Non-animal models

A strategy for maturing Australia’s medical product 
development capabilities

2023

Citation and authorship

CSIRO (2023) Non-animal models: a strategy 
for maturing Australia’s medical product 
development capabilities. CSIRO, Canberra.

This report was authored by Vidip Arora, Nicolás 
González Castro, Rohini Poonyth, Laura Thomas, 
and Greg Williams with input from over 103 
industry, research and government leaders.

CSIRO Futures

At CSIRO Futures we bring together science, technology 
and economics to help governments and businesses 
develop transformative strategies that tackle their biggest 
challenges. As the strategic and economic advisory 
arm of Australia’s national science agency, we are 
uniquely positioned to transform complexity into clarity, 
uncertainty into opportunity, and insights into action.

Accessibility

CSIRO is committed to providing web accessible content 
wherever possible. If you are having difficulties with 
accessing this document, please contact csiro.au/contact.

Acknowledgement

CSIRO acknowledges the Traditional Owners of the 
land, sea and waters, of the area that we live and 
work on across Australia. We acknowledge their 
continuing connection to their culture, and we pay 
our respects to their Elders past and present.

The project team is grateful to the many stakeholders 
who generously gave their time to provide advice 
and feedback on this report through interviews and 
report reviews. In particular, we would like to thank the 
project sponsors who enabled this project to happen.

Copyright notice

© Commonwealth Scientific and Industrial Research 
Organisation 2023. To the extent permitted by law, all rights 
are reserved, and no part of this publication covered by 
copyright may be reproduced or copied in any form or by 
any means except with the written permission of CSIRO.

Disclaimer

CSIRO advises that the information contained in this 
publication comprises general statements based on 
scientific research. The reader is advised and needs to 
be aware that such information may be incomplete or 
unable to be used in any specific situation. No reliance 
or actions must therefore be made on that information 
without seeking prior expert professional, scientific 
and technical advice. To the extent permitted by 
law, CSIRO (including its employees and consultants) 
excludes all liability to any person for any consequences, 
including but not limited to all losses, damages, costs, 
expenses and any other compensation, arising directly 
or indirectly from using this publication (in part or in 
whole) and any information or material contained in it.



Sponsors



Executive summary

This report assesses the potential of emerging non-animal models to complement or 
replace traditional approaches in medical product development over the next 15 years. 
The analysis includes specific opportunities and recommendations for Australia to 
strategically enhance capability in this field, improve research quality and productivity, 
strengthen sovereign capability, and generate new national revenue streams. 

The report defines non-animal models as biological 
models that use human-derived or humanised cells, 
tissues or data. While the scope of this analysis is 
restricted to non-animal model applications across 
the medical product development process, the 
report’s recommendations could benefit applications 
in other fields, such as veterinary and agricultural 
medicines, cosmetic testing, and eco-toxicology. 

The report was informed by consultation with 
103 individuals from 66 organisations across 
industry, research and government.

Why non-animal models? 

The complexity of non-animal models is rapidly 
increasing, equating to or surpassing the performance of 
traditionally used animal models in several applications. 
Due to their enhanced biological relevance, non-animal 
models can increase productivity and reduce costs 
by identifying unsuitable medical products earlier in 
development and re-investing savings in more promising 
candidates. These models also support broader global 
‘3Rs’ objectives to replace, reduce and refine the use 
of animals for research and testing purposes.

Why now? 

Recent policy shifts in the United States and Europe 
encourage the transition away from animal use. These shifts 
are likely to support the already strong growth in the 
global non-animal testing market, valued at USD 1.11 billion 
in 2019, and expected to grow at a compound annual 
growth rate of 10.4% during 2019–2025. At the same time, 
animal model supply chains face increased risks, prompting 
alternative approaches to become more valuable.

Why Australia? 

Australia has comparative global strengths in non‑animal 
models for several organ systems likely to disrupt the 
status quo about the use of animal models over the 
next 15 years, including cardiovascular, respiratory, 
and nervous systems. Australia also possesses key 
foundational capabilities, including existing infrastructure 
(the National Collaborative Research Infrastructure 
Strategy – NCRIS – network), high throughput screening 
capabilities, and internationally recognised capacity for 
induced pluripotent stem cell generation, a key input 
for non-animal model development. These emerging 
models will also be critical to protect and further 
strengthen Australia’s $1.4 billion clinical trials sector.



Australia must act now to secure a key role 
in this emerging capability. 

Despite relevant research and infrastructure strengths, 
Australia is still maturing and coordinating these 
national capabilities. This report seeks to support 
these coordination efforts. It discusses how Australia 
can accelerate non-animal model applications’ 
demonstration, scaling, and commercial success.

The next 15 years will see an increase in the 
use of non‑animal models across all stages of 
the medical product development process.

The most significant growth is likely to come from complex in 
vitro models such as organoids (3D) (estimated $1.28 billion 
in revenue for Australia by 2040) and organ-on-chip (OoC) 
technologies ($310 million in revenue).1 In silico models 
are also anticipated to be more widely applied throughout 
the development process; used in conjunction with in vitro 
models to complement and validate findings. Figure 1 
summarises the expected shifts in non-animal model use 
across stages of the medical product development process.

The organ systems most likely to see non-animal models 
replace the status quo include cardiovascular, respiratory, 
gastrointestinal, skin, eye, and liver.

Figure 1. Expected shifts in the use of models for medical product development

Four national opportunities pair Australian strengths with global needs.

Developing and applying non-animal models within these settings (Figure 2) can benefit the quality of domestic research 
and development (R&D) activities or create revenue streams for non-animal model applications by providing global 
services and partnerships. 

Figure 2. Four national opportunities

FUNDAMENTAL 
RESEARCH

DISCOVERY 
DEVELOPMENT

PRECLINICAL 
DEVELOPMENT

CLINICAL 
DEVELOPMENT

REGULATORY 
APPROVAL AND 
COMPLIANCE

1. Complex in 
vitro models for 
improving the 
R&D productivity 
of national drug 
discovery.


Adoption of more 
complex and 
biologically relevant 
high-throughput 
in vitro models like 
organoids.

2. Organ-specific 
models for 
preclinical 
development


Research and industry 
collaborations in areas 
of national strength 
to support time and 
animal use reductions 
in preclinical testing.

