Sage Journals: Discover world-class research

Abstract

For centuries, animals have been used in research due to their genetic and physiological similarities to humans. However, significant differences exist between humans and animals, which have the potential to confound results obtained from such experiments. These differences result in reduced translatability of animal data to humans, which is a major contributing factor to the 92% failure rate for novel therapies in clinical trials. Advances in scientific research have enabled the development of human-focused New Approach Methodologies (NAMs), which include in silico and 3-D in vitro models. By harnessing these novel approaches, greater predictive power for human biology, human diseases and assessment of novel therapies could be achieved. However, several obstacles remain to their wider adoption, including potential financial constraints, publication bias, and some concerns about the reliability of NAMs due to the novelty of this field, compared to animal studies. Here, we outline the differences between humans and animals used in research, discuss in detail the obstacles to the greater adoption of NAMs in research, and provide recommendations on how to accelerate a shift toward human-focused research.

Keywords

NAMs new approach methodologies replacement organ-on-a-chip

Introduction

Animals have been used in scientific experiments in the UK since the 17th century,¹ as they share the same vital organs and have a high level of genetic homology with humans. However, there are significant physiological and genetic differences between animals and humans, which impact the translation of research results between species.^2–5 One salient consequence of this is the failure of more than 92% of drugs during clinical trials, mostly due to issues with safety and efficacy not predicted by animal data.^3,6 These late-stage failures result in an average cost of USD $1300–$4000 million (£980–£3030 million) to bring a new therapy to market,^7,8 meanwhile some therapies are shelved which could have shown success in humans.⁹ This highlights the importance of progressing towards more predictive, human-relevant research and testing, such as New Approach Methodologies (NAMs). NAMs are defined as in silico, in chemico or in vitro models which could provide greater human relevance than current animal models.^10–13

Despite concerns regarding the human relevance of animal experiments,^14–16 a high level of animal-based research continues worldwide. In the UK, approximately 2.68 million animals were used in research in 2023: 1.47 million within experimental procedures, and the remainder for producing animal lines with genetic mutations.¹⁷ Frameworks are in place to protect animals in research, including Directive 2010/63/EU, which was put in place by the EU and states that where there is a recognised, validated non-animal alternative, animal experiments should not proceed.¹⁸ Additionally, in countries around the world including the UK, USA, EU and Brazil, researchers are encouraged to consider the Three Rs principles — replacement with alternative methods, reduction in the number of animals used, and refinement of experiments to reduce lasting harm — when applying for animal studies licences.¹⁹ However, some argue that the implementation of Directive 2010/63/EU and the Three Rs principles could be improved.¹⁴ Furthermore, focusing on refinement and reduction can hinder focus on replacement, which impacts and potentially delays fulfilment of Directive 2010/63/EU’s final aim of animal replacement.²⁰

This primary focus on refinement and reduction by many researchers may be because they believe that full replacement of animal use is not yet feasible. A survey in the Netherlands regarding a goal set by the Dutch Government, which aimed to increase animal-free research, revealed that approximately 71% of researchers who used animals did not believe that replacement of animals was possible in the near future, though 40% of respondents indicated that they would support replacement.²¹ A 2013 survey in Portugal revealed that approximately 70% of responders believed that some experimental steps may be replaced by NAMs within 50 years.²² Furthermore, a survey of US-based researchers showed that 77% of responders did not believe that NAMs would sufficiently replace animals in their research.²³ However, the survey also revealed awareness of reproducibility issues in animal models due to multiple factors, including lack of detail in animal research publications, and variation between the animals used. Taken together, this highlights that reproducibility within animal models is a significant issue, and that some researchers are open to the idea of NAMs, if these could provide more predictive and human-relevant results compared to animal studies. Hence, this review examines key issues with animal studies, including: the differences between humans and animals; factors affecting the extrapolation of animal data to human disease research; and current NAMs which have the potential to reduce or replace animal studies. Furthermore, obstacles to NAM research will be discussed, and a roadmap to accelerated NAM uptake will be outlined.

What are the main scientific issues associated with animal use in research?

Translational issues due to genetic and physiological differences between humans and animals

Animals are commonly used in research as they are complex organisms which share some physiological and genetic features with humans. For example, mice and humans share 85% gene homology.²⁴ However, although a high proportion of genes are conserved, significant genetic variation exists between the species, and homology within non-coding regions can be lower than 50%.²⁴ Non-coding DNA regions include sequences that can influence transcription of coding regions, which may affect the level and type of gene product.^25,26 Differences in non-coding regions between humans and mice can therefore result in discrepancies in protein structure and function between these species, highlighting that gene regulation and expression are as important as genetic homology. An example of this is the cytochrome P450 (CYP) enzyme superfamily, which is responsible for the metabolism of approximately 80% of all drugs in humans.²⁷ As these genes are highly conserved, small amino acid changes between species may result in significant differences in structure and/or function. For example, the human enzyme CYP2A6 exhibits different functions compared to the murine orthologue.^28,29 Such differences are widespread and negatively impact the translatability of animal data to human safety and toxicity testing.

There are also many physiological differences between humans and mice, including — but not limited to — skin,³⁰ the gastrointestinal (GI) tract,³¹ heart size, output and anatomy,³² metabolic rate,² the immune system,³³ and the nervous system³⁴ (Figure 1). Taking the nervous system as an example, structural differences are present between human and murine brains,³⁴ such as the absence of furrows and ridges on murine brains,⁴² murine motor neurons functioning differently to human motor neurons,^43,44 and differences in pineal and pituitary gland structure and function.^34,45 Despite these differences, approximately 1.9 million mice collectively were used in experiments and breeding in the UK in 2023.¹⁷

Figure 1.

A comparison of mouse and human anatomy, immunology and biochemistry. This diagram outlines key differences between several key organs in mice (left) and humans (right). These organs include (as listed top to bottom): brain and CNS (blue); heart (red); lungs (pink); gastrointestinal tract (tan/brown); pancreas (yellow); and immune system (leukocyte production by bone marrow and lymphatic system (green).^35–41 Image created with BioRender.

Non-human primates (NHPs), such as rhesus macaques and cynomolgus macaques, are often selected for research due to their high level of genetic and anatomical similarity to humans. In fact, over 1800 NHPs were used in experimental procedures in the UK in 2023.¹⁷ However, genetic homology between humans and rhesus macaques is only 90.76%.⁴⁶ Since our evolution from NHPs, human genomes have accumulated many human-specific mutations, including segmental inversions, duplications and translocations,^4,47 which can affect gene expression due to the new position of these genes in relation to promoter and enhancer regions,^4,5,48 and the evolution of new genes in humans and NHPs, accompanied by the loss of less beneficial genes.⁴⁹ This level of genetic variation can lead to significant phenotypic differences, which restricts the reliability of the extrapolation of data from NHP research to humans. An important example of this is the CYP subfamily, as discussed above. CYP amino acid sequences in cynomolgus macaques and marmosets display 68–97% homology to their human orthologues,⁴⁶ but small amino acid changes are known to affect substrate specificity and enzyme activity.⁵ This, and other differences in metabolism, impacts the translation of data from toxicology testing to humans, and as a result the absence of toxicity in NHPs does not translate to an absence of toxicity in humans.^50,51

Furthermore, humans and NHPs display variation with regard to immune cell phenotype, signalling and composition.^52,53 For example, significant differences are present between human and NHP immunoglobulin G and their corresponding Fc receptors, which impact the ability of NHPs to model infectious diseases such as HIV.⁵⁴ This, along with differences in disease progression, confounds HIV research in NHPs, and whilst HIV/AIDS vaccines have been discovered which are effective in NHPs, this has not yet translated to an effective vaccine in humans.⁵⁵ NHPs are also utilised for research into neuroscience and neurological conditions such as Alzheimer’s disease (AD) due to similarities between human and NHP brains and AD-related genes.^56,57 However, whilst age-related amyloid-β accumulation can occur within NHPs, differences remain between human and NHP AD, such as a reduction in phosphorylated tau and reduced cognitive decline compared to humans.^58–60 Whilst humans and animals share the same vital organs, the differences between these species can affect the application of animal data to human disease.

