses of 14C-labeled drug in each of the two periods of the study31 or a relatively high (conventional) dose of 14C-drug in one period (by the therapeutic route) and a microtracer i.v. dose in a separate period,32 there are a range of other designs that are outlined below. Most of these are concomitant different dose route designs, but also a standalone microdose design is included for completeness.

DESCRIPTIONS OF THE STUDY DESIGNS

Non-labeled (12C) i.v. microdose in FIH

Inclusion of a non-labeled microdose i.v. cohort within an early clinical study, perhaps even the FIH study, can add significant additional information that would otherwise be missing during clinical development of an NCE. The authors of this paper are not aware of examples in the published literature of inclusion of a non-labeled microdose in FIH studies for an NCE, except when those have been conducted as phase 0 studies (for a comprehensive review of phase 0, see ref. 28), or in the case of the example detailed below. However, along with other helpful literature providing guidance on which to base a sound decision, including ways to assess the risk of nonlinearity,33 experiences with invoking those studies have helped to demonstrate their feasibility and value, paving the way for inclusion of microdose study arms in conventional phase I FIH ascending dose designs. Historically, use of 14C-radiolabeled drug was seen as being necessary to support studies with doses in the microdose category34 (see section 7.1 of ref. 7), but it has been demonstrated that LC–MS/MS methods supporting studies with non-labeled (or stable-labeled (13C) tracer – see study design 3 below) doses will often be adequate.35-37 It is worth noting that drugs with acidic physicochemical properties are more likely to have reduced distribution beyond the plasma compartment of the systemic circulation and this lower Vd, for any given dose, will improve the likelihood for adequate detection of drug concentrations in plasma using LC–MS/MS. Equally, technical considerations, such as ionization efficiency, signal to noise, and sample preparation play a role in the determination of assay sensitivity by LC–MS/MS. In contrast, detection by AMS is agnostic to such facets of the analyte and so has generally similar utility across all chemical structural types (with specific activity being the main sensitivity determinant).

There are of course many considerations to be taken into account to facilitate this i.v. microdose inclusion, such as whether an appropriately sensitive bioanalytical LC–MS/MS assay will be available to provide the necessary PK information, as alluded to above. In addition, as the PK profile will be based on predicted data, there is an inherent risk herein of a failed study which must be considered as part of the decision criteria for choosing the most appropriate study design and bioanalytical methodology. The main consideration, however, is that discussions behind the decision to progress toward the chosen approach early in clinical development are carried out very early indeed, likely preselection for the lead candidate into development (probably 18 months ahead of conducting the planned FIH study). There has been a recent example of the introduction of a microdose study arm, embedded into the dose escalation phase of an FIH study for GSK3915393 an irreversible inhibitor of the enzyme transglutaminase 2 being investigated for the treatment of celiac disease9 [in press] (see Figure 2 below). This provided several advantages to the study design, including the fact that safety data had been gathered in humans from the much higher magnitude oral dose administrations (e.g., the 60 mg oral dose provided 6,000-fold dose safety cover vs. the 100 μg i.v. dose). By embedding the i.v. dose cohort into the dose escalation sequence, rather than conducting this as the first dose of a phase 0 study design, meant that there was no requirement for a nonclinical study in rodents, as the oral nonclinical and clinical safety data could be used to support this i.v. microdose.7 This sequence of incorporation of the i.v. dose cohort then gave additional time for both the i.v. route specific formulation development and very sensitive bioanalytical assay to be validated. In addition, the i.v. cohort could be conducted in light of the PK data already being available from the preceding oral dose phases. This allowed assessment to be made of the appropriateness of the sampling timepoints as well as the time for the washout between a much higher oral dose administration and the i.v. microdose, as clearly the risk of “carry over” of drug from the oral dose into the i.v. PK assessment has to be appreciated ahead of data generation. The washout period could conceivably have been increased if there was likely to be an issue of this kind (see schematic in Figure 2). In fact, as shown in the schematic, the washout period between periods 2 and 3 of the dose escalation (with the intervening i.v. cohort between) was the same as between periods 1 and 2, thus no time was added to the study overall with the inclusion of the i.v. dose phase.