3. Personalised 
models for trial 
participant and 
clinical treatment 
selection.


Protecting and further 
strengthening the 
competitiveness of 
Australia’s clinical trials 
sector and advancing 
precision medicine 
goals by incorporating 
patient-specific 
models.

4. Onshore production of model components. 


Increasing production capability for critical non-animal model inputs such as human-derived stem cells, 
non-animal derived media, and hydrogels.







Ten recommendations to provide 
Australia with the foundation to 
pursue these opportunities.

While the opportunities were developed considering 
a 15-year time horizon, setting Australia on a path 
towards these opportunities would require actioning the 
recommendations within five years. Within these five years, 
recommendations can be grouped and ordered by themes, 
with those aimed at coordinating and updating existing 
processes considered the most important first steps by 
those consulted (Figure 3). These activities would set a 
strong foundation for the remaining recommendations, 
which aim to integrate local capabilities into medical 
product development before strengthening production 
and commercialisation. Recommendations for non-animal 
model validation data will provide the evidence base to 
generate momentum across all themes. More discrete R&D 
priorities, actioned in parallel to recommendations, will 
act as supporting cross-cutting activities to strengthen 
Australia’s non-animal models’ capabilities further.

Figure 3. Recommendations and R&D priorities

Coordinate and update 
existing processes

1. Establish a national 
consortium that coordinates 
and promotes Australia’s 
non‑animal model capabilities 
2. Develop national data 
collection standards on the 
use of animals in scientific 
research, teaching and testing
3. Align TGA processes and 
industry guidance with new 
FDA procedures for accepting 
non-animal model data


Integrate local capabilities

4. Develop a national biobanking 
and tissue collection network 
5. Integrate outputs from NCRIS 
platforms into a coordinated 
pipeline for non-animal models


Strengthen production 
and commercialisation

6. Facilitate IP management and 
material access for research 
and industry collaborations
7. Enhance commercial 
skillsets across the 
non‑animal model sector
8. Update biomedical R&D 
infrastructure to support 
non‑animal model capabilities


Validation data

9. Conduct retrospective studies that compare animal and non-animal model predictivity
10. Conduct systematic reviews of locally and internationally developed non-animal models 


R&D priorities

• Support the economic case for non-animal models in medical product development
• Improve analytics for increasingly complex in vitro models
• Advance the quality of model inputs and hardware
• Extend the capabilities of in vitro models for a closer recreation of 
human physiology across healthy and diseased states




Glossary

TERM

DEFINITION 

3Rs

Framework for the ethical use of animals for scientific purposes centred on three principles:2 

Replacement is when non-animal alternatives can fully accomplish a project or process goal without animals. 
Reduction is when using fewer animals can accomplish the same goals, or increased information can be 
derived from the same number of animals. 
Refinement is when changes, practices or adaptations in a project or process can minimise the pain and 
distress of the animals used and enhance their well-being. 

Biobank

Collection of biological specimens (cells, tissues, derivative models, blood, bodily fluids, or genetic material, 
most commonly of human origin) and related donor information (demographic background, medically 
relevant history, and clinical data) established and made available to support research activities.

Biological models

Any form of investigation that attempts to replicate animal physiology (including humans).

High-throughput 
screening (HTS)

Simultaneous, automated testing of large compound or gene-targeting libraries in simplified in vitro 
settings to identify a subset (hits) exhibiting a desired physical, chemical, or biological activity. HTS is the 
gold standard for small molecule and compound screening, through which approximately 100,000 drug-like 
molecules can be tested against a target per day to observe reactions and identify hits.

Humanised cells

Cells initially derived from an animal modified to resemble human counterparts regarding genetic 
information, expression profiles, metabolic networks, or overall behaviour.

Induced pluripotent 
stem cells (iPSCs)

Cells with morphology and behaviour resembling that of embryonic stem cells and capable of differentiation 
into most mature cell subtypes. These cells are used widely for disease modelling, regenerative medicine 
research, and drug discovery.

In silico

Category of models that use computational settings to simulate biological systems and their responses to an 
intervention.

In vitro

Category of models using biological components in manufactured, controlled settings (e.g., a culture flask) 
that are exterior to the organism(s) from which the components are derived. 

In vitro to in vivo 
extrapolation (IVIVE)

In silico approach that scales up results obtained in vitro to anticipate a medical product’s behaviour at 
the organ or whole-organism level (e.g., pharmacokinetics/dynamics) and to calculate exposure levels 
associated with an effect (e.g., dosimetry).3

Non-animal model

A subset of biological models that use human-derived or humanised cells, tissues, or data.

Omics

Grouping term encompassing studies, technologies and data associated with genomics, epigenomics, 
transcriptomics, proteomics, metabolomics, and lipidomics.

Precision medicine (or 
personalised medicine)

An approach to designing medical care that optimises efficiency and therapeutic benefit for individual 
patients or patient groups, by using omics and molecular profiling to predict response and guide clinical 
decision-making.

Quantitative systems 
pharmacology (QSP)

In silico approach that models medical product behaviour in an organism by linking together molecular and 
cellular mechanisms, product characteristics, and whole-organism dynamics.

Health digital twin

A computational model of a tissue, organ or entire system that mirrors the physical counterpart found 
within a patient. The models draw information from real-time sources, medical history, and population-wide 
assessments to model personalised health outcomes and support clinical decision-making.4 





2 National Health and Medical Research Council (2019) Information paper: The implementation of the 3Rs in Australia. NHMRC, Canberra, Australia.

3 Chang X, Tan Y-M, Allen DG, Bell S, Brown PC, Browning L, Ceger P, Gearhart J, Hakkinen PJ, Kabadi SV, Kleinstreuer NC, Lumen A, Matheson J, Paini A, 
Pangburn HA, Petersen EJ, Reinke EN, Ribeiro AJS, Sipes N, Sweeney LM, Wambaugh JF, Wange R, Wetmore BA, Mumtaz M (2022) IVIVE: Facilitating the Use 
of In Vitro Toxicity Data in Risk Assessment and Decision Making. Toxics 10(5), 232.

4 Venkatesh KP, Raza MM, Kvedar JC (2022) Health digital twins as tools for precision medicine: Considerations for computation, implementation, and 
regulation. npj Digital Medicine 5(1), 150.