The effects of stress and environment on data from animal experiments

Although animals may experience stressors in their natural environments, the stressors experienced in a laboratory setting are different, unavoidable, and potentially more chronic compared to those experienced in nature.⁶¹ Exposure to these stressors results in prolonged activation of pathways which produce and secrete stress steroid hormones.⁶² This can lead to increased blood pressure and heart rate, perturbation of adaptive and innate immune responses, and subsequent cardiovascular diseases, cancers and GI disorders, all of which can affect the validity of animal experiments.^61,63

Constitutive stress from living outside of their natural habitat, and living in laboratory conditions, can result in epigenetic alterations in animals in laboratories. Offspring can inherit these epigenetic changes, resulting in increased incidence of heart disease, asthma, autoimmune disorders and diabetes, among other disorders.⁶¹ Environmental stressors experienced by animals in captivity, therefore, impact the results obtained through animal experimentation. A computational analysis study of nociceptive sensitivity in mice showed that environmental stressors were responsible for 42% of experimental variability.⁶⁴ Environmental stressors in a laboratory setting include: noise of ventilation systems and other stressed animals;⁶⁵ artificial light;⁶⁶ being handled (and the sex of the person handling them);^67,68 observation of other animals undergoing experimental procedures; the seasons;⁶⁹ and restrictive housing environments such as cages, which do not allow for natural behaviours such as foraging and nesting.^61,66 This can cause many long-term physiological, psychological and epigenetic changes within these animals, limiting the extrapolation of animal data to human disease.^61,64

The application of NAMs to reduce and replace the use of animals in research

Many researchers in academia and industry are developing in silico and human cell-based in vitro models, which could have greater predictive power compared to animal research data, and result in reduced or no animal suffering (Figure 2). In silico techniques use previously generated datasets to predict the outcome of an experiment, utilising mathematical modelling and computer simulations.⁷⁰ These can be used to predict metabolites produced after breakdown of a new drug in vivo,⁷⁹ screen currently available drugs that could be repurposed,⁸⁰ and predict toxicity of chemical compounds.⁸¹

Figure 2.

Examples of the types of NAMs currently used in research. This diagram illustrates the human model systems that are discussed in the main text.^13,70–78 QSAR = Quantitative Structure–Activity Relationship; ADME = Absorption, Distribution, Metabolism and Excretion; PDMS = polydimethylsiloxane. Image created with BioRender.

Components of in vitro human-focused models include: hydrogels, which are 3-D hydrophilic polymer networks that can mimic the 3-D tissue environment; spheroids and organoids; and organ-on-a-chip (OoC). Spheroids can be used to model tumours⁸² or microenvironments such as bone marrow stroma,⁸³ due to their cytokine, oxygen and nutrient gradients, which resemble human tissue more accurately compared to 2-D cell culture.⁷¹ Organoids are produced from stem cells, and can self-organise into organotypic structures, resulting in a higher level of structural complexity compared to spheroids. These can be used to model many processes such as organogenesis, genetic disorders and cancer.^73,84,85 Organoids and spheroids can be incorporated into more complex models, such as OoC.

OoC systems manipulate miniscule volumes of fluids, from 10^–3 to 10^–10 ml, within ‘chips’. These chips are made from materials such as glass, elastomers such as polydimethylsiloxane (PDMS), natural polymers such as alginate, or synthetic polymers such as polystyrene, and contain multiple wells and interconnecting channels.^78,86,87 Cells, hydrogels, spheroids and organoids can be seeded within these chips, replicating single or multiple organs on a chip. OoC present a unique opportunity to model tissue complexity due to the possibility of several cell types being seeded, as well as chemical and physical gradients which are controllable and more analogous to in vivo conditions.⁸⁸ As a result, OoC models could be used to model disease pathophysiology, test chemicals and therapies, and be used in personalised medicine, as primary cells can be cultured within these models⁸⁷ (Table 1). Additionally, these devices have the potential to generate reproducible results, whilst being cost-effective.¹³

Table 1.

A summary of novel organ-on-a-chip (OoC) studies for a range of organs, outlining some of the purposes which the OoC models have been successfully used.

Organ	Purpose	Model	Results	Ref.
GI tract	Host–microbiome interactions within the intestine	Coculture of various bacteria present in the human gut with Caco-2 (intestinal cell line) or primary intestinal cells within a microfluidic device.	Primary intestinal cells exhibited villus morphology in-chip. Cell types and hypoxic gradient present within chip promoted culture of over 200 types of microbes with similar ratios of each compared to human faecal matter.	89
	Drug metabolism within the gut	Caco-2 cultured on top of nitrocellulose membrane and collagen type I within dynamic flow microfluidic device.	In the chip, Caco-2 cells secreted mucus at a higher level compared to 2-D culture. Higher levels of the metabolites present following first-pass metabolism were found in the chip compared to 2-D culture.	90
	Gut inflammation during inflammatory bowel disease	Intestinal cells taken from human intestinal organoids and cocultured with monocyte-derived macrophages within organ-on-a-chip.	Transcriptomic analysis revealed that cells derived from human intestinal organoids in microfluidic culture showed similar genetic signatures compared to fresh human colon cells. Addition of macrophages increased level of proinflammatory cytokines.	91

Brain	Development of brain organoids for long-term (60–70 days) culture	Human cerebrum organoids were developed using induced pluripotent stem cells and cultured within a Matrigel functionalised with brain extracellular matrix (BEM), and cultured within a periodic flow microfluidic device.	BEM supports neural cell spheroid growth to a greater extent than Matrigel. Further, BEM-grown organoids displayed cortical layer development, and increased neurotransmitter response compared to Matrigel-grown organoids. Increased cell proliferation was observed in BEM organoids cultured in periodic-flow OoC compared to within a plate, as well as increased presence of neuronal markers.	92
	Modelling glioblastoma–vasculature interactions	Coculture of glioblastoma (GBM) spheroids and blood vessel models (needle template method used to create a channel in a hydrogel then lined with Human Umbilical Vein Endothelial Cells (HUVECs) within a microfluidic device.	GBM cells from the spheroids migrated toward, grew alongside, but did not enter the blood vessel models within the microfluidic device, therefore mimicking the process of vessel co-option which is observed in GBM.	93
	Modelling glioblastoma–vasculature interactions	Coculture of GBM and brain endothelial cells within 3-D bioprinted hydrogel. 3-D bioprinted structures were then cultured within a microfluidic model of the blood–brain barrier (BBB). These were then exposed to simulated microgravity (μG) to investigate gravity-dependent changes in tumourigenesis.	GBM spheroids within the microfluidic model formed spheroids and were viable for at least 72 hours. The brain endothelial cells within the BBB organised into a linear pattern along the model, creating a perfusable lumen model. μG simulation resulted in reduced function of the BBB and GBM models due to mechanical unloading. The authors state that this could contribute to a potential strategy for passing therapies through the BBB.	94
	Investigating the role of inflammation in a Parkinson’s disease (PD) model	A microfluidic device with three inlets: One for the vasculature model; one for the brain model containing iPSC-derived astrocytes with a mutation found in PD (LRRK2 G2019S); and one for a collagen matrix in between the two models.	Astrocytes containing LRRK2 G2019S mutations did not support development of the BBB, and affected the function and morphology of endothelial cells within the vasculature model. Inhibition of the MEK/ERK pathway resulted in reduced secretion of proinflammatory cytokines and improved structural integrity of the BBB model. Control and PD BBB models were then compared to human postmortem control and PD patients, which revealed similar morphological changes in humans and in the BBB chip.	95

Liver	Predicting drug-induced liver injury	Primary human hepatocytes and human liver sinusoidal endothelial cells were cocultured on fibronectin and collagen-coated surfaces within the microfluidic device, with a Matrigel hydrogel on top.	Cells in the chip behaved similarly to cells in vivo; e.g. a junction-lined bile canaliculi was observed in the sinusoidal endothelial section. Following model characterisation, 27 hepatotoxic and non-toxic drugs were added to the microfluidic devices. The model displayed 80% sensitivity and 100% specificity for detecting toxic and non-toxic compounds, outperforming 3-D hepatic spheroids alone.	11
Liver	Predicting drug-induced liver injury	Microfluidic devices modelling human, rat and dog livers were developed to compare drug-induced liver toxicity. Primary hepatocytes for each species were cultured in between two extracellular matrix-based hydrogels within the devices.	Fialuridine, which is hepatotoxic in humans but not in animals, was assessed. It was uncovered that liver injury markers were present in human liver chips after fialuridine treatment, but not in rat liver chips. A Janssen proprietary compound (JNJ-2), which displayed hepatotoxicity in rats, induced liver injury markers in rat liver chips but not human liver chips. This highlights species-specific differences in drug metabolism.	96

Reproductive organs	Model of human placental xenobiotic exchange	Human trophoblast stem cells and HUVECs were cultured on opposite sides of a collagen-coated membrane to model both the maternal and fetal sides of the placenta in a static and perfusion-based microfluidic systems.	Similar to in vivo placenta, microvilli formed on the trophoblastic barrier section of the model. Placental hormone secretion increased when the trophoblast layer was subjected to flow shear stress, which mimics the maternal blood flow in vivo. A compound that can impact fetal development was then added, resulting in reduction in essential developmental markers such as human chorionic gonadotropin.	97
Reproductive organs	Study of ovarian folliculogenesis and related disorders	Human pre-antral follicles were cultured in a sodium alginate bead within a microfluidic device.	Human follicles were cultured within the sodium alginate bead for up to eight days, and correspondingly the diameter of the follicles increased over the eight-day time frame. Hormones released into the culture medium were examined and it was found that, over the first two days, the concentration of testosterone and oestradiol increased, but from then on, the testosterone levels remained the same and oestradiol levels increased.	98

