Drug discovery and development is the process of generating compounds and evaluating all their properties to convert any new chemical entity to a safe and efficacious drug. This process costs around 1 billion dollars and takes approximately 10 years for the drug to reach the general population¹.

Exhaustive research is needed to discover a drug candidate for a target disease. In addition to this, the manufacturing industry must confirm the efficacy of the compound and determine its toxic potential. As part of this process, researchers have to look at any possible mischievous by-products created during the metabolism of a drug. Active drug metabolites discovered during the course of drug development constitute an important issue of metabolic drug interactions, but still demand unique consideration from both an efficacy and a safety point of view. Improved technology has allowed better identification and quantification of metabolites, raising new issues to be addressed during the course of drug development.

Several new molecular identities recently entering the market have active metabolites and many more are in development. The Metabolites in Safety Testing (MIST) committee of Pharmaceutical Research and Manufacturers of America (PhRMA) has prepared a paper regarding the difficulty in the assessment of metabolites, which has been widely discussed with and by the regulatory authorities over the past few years².

The liver is the primary organ responsible for drug metabolism, and most drugs are substrates for the liver's cytochrome P450 oxidase enzymes. Within the human population, there are P450 variants that give rise to some differences in drug metabolism. But, for the purpose of drug safety testing, the key interest is looking at the differences in metabolism between humans and animals³. Some of the drugs with active metabolite issues include fluoxetine (norfluoxetine), flutamide (HydroxyFlutamide), amifostine (Ethyol), dihydrocodeine (dihydromorphine) and halofantrine (desbutylhalofantrine).

Metabolites of pharmaceutical molecular entities that have biologic activity, which are either therapeutic or adverse, referred to as ‘active metabolites’, represent a difficult and active area in drug development. Active drug metabolites discovered during the course of drug development constitute a subset of the larger issue of metabolic drug interactions, but still demand unique consideration from both an efficacy and a safety point of view⁴.

This may be considered as a distinct subset of the drug interaction issue complicated by the fact that, in this case, each drug provides its own interactions. There are three major cases to be considered for such interactions:

1. Metabolite activity is pharmacologically the same as the parent drug.

2. Metabolite activity is entirely different to that of the parent drug and hazardous. Understanding the mechanism of toxicity of a drug requires the knowledge that whether toxicity is due to the administered parent or a toxicologically active metabolite or both.

3. Metabolites that possess both of the above activities^5,6.

In all three cases, regulatory questions arise regarding how much of the metabolites are produced, is it a major metabolite, how potent they are and the metabolic activities involved and how do they complicate issues of individual susceptibility.

If the metabolite's pharmacological activity is qualitatively the same as the desired activity of the parent, then one must consider whether the parent is only serving as a prodrug perhaps, the metabolite being more attractive for development than the parent, or if the metabolite is potentially a follow-on or next-generation drug and offering some advantages over its parent but not enough to merit aborting the already ongoing development programme.

Considerations in these situations include whether the metabolite has greater potency, greater selectivity of action, poorer bioavailability, a more favourable safety profile or a better pharmacokinetic half-life, as was the case with the metabolites of both terfenadine, (fexofenadine) and penclomedine (4-demethylpenclomedine). Another important point to recognise is that the overall effect of such drugs will consist of those of the administered parent plus the contributions of any active metabolites^7,8

Safety questions associated with an active metabolite are a separate issue. It may be that the toxicity associated with a drug turns out to be overwhelmingly a product of the parent, in which case further development is best pursued with the active metabolite. Conversely, if toxicity is associated with the metabolite, a change in formulation, dosing regimen or route of administration may serve to avoid or significantly reduce the safety issue. If the metabolite, however, is responsible for major toxicity in humans, development of analogues of the parent compound might be able to overcome this drawback².

Drug metabolites can exert toxicity via different mechanisms; however, from a regulatory perspective, it is initially important to determine which type of metabolite is circulating in the body. Metabolites are divided into two broad categories based on their chemical nature and the way they behave in the circulation. They include chemically stable metabolites in the circulation that can reversibly interact with a specific pharmacological target or off-target receptor or enzyme and chemically reactive metabolites that can irreversibly react non-specifically with biological membranes and are observed as downstream products in excreta. The FDA and ICH guidances are focused primarily on the chemically stable metabolites observed in human circulation⁹. Chemically reactive intermediate metabolites will generally not be observed in the circulation in significant concentrations. However, these metabolites can often be observed in excreta as smoking guns, but they are not a major concern¹⁰.