Contents
Sponsors..........................................................................................................................................................i

Executive summary...................................................................................................................................iiGlossary.........................................................................................................................................................vi1 Introduction............................................................................................................................................21.1 What are non-animal models?................................................................................................................................21.2 Why non-animal models?........................................................................................................................................21.3 Why now?.................................................................................................................................................................41.4 Why Australia?..........................................................................................................................................................52 Non-animal models in medical product development.....................................................62.1 Medical product development................................................................................................................................62.2 Current non-animal model use...............................................................................................................................72.3 Non-animal models over the next 15 years.........................................................................................................103 Non-animal models in Australia..................................................................................................153.1 Research strengths.................................................................................................................................................163.2 Clinical strengths....................................................................................................................................................173.3 Industry strengths..................................................................................................................................................173.4 Funding and investment .......................................................................................................................................184 National opportunities and recommendations....................................................................194.1 National opportunities..........................................................................................................................................194.2 Recommendations.................................................................................................................................................274.3 R&D priorities.........................................................................................................................................................345 Appendices..........................................................................................................................................35A.1 Consulted stakeholders.........................................................................................................................................35A.2 Biological model comparisons..............................................................................................................................36A.3 National strengths analysis....................................................................................................................................37A.4 Australian industry capability................................................................................................................................40A.5 Economic analysis methodology...........................................................................................................................43

1 Introduction

1.1 What are non-animal models?

All medical products require research and testing to 
determine their safety and efficacy before human use.5 
These activities are done in biological models, with 
most performed in live animals (animal models). 
Non‑animal models are a subset of biological models 
that use human-derived or humanised cells, tissues 
or data (Table 1). Key differences across existing 
non‑animal model types include their ability to mimic 
tissue features, the spatial layout of cells, and the use of 
controllable biophysical stimuli. Appendix A.2 outlines 
further comparisons across non-animal model types.

Table 1. Non-animal model types

NON-ANIMAL MODEL

DEFINITION

In silico

Computational modelling or simulation of biological systems and their responses to an intervention.6

In vitro 

2D

Cells cultured over a flat surface (well, culture plate or flask).

3D

Cell cultures that interact with an externally provided 3D environment (scaffold) or self-assemble into 3D 
structures (spheroids and organoids), enabling cell migration and interaction.

Organ-on-chip 
(OoC)

Culture in miniature engineered settings (chips) where the location of cells, the physical stimuli delivered 
to them (e.g., stretching), and a controllable flow mimic the characteristics of a tissue, organ or system.7

Tissue explant

Small, dissected fragments of primary tissue that preserve features like cell diversity and 
microenvironment architecture (e.g., tissue biopsies).





1.2 Why non-animal models?

The 3Rs, principles of replacing, reducing and 
refining the use of animals in research, have been a 
fundamental ethical-research framework since the 1960s.8 
While the use of live animals in medical research 
will continue to be an important part of the medical 
product development process for the foreseeable 
future, the Australian code for the care and use of 
animals for scientific purposes requires the application 
of the 3Rs at all stages of animal care and use.9 

5 Medical products are small molecules and biologicals that can be used to detect or treat disease, including diagnostics, therapeutic products, vaccines, and medical devices.

6 Piñero J, Furlong LI, Sanz F (2018) In silico models in drug development: where we are. Current Opinion in Pharmacology 42, 111–121.

7 Leung CM, Haan P de, Ronaldson-Bouchard K, Kim G-A, Ko J, Rho HS, Chen Z, Habibovic P, Jeon NL, Takayama S, Shuler ML, Vunjak-Novakovic G, Frey O, Verpoorte E, 
Toh Y-C (2022) A guide to the organ-on-a-chip. Nature Reviews Methods Primers 2(1), 33.

8 National Health and Medical Research Council (n.d.) The 3Rs. <https://www.nhmrc.gov.au/research-policy/ethics/animal-ethics/3rs> (accessed 10 July 2023).

9 National Health and Medical Research Council (2013) Australian code for the care and use of animals for scientific purposes, 8th edn, Commonwealth of Australia, 
Canberra, Australia. <https://www.nhmrc.gov.au/about-us/publications/australian-code-care-and-use-animals-scientific-purposes#toc__167> (accessed 10 July 2023).



In addition to ethical drivers for using alternatives 
to animal models, mature and emerging non-animal 
models possess a range of characteristics that can 
offer complementary benefits to animal models and, 
in time, may evolve to replace the use of animals in 
several research settings. These benefits include:

• Enhanced biological relevance: Several non-animal 
models offer the potential to better mimic human 
responses compared to animal models, and they can 
more accurately account for human population diversity, 
as they are derived from human cells.10 Limitations 
of animal models include inter-species differences, 
inter-study variability, and low predictivity of toxicity 
in phase 1 clinical trials.11 Whereas non-animal models 
have successfully predicted human clinical findings 
across toxicity, biomarker signals, and drug sensitivity.12 
• Cost savings and increased productivity: Approximately 
90% of medical products fail in clinical development 
(during phase I, II and III clinical trials) despite the 
use of animal tests, in large part due to efficacy or 
safety issues.13 The use of more biologically relevant 
models may identify these issues earlier. These models 
may help products fail where necessary before 
human trials, which account for 69% of R&D costs 
in pharmaceutical development.14 Earlier failure of 
unsuitable products can lower attrition rates during 
human trials and focus investment on promising 
candidates, directly impacting sector productivity. 
For example, an economic impact study estimated that 
widely adopting a single organ-on-chip for a single 
liver toxicity test could generate an additional USD 
3 billion annually for the pharmaceutical sector.15
• Alternative where animal model use is challenging: 
Non-animal models can model rare diseases and 
disorders that would be costlier or more difficult to 
study through traditional approaches (animal models 
and clinical trials), due to statistical limitations and 
lower revenue potential of smaller patient populations.16 
Non-animal models can also be used for research and 
testing where the disease, condition, or toxicity of 
interest cannot be reliably induced in an animal.17
• Applicable to high-throughput screening (HTS): 
Some non-animal models have significant potential 
to interface with HTS. This approach is unsuitable for 
animal models due to the many compounds tested. 
Conservatively, HTS offers a 50- to 200‑fold increase in 
testing efficiency compared to conventional in vitro 2D 
methods.18 HTS allows for faster and potentially more 
cost-effective testing of multiple candidates and doses for 
medical product development and precision medicine. 
• Social license benefits: As the social license for 
animal research and testing diminishes, the demand 
for products developed using alternatives will grow. 
Transitioning away from animal models may also increase 
employee satisfaction across the research community. 