Vasculature	Blood vessels in viral infection	HUVECs were seeded within gelatin-coated microdevice and rocked in order to form a blood vessel model. Primary human monocytes and Puumala virus (PUUV) were added to investigate the interactions between monocytes, virus and blood vessel.	RNA sequencing of PUUV-treated HUVECs revealed upregulation of disease severity markers (e.g. IL-6 and IDO-1). Addition of PUUV alone did not significantly affect HUVEC permeability, but primary monocytes adhered more to PUUV-infected HUVECs within the vessel model, as compared to non-infected HUVECs. The authors postulate that this increased inflammation results in increased vascular permeability upon PUUV infection.	99
	Microvasculature	Tri-compartmental microfluidic model comprising venule (collagen hydrogel), capillary bed (fibroblasts in fibrin gel) and arteriolar regions (smooth muscle cells in collagen), with a HUVEC monolayer spanning all three regions.	The HUVECs in each compartment anastomosed, forming a microvasculature system that mimics in vivo structure and blood flow. Further, smooth muscle cells within the device displayed vasoconstriction, which could be used for further studies, i.e. assessing how this could affect metastatic cancer. Tumour cells were added to the model, and migrated through the device compartments, enabling study of tumour extravasation.	100
	Vasculature pathology in SARS-CoV-2 infection	Coculture of HUVECs with human blood mononuclear cells in a dynamic flow microfluidic device. Human coronavirus HCoV-NL63 and SARS-CoV-2 added.	Presence of both viruses negatively affected endothelial barrier functions in the device, with increased levels of inflammatory cytokine expression, particularly in the SARS-CoV-2 condition. Addition of leukocytes, including monocytes, worsened the endothelial barrier. An angiopoietin-1 peptide was developed and appeared to improve endothelial barrier function in the device.	101

Cancer	Chemotherapeutic response in 3-D in vitro environment	Human chronic myeloid leukaemia cells (K562) were cocultured with mesenchymal stem cell (MSC) spheroids within a microfluidic device, then treated with selected compounds.	Each single and combination therapy induced cell death in K562 cells but not the MSCs, indicating that these therapies are effective. Further, the combination therapies in 2-D and 3-D induced more cell death.	102
	Chemotherapeutic response in 3-D in vitro environment	Colorectal cancer cell spheroids were cultured in Matrigel in a dynamic flow microfluidic device and treated with selected drugs in single and combination therapy, at various dosing schedules.	Colorectal cancer spheroid size and viability decreased when treated with single therapy and decreased further in combination therapy. Reduced cell proliferation and increased apoptosis induction were apparent when spheroids in the devices were treated with single therapy, and these effects were exacerbated in combination therapy. Similar trends were observed in both the device and data from in vivo experiments.	103
	Immune cell exhaustion	Breast tumour cell lines, HUVECs and natural killer (NK) cells tricultured within a collagen hydrogel, in a microfluidic device.	Over seven days in the device, the tumour model environment resulted in NK cell exhaustion, as indicated through downregulation of NK cell survival and proliferation genes. Thus, the NK cells were less able to attack the tumour cells. This phenotype was rescued upon addition of immunotherapies that targeted immune exhaustion markers.	104
	Early detection of cancer	Microfluidic device used in combination with the CRISPR-Cas12a (Clustered Regularly Interspaced Short Palindromic Repeats-Cas12a) system for single-cell telomerase activity measurement, which could allow early cancer detection.	Once optimised, the system was assessed with HUVECs and three cancer cell lines. It was found that HUVECs expressed lower levels of telomerase compared to the cancer cell lines, highlighting that this device holds promise for single-cell early cancer detection.	105

Recently, models of one or more organs have been combined to create a humanised multi-organ-on-a-chip (MOoC).^106,107 Many biological processes rely on signalling from other organs, therefore MOoCs can provide more biologically relevant data when multiple organs are included.^106,108 Additionally, MOoCs enable cell migration to be visualised, enabling the study of metastatic disease.^109,110 The use of MOoCs also permits the assessment of potential off-target effects following treatment with novel therapies, by modelling the organ of interest in combination with other organs.^111–113 Furthermore, the combination of multiple methods — such as in silico mathematical prediction models with in vitro simulations in OoCs — has led to the effective prediction of safe drug dosages that eliminate cancer cells with minimal impact on healthy cells.¹¹⁴ Together, this highlights that novel 3-D models and in silico analysis could have a higher predictive power compared to animal models, and could be used in preclinical testing alongside, or eventually instead of, animals.¹⁰

While NAMs have the potential to reduce or replace animal research and testing, obstacles remain. For example, induced pluripotent stem cells (iPSCs) are frequently selected for in vitro models, as they can differentiate into a myriad of cell types. However, some believe that the process is time-consuming, expensive and has issues with reproducibility.¹¹⁵ Further challenges include: a lack of standardisation; variation between the NAMs used; potential genetic drift within long-term cell culture; and the fact that NAMs may only be able to recapitulate certain aspects of an organ as opposed to the entire organ.^115–117 However, the latter may be overcome by generating several in vitro models, with each model replicating a different aspect. For example, the Organisation for Economic Co-operation and Development (OECD) Test No. 442E utilises three in vitro tests that examine the same aspect of skin sensitisation, but which provide different readouts to determine the effects of the test chemical on the skin.¹¹⁸

What are the barriers to NAM adoption, and how can we address them?

Concerns about the reliability of NAMs to model the efficacy and safety of novel therapeutics

Some NAMs are relatively new, and animal researchers often believe that some of these new methods may not be as reliable as the existing animal-based methods.²¹ Furthermore, many researchers may be reluctant to change to techniques that they are unfamiliar with.¹¹⁶ A survey of researchers in the Netherlands revealed that ‘reliability’ was perceived as the main obstacle to NAMs uptake,²¹ even though animal-based studies may fail to reliably predict whether novel pharmaceuticals will be effective in humans, resulting in 40–50% of clinical trials failing due to a lack of efficacy,^119–121 and 20–30% due to toxicity.^119–121 Animal toxicity data indicating that a new pharmaceutical is safe for use does not necessarily predict a lack of toxicity in humans.^122–124 This not only constitutes a costly clinical trial failure,^125,126 but can result in severe side effects and in some cases human harm.^127,128 Furthermore, some animal studies cannot be replicated, and this can be due to incomplete experimental reporting,¹²⁹ lack of blinding,^130,131 low statistical power and in some cases inadequate statistical analyses.^130,132 Due to these pitfalls, more humanised and standardised alternatives must be considered.

As discussed in the introduction, researchers may have doubts about the ability of NAMs to replace animals in academic research.^21–23 However, surveys have indicated that researchers would be willing to switch to computer-based alternatives, if these were more scientifically accurate and were recommended by their colleagues.¹³³ Several in silico model systems are available, such as GLORY, a software that predicts the structure of metabolites produced when human CYP enzymes metabolise drugs.^79,134 This software could detect potentially toxic metabolic products that may not be present in animals, due to interspecies variations in metabolism. Additionally, there have been promising in vitro models that could be more reliable and predictive compared to animal studies, including an OoC DILI model¹¹ and the MatTek EpiDerm™ model, which is included in an OECD Test Guideline.¹³⁵

Additionally, whilst animals are complex systems, this does not necessarily make them the most effective model, as reflected in the high clinical trial failure rate.¹³⁶ NAMs also have the potential to be scaled up, especially less complex systems that model specific human-focused interactions, which could be useful for producing reproducible models for disease modelling and drug testing.¹³⁶ Recently, NAMs have been used to assess novel therapies, and the resulting data have been included within the clinical trial applications, which were subsequently accepted.^12,137,138 This is highly promising and highlights the potential of NAMs to predict efficacy and safety in humans without relying on animals.

Perceived decrease in likelihood of publication

Despite the scientific concerns around animal use in research, barriers to publication may remain for researchers who aim to transition toward purely NAM-based research. For instance, many researchers have experienced ‘animal methods bias’ during the publication process, whereby reviewers ask the authors to provide data from animal experiments, despite sufficient results having been obtained through NAMs.^139,140 A survey by Krebs et al.¹³⁹ found that 31% of researchers surveyed (n = 68) pre-emptively used animal methods, as they expected reviewers would ask for them, and 44% had been requested to include animal data by reviewers. Despite the relatively small sample size, these results indicate that animal publication bias is present, though further research is required to understand the extent of this. The authors of this review subsequently conducted a workshop, attended by representatives from academia, industry, publishing and animal advocacy groups, to discuss how to mitigate animal methods bias.¹⁴¹ Potential mitigation strategies included amending regulatory requirements, educating reviewers and editors about NAMs, and promoting the validation and publication of NAMs. Following this workshop, animal-free research advocates from various backgrounds formed the Coalition to Illuminate and Address Animal Methods Bias (COLAAB). Together, COLAAB produced an author’s guide to publishing animal-free research.¹⁴⁰ This is a highly useful publication for researchers aiming to publish animal-free research.

Despite the presence of animal methods bias in publishing, research that harnesses in vitro and in silico approaches is becoming increasingly popular. Taylor et al. comprehensively surveyed current trends in animal and NAM studies across seven research areas between 2003 and 2022.¹⁴² They uncovered that the relative number of publications based on the use of NAMs had increased compared to studies that used animals, as well as studies which used both animals and NAMs. These data are highly encouraging for researchers aiming to utilise NAMs in their research, as it indicates that despite animal methods bias, a high volume of NAM-based research has been, and is being, published.