Details are in the caption following the image
Schematic of inclusion of a non-labeled i.v. microdose in an FIH dose escalation study. Note: Planned dose level 4 was not conducted – it was determined that sufficiently high exposures to provide efficacy would have already been achieved at lower doses. FIH, first-in-human.

In addition to the use of an i.v. microdose to establish the definitive systemic PKs of the drug under study (including absolute bioavailability), an i.v. administration was further utilized in assessments of the impact of inhibition of CYP3A4 on the metabolism of GSK3915393 through co-administration of i.v. and oral administrations with grapefruit juice or itraconazole. This adaptive study design and iterative use of i.v. administration along with the oral dose administrations, facilitated greater understanding of the drug–drug interaction potential for this NCE but also provided insights into the likely target engagement for parent drug in the intestinal lamina propria to help to inform on future studies in patients.

It has to be recognized that patients will get no direct therapeutic benefit from such an i.v. microdose administration and that the overall study duration may be lengthened through inclusion of this additional study group. There may be therefore less acceptability of using such a design in an FIH where the participants are, for example, oncology patients. However, potential societal benefit may be increased through bringing a medicine to the broader patient population more rapidly, so this option should not be automatically ruled out. All of that said, the concomitant two route dosing microtracer approach may still often be the more acceptable approach for such patient groups vs. a simpler microdose design (see Table 2 and ref. 26).

Table 2. Summary of options for consideration to inclusion of an i.v. microdose/microtracer
Non-labeled (12C) Stable Labeled (13C) Radiolabeled (14C)
Pros Cons Pros Cons Pros Cons
No additional ILAPI req'd i.v. microdose PK (i.e., not a microtracer design) Microtracer approach can be used ILAPI req'd (££) Microtracer approach can be used ILAPI req'd (£££)
Standard LC–MS/MS bioanalytical method

Sufficient

LC–MS/MS sensitivity req'd (~ 10 pg/mL typical)

No radiolicences req'd End points limited to PK (no metabolism or excretion data) Maximal assay sensitivity “guaranteed” LC + AMS method required (as well as LC–MS/MS)
No radiolicences req'd Standalone microdose req'd (separate cohort) Further sensitive isotopic LC–MS/MS bioanalytical method req'd ADME endpoints beyond PK are possible Radiolicences may be req'd in some territories/facilities
  • ADME, absorption, distribution, metabolism, and excretion; AMS, accelerator mass spectrometry; ILAPI, isotopically labeled active pharmaceutical ingredient (extra cost and preparation time); LC-MS/MS, liquid-chromatography tandem mass spectrometry; PK, pharmacokinetic.

This non-labeled i.v. microdose approach as an addition to the FIH study can be viewed as an “easy sell” because there is no requirement for additional label incorporation (whether that be stable or radiolabel). However, to ensure appropriate delivery of resources and enabling activities in an acceptable timeframe, it does require project teams to plan for it early in development (or in the late discovery phase). The discussion is often easier to have with project teams compared with those involving phase 0 approaches, largely due to organizational, educational, and culture alignment issues.28 As a result, inclusion of the non-labeled i.v. microdose in the FIH dose escalation study is now a default consideration for at least one major pharmaceutical company.

Concomitant doses of non-labeled therapeutic oral +14C microtracer (i.v.)