As emerging non-animal models continue advancing 
through validation processes, which require the support 
of animal models, the short to medium term will likely 
see non-animal models complement and rationalise 
animal model use rather than fully replace it. 

10 Additional comparisons of biological resemblance, technological maturity and other considerations can be found in Appendix A.2.

11 Atkins JT, George GC, Hess K, Marcelo-Lewis KL, Yuan Y, Borthakur G, Khozin S, LoRusso P, Hong DS (2020) Pre-clinical animal models are poor predictors of 
human toxicities in phase 1 oncology clinical trials. British Journal of Cancer 123(10), 1496–1501.

12 Dowden H, Munro J (2019) Trends in clinical success rates and therapeutic focus. Nature Reviews Drug Discovery 18(7), 495–496; Grossman JE, Muthuswamy 
L, Huang L, Akshinthala D, Perea S, Gonzalez RS, Tsai LL, Cohen J, Bockorny B, Bullock AJ, Schlechter B, Peters MLB, Conahan C, Narasimhan S, Lim C, Davis 
RB, Besaw R, Sawhney MS, Pleskow D, Berzin TM, Smith M, Kent TS, Callery M, Muthuswamy SK, Hidalgo M (2022) Organoid Sensitivity Correlates with 
Therapeutic Response in Patients with Pancreatic Cancer. Clinical Cancer Research 28(4), 708–718; Jang K-J, Otieno MA, Ronxhi J, Lim H-K, Ewart L, Kodella 
KR, Petropolis DB, Kulkarni G, Rubins JE, Conegliano D, Nawroth J, Simic D, Lam W, Singer M, Barale E, Singh B, Sonee M, Streeter AJ, Manthey C, Jones B, 
Srivastava A, Andersson LC, Williams D, Park H, Barrile R, Sliz J, Herland A, Haney S, Karalis K, Ingber DE, Hamilton GA (2019) Reproducing human and cross-
species drug toxicities using a Liver-Chip. Science Translational Medicine 11(517), eaax5516; Ronaldson-Bouchard K, Teles D, Yeager K, Tavakol DN, Zhao Y, 
Chramiec A, Tagore S, Summers M, Stylianos S, Tamargo M, Lee BM, Halligan SP, Abaci EH, Guo Z, Jacków J, Pappalardo A, Shih J, Soni RK, Sonar S, German 
C, Christiano AM, Califano A, Hirschi KK, Chen CS, Przekwas A, Vunjak-Novakovic G (2022) A multi-organ chip with matured tissue niches linked by vascular 
flow. Nature Biomedical Engineering 6(4), 351–371.

13 Dowden H, Munro J (2019) Trends in clinical success rates and therapeutic focus. Nature Reviews Drug Discovery 18(7), 495–496; Harrison RK (2016) 
Phase II and phase III failures: 2013–2015. Nature Reviews Drug Discovery 15(12), 817–818; Sun D, Gao W, Hu H, Zhou S (2022) Why 90% of clinical drug 
development fails and how to improve it? Acta Pharmaceutica Sinica B 12(7), 3049–3062.

14 Congressional Budget Office (2021) Research and Development in the Pharmaceutical Industry, United States. <https://www.cbo.gov/publication/57126> 
(accessed 10 July 2023).

15 Ewart L, Apostolou A, Briggs SA, Carman CV, Chaff JT, Heng AR, Jadalannagari S, Janardhanan J, Jang K-J, Joshipura SR, Kadam MM, Kanellias M, Kujala 
VJ, Kulkarni G, Le CY, Lucchesi C, Manatakis DV, Maniar KK, Quinn ME, Ravan JS, Rizos AC, Sauld JFK, Sliz JD, Tien-Street W, Trinidad DR, Velez J, Wendell 
M, Irrechukwu O, Mahalingaiah PK, Ingber DE, Scannell JW, Levner D (2022) Performance assessment and economic analysis of a human Liver-Chip for 
predictive toxicology. Communications Medicine 2(1), 154.

16 Blumenrath SH, Lee BY, Low L, Prithviraj R, Tagle D (2020) Tackling rare diseases: Clinical trials on chips. Experimental Biology and Medicine 245(13), 1155–
1162; Roth A, MPS-WS Berlin 2019 (2021) Human microphysiological systems for drug development. Science 373(6561), 1304–1306.

17 National Center for Advancing Translational Sciences (2022) Researchers Create 3-D Model for Rare Neuromuscular Disorders, Setting Stage for Clinical Trial. 
<https://ncats.nih.gov/news/releases/2022/researchers-create-3-D-model-for-rare-neuromuscular-disorders-setting-stage-for-clinical-trial> (accessed 10 July 
2023); Rumsey JW, Lorance C, Jackson M, Sasserath T, McAleer CW, Long CJ, Goswami A, Russo MA, Raja SM, Gable KL, Emmett D, Hobson-Webb LD, Chopra 
M, Howard JF, Guptill JT, Storek MJ, Alonso-Alonso M, Atassi N, Panicker S, Parry G, Hammond T, Hickman JJ (2022) Classical Complement Pathway Inhibition 
in a “Human-On-A-Chip” Model of Autoimmune Demyelinating Neuropathies. Advanced Therapeutics 5(6), 2200030.

18 Based on HTS 96 well format. Hosseinzadeh S (n.d.) Navigating Drug Discovery with High-Throughput Screening – BIT 479/579 High-throughput Discovery. 
<https://htds.wordpress.ncsu.edu/topics/novel-high-throughput-micro-nanofluidic-technology/> (accessed 10 July 2023).