Reduced availability of funding

Some researchers may be hesitant to move away from animal models because they are concerned that less funding is available for NAMs-based research. This is a valid concern. In the UK, the main research funding body, United Kingdom Research and Innovation (UKRI), has invested £50 million over the last five-year period toward NAM research;¹⁴³ however, in 2024, the UK government’s entire research and development budget was £20,400 million¹⁴⁴ (highlighting that more funding is provided to animal research compared to NAM research). The Netherlands has increased its funding toward animal alternatives, but even then, a study revealed that NAMs received approximately half of the funding compared to animal research.¹⁴⁵

Whilst NAM research typically receives less funding compared to animal research, the level of NAM funding is increasing, particularly in the UK, the EU and the USA. For example, a 2013 survey disclosed that only seven EU member states had set aside a specific budget for NAM-based projects.¹⁴⁶ However, a more recent assessment of EU-funded studies on Alzheimer’s disease, breast cancer, and prostate cancer found that 60% of studies did not use animals, and that €1200 million (56%) of funding was awarded to animal-free research.¹⁴⁷ This highlights an increase in NAM research funding within these research areas. Furthermore, in 2024, Andrew Griffith (the UK Minister for Science, Research and Innovation) declared that more funding will be designated for the development, validation and dissemination of NAMs.¹⁴⁸ The UK is currently part of a funding programme with the EU known as Horizon Europe, which has pledged €270 million toward NAM research between 2021–2027.^149,150 Whilst this is substantial funding, it is approximately 0.28% of Horizon Europe’s total budget of €95,500 million, and more money would aid a greater shift toward NAM development and uptake.¹⁵⁰ One key issue for worldwide adoption of NAMs is that in lower-and-middle income countries, there may not be a specific budget set aside for NAMs by the government, and the support may not be available for researchers working with non-animal methods.^151,152 For example, for the four years up to 2020, 0.2% of India’s Department of Biotechnology funding was designated for NAMs.¹⁵³ However, in 2019, the Indian Council of Medical Research (ICMR) announced the development of a Centre of Excellence in Human-Pathway-Based Biomedicine and Risk Assessment, which will conduct NAM research. This could then provide useful data for NAM researchers worldwide.¹⁵³

State-funded research councils, independent research charities and animal advocacy groups, such as Animal Free Research UK, also fund research that advances NAM use in research.^154–157 Recently, the National Centre for the Replacement, Reduction & Refinement of Animals in Research (NC3Rs), the Biotechnology and Biological Sciences Research Council (BBSRC) and UKRI collaborated to provide £4.7 million funding toward NAM development.¹⁵⁸ Further, the Department for Science, Innovation and Technology has provided £4.85 million toward projects that increase uptake and validation of NAMs.¹⁵⁹ A similar increase in funding has been observed in the USA. The National Institutes of Health (NIH) has commenced a programme known as Complement Animal Research In Experimentation (Complement-ARIE), which has pledged approximately USD $18 million per year to accelerate the development, standardisation and validation of NAMs.^160–163 However, as the NIH’s budget is almost $48,000 million, with an estimated $19,600 million designated for animal studies,¹⁶⁴ more substantial change could be achieved with greater funding.¹⁶⁵ Further, the Foundation for the National Institutes of Health (FNIH) has created public–private partnerships, which bring together NAM developers and regulatory bodies in order to accelerate NAM acceptance and/or validation.¹⁶⁶ Together, not only will this funding support researchers currently developing NAMs but, if enough funding is available, this could also increase the rate at which NAMs are validated and standardised. In turn, this could motivate more researchers to select NAMs instead of animals.

Additionally, many industries choose to invest in NAMs. Collaborations between industry and academia could lead to the scaling up and commercialisation of NAMs,¹⁶⁷ potentially resulting in more standardised and cost-effective preclinical research. Market research has highlighted a shift in the market toward 3-D cell culture models, estimating that it will grow at a mid-teen compound annual growth rate from 2021–2028.^168,169 Further projections estimate that the global 3-D cell culture market will be worth USD $3600 million (£2860 million) by 2030,¹⁶⁹ with the organ-on-chip sector worth $303.6 million (£238.8 million) in 2026,¹⁷⁰ rising to $630 million (£495.6 million) by 2029.¹⁷¹ This further highlights the increasing popularity of 3-D in vitro models, particularly organ-on-chip, which could promote greater adoption within research.

Future directions

Replacing animal studies with NAMs has the potential to increase the rate of progress within research, and could provide economic benefits.¹⁷² However, barriers remain: many NAMs are in early stages of research, and it remains that there are few standardised, validated NAMs. Here, we suggest some solutions to overcome existing barriers for academic and regulatory NAM use (Figure 3). Four main considerations for NAM adoption are knowledge, funding, collaboration, and standardisation and regulatory approval.

Figure 3.

An outline of the necessary steps toward establishing and validating NAMs for use in basic research and regulation. (a) A schematic of the four main themes identified in this review, and ways to promote NAMs under the respective themes. (b) A tri-phasic approach to improving and disseminating NAMs, with some aspects that would be important throughout these phases.^{115,116,173–176} Image created with BioRender.

Knowledge

To promote confidence in NAMs, further education into NAMs and their benefits must be provided. Acceptance of NAMs may occur through publication of successful methods and presenting these findings at conferences. More widespread uptake of NAMs could be driven by educating current and new researchers. An interdisciplinary workshop in the Collegium Helveticum in Switzerland discussed four aspects which could further promote education about NAMs: statements from universities about NAMs; knowledge transfer through teaching about NAMs; a push toward NAMs in scientific research; and implementation of ‘transition units’, to ensure smooth transition toward NAMs.¹⁷³ One of these areas, knowledge transfer, is currently being addressed by the Karolinska Institutet in Sweden. Here, a ‘ToxMaster’ programme is offered, which covers topics including in silico toxicology techniques and the Three Rs.¹⁷⁷ Universities worldwide could organise similar courses, or assimilate this knowledge into pre-existing courses, allowing the next generation of researchers to drive NAM research. Further familiarisation with NAMs could include organising workshops, webinars, and in-person training sessions such as ‘helpathons’. For example, Animal Free Research UK ran an organ-on-chip Masterclass, which enabled researchers in academia and industry to gain relevant knowledge and experience.¹⁷⁸ People for the Ethical Treatment of Animals (PETA) ran successful workshops and webinars that focused on animal alternatives.¹⁷⁹ Furthermore, in the USA, the 2024 Summer Immersion on Innovative Approaches in Science gathered early career researchers across the globe to provide education and training in NAMs.¹⁸⁰ These learning opportunities could inform current NAM researchers and encourage researchers who do not use NAMs to consider them in their research.

Funding

Greater funding of NAMs development and validation, in both academia and industry, would facilitate the shift away from animal research, as it would enable researchers to develop a greater volume of human-based non-animal methods. Funding programmes (such as Complement-ARIE and public–private partnerships) will help to establish collaborations between academia, industry and regulatory bodies, which could then accelerate development and validation of NAMs in research. However, further NAM-specific funding incentives could drive researchers to incorporate NAMs into their research.

Collaboration

Fostering collaborations between academia, industry, regulatory and government bodies would enable greater transfer of knowledge and resources between each sector, therefore driving the progress of NAMs. For example, the NIH in collaboration with the FNIH is constructing a Validation and Qualification Network, which will allow public and private sector partners to collaborate, with the goal of accelerating the validation and adoption of NAMs.¹⁸¹ Furthermore, the Indian Council of Medical Research has established the Collaborating Centre of Excellence, which promotes collaboration of research institutions and organisations to advance biomedical research.¹⁸² Through these collaborations, validated human-relevant non-animal models could be developed, which could promote changes in policy to enable these to be used in basic research and regulatory testing.

Standardisation and regulatory approval

Standardising and validating NAMs may encourage more researchers to switch to these methods, as it could increase their perceived reliability. These could then be scaled up and used in drug and chemical toxicity testing if successful. An example of this is organ-on-chip technology, which demonstrates great potential for drug screening and testing, but its ability to reach this potential is hampered by the absence of a clear regulatory pathway in basic research.^13,183 There is a lack of standardised NAMs for use in medical research, which could be because after a new method has been developed, the funding and time required for its validation may be unavailable.¹⁸⁴ Choosing to fund NAM-based studies over animal studies may promote the uptake of NAMs in research and regulation. An audit of all available NAMs would enable a full assessment of NAM progress and inform decisions on the optimal strategy to proceed.¹⁷⁴ This has already been completed to an extent in Belgium, where the ‘RE-place’ project has collected all NAMs in one open access database.¹⁸⁵ Ideally, NAM research worldwide could be included on an online, open access database, which would either be free or low-cost to access. This would not only aid the standardisation of NAMs due to the transparency of research, but would also inform low- and medium-income countries on the current available NAMs, therefore promoting international collaborations.