As indicated by the poll carried out across the companies within this working group, this design is probably the most regular approach to provision of i.v. data in a microtracer design, whether that be as part of a human ADME study,10 as a standalone single period study,17 or even built into the FIH study.18 The main advantages of this design over the non-labeled microdose (design 1 above) or stable microtracer (design 3 below) includes the “guaranteed” maximal sensitivity that is conferred by the 14C-label and detection by AMS through use of a high specific activity radiolabeled drug. Furthermore, because this approach includes an oral therapeutic dose there are no concerns in terms of nonlinear PK issues.33

As an aside, the study for vismodegib (a hedgehog signaling pathway antagonist for the treatment of basal cell carcinoma), described in Lappin 2016, used an i.v. pulse dose of 14C-labeled drug on day 1 and again on day 7 of repeated oral therapeutic dose administration to investigate nonlinear PKs and dose-dependent absorption and concentration-dependent plasma protein binding of vismodegib. Clearance and Vd of vismodegib were found to increase with repeated administration. This study nicely showed the insights that can be obtained through use of a microtracer for i.v. PK data generation in repeat dose administration of a therapeutic drug, which is of course most relevant for medicines intended for chronic use.

A NOTE ON DATA DECONVOLUTION

A complexity that arises, is the need to consider deconvolution of data sets as a result of co-administration of more than one dose, where contributions to measured systemic exposures of the drug analyzed, originate from both doses. This is well explained in an example human ADME study of the non-steroidal glucocorticoid receptor modulator velsecorat,38 where the 14C-labeled drug in the i.v. dose also contained non-labeled drug which would be measured (by the bioanalytical LC–MS/MS assay) along with the non-labeled drug from the inhaled dose that was co-administered. The magnitude of this phenomenon will vary from compound to compound, but is more marked when the absolute bioavailability (F) from the extravascular route dose administration is very low (as was the case for the dual pharmacophore (muscarinic acetylcholine receptor antagonist/β2-adrenoreceptor agonist) batefenterol10 with F of only 1.3% following inhaled administration). Even though the i.v. dose was only 4 μg, whereas the inhaled dose was 300-fold higher at 1200 μg, the non-labeled drug from the i.v. dose contributed significantly to the systemically available drug concentrations which were measured by the LC–MS/MS assay and thus deconvolution was required to provide systemic drug concentration data for drug arising from the inhaled dose only. The data from the analysis of samples by AMS, was used, along with the known specific activity of the i.v. dosed radioactive drug to provide the concentration of non-labeled drug arising from the i.v. dose. This was then subtracted from the LC–MS/MS data to provide the resultant concentration that arose from the inhaled dose.

Concomitant doses of non-labeled therapeutic oral +13C microtracer (i.v.)

This design has found favor in situations where the focus of the investigation is to establish the PKs of the drug following i.v. administration, without the need for 14C-drug and the consequential support by AMS.19-22

By its very nature, as a stable (non-radiolabel) isotope, the use of 13C as the tracer negates any considerations around shipment, storage, and handling of samples with respect to those that have to be considered for source materials and samples containing radioactivity, albeit that in some territories there may be a cutoff below which samples are considered to be non-radioactive even when they contain measurable levels of radioactivity above the natural background.

There are many advantages to the use of the stable label microtracer approach, as well as considerations, such as restrictions around analytical sensitivity and the idiosyncrasies of the synthesis route for 13C-labeled drug, which will be different from that used as the stable label internal standard for the bioanalytical assay and can lead to technical difficulties in the synthesis. Both of these latter aspects were mentioned in a recent paper39 where the 14C-labeled drug was used instead of stable labeled drug for the i.v. microtracer dose administration. Similarly complicated approaches toward the use of analytical internal standard at, for example, M + 4 vs. parent drug and M + 8 for the dosed stable labeled drug were adopted for another study.40 The greater the change in molecular weight vs. parent drug, the greater the likelihood for a significant impact on the physicochemical properties of the dosed material, potentially creating misleading data. This has, in some cases, led to additional animal work being conducted to attempt to mitigate against the likelihood of this risk occurring.41 This mass-dependent isotope effect is bigger for elements with lower mass and, thus, the highest for deuterium labeling with a 100% increase in mass over hydrogen. The impact of deuteration on the physicochemical properties of a molecule can sometimes be observed by a lower metabolic clearance rate or earlier retention in reversed-phase chromatography (less hydrophobic).42 In addition, 15N-labeled compounds can sometimes be separated from their 14N isotopomers due to an effect on pKa.43 It is, therefore, good practice to keep the number of non 13C-labels to a minimum or, if this is not feasible, to evaluate the similarity of the stable labeled version in a comparative in vitro metabolism experiment. A nice methodology to calculate the minimum number of labels required in support of microdose absolute bioavailability studies was reported44 mitigating isotopic interferences, taking into account selective reaction monitoring transitions and carefully selecting the position and number of labels used. From resourcing and cost standpoints, the synthesis of the stable isotope requires new chemistry and is not always guaranteed to succeed. The positions and number of stable labels needs to be carefully considered to minimize any “cross talk” among the LC–MS/MS transitions of the non-labeled drug product, the stable label internal standard, and the stable labeled i.v. drug product.

