Answers are grounded in this paper and cite their sources.
Drug development is expensive. Bringing a new drug to market costs between 300 million and 2.8 billion US dollars, depending on the therapeutic area.1 Despite continuous biomedical and technological advances across all sectors of drug development, success is still low and economic risks remain high. Early-phase clinical pharmacology (CP) studies are critical in defining how humans handle (pharmacokinetics) and tolerate (pharmacodynamics) potential new drugs. They are relatively simple compared with later-phase clinical trials but provide essential data to derisk development, to inform dosing, and to help with “go/no-go” decisions. Examples include first-in-human (single and multiple ascending doses), food effect, bioequivalence or relative bioavailability, thorough QT, and pharmacokinetic drug–drug interaction (DDI) and gastric pH–dependency studies and, increasingly, small patient cohorts (Phase 1b) for an early read on potential efficacy. Clinical pharmacology studies typically recruit healthy subjects at one research site, and should be logistically straightforward, rapid, and relatively inexpensive. The cost of CP studies has increased in parallel with overall drug development cost due to an increased complexity of studies and greater regulatory oversight following the TeGenero (2006) and Bial (2016) Phase 1 trial disasters. The COVID-19 pandemic also intensified competition for study subjects and for research sites less affected by the infection. Notably, the clinical research organizations (CROs) that run CP studies for sponsors are subject to intense regulatory scrutiny, which understandably manifests as a logistical focus for compliance rather than the pursuit of novel approaches. Efficiency in the planning, execution, monitoring, and reporting of CP studies has not kept pace with innovation in other sectors of drug development, for example, model-informed drug development (MIDD) and electronic regulatory submissions. Thus, CP studies are ripe for disruptive innovation to accelerate drug development, allowing for earlier identification of drugs with low potential for success—the “kill fast” concept that saves money.2 In this commentary, 10 suggestions are made to improve the efficiency of CP studies. Many protocols for CP studies are unnecessarily long, overcomplicated, and time consuming to write. Where possible, protocols should be shortened without compromising participant safety or increasing overall risk. In most cases, study subjects are healthy volunteers, which, by definition, are people without clinically significant past or current medical histories. Extensive lists of inclusion and exclusion criteria to cover every permutation of “healthy” are unnecessary when doctors conduct robust screening of volunteers to identify preexisting medical conditions. Other key points include to summarize rather than duplicate information provided in the investigator brochure; schedule of assessments in tabulated form only; inclusion/exclusion criteria essential for study conduct; reference standard definitions, such as those for adverse effects, rather than providing details in full; and standardization of on-study routine clinical data such as vital sign and electrocardiogram ranges. In first-in-human ascending-dose studies, the inclusion of subjects receiving placebos and the blinding of subjects and/or study staff significantly increases cost and complexity. Arguably, both are insufficiently justified from a scientific perspective.3 Cohorts of subjects are small and therefore underpowered to adequately determine significant differences in effects between placebo and treatment groups (typically a 2:6 placebo-to-drug ratio in ascending-dose studies). Importantly, as discussed by Parasrampuria and Benet, placebos and blinding add little value to decision making by data monitoring boards, for whom the primary question is whether to escalate to a higher dose or not.3 It may also be unethical to submit subjects on placebos to unnecessary procedures and the potential harms of study participation. Open-label study designs without placebo controls should become the standard for CP studies unless adequately justified. In some cases, multiple CP objectives can be achieved under the same protocol to replace additional studies and more efficiently use resources. Examples include crossover of subjects to a food effect arm following fasting in a single dose cohort (at a dose relevant for later development); rollover of subjects after a single dose into multiple dosing; matched timing of vital signs and electrocardiograms to pharmacokinetic blood samples for cardiovascular safety, particularly QT assessment; and study of the new drug as a perpetrator of pharmacokinetic DDIs toward the end of multiple dosing using cytochrome P450 enzyme selective substrate cocktails.4 Some aspects of CP studies, such as male-only participation, are relics from the past that should change to accelerate enrollment and to more broadly understand pharmacokinetics and tolerability.5 Suggestions are greater inclusion of females of nonchildbearing potential and females with adequate contraception, increased upper age limits to allow participation of healthy older adults (e.g., up to 65–70 years of age), active recruitment of traditionally underrepresented groups (e.g., racial/ethnic backgrounds, rural locations), and increased upper weight limits to reflect heavier populations (e.g., body mass indexes up to 32 kg/m2). Pharmacokinetic and exposure–response models from animal data are extrapolated to determine dose levels in first-in-human studies based on pharmacological approaches (e.g., minimum anticipated biological effect level), rather than more crude