1.3 Why now?

The biological model landscape is changing rapidly, and 
several factors are driving an escalation in the pace of 
development and adoption of non-animal models, including:

• Limited and vulnerable access to animals: There are 
constraints on the availability of non-human primates 
for medical research due to increased demand 
from infectious disease research spurred by the 
COVID-19 pandemic, international transport issues, 
and limitations imposed by major suppliers in Asia.19 
There is a potential 5-year waiting list in Australia 
at the National Non‑Human Primate Breeding and 
Research Facility, arising from the inherent difficulty 
in forecasting and matching future demand.20 
Mice, which comprise the majority of animals used for 
human or animal biology research purposes in Australia, 
also face local supply challenges, as demonstrated 
by the 2021 near closure of the Western Australia 
Animal Resources Centre.21 Further, importing animal 
models not produced locally (e.g., guinea pigs and 
golden hamsters) involves a logistically demanding 
process with strict biosecurity regulations.22
• International policy changes: International policy, 
regulation and legislation changes encourage (and 
sometimes mandate) the development and adoption 
of non-animal models.23 For example, European Union 
(EU) law requires non-animal models and approaches 
wherever possible, and the European Parliament has 
passed a resolution requesting an EU-wide action 
plan for ending animal model testing by 2030.24 In the 
United States (US), the Environmental Protection Agency 
announced plans to eliminate requests for and funding of 
all mammalian studies by 2035.25 Further, the US Congress 
passed the ‘FDA Modernization Act 2.0’ in 2022, 
modifying pre-existing language explicitly to allow the 
use of non-animal models to fulfil testing requirements.26
• Global capability is rapidly maturing: Leading 
jurisdictions, such as the US, EU, and the United Kingdom 
(UK),27 have created non-animal model strategies 
and established institutions dedicated to supporting 
non-animal model development and validation. 
Examples include the US National Center for Advancing 
Translational Sciences (NCATS) and its Tissue Chip 
consortium, the UK National Centre for the Replacement 
Refinement and Reduction of Animals in Research 
(NC3Rs), the EU Organ-On-Chip Society (EUROoCs) and 
Reference Laboratory for alternatives to animal testing 
(EURL ECVAM), and the Netherlands Human Organ and 
Disease Model Technologies consortium (hDMT).


19 Einhorn B, Lew L (28 September 2022) Lab Monkeys Are the Latest Covid Shortage. Bloomberg.com. <https://www.bloomberg.com/news/
articles/2022-09-28/research-monkey-shortage-boosts-china-s-vaccine-development> (accessed 10 July 2023); Ramos KS, Downey A, Yost OC (Eds) (2023) 
Nonhuman Primate Models in Biomedical Research: State of the Science and Future Needs. National Academies Press, Washington, D.C.; Roth A, MPS-WS 
Berlin 2019 (2021) Human microphysiological systems for drug development. Science 373(6561), 1304–1306.

20 Monash Animal Research Platform, consultation (2023).

21 Animal Welfare Victoria (2022) Statistics of animal use in research and teaching, Victoria: 1 January 2020 – 31 December 2020, Department of Jobs, 
Precincts and Regions, Melbourne, Australia. <https://agriculture.vic.gov.au/__data/assets/pdf_file/0009/880047/2020-Statistics-of-animal-use-in-research-
and-teaching-report_FINAL.pdf> (accessed 10 July 2023); Department of Natural Resources and Environment Tasmania (2022) Animal Research Statistics 
Tasmania: Annual Report. State of Tasmania, Hobart, Australia. <https://nre.tas.gov.au/Documents/Animal%20Research%20Report%20Number%2026%20
for%202021.pdf> (accessed 10 July 2023); Mallapaty S (2021) Loss of Australia’s largest lab animal supplier will leave “huge gap”. Nature, 10 July; NSW 
Government Department of Primary Industries (2021) NSW 2020 Animal Use in Research Statistics. NSW Government, Sydney, Australia. <https://www.
animalethics.org.au/__data/assets/pdf_file/0007/1395466/INT21-148540-2020-Animal-use-in-research-statistics-report.pdf> (accessed 10 July 2023); Turner L 
(2021) DAF Animal Ethics Committees Annual Report Summary for the period 1 July 2020 to 30 June 2021. State of Queensland, Brisbane, Australia. <https://
www.publications.qld.gov.au/ckan-publications-attachments-prod/resources/fa04f14f-f5b3-48bc-a7ca-136e93b69961/summary-2020-21-aec-annual-report.
pdf?ETag=07b43ba4a0e56f916a5f99c147747aa8> (accessed 10 July 2023).

22 Australian Government Department of Agriculture, Fisheries and Forestry (2020) Conditions for importing live animals into Australia for laboratory research. 
<https://www.agriculture.gov.au/biosecurity-trade/import/goods/live-animals/laboratory-animals> (accessed 10 July 2023).

23 Handley E (2023) Phasing out the use of animals in science. Open Access Government. <https://www.openaccessgovernment.org/animals-science-testing-
experiments-research-2/160677/> (accessed 10 July 2023).

24 European Parliament and the Council of the European Union (2010) Directive 2010/63/EU of the European Parliament and of the Council of 22 September 
2010 on the protection of animals used for scientific purposes. <https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32010L0063&from=EN> 
(accessed 10 July 2023); Hazekamp A (2021) MOTION FOR A RESOLUTION on a coordinated Union-level Action Plan to facilitate the transition to innovation 
without the use of animals in research, regulatory testing and education | B9-0427/2021. European Parliament. <https://www.europarl.europa.eu/doceo/
document/B-9-2021-0427_EN.html> (accessed 10 July 2023).

25 US Environmental Protection Agency (2019) Administrator Wheeler Signs Memo to Reduce Animal Testing, Awards $4.25 Million to Advance Research on 
Alternative Methods to Animal Testing. <https://www.epa.gov/newsreleases/administrator-wheeler-signs-memo-reduce-animal-testing-awards-425-million-
advance> (accessed 10 July 2023).

26 117th United States Congress (2022) FDA Modernization Act 2.0. <https://www.congress.gov/117/bills/s5002/BILLS-117s5002cps.pdf> (accessed 10 July 
2023).

27 Interagency Coordinating Committee on the Validation of Alternative Methods (2018) A Strategic Roadmap for Establishing New Approaches to Evaluate the 
Safety of Chemicals and Medical Products in the United States. National Toxicology Program (NTP).



