How viable are alternatives to animal testing in determining the toxicities of therapeutic drugs?

KEYWORDS:

• Alternative test methods
• animal tests
• drug efficacy
• drug safety
• human variation
• integrated testing
• interspecies differences

1. The problem

There is a very high rate of withdrawal of new drugs during or even after clinical trials, due to lack of efficacy or adverse side-effects not identified during expensive and time-consuming preclinical testing, much of it involving laboratory animals. For example, BioMedTracker analyzed the approval success rate between 2006 and 2015 of 9,985 clinical new drugs from 7,455 development programs in 1,103 companies, and found that only 9.6% of them progressed from Phase I via Phases II and III to FDA approval [1]. Toxicity has been suggested as the reason for failure in approximately 20-25% of cases [2].

2. The reasons

Models can only be useful, if enough is known about what is being modeled and about the models themselves – our current understanding of mechanisms of toxicity is inadequate, so it cannot be known if, and if so, when, laboratory animals can be acceptable models for humans.

In addition, the danger of the high fidelity fallacy is poorly appreciated – i.e. that despite general similarities between animals and humans, major interspecies differences inevitably mean that data from animal tests cannot be a relevant and reliable basis for predicting effects in humans [3,4]. Crucially, this applies to the most important enzymes involved in drug metabolism [5,6].

The current reliance on animal procedures also limits progress on understanding the human diseases which need to be treated. Again, not enough is known about particular human conditions or about the animal disease models purported to be equivalent to them.

Also, there is no such thing as a typical ‘human’ – most testing involves healthy, young male volunteers, but most of the population needing therapies are not healthy, young or male. Insufficient attention is paid to the significance of age, sex, race, occupation, lifestyle, and previous or concurrent illness or therapies – against a background of immense human genetic variation.

For these reasons, when we analyzed the toxicity data for over 2,000 drugs, which had been collected and classified independently by Instem Limited, we were not surprised to find that the calculated Likelihood Ratios showed that the absence of toxicity in rats, mice, rabbits, dogs and non-human primates provided little or virtually no evidential weight that adverse drug reactions would also be absent in humans. We also found that the absence of toxicity in one species provided no evidential weight with regard to toxicity in any other species [7].

3. The solution

A revolution in thinking and practice is needed, involving an escape from relying on animal tests, and resulting in a strong and convincing focus on new (‘alternative’) methodology based on modern and advanced cell and molecular biology, to provide data of direct relevance to humans in real-world human situations, in relation to the diseases to be tested and the therapies, including new drugs, which could prevent, relieve or cure them.

4. The alternatives

The range of alternatives includes the following: a) in vitro preparations of many of the major tissues of the human body; b) microphysiological systems, such as single organ and multi-organ chips, with the ultimate goal of creating human-on-a-chip (a human-like mini-organism in vitro), and even ‘you-on-a-chip’ (focusing on the individual patient); c) ‘omics’ approaches based on the presence, expression and regulation of genes, including genomics, epigenomics, toxicogenomics, transcriptomics, proteomics and metabolomics; d) in silico approaches based on the quantitative structural and activity properties of candidate compounds and the modeling of drug-receptor interactions; e) the computer-based storage, retrieval and analysis of data; and f) the acceptably safe and ethical use of human volunteers, including microdose and low-dose exposure to new drugs, clinical imaging and biomonitoring with biomarkers of exposure and effect.

There should be particular emphasis on effects on the liver, the cardio-vascular system, the nervous system and the respiratory system, which are major targets. Special emphasis should also be placed on solving particular difficulties related to carcinogenicity and reproductive toxicity testing.

More attention needs to be paid to how these new technologies could be applied to answer specific questions, not only about the mechanisms underlying important diseases and the clinical value of new drugs, but also those underlying possible adverse side-effects.

One advantage with therapeutic drugs, which needs to be more-effectively exploited, is that, by contrast with studies on the potential toxicity of industrial chemicals and products such as pesticides, human data on their effects are required in the development and regulatory process, and also after their acceptance for use. If animal tests cannot reveal what needs to be known about new drugs, what confidence can there be that they can provide relevant and reliable information about the risks involved in human exposure to these other products?

These issues and the alternative approaches are discussed in detail in The History of Alternative Test Methods in Toxicology, published in 2018 [8].

5. The strategy

There will be an increasing number and variety of sophisticated alternative methods capable of providing answers to specific questions. They will offer the possibility of high-throughput screening to identify potential cell, tissue and system targets and target mechanisms.

No one method will be able to replace a particular animal test, and that should not be the aim. Instead, the alternatives will need to be used selectively and in combination, in stepwise schemes and with major decision points, according to intelligent strategies tailor-designed for specific purposes and situations, taking into account the molecular and mechanistic basis of human diseases, desirable drug actions and undesired adverse effects, along with a wide range of human variables.

Attention must also be paid to the possibility of unexpected adverse effects not predicted in more-focused testing strategies. Low-dose testing and Phase 0 (microdosing) studies in human volunteers could play a vital role here, and there will always be a need for high-quality post-marketing surveillance. The crucial question, in specific circumstances, is what information would be needed, without animal testing, before testing in human volunteers would be considered to be scientifically worthwhile and ethically acceptable.

It is unlikely that any one laboratory will be able to have and maintain the expertise necessary for performing the full range of procedures, so contract laboratories will specialize in providing a limited number of tests for a wide range of clients.

The study directors within pharmaceutical companies will need to have the skills required to select an appropriate range of tests, arrange for their performance in-house or by experienced contractors, then collect and integrate the results obtained, in order to make a decision about the likelihood that a given compound will be efficacious and tolerable in the patients for whom it is intended.

Regulators will also have to be first-class scientists, able to consider effectively the complex scenarios reported to them in accordance with paradigms fundamentally different from those applied today.

6. Expert opinion

The revolution in thinking and practice that we seek is necessary, because the current approach to the preclinical testing of new pharmaceuticals simply does not work. It is like trying to do a jigsaw puzzle without having a picture on the box to show what is to be created, and without knowing whether the box contains the right pieces and the right number of pieces. The chances of success are slim.

There needs to be an end to the check-list approach, one drug/one dose-suits all thinking, and reliance on animal tests which suffer from such intractable problems that they cannot provide a relevant and reliable basis for making decisions which profoundly affect the lives of human patients. The challenge is enormous, but it must be met, for the benefit of all concerned.

Declaration of interest

J Bailey is an employee of Cruelty Free International, London, UK – an organization that campaigns for the replacement of animal testing with human-specific methods

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.