Q&A: Drs. Punam Sandhu, Dennis C. Dean, and Thomas A. Baillie, Drug Metabolism, Merck

Merck: We conducted a microdosing study in dogs utilizing AMS detection to examine the linearity of the pharmacokinetics (PK) of a novel nucleoside antiviral agent across sub-pharmacological and pharmacological doses. The study was designed primarily to validate in-house the technology associated with microdosing and AMS with a view to potentially conducting such studies in human volunteers.

Merck: Presently at Merck, we view human microdosing as a useful method to address specific compound/program issues. For example, an early understanding of key PK properties of a drug candidate in humans could be useful information for design of appropriate follow up studies in preclinical species and humans. Thus far, we have not employed microdosing studies in situations where multiple closely related compounds are assessed in humans using a microdose to select a lead candidate (i.e., the use of microdosing studies in a screening mode). While there may be situations in which a preemptive human PK screening strategy via microdosing could prove valuable, we do not see this approach as having general utility considering the time/resource requirements and the limitations of a microdose to predict therapeutic dose PK in certain situations.

Merck: As discussed earlier, we do not believe that microdosing studies, in the form of a general screening tool, represent a viable, value-added paradigm for the large majority of drug discovery and development program. It is important to recognize that the majority of failures in drug development today occur as a result of preclinical toxicity, poor clinical tolerability, or a lack of therapeutic efficacy, and not due to inappropriate PK characteristics (as was the case in the 1970s and 1980s). In some situations, however, microdosing may facilitate early failure of compounds. There will also be situations where the main objective of a microdosing study is to design appropriate follow up studies, and not necessarily to terminate the further development of a specific compound.

Where we believe microdosing studies can, however, be useful are for compounds with varying or unusual pharmacokinetics in preclinical species, such as molecules with long terminal elimination half-lives or those that sequester in tissues. An early human microdosing study in such a scenario may be beneficial in obtaining an initial assessment of the elimination and distribution properties of the compound in humans, provided that the PK across the "micro" and "therapeutic" dose range prove to be linear.

Merck: Compared to LC-MS/MS, a limitation of positron emission tomography (PET) and AMS is that information on the identity of metabolites cannot be obtained directly from these techniques. Furthermore, since both techniques measure total radiotracer levels, it is not possible to discern whether the circulating chemical moieties are parent compound and/or metabolites. Both PET and AMS provide much greater sensitivity than LC-MS/MS, thus allowing the quantitation of very low drug levels in plasma after microdosing.

While PET tracers are very useful and widely used as biomarker tools for determining receptor occupancy (in which the PET tracer is used as a microdose along with a more relevant therapeutic dose range of unlabelled drug candidate), use in PK/distribution studies in which the drug candidate is labeled with a positron-emitting isotope are limited by the very short half-life of such tracers. PET studies do have the unique advantage of being able to provide a real-time image of drug distribution, usually when administered via intravenous dosing.

From a methodological standpoint, microdose/AMS or PET studies are labor and resource intensive compared to LC-MS/MS. In the case of AMS, sample collection and handling issues introduce complexities associated with the need to prevent contamination from extraneous sources of ¹⁴C. However, when appropriate procedures are put in place, excellent concordance can be obtained between measurements conducted by AMS and LSC. Another limitation of AMS is lack of high-throughput sample analysis due to the need for samples to be graphitized prior to analysis. PET studies are often limited by access to the PET tracer of a drug candidate as a result of synthesis and handling issues (again half-life is a constraint), as well as the high cost and inconvenience of conducting the studies in appropriate PET clinical centers.

Merck: For compounds that display non-linear kinetics, it would not be possible to obtain accurate predictions of PK properties at the therapeutic dose levels from microdosing studies. Unfortunately, there is very little information in the literature on pharmacokinetic properties of pharmaceutical compounds whereby linearity has been assessed using sub-pharmacological and pharmacological doses. Also, each drug candidate will be case-specific so that establishing linear PK across the dose range in some animal model would likely be required.

Merck: Though both documents agree on the definition of a microdose, there are significant differences in the animal toxicology study packages required to support human microdosing studies. According to the FDA guidance, genetic toxicity data is not essential for initiating human microdosing studies, yet in vitro genotoxicity studies are an integral requirement in the EMEA guidance. Further, in the FDA guidance, data in animal species are required only for the intended route of administration of compound in humans. In the EMEA guidance, two routes of administration are generally required; intravenous as well as the intended clinical route. It would be helpful to have some harmonization between the two sets of guidance.

The EMEA guidance allows for simultaneous administration of more than one compound from a structurally related series as long as the total amount of test compound(s) does not exceed 100 μg. It is not clear from the FDA guidance whether it is permissible to administer more than one compound to human volunteers at any given time.

Finally, neither document offers any guidance on the amount of ¹⁴C radiotracer that can be used to classify microdosing studies utilizing radiotracer compounds as exempt from regulatory requirements that govern traditional radiotracer studies. The studies reported thus far in the literature utilizing nanoCurie levels of radiolabel in microdosing studies have employed amounts ranging from ~3-200 nCi/subject.