toxicological methods (e.g., no observable adverse effects level).10 A “predict, learn, and confirm” cycle is used to accelerate dose escalation in first-in-human studies based on model-predicted exposures calculated during the study.11 Adaptive study designs allow fewer cohorts to attain target concentrations that are meaningful for later testing in proof-of-activity Phase 2 studies. Clinical pharmacokinetic DDI studies are replaced by physiologically based pharmacokinetic predictions for “victims” and “perpetrators” of cytochrome P450–mediated interactions.12 Virtual bioequivalence and food-effect studies for new formulations are used to replace clinical studies.13 Many institutional review board (IRB) members, although well intentioned, lack training and experience in CP, the translational nonclinical-to-clinical interface, and the realities of drug development, resulting in ill-guided reviews or reviews narrowly focused on areas of personal academic/clinical interest. At best, critiques that are irrelevant to the benefit–risk assessment are frustrating but easily refuted, and at worst, they are unnecessarily obstructive and detrimental to timelines and costs. Expertise in translational CP is required to adequately assess the benefit–risk of CP studies, particularly first-in-human studies. Those with hands-on experience in this area, most of whom work in industry rather than academia, should be incentivized to join IRBs and be remunerated appropriately. The monitoring of CP studies by CROs in “real time” is unnecessarily labor and resource intensive for study sites and investigators. An example is the reporting of all adverse effects, independent of clinical significance, shortly after they commence back to sponsors via online platforms. This is meant to provide additional clinical oversight, but it regularly causes unnecessary concern for sponsors and wasteful follow-up work. Some adverse effects take time to evolve and medically assess correctly, for example, elevated liver enzymes. Furthermore, clinical significance, severity, relatedness to treatment, and any dose–response relationship, may become apparent only following the completion of multiple cohorts. When significant adverse effects do occur, sponsors should be immediately contacted by investigators directly to provide a detailed narrative, review management plans, and discuss consequences for the study. During the running of CP studies by CROs, large volumes of data are often subject to unthoughtful quality control that is procedural in nature rather than practical and informative for decision making. Obsessively curating data judged as not clinically significant by investigators, such as parameters slightly outside reference ranges, and the repetitive manual signing of paper documentation for not clinically significant results, are examples of time wasting that dramatically increase quality control burden. Clinical units that specialize in CP studies therefore require many employees who, although often well educated, do relatively menial work, losing the excitement of drug development. Their potential is better realized elsewhere within the system, such as the accurate monitoring and recording of adverse effects (high-quality rather than high-quantity data). Investment in clinical trial assistants and monitors with previous health care experience helps with the recognition of clinically important data and improves overall quality. An increasing trend in first-in-human studies is to include a very small patient cohort (less than 20) following the dose escalation parts. The argument is to achieve an early read on efficacy to inform Phase 2 and raise capital. However, since these cohorts are so underpowered, it can be an expensive, time-consuming, and uninformative high-risk strategy unless success is predefined by highly sensitive biomarkers of potential efficacy. Removing patient cohorts entirely from first-in-human protocols is the preferred option in many cases, allowing more rapid achievement of Phase 1 objectives (safety, tolerability, and pharmacokinetics), writing of the clinical study report, and progress to adequately powered Phase 2 studies. While a few countries have dedicated specialist training programs for doctors in pharmaceutical medicine (e.g., Faculty of Pharmaceutical Medicine in the United Kingdom), most do not, and rely on doctors trained in other specialties to act as principal and subinvestigators for CP studies. Many of these doctors have little or no training in CP and lack the knowledge and experience to fully engage in drug development programs with sponsors. There is considerable scope for more pharmaceutical medicine training programs globally.14 Doctors with experience in translation CP should teach and mentor junior colleagues to be intellectual rather than purely operational. Given the huge expense of drug development, this commentary makes suggestions to improve the efficiency of CP studies to lower cost. Smarter study designs and streamlined protocols are encouraged. Data should be high quality to inform decision making, rather than just high quantity. Closer direct collaboration between sponsors and investigators is urged to increase flexibility and to fully realize the benefits of MIDD. The education and training of drug development medical specialists is critical. Flexible, rapid, and inexpensive CP studies will accelerate drug development so that new drugs reach patients more rapidly at the right doses with manageable cost. Thomas M. Polasek is an employee of Certara, and Virna Schuck is an employee of Ribon Therapeutics Inc. There was no specific funding for this article.
No takes yet. Share an insight, caveat, or question.
Polasek et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: