Diversity in clinical trials: Optimizing drug development strategy, leveraging data and increasing efficiency with modeling and simulation.

By Lisa Benincosa and Katy Moore

Clinical trials have evolved over the years with inclusion of broader patient populations; yet, additional efforts are needed to better understand the safety and efficacy across the diversity of patients who will be using the medicine. Factors such as race, gender, preexisting conditions, and age can all impact how a drug will be metabolized, absorbed, and eliminated out of the body. Diversity within the patient population is important in drug development because drugs can behave differently from one population to another.

There are many approaches to increasing diversity in clinical trials. Drug developers, regulatory authorities, patients, and the medical community all have a role in achieving diversity in clinical research. Patient-centric bioanalytics and digital health technologies can help increase trial accessibility, while the use of real-world evidence and innovative study design can expand the utility of data sources.

Model-informed drug development (MIDD) where the integration of data and models from the non-clinical and clinical programs as well as data from other relevant external research resources can also provide insights into recommended dosing for broader patient populations or even in patients that may not have been studied explicitly (e.g. adolescents). The potential arises to improve efficiency by extrapolating an understanding of one population’s characteristics and applying that to others with known characteristics.

Setting the framework for a clinical study: Clinical pharmacology development plan

Early in the drug development process, defining the target product profile for the specific indication and patient population sets the framework for the clinical development study strategy. Who will be using this product, and what characteristics are likely to affect their response? How will data be collected to ensure proper representation of the target population? These questions are fundamental to protocol design.

Further, thinking ahead about commercialization and local markets from the outset is essential. For example, if the intent is to conduct clinical trials in the U.S. and submit a U.S. NDA, that evidence package might be inadequate for later plans to market in APAC. In this case, considering characteristics of the Asian population compared to U.S. subjects should be foremost in the study.

As part of drug development, the main principles of the clinical pharmacology development plan are to understand what the body does to the drug (pharmacokinetics [PK] or exposure) and what the drug does to the body (pharmacodynamics [PD] or response). Together, PK and PD inform the dose for a patient population. The clinical pharmacology development plan includes a strategy for collecting PK and PD data in various patient populations and a modelling plan to maximize the value of the diverse clinical data collected.

Integrating data across clinical studies and using modeling to understand patient variability continues to be the standard in defining the dose across a broad range of patient factors. These factors are both intrinsic (e.g., sex, race, weight, disease factors, genetic factors, hepatic, or renal impairment) and extrinsic (e.g., concomitant medications, diet, smoking).

Using PK/PD modeling tools to support trial designs, dose, and diversity in clinical trials

Using modeling and simulation approaches from all phases of the drug development program can facilitate a better understanding of patient variability from diverse populations. These approaches include population PK modeling, physiologically based PK (PBPK) modeling, allometric scaling, quantitative systems pharmacology (QSP), simulation of various clinical scenarios, and more.

To assess the probability of success in developing a medicine, leveraging a range of quantitative approaches are used to inform decision-making, including extrapolating data from large populations with similar characteristics to provide supporting evidence for safety, effectiveness, and dosing.

MIDD includes bridging and extrapolation approaches of the scientific and clinical data to recommend dosing to broader patient populations or even in patients who may not have been studied explicitly. This way, MIDD can greatly improve the efficiency of the development program. In addition, MIDD can be brought in midway in the study to make evidence generation more robust.

For example, MIDD has supported optimal trials designs in pediatric drug development. MIDD approaches help drug developers understand growth-related factors that can influence a drug’s pharmacology, thus optimizing the number of patients and duration of a study, minimizing the sampling burden for PK, biomarkers and/or endpoints, and much more.

In short, MIDD and prospective clinical pharmacology and modeling planning can help researchers better understand the science and optimize development programs. MIDD supports sponsors striving to increase diversity, not only to meet industry mandates and develop effective treatments efficiently but to be responsible to communities and patients needing access to treatments.