• Leading jurisdictions are also establishing specific 
funding and translational projects for non-animal 
models, such as the NCATS’ Tissue Chip for Drug 
Screening and past EU projects from Horizon 2020 like 
ORCHID,28 OrganTrans,29 and PATROLS.30 In 2022 the 
US National Institutes of Health (NIH) provided USD 89 
million to 214 organoid-related projects and USD 256 
million to 608 organ-on-chip-related projects.31 Similarly, 
the EU will contribute up to EUR 37 million of dedicated 
funding in 2024 for non-animal model research.32
• Industry shifts: Pharmaceutical companies are 
increasingly focusing on improved data collection and 
transparency around animal testing to drive innovation 
and advance the 3Rs.33 Pharmaceutical companies 
are also leveraging non-animal models to support 
preclinical research and reduce the number of animals 
used, as demonstrated by Roche’s investments in 
human‑relevant models that have seen a 40% reduction 
in experimental animal use over the past ten years.34
• Growing global market: There is a growing 
international and domestic market of alternatives 
to animal testing.35 The global non-animal testing 
market’s value was USD 1.11 billion in 2019, and is 
expected to grow at a compound annual growth 
rate (CAGR) of 10.4% during 2019–2025.36


1.4 Why Australia?

Australia has comparative global strengths in various 
non-animal model applications (see Section 3). 
World‑class Australian researchers across diverse 
organ systems support these. Australia also boasts a 
globally competitive clinical trials sector – valued at 
$1.4 billion annually – that can be protected and further 
strengthened by applying non-animal models.37

Building non-animal model capabilities can also 
provide flow-on benefits to other nationally important 
sectors (e.g., through applications in veterinary and 
agricultural medicines development), to valuable 
national environmental assets (e.g., through 
applications in eco-toxicology), and to ancillary 
technologies (e.g., microfabrication, data capture 
and analysis, and biological supplies). 

Australia possesses a foundation of non-animal 
model‑related expertise, academic output, and research 
infrastructure. However, it is yet to leverage these 
characteristics into a comprehensive national capability 
for medical product development. Through nationally 
coordinated investments into non-animal model 
capabilities, Australia can pursue the opportunities 
outlined in this report (Section 4.1) to improve 
domestic R&D and generate novel revenue streams 
by providing global services and partnerships.

28 Mastrangeli M, Millet S, Mummery C, Loskill P, Braeken D, Eberle W, Madalena C, Fernandez L, Graef M, Gidrol X, Picollet-D’Hahan N, van Meer B, Ochoa I, 
Schutte M, van den Eijnden-van Raaij J (2019) Organ-on-Chip In Development ORCHID Final Report. <https://h2020-orchid.eu/> (accessed 10 July 2023).

29 OrganTrans Consortium (2020) OrganTrans: Process for organoids transplantation. <https://organtrans.eu/#aboutproject> (accessed 10 July 2023).

30 PATROLS - Physiologically Anchored Tools for Realistic nanOmateriaL hazard aSsessment (2021) PATROLS: Advanced Tools for NanoSafety Testing. <https://
www.patrols-h2020.eu/about-us/index.php> (accessed 10 July 2023).

31 United States National Institutes of Health (2023) Modernized NIH RePORTER version 2020.9 (02/04/2023 data). <https://reporter.nih.gov/> (accessed 2 April 
2023).

32 European Commission - Single Electronic Data Interchange Area (SEDIA) (2023a) Gaining experience and confidence in New Approach Methodologies 
(NAM) for regulatory safety and efficacy testing – coordinated training and experience exchange for regulators, Funding & tender opportunities. <https://
ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/topic-details/horizon-hlth-2024-ind-06-09> (accessed 10 July 2023); European 
Commission - Single Electronic Data Interchange Area (SEDIA) (2023b) Innovative non-animal human-based tools and strategies for biomedical research, 
Funding & tender opportunities. <https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/topic-details/horizon-hlth-2024-
tool-05-06-two-stage;callCode=HORIZON-HLTH-2024-TOOL-05-two-stage> (accessed 10 July 2023).

33 Novartis (2023) Novartis in Society Integrated Report 2022. <https://www.novartis.com/sites/novartiscom/files/novartis-integrated-report-2022.pdf> 
(accessed 10 July 2023).

34 F. Hoffmann-La Roche (2023) Reducing animal testing, Animal Research at Roche. <https://live.roche.com/innovation/ethical-standards/animal-research/
alternatives> (accessed 10 July 2023).

35 Vitika V, Surya N (2022) Non-Animal Alternative Testing Market, Allied Market Research. <https://www.alliedmarketresearch.com/non-animal-alternative-
testing-market-A25675> (accessed 10 July 2023).

36 Global non-animal testing market includes academic research institutions and various industries (such as pharmaceuticals, medical devices, chemicals & 
pesticides and cosmetics) performing animal testing to conduct basic research, toxicology profiling, and others. The Business Research Company (2023) 
Global Animal Testing And Non-Animal Alternative Testing Market Report And Strategies To 2032. <https://www.thebusinessresearchcompany.com/report/
animal-testing-and-non-animal-alternative-testing-market> (accessed 10 July 2023). 

37 MTPConnect (2021) Australia’s Clinical Trials Sector: Advancing innovative healthcare and powering economic growth. <https://www.mtpconnect.org.au/
images/MTPConnect_Australia's%20Clinical%20Trials%20Sector%20report%202021.pdf> (accessed 10 July 2023).



2 Non-animal models in medical 
product development

2.1 Medical product development

Medical products must follow a development process before 
being approved for human use and commercialisation 
(Table 2). Each stage of development aims to advance 
the understanding of a product’s interactions within the 
human body, refine its characteristics to maximise efficacy 
and minimise safety concerns. Product development 
does not necessarily follow a sequential path 
through these stages, as regulatory and validation 
requirements can vary by product type and jurisdiction. 
However, the progression through these stages relies 
on satisfying mandatory primary activities before 
advancing to human trials in clinical development.

Table 2. Medical product development stages

DEVELOPMENT 
STAGE

PURPOSE

PRIMARY ACTIVITIES

Fundamental 
research

Typically follows one of two objectives: 

• Understanding the mechanisms of 
organ and tissue function.
• Modelling disease phenotypes to 
observe cell behaviour and interaction.


• Basic science
• Disease modelling
• Model development and validation


Discovery 
development

Selecting a potential medical product 
(candidate) for preclinical development. 

• Target identification and validation
• Compound screening and lead optimisation 


Preclinical 
development

Determining the safety and efficacy of 
the medical product candidate prior to 
studies in humans.

• Safety – Pharmacodynamics, pharmacokinetics, and toxicology
• Efficacy – Biopharmaceutics and clinical efficacy-supporting studies


Clinical 
development

Evaluating the efficacy of new medical 
products and the application of novel 
clinical techniques directly in humans. 