The concept of the concomitant microtracer approach is summarized in Figure 3.

Details are in the caption following the image
Concomitant oral and intravenous dosing microtracer concept. Adapted from Figure 1 of ref. 1. LC-MS, liquid-chromatography mass spectrometry; PK, pharmacokinetic.

Concomitant doses of 14C therapeutic oral (human ADME) + 13C microtracer (i.v.)

This study design is a very elegant and efficient way of combining the conventional human ADME study design with an assessment of i.v. PK in one study period.23-25, 27 It does have the downside that no assessment of the impact of first pass metabolism can be made in the range of ways that can be estimated via administration of the radiolabeled drug by both routes of administration31, 45 but equally there are efficiencies to be made through this single period design. In particular, where patients are the study participants this can facilitate their more rapid release from the clinical pharmacology study to a therapeutic regimen protocol.26 A summary of the options for consideration during discussions of whether to use non-labeled drug as a microdose, stable, or radiolabeled drug in a microtracer scenario to provide i.v. PK in humans for an NCE, is provided in Table 2.

CONCLUSIONS AND FUTURE DIRECTIONS

The premise of this paper is to assist industry clinical pharmacokineticists through increased awareness of the options available to them to provide PK data in humans following i.v. administration of NCEs. As outlined, there are complexities involved in making the appropriate choice as to which study design and technical approach to use to deliver these data, which is becoming ever more useful in drug development, not least through verification and further utility of physiologically-based PK models. A discussion of the approaches that have been used in recent years has been provided, addressing some of the technical complexities associated with the various approaches from microdose to microtracer, with or without the inclusion of isotopic labels. The key message is to ensure that whatever route is taken to provision of i.v. PK data in humans, the most efficient design for any particular NCE should be used and that design should be chosen in an unbiased and informed manner. The most recent development in this area is the move, in one company among those in this working group, toward inclusion of a non-labeled i.v. microdose as a default approach in the FIH study, or at least consideration of doing so early enough in project progression to take advantage of this option, if warranted.

A recent output of the International Consortium for Innovation and Quality in Pharmaceutical Development (IQ) Human PK Prediction Working Group46 included an inference that 20% of the respondents (5 of the 25 surveyed) use an i.v. (radiolabeled) microtracer in early phase I to provide i.v. PK data. Taking this at face value supports the fact that a trend toward generation of such data early in clinical development is already underway. It will be interesting to see whether the choice of design and timing within “early phase I,” perhaps to the FIH study itself, is settled upon as a preferred/default approach. Nevertheless, the paper did indicate that the authors suggest “Microdose approaches to gather early IV PK data would allow improved confidence or refinement of predictions of clearance….”. It does seem entirely feasible that soon, the routine generation of intravenous route data for all NCEs in development will be a reality, thereby turning back the clock by around 30 years such that these data are available to aid in decision making at the earliest stages of clinical development, facilitated by modern techniques and strategic approaches.

ACKNOWLEDGMENTS

The authors thank Dr. Matthew Hoffmann (formerly of Bristol-Myers Squibb, New Jersey) and all reviewers from the author's companies for their considered review of this paper.

    FUNDING

    No funding was received for this work.

    CONFLICT OF INTEREST

    All authors are employees of pharmaceutical companies listed on the title page and may own stocks in those companies.