Conclusions

For decades, the ‘gold standard’ in research was considered to be animal studies. However, due to significant differences between humans and animals in genetics, physiology, pathophysiology and more, animal models fail to accurately recapitulate human disease, which could hinder our understanding of these diseases and our ability to develop effective therapies for them. This is further reflected in the high proportion of drugs that fail in clinical trials, having appeared promising in terms of efficacy and safety in animal tests. Human-specific NAMs, including in vitro and in silico models, are gaining popularity as researchers and funders realise that they have the potential to provide cheaper and more accurate data compared to animal studies.⁹ However, more must be done to develop and promote these technologies. There remain several barriers to wide utilisation of NAMs, such as lack of validation and standardisation, reduced levels of funding compared to animal studies and some concerns about their reliability. More research must be completed to optimise NAMs, and more funding and greater collaboration within and between countries is essential for their widespread uptake. If the actions outlined in this review are completed, then we can increase the reliability and use of NAMs, which could provide more human-relevant data at an overall cheaper cost compared to animal data.

Footnotes

Acknowledgments

We thank Dr Ross Dobie and Oscar Lavery for their insightful feedback on this review article.

ORCID iD

Lauren Hope

Funding

This work was commissioned and funded by Animal Free Research UK.

Declaration of conflicting interests

We have no conflicts of interest to disclose.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analysed in this study. *

References

Ribatti

. William Harvey and the discovery of the circulation of the blood. J Angiogenesis Res 2009; 1: 3.

Perlman

. Mouse models of human disease: An evolutionary perspective. Evol Med Public Health 2016; 2016: 170–176.

Lansdowne LE . Animal-free methods for drug safety testing. Technology Networks. https://www.technologynetworks.com/drug-discovery/blog/animal-free-methods-for-drug-safety-testing-347431 (2021, accessed 20 May 2025).

Bailey

. Lessons from chimpanzee-based research on human disease: The implications of genetic differences. Altern Lab Anim 2011; 39: 527–540.

Bailey

. Monkey-based research on human disease: The implications of genetic differences. Altern Lab Anim 2014; 42: 287–317.

Thomas

Chancellor

Micklus

, et al. Clinical development success rates and Contributing Factors 2011–2020. BIO. https://www.bio.org/clinical-development-success-rates-and-contributing-factors-2011-2020 (2021, accessed 20 May 2025).

Wouters

McKee

Luyten

. Estimated research and development investment needed to bring a new medicine to market, 2009–2018. JAMA 2020; 323: 844–853.

Herper

. The truly staggering cost of inventing new drugs. Forbes. https://www.forbes.com/sites/matthewherper/2012/02/10/the-truly-staggering-cost-of-inventing-new-drugs/ (2012, accessed 20 May 2025).

Van Norman

. Limitations of animal studies for predicting toxicity in clinical trials. JACC Basic Transl Sci 2019; 4: 845–854.

10.

Levner

Ewart

. Integrating Liver-Chip data into pharmaceutical decision-making processes. Expert Opin Drug Discov 2023; 18: 1313–1320.

11.

Ewart

Apostolou

Briggs

, et al. Performance assessment and economic analysis of a human Liver-Chip for predictive toxicology. Commun Med 2022; 2: 1–16.

12.

Ito

Morimoto

Takahashi

, et al. Maiden voyage: Induced pluripotent stem cell-based drug screening for amyotrophic lateral sclerosis. Brain 2023; 146: 13–19.

13.

Elvira

. Microfluidic technologies for drug discovery and development: Friend or foe? Trends Pharmacol Sci 2021; 42: 518–526.

14.

Balls

Combes

. Questionable progress in the application of the Three Rs to improve science, human well-being and animal welfare. Altern Lab Anim 2018; 46: 55–57.

15.

Marshall

Bailey

Cassotta

, et al. Poor translatability of biomedical research using animals — A narrative review. Altern Lab Anim 2023; 51: 102–135.

16.

Bailey

Balls

. Clinical impact of high-profile animal-based research reported in the UK national press. BMJ Open Sci 2020; 4: 1–12.

17.

Home Office . Annual statistics of scientific procedures on living animals, Great Britain 2023. https://www.gov.uk/government/statistics/statistics-of-scientific-procedures-on-living-animals-great-britain-2023/annual-statistics-of-scientific-procedures-on-living-animals-great-britain-2023 (2024, accessed 20 May 2025).

18.

Balls

Why are validated alternatives not being used to replace animal tests?

Altern Lab Anim 2018; 46: 1–3.

19.

NC3Rs . The 3Rs. https://nc3rs.org.uk/the-3rs (2023, accessed 20 May 2025).

20.

Bailey

. It’s time to review the Three Rs, to make them more fit for purpose in the 21st century. Altern Lab Anim 2024; 52: 155–165.

21.

Bressers

van den Elzen

Gräwe

, et al. Policy driven changes in animal research practices: Mapping researchers’ attitudes towards animal-free innovations using the Netherlands as an example. Res Integr Peer Rev 2019; 4: 8.

22.

Franco

Olsson

. Scientists and the 3Rs: Attitudes to animal use in biomedical research and the effect of mandatory training in laboratory animal science. Lab Anim 2014; 48: 50–60.

23.

Walker

Saylor

Waltz

, et al. Translational science: A survey of US biomedical researchers’ perspectives and practices. Lab Anim 2021; 51: 22–35.

24.

National Human Genome Research Institute . Why mouse matters. https://www.genome.gov/10001345/importance-of-mouse-genome (2010, accessed 20 May 2025).

25.

Blencowe

. Alternative splicing: New insights from global analyses. Cell 2006; 126: 37–47.

26.

Kelemen

Convertini

Zhang

, et al. Function of alternative splicing. Gene 2013; 514: 1–30.

27.

Sutton

Sutherland

Shnyder

, et al. Improved preparation and detection of cytochrome P450 isoforms using MS methods. Proteomics 2010; 10: 327–331.

28.

Hammer

Schmidt

Marx-Stoelting

, et al. Cross-species analysis of hepatic cytochrome P450 and transport protein expression. Arch Toxicol 2021; 95: 117–133.

29.

Honkakoski

Negishi

. The structure, function, and regulation of cytochrome P450 2A enzymes. Drug Metab Rev 1997; 29: 977–996.

30.

Zomer

Trentin

. Skin wound healing in humans and mice: Challenges in translational research. J Dermatol Sci 2018; 90: 3–12.

31.

Hugenholtz

De Vos

. Mouse models for human intestinal microbiota research: A critical evaluation. Cell Mol Life Sci 2018; 75: 149–160.

32.

Laflamme

Sebastian

Buetow

. 10 — Cardiovascular. In: Treuting

Dintzis

(eds). Comparative anatomy and histology. San Diego: Academic Press, 2012, pp. 163–189.

33.

Mestas

Hughes

CCW

. Of mice and not men: Differences between mouse and human immunology. J Immunol 2004; 172: 2731–2738.

34.

Hagan

Bolon

Keene

. 20 — Nervous system. In: Treuting

Dintzis

(eds). Comparative anatomy and histology. San Diego: Academic Press, 2012, pp. 403–444.

35.

Hodge

Bakken

Miller

, et al. Conserved cell types with divergent features in human versus mouse cortex. Nature 2019; 573: 61–68.

36.

Ghosh

Singh

Aruna

, et al. The genomic organization of mouse resistin reveals major differences from the human resistin: Functional implications. Gene 2003; 305: 27–34.

37.

Hamlin

Altschuld

. Extrapolation from mouse to man. Circ Cardiovasc Imaging 2011; 4: 2–4.

38.

Dolenšek

Rupnik

Stožer

. Structural similarities and differences between the human and the mouse pancreas. Islets 2015; 7: e1024405.

39.

Rydell-Törmänen

Johnson

. The applicability of mouse models to the study of human disease. Mouse Cell Cult 2018; 1940: 3–22.

40.

Pan

Deutsch

Wert

, et al. Comprehensive anatomic ontologies for lung development: A comparison of alveolar formation and maturation within mouse and human lung. J Biomed Semant 2019; 10: 18.

41.

Martignoni

Groothuis

GMM

de Kanter

. Species differences between mouse, rat, dog, monkey and human CYP-mediated drug metabolism, inhibition and induction. Expert Opin Drug Metab Toxicol 2006; 2: 875–894.

42.

Ruberte

Schofield

Sundberg

, et al. Bridging mouse and human anatomies; a knowledge-based approach to comparative anatomy for disease model phenotyping. Mamm Genome 2023; 34: 389–407.

43.

Manuel

Chardon

Tysseling

, et al. Scaling of motor output, from mouse to humans. Physiology 2019; 34: 5–13.

44.

Sieck

. Physiology in perspective: Of mice and men. Physiology 2019; 34: 3–4.

45.

La Perle

KMD

Dintzis

. 15 — Endocrine system. In: Treuting

Dintzis

Montine

(eds). Comparative anatomy and histology. 2nd ed. San Diego: Academic Press, 2018, pp. 251–273.

46.

Gibbs

Rogers

Katze

, et al. Evolutionary and biomedical insights from the Rhesus macaque genome. Science 2007; 316: 222–234.

47.

Dennis

Harshman

Nelson

, et al. The evolution and population diversity of human-specific segmental duplications. Nat Ecol Evol 2017; 1: 1–10.

48.

