Is it time for a new tool box for drug research and development?

Prof James Fildes and Amy Stewart

In this blog, Professor James Fildes, CEO and Chief Scientist and Amy Stewart, Research Technician at Pebble Biotechnology Laboratories argue that the significant difference in human to mice gene expression in various tissues could explain why many drugs that are successful in rodents fail in humans.

Investment in pharmaceutical R&D is at a record high, with $238 billion spent in 2021 alone. Yet, despite the increased investment, the number of new drugs approved per billion US dollars is halving every 9 years.

There is increasing recognition - and some very significant evidence - that the R&D process has been obstructed by decades of unsuitable laboratory models, impeding clinical translation. Small animals offer a high-throughput and cost-effective model for preclinical testing. However, many reveal fundamental human differences which limit their use in medical research. Compared to humans, rodents exhibit distinct differences in drug absorption, metabolism, biodistribution and excretion.

Drug toxicity is also non-comparable as rodents may tolerate drugs that elicit significant toxic responses in humans. Conversely, various drugs have elicited toxic responses in rodents, whilst demonstrating safety in humans.

There are also major immunological and inflammatory disparities, with distinct divergence in rodent vs human immune systems. Despite this, mice and rats are routinely used in immunological research and drug development. The situation is further exacerbated at the genomic level. Using the kidney as an example, only 10% of the 4,644 kidney-related human genes are conserved in the mouse. In fact, in all tissues, gene expression differs significantly between mice and humans. Collectively, these commonly overlooked differences help explain why so many drugs that are successful in rodents ultimately fail in humans. The perception that ‘an inadequate model is better than no model at all’ is based on scientific familiarity and a reluctance to move away from poor-quality small animal models.

Large animal models, including non-human primates and pigs, offer superior scientific fidelity to humans but are prohibitively expensive, time-consuming, and ethically challenging. The pharmaceutical industry is aware that the current approach is contributing to late-stage, expensive trial failures, and that these inadequate testing models are a lost opportunity in progressing therapies into clinic, ultimately improving patients' lives. There needs to be a monumental shift in the R&D culture, starting with the development of a new research toolbox.

Living Organ System Animation
Living Organ Systems - PBL (

Developing and validating in-silico and advanced organ-on-a-chip systems could play a critical role in de-risking therapies prior to clinical trial commencement. More recently, physiological ex-vivo systems have been developed, that have the potential to improve translation and accelerate drug development.

Ex-vivo organ perfusion systems were first developed to improve organ preservation prior to transplantation. More recently, another application for this technology has become apparent; its use as a testing platform to accelerate the development of new therapies, protocols and medical devices, without the need for laboratory animals.

Several groups have developed ex-vivo perfusion protocols using pig organs and tissues sourced from the food industry. Organs are placed into organ chambers and connected to a circuit that recreates the in-vivo environment. As oxygenated, nutrient-rich blood is pumped across the vasculature, organ function and metabolism are restored. For example, kidneys perform gluconeogenesis, converting metabolic waste products to glucose and bicarbonate and excrete toxins via urination; Livers produce enzymes and bile; lungs perform gaseous exchange; hearts beat; the limb consumes oxygen and ATP, and releases metabolic waste products.

Most organs can be maintained in optimal health for 24-48 hours, providing sufficient time to test new therapies and drugs, without the need for regulatory approvals. Arterial and venous blood biochemistry, co-oximetry, metabolic state and electrolyte balance can be easily monitored, generating highly translational and clinically relevant data. As full organ function is restored with a complete immune compartment, the immunological and inflammatory response to a drug or intervention can be evaluated in real-time. Furthermore, the use of whole isolated organs offers a far superior physiological model than organoids and organ-on-a-chip approaches, with native gene expression and entire cellular structures in place.

The Pebble team have been developing ex-vivo perfusion protocols (initially for organ transplantation) for over 20 years. Their unique holistic approach aims to recreate the in-vivo environment as closely as possible, with the goal of ‘tricking’ the tissue to behave as if it is still present inside the body. Multiple organs can be combined with one blood supply (a video of the system – referred to as LIVING-ORGAN systems). This recapitulates the dynamic interactions and crosstalk between organs and the immune system. Pebble’s systems can be used to evaluate safety, toxicity and efficacy and enable immune profiling, dosing, biodistribution, pharmacokinetic and dynamics.

Using this approach, Pebble can accelerate innovations into the clinic, without the use of laboratory animals. These next-generation platforms support and accelerate preclinical R&D at any stage of the pipeline, from de-risking assets, accelerating internal R&D via non-IND reportable data, and creating FDA/MHRA/IND submission-ready data packs to support regulatory approval for a clinical trial.

The scientific community are beginning to recognise that the time has come to throw away the rusty old R&D toolbox and replace it with next-generation tools. Ex-vivo approaches are not the sole solution to drug development but should be considered an essential tool within the new R&D drug development toolbox.

More within