Janet Kelly
Janet Kelly

Biomedical research seeks to understand human physiology and disease risk, as well as myriad biological processes in humans—from the genetic level on up. Biomedical research also informs drug development. Especially valuable findings may emerge from in vivo human studies. However, relying exclusively on human in vivo studies is neither morally ethical nor scientifically feasible. Often, scientists must turn to animal models.

Drugs used to treat Alzheimer’s, cancer, diabetes, hepatitis B and C, HIV infection, Parkinson’s disease, leukemia, and various mental health conditions would not be available without animal research.1 This is only a shortlist. Indeed, 83% of Nobel Prizes awarded for “Outstanding Contributions to Medicine” since the award’s inception have involved using animals for research.2

Mice, rats, minipigs, dogs, and nonhuman primates are some of the animals that play invaluable roles in developing drugs that support human health and well-being. Research that relies on animal studies has contributed to countless medical breakthroughs, and as technology and scientific knowledge stands today, the reliance on animal studies will not end soon. Nonetheless, scientists are increasingly concerned about animal welfare. They are looking for ways to continue their research while supporting a commitment to refining, reducing, and replacing animals in research.

Species selection 101

Regulatory guidelines typically require one rodent and one nonrodent species when testing drug candidates. However, these guidelines often accommodate exceptions. (For example, large molecule biotherapeutic candidates may be capable of engaging relevant targets in nonhuman primate models, but not in ordinary rodent models.) In any case, the species suitable for any given study will depend on several factors.

Investigators lean on drug metabolism and pharmacokinetic characteristics to determine which animal species to use. These (mostly) in vitro investigations try to identify similarities in reactions to test materials at the cellular level. For example, if compound X produces metabolite Y in a human being, it is essential to know which animal species produce the same metabolite.

Other critical dosing information is also derived from in vitro testing. For example, investigators need to know if biological mechanisms within certain species could potentially inhibit or increase absorption, thereby increasing human safety and/or toxicological concerns.

Physiological characteristics also contribute to species selection. Scientists know that physiologically, skin and heart of minipigs are more like those of human beings than are those of other species, making minipigs the best choice for testing dermally applied compounds or cardiac drugs. Similarly, a nonhuman primate’s immune system is closer to a human’s immune system than is a minipig’s or canine’s immune system; thus, nonhuman primates are frequently used in testing drugs that act on the immune system.

The bottom line is that investigators are deliberate when conducting animal and human testing. They employ a step-by-step process that applies lessons learned during testing—whether it involves in vitro studies or Phase I trials. The idea is to develop refined versions of the original tests, and then to run them on test populations in subsequent testing phases.

Respectful animal research

Drug development continues to require animal research, but all researchers now realize the importance of prioritizing animal welfare. Refinement of animal experiments is where big differences in welfare can be easily made by all researchers.

Housing and handling of animals has changed significantly in the last two decades, and refinements are introduced every day. Housing animals together (that is, group housing) is the global standard, but enclosures should be constructed based on the needs of the species that are being housed. The construction (materials) and dimensions (design and arrangement) of the enclosures must allow natural movement and behavior; enclosures that are too small or stacked to maximize capacity are unacceptable and contrary to animal welfare guidelines.

Multilevel or stratified enclosures may provide more natural environments for nonhuman primates or rodents, but the same cannot be said for dogs or minipigs. Likewise, gridded flooring may be helpful for maintaining hygiene, but it discourages natural movement and is quite uncomfortable for most animals. Having solid floors lined with sawdust or other absorbent materials is one way to soften the environment and support animal welfare.

Similarly, training staff to handle test animals in ways that minimize discomfort during testing is paramount. Training the animals to cooperate with the technician for a scientific procedure leads to a less stressful experience for both the animal and the technician. Reducing the number of invasive procedures is another way to reduce animal stress and improve animal welfare.

Let’s say a laboratory needs to conduct an experiment that requires multiple blood samples (an invasive technique) to track a compound over time. Instead of inserting a needle every 30 minutes, a catheter is inserted to aid in withdrawing blood as often as needed. That means 1 needle stick instead of 10.

There are also analytical techniques today that were not previously available. Identifying and monitoring certain biomarkers provides a clearer picture of the minuscule changes happening inside the organism, but at far lower dose levels than would be needed for these changes to be found via histology or necropsy. Animal welfare technicians who work with the animals daily are also ideally placed to recognize subtle, seemingly insignificant changes that a research scientist or investigator might miss. And enriching feeding time by creating foraging scenarios or providing frozen fruit can offer test animals additional stimuli and promote their natural behavior.

