Human-Based Platforms in Biomedical Research

Human-Based Platforms in Biomedical Research

Exploring alternatives to animal research to understand biological processes in human health and disease

Michelle Dotzert, PhD

Animal research has contributed greatly to our understanding of biological processes in health and disease, and has led to numerous medical discoveries—from endocrine effects of exogenous insulin to the curative effects of penicillin and the polio vaccine, among numerous others. As such, animal models have become the gold standard approach in biomedical research. However, these studies are costly and time consuming, and a large portion of pre-clinical research does not translate successfully into clinical trials. Alternatives to animal methods, including a combination of in vitro, in silico, and in vivo methods, are being employed to create more accurate models of human disease, reduce the number of animal lives lost, and accelerate drug development.

Understanding the challenges and limitations of animal research

According to Charu Chandrasekera, PhD, founder and executive director of the Canadian Centre for Alternatives to Animal Methods, some animal research has undeniably been a valuable contribution to the life sciences, but many studies have actually misled scientists in their efforts to understand and find treatments for disease in humans. “Given how many genes and pathways discovered in animal models have not played a significant role in a human physiological or pathophysiological context, it is time to change the way we unravel human biology,” says Chandrasekera.

It can take 10 to 12 years for a new drug candidate to progress from initial testing to final FDA approval and launch on the market. Despite increased spending on drug development, including costly and time-consuming animal studies, the number of substances that reach market launch is decreasing, and the pre-clinical to clinical translation failure rate is 95 percent.

According to Chandrasekera, “these translational discrepancies are primarily due to immutable interspecies differences that occur at every level of biological action—from genes to whole organism level—further confounded by biological variability, and even differences in housing and husbandry practices, to say nothing of experimental conditions.”

Another challenge lies in selecting an animal species that will mimic the human health or disease state, as well as the response to therapeutics, as some species are resistant to diseases to which humans are susceptible. The SARS-CoV-2 virus is one such example. Preliminary findings suggest susceptibility to the novel coronavirus does not match phylogenetic relationships. Mice, for example, are not susceptible to SARS-CoV-2 infection, despite being phylogenetically more similar to susceptible species (including humans) than non-susceptible species.

There are also a wide variety of real-world factors that influence disease pathogenesis, which are not reflected in animal models. According to Chandrasekera, without factoring in genetic variability, co-morbidities, medications, lifestyle, socio-economic factors, and even environmental and geographic variations, “we are only capturing a snapshot of a human disease-like state.”

Creating a more accurate model of human disease

There is no single approach that can replace animal models with an ideal human model. Creating a more accurate model of human disease requires “integrative, human biology-based hypothesis generation and experimental approaches—starting with human clinical observations—to make humans the quintessential animal model,” says Chandrasekera. In this way, answers are obtained in segments, and integrated to create a complete picture.

This approach includes a combination of in vitro, ex vivo, in silico, and in vivo methods that may be used to determine variations and resilience at the -omic, pathway, cellular, tissue, organ, and systems level.

Organoids, for example, are in vitro three-dimensional cell aggregates derived from primary tissue, embryonic stem cells, or induced pluripotent stem cells that are capable of self-organization and demonstrate similar function to the tissue of origin. They have been used in disease modeling, drug screening, and efficacy testing, among other applications.

“For gene function in particular, you can utilize patient-derived stem cell-based approaches to study ‘human knockouts’ (with naturally occurring mutations), and by recreating disease-in-a-dish amenable to genetic manipulation to make relevant discoveries that will advance effective therapeutics for humans,” explains Chandrasekera.

Other alternatives for modeling human disease in vitro include three-dimensional bioprinting, and organ-on-a-chip microphysiological systems. Three-dimensional bioprinting is a technique by which living cells and their extracellular matrices are carefully positioned or “printed” layer-by-layer or in a mixture to create a construct with cytoarchitectural, biochemical, physiological, and mechanical properties similar to natural tissues, with the potential to provide more relevant, detailed insights into disease mechanisms and drug effects. Organ-on-a-chip is a minimally functional unit of a complex organ modeled on a platform the size of a thumb drive where living cells are cultured and maintained in one or more micrometer-sized chambers with fluidic channels that reproduce blood and/or air flow just as in the human body. “This technology is revolutionizing the biomimicry of physiological processes at the single organ level (e.g., liver, kidney, lung, skin, intestine)—poised to bridge the in vitro-to-in vivo data gaps when multiple organs are interconnected as human-on-a-chip,” explains Chandrasekera.

There are also multiple in silico methods emerging as important tools in drug development. Physiologically based pharmacokinetic modeling, for example, integrates drug and physiological information into a mathematical model to predict absorption, distribution, metabolism, and excretion. Similarly, in vitro to in vivo extrapolation enables the transposition of in vitro observations such as concentration-effect curves, into in vivo dose-response curves to predict a response in a biological organism. Artificial intelligence technologies are also being used to identify potential candidates in drug discovery, and predict toxicity and other properties of novel molecules.

Several in vivo techniques such as non-invasive imaging, physiological studies, microdosing, and epidemiological studies are also valuable alternative methods.

While powerful, these methods may not be sufficient to serve as standalone alternatives to animal research at this time. Organ-on-a-chip alone, for example, may not yet be sufficient to replace animal models, but when combined with complex tissue models, in silico modeling, and other methods, it is possible to significantly reduce the number of animals used at this stage. “In essence, what it requires is a different mindset, other than the predominant ‘let’s just knock it out and see what happens’ line of scientific inquiry,” says Chandrasekera.

Another notable benefit of this integrated approach is that the information gained can be supplemented with growth, development, exposure, and lifestyle factor information. This supports a more complete understanding of the molecular and physiological mechanisms underlying human disease.

The future of alternatives to animal research

The adoption of alternative methods may be challenging, but is not impossible. Chandrasekera explains that it is not about finding a single, all-encompassing method or system that applies to every disease condition; rather, it is about being creative with animal-free methods to formulate and test hypotheses, fit for purpose. “The single biggest challenge is the cultural mindset…we have a scientific culture ingrained in animal research,” she says.

Laboratory animals are used for drug testing, developing medical procedures, vaccine production, diagnostics, and more. As their use is widespread and well established, the adoption of alternative methods will necessitate robust, reliable options. Some of these options are already available, and others are in development.

Chandrasekera believes that these methods must also be utilized to change our view of human biology. “If we utilize alternative methods to view human biology as an integrative framework—from genes to organism—then it is possible to predict human outcomes with higher relevance and accuracy,” she says.

Numerous companies and organizations are working toward implementing alternatives to animal methods, and the United Kingdom and United States have national road maps and legislation (including the Environmental Protection Agency’s amendment on the Toxic Substances Control Act, and the European Union Directive 2010/63 that calls for ultimate replacement) for developing and adopting non-animal technologies to shift away from animal testing. However, the use of animals for research purposes is a highly contentious subject, and some express concern at the prospect of phasing out animal studies altogether.

In discussing the challenges and limitations associated with animal research, Chandrasekera provides some food for thought: “Honestly, when you need 47 mouse lines to recreate human Ebola, that is not a huge scientific achievement to be touted in a high-profile journal; it is a compelling testament to our misguided approach.”

The topic of animal methods in research is complex, and requires both scientific and ethical considerations. Regardless of whether one is in favor of or against animal methods, perhaps a more pertinent question is whether there are better ways to understand human disease pathogenesis and develop effective treatments.