In 1999, PhRMA commissioned a body composed of members of the Drug Metabolism, Clinical Pharmacology and Safety Assessment, PhRMA subcommittees, to review the topic and to present their findings at a workshop organised jointly by PhRMA member companies and the USFDA. Many of the issues were discussed at this workshop, which was held in Bethesda, Maryland, on 14–15 November 2000, together with the deliberations of the MIST committee¹¹. Multiple discussions were conducted regarding the problems of safety concerns of the drug metabolites and ultimately now the final FDA guidelines are issued for the metabolites in safety testing. Proposals have included metabolites with exposures that are:

➢ >25% of the exposure to the circulating drug-related material

➢ >1 0% of the administered dose or systemic exposure

➢ >1 0% of the systemic exposure of the parent drug at a steady state

➢ >10% of the total drug-related exposure and at significantly greater levels in humans than the maximum exposure seen in the toxicity studies⁹.

Issues pertaining to drug metabolites in safety testing are dealt with according to three major phases of drug development: the preclinical phase leading up to first human trials, the initial evaluation of the drug candidate in human volunteers for safety, tolerability and efficacy and the conduct of large-scale clinical trials in patient populations leading up to product registration. For each phase, the objective of the safety assessment programme is defined, the types of drug metabolism studies relevant to safety assessment are presented and the implications of drug metabolism for the safety programme are discussed.

The objective of safety studies is the preliminary evaluation of the toxicity profile of the candidate drug in two animal species, one rodent and one non-rodent, and to evaluate the potential for genotoxicity in support of Phase I safety and tolerability studies in healthy human volunteers. Before doing clinical trials, some information is usually obtained on metabolic pathways through a combination of in vitro and in vivo studies in animals and in vitro studies with human tissue preparations. Attempts are often made in the preclinical phase of drug development to identify the major metabolites of the candidate drug in various studies and to synthesise these metabolites for pharmacological testing. This type of information may provide guidance in deciding whether it is necessary to monitor certain metabolites, in addition to the parent drug, in preclinical safety assessment as well as in selected studies in the clinical development programme⁹.

For regulatory toxicology studies, conducted under good laboratory practices regulations designed to start the first human trials, it is customary to measure the parent compound in order to demonstrate dose-related exposure and to calculate potential safety margins.

Assessment of metabolite exposure in these studies is not routinely required and is generally impractical as only preliminary metabolism information is available prior to first introduction to man. The value of monitoring circulating metabolites pre-Investigational New Drug (IND) versus the resources required to do so before knowing the human metabolite exposure is questionable.

Pharmacokinetic data for the drug candidate in the animal species to be used for safety testing are important for designing appropriate toxicokinetic studies that aim to relate dose of the parent to systemic exposure. In determining how much metabolism data should be acquired to address drug safety, particularly with a compound that undergoes extensive biotransformation, the key consideration is whether the toxicology species are being exposed qualitatively to the same types of metabolites as seen in humans.

Attempts usually are made to predict metabolites that would be expected to appear in the human circulation before the start of clinical trials, based on the findings of in vitro experiments with human liver microsomes, slices or hepatocytes. Although the extent of systemic exposure to metabolites typically will vary across species, it is rare that drugs form truly unique metabolites in humans that are not detected in preclinical species.

However, if a metabolite is formed in the human liver preparations in vitro but is not observed in corresponding experiments with animal preparations, some consideration should be given to performing a genetic toxicology assessment of the metabolite. In some cases, this may include testing in an in vitro system in which the metabolite is present before conducting the first human trials.

Finally, in vitro genetic toxicity studies usually are completed on the parent compound prior to the first human studies¹². The capability of human-derived in vitro systems to adequately generate the metabolites that are important to humans in vivo ranges upto 70% and hence it is possible to miss 30% of the important metabolites¹³.

