Common Clinical Pharmacology Mistakes to Avoid in Drug Development

Drug development is a complex process where clinical pharmacology data must be generated to support regulatory approval and marketing authorization. Each drug program has unique challenges and special circumstances, but programs often make the same types of mistakes. These mistakes can range from simple errors to large missteps that can threaten a drug’s chance for approval. This is particularly evident regarding clinical pharmacology, pharmacokinetic (PK), and pharmacodynamic (PD) data. One of the most common mistakes in clinical pharmacology drug development is simply not taking full advantage of what the field has to offer. Clinical pharmacology/PK/PD is sometimes thought of as a bump in the road or a town you must drive through to get to your final destination of collecting safety and efficacy data in late-stage studies. However, if done properly, clinical pharmacology can be a roadmap to guide the rest of the development program to a successful regulatory approval. Having supported hundreds of drug development programs over the years, below we provide an overview of common clinical pharmacology mistakes and errors to avoid in drug development.

General Clinical Pharmacology Mistakes

1. Assuming that a complete clinical pharmacology program only consists of first-in-human (FIH) single ascending dose (SAD) and multiple ascending dose (MAD) studies. There are many other clinical pharmacology studies needed to support an NDA or BLA.
2. Not considering the need for certain clinical pharmacology studies such as special population, drug-drug interaction (DDI), human mass balance (ADME), or food effect studies and not conducting them at the right time during drug development.
3. Evaluating PK in only 1 or 2 dose levels rather than a wider dose range.
4. Having sub-optimal Phase 1 dose-escalation stopping criteria – too strict or not strict enough.
5. Having a dose selection strategy that is not backed up by clinical pharmacology data.
6. Not determining human metabolite profiling early enough in Phase 1 studies to inform PK data collection and analysis in Phase 2 or 3 studies.
7. Missing opportunities in Phase 1 clinical pharmacology studies to support modeling and simulation/population PK in later phase studies. Population PK analyses conducted in Phase 2 and 3 studies are built upon the data collected in Phase 1 studies.

Pharmacokinetic and Pharmacodynamic Mistakes

1. Having an inadequate PK sampling schedule which can lead to inaccurate PK parameters (such as half-life) and the possibility of sub-optimal dose selection for future clinical studies.
2. Missing an opportunity to collect urine for PK parameters early in development.
3. Not considering relevant pharmacodynamic biomarkers (PD) early enough in the development program. Biomarkers help define the PD profile of the drug and early identification helps to determine if biomarkers need to be followed in Phase 2 and 3 studies.
4. Having an inadequate dose range by not dosing low enough or high enough and not exploring more doses overall.
5. Not collecting enough PK/PD data in a given population. For example, restricting the PK sampling schedule in an oncology program.
6. Missing opportunities to collect PK/PD data early in development to help avoid having to conduct studies later. For example, collecting rich time-matched PK and ECG data over a sufficient dose range in SAD/MAD studies may eliminate the need to conduct a full thorough QT study later in development by using a modeling approach called concentration-QT analysis.
7. Not having the proper blood sampling schedule for IV administered drugs by not collecting a sample at the end of IV infusion. The best practice is to collect a PK blood sample just prior to the end of infusion (e.g., within minutes before turning off the pump), and collecting multiple blood samples shortly after the end of IV infusion is also important. The PK parameters (i.e., Cmax) will not be a true representation of the drug if these best practices are not followed.

Modeling and Simulation Mistakes

1. Not including sparse sampling of PK and/or PD data in Phase 2 and 3 studies which can be used for population PK analysis.
2. Failing to collect rich PK and PD data in Phase 1 studies to support model development in later phase clinical studies.
3. Failing to obtain actual dose and PK/PD sampling times in CRFs for Phase 2 and 3 studies.
4. Not taking advantage of concentration-QT analysis by failing to enrich already planned Phase 1, 2, and 3 studies with time-matched PK and ECG samples.

Human Mass Balance (ADME) Study Mistakes

1. Failing to conduct an initial human metabolite ID screen in a Phase 1 study before going to Phase 2. It is important to have an initial assessment of the metabolite profile early in the development program.
2. Conducting a human mass balance/radiolabeled ADME study too late in a development program. It is important to identify active metabolites that may need to be followed in Phase 2 and 3 studies.
3. Starting the process to radiolabel the drug too late in the development program. Initial planning discussions for radiolabeling should start around the time of an IND filing.

Renal and Hepatic Impairment Study Mistakes

1. Failing to consider how model-based approaches can satisfy the renal and hepatic impairment dosing requirement. For example, a standalone renal impairment or hepatic impairment study may be unnecessary if data already being collected during Phase 2 or 3 studies are enriched for those populations over a sufficient dose range.
2. Not following the FDA guidance documents regarding reduced study designs for hepatic and renal impairment studies.
3. Discounting possible Phase 4 post-approval strategies.
4. Failing to monitor for metabolites in renal and hepatic impairment studies. For example, only evaluating PK for the parent drug without evaluating PK for active metabolites is a common mistake.

Drug-Drug Interaction (DDI) Study Mistakes

1. Not having sufficient in vitro CYP, transporter assays, or DMPK studies to inform potential clinical DDIs.
2. Not considering how DDIs might impact dose selection in efficacy and safety Phase 2 and 3 studies. For example, a DDI from a concomitant medication could result in lower or higher drug exposures, a negative efficacy signal, or create a safety issue.
3. Not building an overall DDI strategy using model-based methods from Phase 2 and 3 studies.
4. Not properly accounting for polypharmacy patients in the development program.

Conclusions

If utilized appropriately, clinical pharmacology data can be used as a road map to guide the overall clinical development program. It is important to generate the right amount of clinical pharmacology data at the right time. Having a strong clinical pharmacology strategy can not only save valuable time and resources but is also vital to achieving regulatory approval. Clinical pharmacology mistakes are common but are avoidable. The collection of relevant and high-quality clinical pharmacology data requires both robust protocol design and experienced study oversight. Careful planning with the help of a clinical pharmacologist or pharmacokineticist is crucial to avoid missteps. Allucent’s team of clinical pharmacology experts can help avoid mistakes and address gaps in a clinical pharmacology program. Contact us to learn how Allucent can help support your clinical pharmacology strategy and other drug development needs.