Teichmann

Babu

. The impact of genomic neighborhood on the evolution of human and chimpanzee transcriptome. Genome Res 2009; 19: 785–794.

49.

Demuth

Bie

Stajich

, et al. The evolution of mammalian gene families. PLoS One 2006; 1: e85.

50.

Bailey

Thew

Balls

. Predicting human drug toxicity and safety via animal tests: Can any one species predict drug toxicity in any other, and do monkeys help? Altern Lab Anim 2015; 43: 393–403.

51.

Hudson

Bhogal

Balls

. The use of non-human primates in regulatory toxicology: Comments submitted by FRAME to the Home Office. Altern Lab Anim 2005; 33: 529–540.

52.

Bjornson-Hooper

Fragiadakis

Spitzer

, et al. A comprehensive atlas of immunological differences between humans, mice, and non-human primates. Front Immunol 2022; 13: 867015.

53.

Bjornson-Hooper

Fragiadakis

Spitzer

et al. A comprehensive atlas of immunological differences between humans, mice and non-human primates. Front Immunol. 2022; 13: 867015.

54.

Crowley

Ackerman

. Mind the gap: How interspecies variability in IgG and its receptors may complicate comparisons of human and non-human primate effector function. Front Immunol 2019; 10: 697.

55.

Bailey

. An assessment of the role of chimpanzees in AIDS vaccine research. Altern Lab Anim 2008; 36: 381–428.

56.

Zhang

Qin

. Current state of research on non-human primate models of Alzheimer’s disease. Anim Models Exp Med 2019; 2: 227–238.

57.

Okano

Kishi

. Investigation of brain science and neurological/psychiatric disorders using genetically modified non-human primates. Curr Opin Neurobiol 2018; 50: 1–6.

58.

Heuer

Rosen

Cintron

, et al. Nonhuman primate models of Alzheimer-like cerebral proteopathy. Curr Pharm Des 2012; 18: 1159–1169.

59.

Lemere

Beierschmitt

Iglesias

, et al. Alzheimer’s disease Aβ vaccine reduces central nervous system Aβ levels in a non-human primate, the Caribbean Vervet. Am J Pathol 2004; 165: 283–297.

60.

Elfenbein

Rosen

Stephens

, et al. Cerebral β-amyloid angiopathy in aged squirrel monkeys. Histol Histopathol 2006; 22: 155–167.

61.

Bailey

. Does the stress of laboratory life and experimentation on animals adversely affect research data? A critical review. Altern Lab Anim 2018; 46: 291–305.

62.

Murgatroyd

Spengler

. Epigenetic programming of the HPA axis: Early life decides. Stress 2011; 14: 581–589.

63.

Bailey

Does the stress inherent to laboratory life and experimentation on animals adversely affect research data?

Altern Lab Anim 2017; 45: 299–301.

64.

Gaskill

Garner

. Stressed out: Providing laboratory animals with behavioral control to reduce the physiological effects of stress. Lab Anim 2017; 46: 142–145.

65.

Brouček

. Effect of noise on performance, stress, and behaviour of animals. Slovak J Anim Sci 2014; 47: 111–123.

66.

Akhtar

The flaws and human harms of animal experimentation. Camb Q Healthc Ethics 2015; 24: 407–419.

67.

Chesler

Wilson

Lariviere

, et al. Identification and ranking of genetic and laboratory environment factors influencing a behavioral trait, thermal nociception, via computational analysis of a large data archive. Neurosci Biobehav Rev 2002; 26: 907–923.

68.

Sorge

Martin

Isbester

, et al. Olfactory exposure to males, including men, causes stress and related analgesia in rodents. Nat Methods 2014; 11: 629–632.

69.

Meyer

Caston

Mensah-Nyagan

. Seasonal variation of the impact of a stressful procedure on open field behaviour and blood corticosterone in laboratory mice. Behav Brain Res 2006; 167: 342–348.

70.

Heller

Lockwood

Janes

, et al. Technologies for measuring pharmacokinetic profiles. Annu Rev Anal Chem 2018; 11: 79–100.

71.

Zanoni

Pignatta

Arienti

, et al. Anticancer drug discovery using multicellular tumor spheroid models. Expert Opin Drug Discov 2019; 14: 289–301.

72.

Choudhary

Malik

Tomar

. Identification of SARS-CoV-2 cell entry inhibitors by drug repurposing using in silico structure-based virtual screening approach. Front Immunol 2020; 11: 1664.

73.

Gunti

Hoke

ATK

, et al. Organoid and spheroid tumor models: Techniques and applications. Cancers 2021; 13: 874.

74.

Ramme

Koenig

Hasenberg

, et al. Autologous induced pluripotent stem cell-derived four-organ-chip. Future Sci OA 2019; 5: FSO413.

75.

Zieba

. What are organoids and how are they made? The Scientist. https://www.the-scientist.com/mini-organs-in-a-dish-the-versatility-and-applications-of-organoids-70354. (2022, accessed 20 May 2025).

76.

Kim

Gwon

Park

, et al. Therapeutic strategies of three-dimensional stem cell spheroids and organoids for tissue repair and regeneration. Bioact Mater 2023; 19: 50–74.

77.

Caleffi

Aal

MCE

Gallindo

HDEOM

, et al. Magnetic 3D cell culture: State of the art and current advances. Life Sci 2021; 286: 120028.

78.

Cao

UMN

Zhang

Chen

, et al. Microfluidic organ-on-a-chip: A guide to biomaterial choice and fabrication. Int J Mol Sci 2023; 24: 3232.

79.

de Bruyn Kops

Stork

Šícho

, et al. GLORY: Generator of the structures of likely cytochrome P450 metabolites based on predicted sites of metabolism. Front Chem 2019; 7: 1–15.

80.

Choudhary

Malik

Tomar

. Identification of SARS-CoV-2 cell entry inhibitors by drug repurposing using in silico structure-based virtual screening approach. Front Immunol 2020; 11: 1–14.

81.

Carstens

Carpenter

Martin

, et al. Integrating data from in vitro new approach methodologies for developmental neurotoxicity. Toxicol Sci 2022; 187: 62–79.

82.

Jiang

Munguia-Lopez

, et al. Engineering bioprintable alginate/gelatin composite hydrogels with tunable mechanical and cell adhesive properties to modulate tumor spheroid growth kinetics. Biofabrication 2019; 12: 015024.

83.

Busch

Gulliver

Mulholland

, et al. The role of the BMP pathway in sustaining CML stem cells in the bone marrow niche. Blood 2019; 134: 3732.

84.

Inak

Rybak-Wolf

Lisowski

, et al. Defective metabolic programming impairs early neuronal morphogenesis in neural cultures and an organoid model of Leigh syndrome. Nat Commun 2021; 12: 1929.

85.

Saltsman

Hammond

Narayan

NJC

, et al. A human organoid model of aggressive hepatoblastoma for disease modeling and drug testing. Cancers 2020; 12: 2668.

86.

Whitesides

. The origins and the future of microfluidics. Nature 2006; 442: 368–373.

87.

Centre for Human Specific Research . Organ-on-chip. https://humanspecificresearch.org/organ-on-a-chip/ (2024, accessed 20 May 2025).

88.

Mansoorifar

Gordon

Bergan

, et al. Bone-on-a-chip: Microfluidic technologies and microphysiologic models of bone tissue. Adv Funct Mater 2021; 31: 2006796.

89.

Jalili-Firoozinezhad

Gazzaniga

Calamari

, et al. A complex human gut microbiome cultured in an anaerobic intestine-on-a-chip. Nat Biomed Eng 2019; 3: 520–531.

90.

Guo

, et al. A biomimetic human gut-on-a-chip for modeling drug metabolism in intestine. Artif Organs 2018; 42: 1196–1205.

91.

Beaurivage

Kanapeckaite

Loomans

, et al. Development of a human primary gut-on-a-chip to model inflammatory processes. Sci Rep 2020; 10: 21475.

92.

Cho

Jin

, et al. Microfluidic device with brain extracellular matrix promotes structural and functional maturation of human brain organoids. Nat Commun 2021; 12: 4730.

93.

Bae

Kim

Han

, et al. Development of tumor-vasculature interaction on chip mimicking vessel co-option of glioblastoma. BioChip J 2023; 17: 77–84.

94.

Silvani

Basirun

, et al. A 3D‐bioprinted vascularized glioblastoma‐on‐a‐chip for studying the impact of simulated microgravity as a novel pre‐clinical approach in brain tumor therapy. Adv Therap 2021; 4: 2100106.

95.

de Rus Jacquet

Alpaugh

Denis

, et al. The contribution of inflammatory astrocytes to BBB impairments in a brain-chip model of Parkinson’s disease. Nat Commun 2023; 14: 3651.

96.

Jang

Otieno

Ronxhi

, et al. Reproducing human and cross-species drug toxicities using a liver-chip. Sci Transl Med 2019; 11: eaax5516.

97.

Cao

Wang

Liu

, et al. Self-assembled human placental model from trophoblast stem cells in a dynamic organ-on-a-chip system. Cell Prolif 2023; 56: e13469.

98.

Aziz

AUR

Deng

, et al. A microfluidic device for culturing an encapsulated ovarian follicle. Micromachines 2017; 8: 335.