The overall objective of the preclinical safety programme in this phase of drug development, which typically may take up to 3 years to complete, is to support Phase I (safety and tolerability) and Phase II (efficacy) clinical trials. These studies normally are conducted in both normal volunteers and patient populations. Additional studies conducted during this phase of development include:

1. Genetic toxicology studies

2. Completion of appropriate dose-ranging studies that are needed to allow initiation of carcinogenicity studies in rodents

3. Reproductive toxicology studies

4. Assessment of the reversibility of any drug-induced changes.

The drug metabolism studies during the initial clinical evaluation of a drug encompasses a broad range of studies including synthesis and pharmacological evaluation of abundant metabolites identified in late preclinical and/or Phase I studies, in vitro evaluation of the human drug-drug interaction potential of the parent compound to facilitate decisions on the need for specific clinical drug interaction studies, cytochrome P-450 enzyme typing experiments to identify the isoform responsible for the formation of major (oxidative) metabolites and the development and implementation of sensitive and specific assays to support quantitative analysis of the parent drug and any metabolites that are considered important from either a pharmacological or a toxicological standpoint¹⁴.

Initial human studies are not typically conducted with radiolabelled material and thus biological fluids usually are evaluated only for the presence of metabolites that are suspected to be present on the basis of the results of the human in vitro findings or in vivo animal data. Perhaps, the most valuable information provided by drug metabolism at this stage of development with respect to the safety programme derives from the absorption, distribution, metabolism and excretion (ADME) studies. Human ADME studies provide similar information on routes of elimination, mass balance and metabolic profiles in plasma and urine. Results of preclinical and clinical ADME studies allow comparisons of:

This information will tell us whether there is any possibility of a unique metabolite in humans. It should also be emphasised that only major human metabolites of the drug candidate are important in human safety evaluation. The radiochromatographic profiles from human plasma will provide information of a quantitative nature on circulating metabolites. Metabolites requiring further studies will be decided according to the predefined criteria¹⁵.

For further evaluation of the metabolites, samples can be obtained from a limited number of patients in Phase II and/or III clinical trials also. When a metabolite is deemed sufficiently important to monitor in human studies, it also should be characterised in the animals used for safety evaluation. Whatever the quantitative definition, a major metabolite, if pharmacologically inactive, would not need to be monitored on a routine basis, but should at least be identified in each of the toxicology species and in humans¹⁶.

Long-term toxicity evaluation of the drug candidate, including carcinogenicity studies, needs to be completed in support of the New Drug Application submission for marketing approval. If it is necessary, based on the results of the Phase I and II clinical programmes, to monitor one or more metabolites of the drug candidate, appropriate bioanalytical assays should be developed to quantify both parent drug and metabolite in biological fluids from subjects in Phase III trials.

Topics of particular interest with respect to important active metabolites include inter- and intra-subject variability in metabolite exposure and alterations in metabolite area under cover (AUC) as a function of age, gender, renal and/or hepatic insufficiency, target disease state and drug-drug interactions. For drugs that exhibit moderate-to-high variability in pharmacokinetics or pharmacodynamics, genotype and phenotype information with respect to metabolite formation may be useful in fully understanding the behaviour of the drug candidate in the target patient population¹⁷.

While it would be somewhat unusual, based on the current trend to perform clinical ADME studies relatively early in development, to detect an abundant unique human metabolite during Phase III, it is nevertheless possible that the identification of all significant metabolites in plasma from the ADME study is not complete prior to the initiation of Phase III. In such a case, e.g. where the metabolic profile in human plasma is highly complex, the considerations discussed in the previous section would apply.

As discussed above, contemporary drug metabolism studies are conducted progressively earlier in drug development so that drug metabolism activities in Phase III tend to focus on completion of the clinical pharmacology programme required for registration. In this regard, drug metabolism data probably have less of an impact on the safety programme in Phase III than in the earlier stages of the overall development process.

Finally, while developmental and reproductive toxicology studies usually are performed only with the parent drug, it would appear prudent to establish that all major circulating human metabolites also are present in the plasma of at least one of the animal species, typically rats and rabbits used for such studies¹³.

There are many examples of reactive drug metabolites that have induced clinical adverse effects and led to drug withdrawals. The guidance suggest that, if possible, appropriate species should be found that will generate sufficient plasma concentrations of the metabolite to conduct safety testing or the metabolite will need to be synthesised and tested in appropriate preclinical species with associated toxicokinetic data¹⁶. The need for efficient identification of circulating human metabolites that may require separate non-clinical safety assessment early in the clinical development is a significant scientific challenge.

References