• Clinical trials – Phase I–III
• Clinical applications, e.g., precision medicine


Regulatory 
approval and 
compliance

Obtaining approval before initiating 
human studies in clinical development, 
when progressing through phases 
of clinical trials, and when obtaining 
marketing authorisation prior to 
commercialisation.

Approval

• Safety assessments of complex or novel therapeutic goods
• Validation and qualification of models for specific contexts of use


Compliance

• Batch release testing for well-known pharmaceutical ingredients
• Categorisation of hazards for product labelling of well‑established 
and low-complexity therapeutic goods








2.2 Current non-animal model use

A range of non-animal models are currently used in medical 
product development. However, the application of specific 
types varies by development stage. Investigation of all 
non-animal model types for their potential use in medical 
product development is underway; however, the inclusion 
of a model type in the current state analysis was determined 
by whether the model type is already used by the 
industry for internal decision-making or in the regulatory 
approval of a novel medical product (see Figure 4).

Figure 4. Current models used for medical product development

Fundamental research

Animals are the most common biological model used 
for fundamental research. The latest comparable 
Australian State and Territory data indicates that the 
use of animals for ‘understanding human or animal 
biology’ accounted for 20.2% of all reported research 
and teaching uses in Victoria in 2020, 12.1% in New 
South Wales in 2020, and 12.6% in Tasmania in 2021.38

In vitro 2D is the most common non-animal model 
type used in fundamental research, providing insight 
into cell behaviour, gene activity and response to 
external compounds for many diseases. For instance, 
the Cancer Cell Line Encyclopedia was a project 
that performed an extensive pharmacological 
profiling of cancer cell lines using the traditional 2D 
culture setting.39 In recent years, in silico and more 
complex in vitro models (3D, organ‑on‑chip and 
tissue explants) are becoming increasingly common 
in fundamental research, both in replacing animal 
models and providing new sources of insight.40 



38 Animal Welfare Victoria (2022) Statistics of animal use in research and teaching, Victoria: 1 January 2020 – 31 December 2020, Department of Jobs, 
Precincts and Regions, Melbourne, Australia. <https://agriculture.vic.gov.au/__data/assets/pdf_file/0009/880047/2020-Statistics-of-animal-use-in-research-
and-teaching-report_FINAL.pdf> (accessed 10 July 2023); Department of Natural Resources and Environment Tasmania (2022) Animal Research Statistics 
Tasmania: Annual Report. State of Tasmania, Hobart, Australia. <https://nre.tas.gov.au/Documents/Animal%20Research%20Report%20Number%2026%20
for%202021.pdf> (accessed 10 July 2023); NSW Government Department of Primary Industries (2021) NSW 2020 Animal Use in Research Statistics. NSW 
Government, Sydney, Australia. <https://www.animalethics.org.au/__data/assets/pdf_file/0007/1395466/INT21-148540-2020-Animal-use-in-research-
statistics-report.pdf> (accessed 10 July 2023); Turner L (2021) DAF Animal Ethics Committees Annual Report Summary for the period 1 July 2020 to 30 June 
2021. State of Queensland, Brisbane, Australia. <https://www.publications.qld.gov.au/ckan-publications-attachments-prod/resources/fa04f14f-f5b3-48bc-
a7ca-136e93b69961/summary-2020-21-aec-annual-report.pdf?ETag=07b43ba4a0e56f916a5f99c147747aa8> (accessed 10 July 2023).

39 Broad Institute (n.d.) Cancer Cell Line Encyclopedia (CCLE). <https://sites.broadinstitute.org/ccle/> (accessed 11 July 2023).

40 Blinova K, Dang Q, Millard D, Smith G, Pierson J, Guo L, Brock M, Lu HR, Kraushaar U, Zeng H, Shi H, Zhang X, Sawada K, Osada T, Kanda Y, Sekino 
Y, Pang L, Feaster TK, Kettenhofen R, Stockbridge N, Strauss DG, Gintant G (2018) International Multisite Study of Human-Induced Pluripotent Stem 
Cell‑Derived Cardiomyocytes for Drug Proarrhythmic Potential Assessment. Cell Reports 24(13), 3582–3592; Cimen Bozkus C, Bhardwaj N (2021) Tumor 
organoid‑originated biomarkers predict immune response to PD-1 blockade. Cancer Cell 39(9), 1187–1189; Lawrence CL, Pollard CE, Hammond TG, Valentin 
J-P (2008) In vitro models of proarrhythmia. British Journal of Pharmacology 154(7), 1516–1522.



Discovery development

Target identification uses in vitro 2D models due to the 
large number of compounds and experimental conditions 
to evaluate, as well as the high reproducibility required. 
On the other hand, target validation studies use animal 
models (mainly mice) due to their physiological complexity, 
lower cost, short lifespan, and relative ease of breeding. 

Target identification and validation activities are sometimes 
complemented by in silico models or complex in vitro 
models such as organoids or organs-on-chips. In silico 
models can incorporate and analyse primary data to 
identify, and at times even select and prioritise, potential 
targets.41 At the same time, organoid and organ-on-chip 
models pose value due to their higher complexity across 
cellular interaction and structure, increased biological 
relevance, and their amenability to genetic modification.42 
The increasingly complementary role of non-animal models 
stems from limitations of disease modelling using animals. 
For instance, mice are often used as genetic disease models 
in screenings to validate identified therapeutic targets.43 
However, they can be inaccurate in fully recreating disease 
hallmarks. This is the case for Alzheimer’s disease (AD), 
with a complex ‘brain in a dish’ organoid model more 
closely demonstrating anatomical and physiological 
properties of AD-impacted human brain tissue.44

As with target identification, compound screening 
activities require models that can be used at high 
speeds with minimal variation (i.e., highly reproducible) 
and so are commonly performed using in vitro 2D 
models. In recent years higher complexity in vitro 3D 
models have also been used for compound screening 
when a lower throughput is acceptable, as they can 
be more representative of human physiology and 
potentially more accurate in predicting responses.

Lead optimisation, which involves making 
chemical modifications to the most promising 
molecules’ characteristics (i.e., their structure) 
to generate refined drug-like compounds, is 
informed by both in vitro and in silico models.

