Mouse Vs Man – The Importance of Human-Based Approaches in Biomedical Research

Incorporating more relevant, human-based approaches (in vitro and in silico) in biomedical and toxicological research is a hot topic these days (see recent conferences organized by SOT, BioMed21, WC10, SACATM, and the ASCCT). While those of us developing in vitro systems including primary cell-based models, 3D assays, bioprinted tissues and microphysiological-based systems or “tissue-chips” (see Vincent 2015, Horvath 2016, Sistare 2016, Ingber 2016 and Berg 2017) have been convinced of the value of focusing on human-based systems, others may not be as aware of the limitations of animal models for understanding human disease biology.

This is important since these differences contribute to higher costs of pharmaceutical discovery, clinical failures, and delays of new medicines to patients. Areas of biology impacted by this include (1) response to DNA damage; (2) control of blood flow (vascular hemostasis); and (3) immune responses.

Response to DNA Damage.  Mice and rats, the favored species used in biomedical research, have a lifespan of less than 2 years, while humans live more than 70 years. Living for such a short length of time, rodents are less dependent on mechanisms supporting genome maintenance and integrity. Indeed, DNA mutation rates are higher in mouse than humans (Milholland, 2017) and expression of DNA repair genes is lower (MacRae, 2015). Humans are also more reliant on the immune system for constant surveillance to identify and eliminate cells containing genomic mutations. That the systems for managing DNA repair and controlling mutated cells in people are so different from rodents makes it hard to justify expensive and time-consuming two-year carcinogenicity studies often performed for new product regulatory submissions.

Control of Blood Flow (Vascular Hemostasis). The vascular system is responsible for bringing oxygen and nutrients to tissues via the blood, with the flow of blood controlled by the process of hemostasis. Blood flow is regulated by clotting (coagulation), fibrinolysis (clot lysis or resolution) as well as by constriction or dilation of the blood vessel wall, and is mediated by platelets, serum factors and enzymes within blood, as well as by smooth muscle cells and endothelial cells that line the vascular wall. The architecture of the vasculature of rodents and people are very different. Rodents are very small and their vascular beds are densely arranged, with characteristically short distances between the average tissue resident cells and the closest blood vessel (see here and here).   In contrast, the vasculature in human tissues is less dense and contains larger vessels that are subject to significantly higher vascular wall shear stresses.  Unlike rodents, humans are more vulnerable to blocks in blood vessel flow (heart attacks, pulmonary thrombosis, stroke) than are rodents. Thus, it is not surprising that rodent studies are not very good for predicting cardiovascular outcomes in people. Cardiovascular toxicity continues to be one of the leading causes of clinical failures (and a few years ago reported by AstraZeneca to account for 24% of clinical trial failures.

Immune Responses. Humans live in a challenging environment – exposed to infectious agents, bacteria and viruses from our interactions with the people and pets around us, the food we eat, and our external environment (taken a swim in a pool recently anyone?). In contrast, research animals are housed in highly standardized, clean facilities with filtered water and specialized food. In such facilities, protected from the external world, rodent immune systems remain naïve. So-called memory immune cells, the white blood cells that are generated upon exposure to antigens or infectious agents to fight and kill these invaders are more frequent in humans, while naïve immune cells predominate in laboratory mice. (See Mestas et al. for a laundry list of differences between mouse and human immune systems and also the transcriptomics studies by Seok et al.). In the field of cancer research, one can argue that the recent burst of activity in the area of immuno-oncology might have occurred many years earlier, but for the heavy bias of the research community towards the use of mouse cancer models. These xenograft models, in vivo transfer models whereby human cancer cell lines are adoptively transferred into immunocompromised mice, have been the accepted standard for demonstrating preclinical efficacy in oncology for many years. Obviously, since immunocompromised mice have no immune systems, they do not capture the relevant biology for “immuno-oncology”.

Given the many fundamental differences in the biology of rodents and people (not to mention the CNS), it is clear that new approaches are needed to move us beyond our dependence on animal studies if we are to improve pharmaceutical drug discovery and chemical safety testing. While research in mice and rats has its value, we owe it to patients and research funders to support approaches that provide better translation into human benefit. In vitro systems can and have been developed to model aspects of human-specific mechanisms. Let’s continue to support and expand these new innovations. Combining these with in silico methods through pathway-based frameworks such as the OECD’s Adverse Outcome Pathway effort, as discussed at the recent BioMed21 conference, will permit integration of these methods into decision making processes, helping us to achieve the goal of more successful and safer products at lower costs.

It will take a village to accomplish this. In the US, the NIEHS is on the right track with the Strategic Roadmap initiative (see here for an update from the most recent SACATM meeting).  This initiative, governed by the ICCVAM committee representing 16 federal agencies and supported by NTP’s NICEATM center, is “driving a national strategy to ensure the safe, effective, and timely implementation of human-based, predictive approaches in toxicity testing”. This effort to develop and implement data-driven approaches to provide better information at lower cost is supported by all stakeholders including industry, regulatory and academic groups as well as advocates of animal welfare.  To find out more, check out the Human Toxicology Project Consortium, the Tissue Chip program at NCATS, the Center for Alternatives to Animal Testing at Johns Hopkins (CAAT), the Computational Toxicology Communities of Practice at EPA or visit AltTox.org.