99.

Noack

Van Haperen

Van den Hout

MCGN,

et al. A three-dimensional vessel-on-chip model to study Puumala orthohantavirus pathogenesis. Lab Chip. 2024; 24: 5347–5359.

100.

Chen

Blazeski

Zhang

, et al. Development of a perfusable, hierarchical microvasculature-on-a-chip model. Lab Chip 2023; 23: 4552–4564.

101.

RXZ

Lai

BFL

Rafatian

, et al. Vasculature-on-a-chip platform with innate immunity enables identification of angiopoietin-1 derived peptide as a therapeutic for SARS-CoV-2 induced inflammation. Lab Chip 2022; 22: 1171–1186.

102.

Busch

Mulholland

Zagnoni

, et al. Overcoming BCR::ABL1 dependent and independent survival mechanisms in chronic myeloid leukaemia using a multi-kinase targeting approach. Cell Commun Signal 2023; 21: 342.

103.

Petreus

Cadogan

Hughes

, et al. Tumour-on-chip microfluidic platform for assessment of drug pharmacokinetics and treatment response. Commun Biol 2021; 4: 1–11.

104.

Ayuso

Rehman

Virumbrales-Munoz

, et al. Microfluidic tumor-on-a-chip model to evaluate the role of tumor environmental stress on NK cell exhaustion. Sci Adv 2021; 7: eabc2331.

105.

Jiang

Wang

Luo

, et al. Detecting telomerase activity at the single-cell level using a CRISPR-Cas12a-based chip. Lab Chip 2024; 25: 49–56.

106.

Picollet-D’hahan

Zuchowska

Lemeunier

, et al. Multiorgan-on-a-chip: A systemic approach to model and decipher inter-organ communication. Trends Biotechnol 2021; 39: 788–810.

107.

Wang

Han

Tang

, et al. Revealing transport, uptake and damage of polystyrene microplastics using a gut-liver-on-a-chip. Lab Chip 2024; 25: 1656–1668.

108.

Srivastava

Foo

Aggarwal

, et al. Organ-on-chip technology: Opportunities and challenges. Biotechnol Notes 2024; 5: 8–12.

109.

Guo

, et al. Design and construction of a multi-organ microfluidic chip mimicking the in vivo microenvironment of lung cancer metastasis. ACS Appl Mater Interfaces 2016; 8: 25840–25847.

110.

Wang

, et al. Metastasis-on-a-chip mimicking the progression of kidney cancer in the liver for predicting treatment efficacy. Theranostics 2020; 10: 300–311.

111.

Qiao

Chang

, et al.‘Multi-organ-on-a-chip’ for drug testing applications. Authorea. https://www.authorea.com/users/775355/articles/876129--multi-organ-on-a-chip-for-drug-testing-applications (2024, accessed 20 May 2025).

112.

Ingber

. Human organs-on-chips for disease modelling, drug development and personalized medicine. Nat Rev Genet 2022; 23: 467–491.

113.

Skardal

Aleman

Forsythe

, et al. Drug compound screening in single and integrated multi-organoid body-on-a-chip systems. Biofabrication 2020; 12: 025017.

114.

Palacio-Castañeda

Dumas

Albrecht

, et al. A hybrid in silico and tumor-on-a-chip approach to model targeted protein behavior in 3D microenvironments. Cancers 2021; 13: 2461.

115.

Turner

Pound

Owen

, et al. Incorporating new approach methodologies into regulatory nonclinical pharmaceutical safety assessment. ALTEX 2023; 40: 519–533.

116.

Sewell

Alexander-White

Brescia

, et al. New approach methodologies (NAMs): Identifying and overcoming hurdles to accelerated adoption. Toxicol Res 2024; 13: tfae044.

117.

Serafini

Sepehri

Midali

, et al. Recent advances and current challenges of new approach methodologies in developmental and adult neurotoxicity testing. Arch Toxicol 2024; 98: 1271–1295.

118.

OECD . Test No. 442E: In vitro skin sensitisation: In vitro skin sensitisation assays addressing the key event on activation of dendritic cells on the adverse outcome pathway for skin sensitisation. OECD Guidelines for the Testing of Chemicals, Section 4. Paris: OECD Publishing. https://doi.org/10.1787/9789264264359-en

119.

Harrison

. Phase II and phase III failures: 2013–2015. Nat Rev Drug Discov 2016; 15: 817–818.

120.

Dowden

Munro

. Trends in clinical success rates and therapeutic focus. Nat Rev Drug Discov 2019; 18: 495–496.

121.

Sun

Gao

, et al.

Why 90% of clinical drug development fails and how to improve it?

Acta Pharm Sin B 2022; 12: 3049–3062.

122.

Bailey

Balls

. Recent efforts to elucidate the scientific validity of animal-based drug tests by the pharmaceutical industry, pro-testing lobby groups, and animal welfare organisations. BMC Med Ethics 2019; 20: 16.

123.

Clark

Steger-Hartmann

. A big data approach to the concordance of the toxicity of pharmaceuticals in animals and humans. Regul Toxicol Pharmacol 2018; 96: 94–105.

124.

Bailey

Thew

Balls

. An analysis of the use of animal models in predicting human toxicology and drug safety. Altern Lab Anim 2014; 42: 181–199.

125.

Huss

. The high price of failed clinical trials: Time to rethink the model. Clinical Leader. https://www.clinicalleader.com/doc/the-high-price-of-failed-clinical-trials-time-to-rethink-the-model-0001 (2016, accessed 20 May 2025).

126.

Fogel

. Factors associated with clinical trials that fail and opportunities for improving the likelihood of success: A review. Contemp Clin Trials Commun 2018; 11: 156–164.

127.

Stebbings

Findlay

Edwards

, et al. “Cytokine storm” in the phase I trial of monoclonal antibody TGN1412: Better understanding the causes to improve preclinical testing of immunotherapeutics. J Immunol 2007; 179: 3325–3331.

128.

Roth-Cline

. Clinical trials in the wake of vioxx. Circulation 2006; 113: 2253–2259.

129.

Omary

Cohen

El-Omar

, et al. Not all mice are the same: Standardization of animal research data presentation. Gastroenterology 2016; 150: 1503–1504.

130.

Robinson

Krieger

Khan

, et al. The current state of animal models in research: A review. Int J Surg 2019; 72: 9–13.

131.

Pound

Ebrahim

Sandercock

, et al.

Where is the evidence that animal research benefits humans?

BMJ 2004; 328: 514–517.

132.

Munafò

Nosek

Bishop

DVM

, et al. A manifesto for reproducible science. Nat Hum Behav 2017; 1: 1–9.

133.

Badyal

Desai

. Animal use in pharmacology education and research: The changing scenario. Indian J Pharmacol 2014; 46: 257–265.

134.

Ahmed

Moni

Sonawane

, et al. A comprehensive in silico exploration of pharmacological properties, bioactivities and COX-2 inhibitory potential of eleutheroside B from Eleutherococcus senticosus (Rupr. & Maxim.) Maxim. J Biomol Struct Dyn 2021; 39: 6553–6566.

135.

OECD . Test No. 439: In vitro skin irritation: Reconstructed human epidermis test method. OECD Guidelines for the Testing of Chemicals, Section 4. Paris: OECD Publishing, 2021. https://doi.org/10.1787/9789264242845-en.

136.

Dobie

. Hurdles preventing the adoption of human-specific research. Centre For Human Specific Research. https://humanspecificresearch.org/hurdles-preventing-the-adoption-of-human-specific-research/ (2024, accessed 20 May 2025).

137.

Rumsey

Lorance

Jackson

, et al. Classical complement pathway inhibition in a ‘human-on-a-chip’ model of autoimmune demyelinating neuropathies. Adv Ther 2022; 5: 2200030.

138.

Hesperos . Human-on-a-chip data enables clinical trial (NCT04658472) highlighting potential for in vitro approach in lieu of animal studies for rare neuromuscular disorders. Hesperos Inc. https://hesperosinc.com/efficacy-study-for-rare-neuromuscular-disorders/ (2022, accessed 20 May 2025).

139.

Krebs

Lam

McCarthy

, et al. A survey to assess animal methods bias in scientific publishing. ALTEX 2023; 1–13.

140.

Krebs

Camp

Constantino

, et al. Author guide for addressing animal methods bias in publishing. Adv Sci 2023; 10: 2303226.

141.

Krebs

Camp

Constantino

, et al. Proceedings of a workshop to address animal methods bias in scientific publishing. ALTEX 2022; 1: 1–12.

142.

Taylor

Modi

Bailey

. An analysis of trends in the use of animal and non-animal methods in biomedical research and toxicology publications. Front Lab Chip Technol 2024; 3: 1426895.

143.

UK Parliament . Question for Department for Science, Innovation and Technology, UIN 4523, tabled on 6 September 2024. https://questions-statements.parliament.uk/written-questions/detail/2024-09-06/4523/ (2024, accessed 2025).

144.

DSIT . Government backs UK R&D with record £20.4 billion investment at Autumn Budget. https://www.gov.uk/government/news/government-backs-uk-rd-with-record-204-billion-investment-at-autumn-budget (2024, accessed 20 May 2025).