Researching with care and compassion is not only ethical, but also good science. The reality is that poor conditions can significantly impact data quality due to their physiological effects on animals. For example, stress can change cortisol levels, impact white blood cell counts, and inhibit food consumption, which can affect weight and behavior. Add a drug candidate to that scenario, and it has a knock-on effect.

Data derived from unhappy, stressed, or poorly treated animals will be inherently skewed as it does not reflect that animal’s normal physiology. This can significantly impact absorption, distribution, metabolism, and elimination data and call into question any safety/toxicological testing based on that data.

Stopping animal use altogether

Biomedical research as a field of study would crumble if it were to discontinue using animals completely today. The foundation of animal welfare in drug development is built on replacing animal studies, but it also places great value on overall reduction and refining existing techniques. The industry is taking great strides to reduce and refine animal use; however, replacing animal studies altogether will be far more challenging.

In silico techniques could one day help replace animal studies, but they have a long way to go. These techniques aid drug development by allowing scientists to answer pharmacological questions using databases, data analysis, and machine learning.

Put simply, in silico techniques use computers to model pharmacokinetic responses. The concept has been around since the 1990s and continues to provide exciting breakthroughs, but it is fundamentally limited based on the availability of large datasets and the accuracy of bioinformatic algorithms. However, in silico technology has proven to be valuable during early development as a way to identify and characterize promising compounds.

Organ-on-a-chip (OOAC) is also an emerging technology that could eventually help replace animal studies, but it remains a distant goal. OOAC uses a combination of cell biology,
engineering, and biomaterial to mimic human tissue’s structural and functional characteristics and to enable predictions of human response to drug compounds and other stimuli.3 OOAC has already been used to study the physiology of the human heart, liver, lungs, intestines, and kidneys, but integrating multiple organs—that is, an entire human body—into a single chip for analysis has presented scientific and financial challenges.

Both in silico and OOAC technology demonstrate great promise in replacing animal studies in drug development in the coming years. For now, however, the traditional combination of in vitro and in vivo studies and human clinical trials provides unmatched insight into drug response and development complexities.

As technologies evolve and data accumulate, it is expected that animal tests will be refined, and that significantly fewer animals will be used.

Holding drug developers and sponsors accountable

Regulators in Europe, the United States, China, and Japan all have minimum standards for animal welfare, encompassing the principles of the 3Rs—reducing, refining, and replacing animals in drug development. Different regulatory bodies may interpret animal welfare differently and issue different guidance documents, but these bodies are sure to ask drug developers and sponsors how and why they used specific animal models.

IND/NDA applications that lack proper, science-based justifications for animal use will likely be rejected or placed on hold until acceptable rationales are provided. Such is the state of the regulatory landscape on this topic.

While most laboratories today have social responsibility directives to support and shareholders to satisfy, drug sponsors and developers also hold each other accountable. Their companies may compete in the marketplace, but sponsors talk to each other at industry events, conferences, and meetings. Laboratories that eschew regulatory guidance or provide substandard conditions for animals can—in addition to receiving regulatory sanctions—earn the ire of their industry colleagues and lose their customer base.

Final word on animal welfare

The time and resources needed to train staff and improve the living conditions of research animals can be significant, but the potential for scientific advancement far outweighs the initial cost. When a drug candidate is held up or fails due to poor quality data, everyone loses. That is why the industry’s participants—sponsors, developers, scientists, and technicians—need to do everything it can to deliver the highest standards of animal welfare to ensure that safe and effective therapies make it to market. Animal welfare is a small but fundamental part of drug development.


1. Foundation of Biomedical Research. Animal Testing and Research Achievements. Accessed February 7, 2023.
2. National Association for Biomedical Research. The Importance of Animal Research. Accessed February 7, 2023.
3. Wu Q, Liu J, Wang X, et al. Organ-on-a-chip: recent breakthroughs and future prospects. Biomed. Eng. Online 2020; 19(1): 9. DOI: 10.1186/s12938-020-0752-0.


Janet Kelly has spent more than 30 years in scientific and operational roles in toxicology at multinational contract research organizations, culminating in a year in China overseeing the construction, commissioning, and opening of a new nonclinical facility.




Previous articleBook Review: The Exceptions
Next articleSingle-Use Pumps in Biopharmaceutical Manufacturing