145.

Technopolis Group . Study into the finances of animal testing and animal-free research innovations. https://www.technopolis-group.com/report/study-into-the-finances-of-animal-testing-and-animal-free-research-innovations/ (2020, accessed 20 May 2025).

146.

Taylor

EU member state government contribution to alternative methods. ALTEX 2014; 31: 215–218.

147.

Deceuninck

Gastaldello

Mennecozzi

, et al. Exploring the relationship between the societal impact of EU-funded biomedical research in selected disease areas and the use of animal and nonanimal based approaches. Luxembourg: Publications Office of the European Union, 2024. DOI: 10.2760/110980JRC138356.

148.

UK Parliament . Animal testing, Volume 745: Debated on Monday 19 February 2024. https://hansard.parliament.uk/Commons/2024-02-19/debates/0C64E6BD-AE31-4DD7-80BB-3CEDC6272C2E/a (2024, accessed 20 May 2025).

149.

European Commission . Horizon Europe. https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-europe_en (2023, accessed 20 May 2025).

150.

Marshall

Constantino

Seidle

. Phase-in to phase-out — Targeted, inclusive strategies are needed to enable full replacement of animal use in the European Union. Anim Open Access J 2022; 12: 863.

151.

Fakoya

. Who is concerned about animal care and use in developing countries? ALTEX Proc 2012; 1: 265–269.

152.

Sharma

Verhagen

Elkins

, et al. Research from low-income and middle-income countries will benefit global health and the physiotherapy profession, but it requires support. Phys Ther 2023; 103: pzad081.

153.

Parvatam

Kaushik

. Why infection research in India must become relevant to humans. Nature India. https://www.nature.com/articles/nindia.2020.30 (2020, accessed 20 May 2025).

154.

Animal Free Research UK . Animal Free Research UK research funding. https://www.animalfreeresearchuk.org/funding/ (2017, accessed 20 May 2025).

155.

Humane Research Trust . Apply for a research grant. https://humaneresearch.org.uk/apply-for-funding/ (2024, accessed 20 May 2025).

156.

LifETIME

CDT. EPSRC-SFI

Joint Centre for Doctoral Training in Engineered Tissues for Discovery. Industry and Medicine. https://lifetime-cdt.org/ (2023, accessed 20 May 2025).

157.

NC3Rs . Research and innovation funding schemes. https://nc3rs.org.uk/our-funding-schemes (2024, accessed 20 May 2025).

158.

NC3Rs . BBSRC, NC3Rs and UKRI invest £4.7M in the development of non-animal technologies. https://nc3rs.org.uk/news/bbsrc-nc3rs-and-ukri-invest-ps47m-development-non-animal-technologies (2022, accessed 20 May 2025).

159.

NC3Rs . £4.85M to accelerate the use of non-animal approaches in research. https://nc3rs.org.uk/news/ps485m-accelerate-use-non-animal-approaches-research (2024, accessed 20 May 2025).

160.

National Institutes of Health . Funding opportunities. https://commonfund.nih.gov/complementarie/fundingopportunities (2024, accessed 20 May 2025).

161.

ALTEX . NIH funding for NAM development announced. https://www.altex.org/index.php/altex/announcement/view/570 (2024, accessed 20 May 2025).

162.

ALTEX . NIH offers funding for NAMs data hub and coordinating center. https://www.altex.org/index.php/altex/announcement/view/579 (2024, accessed 20 May 2025).

163.

National Toxicology Program . ICCVAM agency activities. https://ntp.niehs.nih.gov/whatwestudy/niceatm/iccvam/agency-activities (2025, accessed 20 May 2025).

164.

PETA.

The numbers don’t lie: NIH is financially bloated, ineffective, cruel. https://www.peta.org/blog/nih-numbers-bloated-ineffective-cruel/ (2020, accessed 20 May 2025).

165.

National Institutes of Health . Grants & funding. https://www.nih.gov/grants-funding (2025, accessed 20 May 2025).

166.

Foundation for the National Institutes of Health . Research support. https://fnih.org/our-programs/powering-science/research-support/ (2024, accessed 20 May 2025).

167.

NC3Rs . Launch of a roadmap for non-animal technologies. https://nc3rs.org.uk/news/launch-roadmap-non-animal-technologies (2023, accessed 20 May 2025).

168.

Wood

. Global 3D cell culture global markets, 2021–2028 — Growing focus towards the development of personalized medicine & regenerative medicine. PR Newswire. https://www.prnewswire.com/news-releases/global-3d-cell-culture-global-markets-2021-2028---growing-focus-towards-the-development-of-personalized-medicine--regenerative-medicine-301499941.html (2022, accessed 20 May 2025).

169.

Wood

. Global 3D cell culture market research report 2023: Market to reach $3.6 billion by 2030 — cancer and stem cell research leads the market. PR Newswire. https://www.prnewswire.com/news-releases/global-3d-cell-culture-market-research-report-2023-market-to-reach-3-6-billion-by-2030---cancer-and-stem-cell-research-leads-the-market-301820754.html (2023, accessed 20 May 2025).

170.

Valuates Reports . Organ-On-Chip (OOC) market size is USD 303.6 million by 2026 at CAGR 39.9%. PR Newswire. https://www.prnewswire.com/in/news-releases/organ-on-chip-ooc-market-size-is-usd-303-6-million-by-2026-at-cagr-39-9-valuates-reports-856995001.html (2021, accessed 20 May 2025).

171.

MarketsandMarkets. Organ-on-chip market size, share & trends. https://www.marketsandmarkets.com/Market-Reports/organs-on-chips-market-144117291.html (2024, accessed 20 May 2025).

172.

Hutchinson

Owen

Bailey

. Modernizing medical research to benefit people and animals. Animals 2022; 12: 1173.

173.

Deckha

Michel

Azilagbetor

, et al. Accelerating Animal Replacement: How Universities Can Lead — Results of a one-day expert workshop in Zurich, Switzerland. Altern Lab Anim 2025; 53: 106–118.

174.

Animal Aid . Animal Aid’s roadmap — Phasing out animal testing. https://www.animalaid.org.uk/wp-content/uploads/2024/11/Roadmap-to-phase-out-animal-testing_Amended-Feb-2025.pdf (2024, accessed 20 May 2025).

175.

Walder

Pallocca

Bastos

, et al. EU roadmap for phasing out animal testing for chemical safety assessments: Recommendations from a multi-stakeholder roundtable. ALTEX 2025; DOI: 10.14573/altex.2503241.

176.

Westmoreland

Bender

Doe

, et al. Use of New Approach Methodologies (NAMs) in regulatory decisions for chemical safety: Report from an EPAA Deep Dive Workshop. Regul Toxicol Pharmacol 2022; 135: 105261.

177.

Hanberg

. Academic training to promote the implementation of alternative methods and the Three Rs in the field of toxicology. Altern Lab Anim 2021; 49: 306–308.

178.

Modi

. Animal Free Research UK holds its first organ-on-chip training masterclass. https://www.animalfreeresearchuk.org/first-organ-on-chip-training-masterclass/ (2024, accessed 20 May 2025).

179.

Melbourne

Bishop

Brown

, et al. A multi-faceted approach to achieving the global acceptance of animal-free research methods. Altern Lab Anim 2016; 44: 495–498.

180.

McCarthy

Bailey

Baron

, et al. Creating training opportunities in new approach methodologies for early-career researchers. NAM J 2025; 1: 100006.

181.

Foundation for the National Institutes of Health . Validation & qualification network design phase. https://fnih.org/our-programs/validation-qualification-network-design-phase/ (accessed 20 May 2025).

182.

Indian Council of Medical Research . ICMR Collaborating Centre of Excellence (ICMR-CCoE). https://www.icmr.gov.in/icmr-collaborating-centre-of-excellence-icmr-ccoe?q=centre_of_excellence (accessed 20 May 2025).

183.

Reyes

Van Heeren

Guha

, et al. Accelerating innovation and commercialization through standardization of microfluidic-based medical devices. Lab Chip 2021; 21: 9–21.

184.

Herrmann

Jayne

(eds). Animal experimentation: Working towards a paradigm change. Leiden: Brill, 2019, 752 pp.

185.

Van Mulders

Liodo Missigba

Mertens

, et al. RE-place: A unique project collecting expertise on New Approach Methodologies. Front Pharmacol 2022; 13: 930148.

Breaking down the barriers to animal-free research

Abstract

Keywords

Introduction

What are the main scientific issues associated with animal use in research?

Translational issues due to genetic and physiological differences between humans and animals

The effects of stress and environment on data from animal experiments

The application of NAMs to reduce and replace the use of animals in research

What are the barriers to NAM adoption, and how can we address them?

Concerns about the reliability of NAMs to model the efficacy and safety of novel therapeutics

Perceived decrease in likelihood of publication

Reduced availability of funding

Future directions

Knowledge

Funding

Collaboration

Standardisation and regulatory approval

Conclusions

Footnotes

Acknowledgments

ORCID iD

Funding

Declaration of conflicting interests

Data Availability Statement

References