Preclinical development

While a large amount of preclinical data is generated 
using simple, lower-cost in vitro 2D models, many 
pharmacology studies in preclinical development require 
disease-specific and physiologically complex models to 
gain a more accurate insight into systemic effects. In these 
cases, animal models are commonly used alongside in 
silico models, such as quantitative systems pharmacology 
(QSP) and physiologically based pharmacokinetic (PBPK) 
models. In support of the transition away from animal 
use in pharmacology studies, the US Food and Drug 
Administration (FDA) has waived the requirement for clinical 
drug-drug interaction studies in animal models where in 
vitro tests have demonstrated strong in vitro-in vivo human 
correlation and validated their human concordance.45

Preclinical toxicology studies are highly organ-specific, 
and regulators often require these studies in two animal 
models (a rodent and a non-rodent species), complemented 
by in vitro 2D models. However, more complex in vitro 
models (i.e., organoids and organs‑on‑chips) and in silico 
models have begun to demonstrate higher accuracy in 
assessing organ-specific toxicology.46 As an illustration, 
Emulate (US), a spin-out from Harvard University, 
produces organ-on-chip models across multiple preclinical 
applications and organs.47 In 2019, their liver-chip model 
demonstrated a more accurate prediction of human 
toxicity than animal model equivalents, sometimes 
flagging compounds that animal models missed.48 

41 Agamah FE, Mazandu GK, Hassan R, Bope CD, Thomford NE, Ghansah A, Chimusa ER (2020) Computational/in silico methods in drug target and lead 
prediction. Briefings in Bioinformatics 21(5), 1663–1675.

42 Shamshirgaran Y, Jonebring A, Svensson A, Leefa I, Bohlooly-Y M, Firth M, Woollard KJ, Hofherr A, Rogers IM, Hicks R (2021) Rapid target validation in a 
Cas9‑inducible hiPSC derived kidney model. Scientific Reports 11(1), 16532.

43 Buchovecky CM, Turley SD, Brown HM, Kyle SM, McDonald JG, Liu B, Pieper AA, Huang W, Katz DM, Russell DW, Shendure J, Justice MJ (2013) A suppressor 
screen in Mecp2 mutant mice implicates cholesterol metabolism in Rett syndrome. Nature Genetics 45(9), 1013–1020; Landrette SF, Xu T (2011) Somatic 
Genetics Empowers the Mouse for Modeling and Interrogating Developmental and Disease Processes. PLoS Genetics 7(7), e1002110.

44 Blanchard JW, Victor MB, Tsai L-H (2022) Dissecting the complexities of Alzheimer disease with in vitro models of the human brain. Nature Reviews 
Neurology 18(1), 25–39.

45 Zhang D, Luo G, Ding X, Lu C (2012) Preclinical experimental models of drug metabolism and disposition in drug discovery and development. 
Acta Pharmaceutica Sinica B 2(6), 549–561.

46 Jean D, Naik K, Milligan L, Hall S, Mei Huang S, Isoherranen N, Kuemmel C, Seo P, Tegenge MA, Wang Y, Yang Y, Zhang X, Zhao L, Zhao P, Benjamin J, 
Bergman K, Grillo J, Madabushi R, Wu F, Zhu H, Zineh I (2021) Development of best practices in physiologically based pharmacokinetic modeling to support 
clinical pharmacology regulatory decision-making—A workshop summary. CPT: Pharmacometrics & Systems Pharmacology 10(11), 1271–1275; Matsui T, 
Shinozawa T (2021) Human Organoids for Predictive Toxicology Research and Drug Development. Frontiers in Genetics 12, 767621.

47 Emulate (n.d.) About Us – A Culture of Innovation. <https://emulatebio.com/about/> (accessed 11 July 2023). 

48 Jang K-J, Otieno MA, Ronxhi J, Lim H-K, Ewart L, Kodella KR, Petropolis DB, Kulkarni G, Rubins JE, Conegliano D, Nawroth J, Simic D, Lam W, Singer M, Barale 
E, Singh B, Sonee M, Streeter AJ, Manthey C, Jones B, Srivastava A, Andersson LC, Williams D, Park H, Barrile R, Sliz J, Herland A, Haney S, Karalis K, Ingber DE, 
Hamilton GA (2019) Reproducing human and cross-species drug toxicities using a Liver-Chip. Science Translational Medicine 11(517), eaax5516.



Clinical development

While clinical development involves trialling medical 
products directly in humans, the use of organoids 
supports personalised medicine in clinical settings and 
can help investigate the mechanisms of action behind 
unexpected responses.49 These patient-specific models can 
predict an individual’s response to an already approved 
medical product without bearing the risk of an adverse 
reaction or treatment resistance. For example, intestinal 
organoid models have successfully been used in the 
Netherlands for several years to inform therapeutic 
regimens for cystic fibrosis patients.50 Similarly, several 
organisations within Australia are exploring the use of 
patient‑derived organoids to predict response to treatment 
for colorectal, pancreatic and breast cancer patients.51

Regulatory approval and compliance 

The formal consideration of non-animal model 
data in submission packages for clinical trial 
authorisation and regulatory compliance activities 
depends on that non-animal model’s validation 
and acceptance for a specific application. 

The FDA, the European Medicines Agency (EMA), 
and organisations like the Organisation for 
Economic Co‑operation and Development (OECD) 
Test Guidelines Programme and the International 
Council for Harmonisation of Technical Requirements 
for Pharmaceuticals for Human Use (ICH), have 
already validated and adopted protocols on a range of 
non‑animal models. The internationally harmonised 
OECD Test Guidelines now include 22 protocols for 
non-animal models and corresponding applications.52 

Leading international regulators are also adapting to 
the growing application of non-animal models through 
dedicated institutional programs, updated guidelines, 
analyses on the use of non-animal models, and in-house 
model testing.53 For example, the FDA has established the 
Alternative Methods Group and the Innovative Science 
and Technology Approaches for New Drugs (ISTAND) pilot 
program, while the EMA has established the Innovation 
Task Force (ITF).54 The FDA has already evaluated a range 
of medical product regulatory submissions including 
data from non-animal models: 115 involving 3D skin 
models, 760 involving spheroids, 83 involving organoids 
and 178 involving induced pluripotent stem cells.55 

49 Li L, Knutsdottir H, Hui K, Weiss MJ, He J, Philosophe B, Cameron AM, Wolfgang CL, Pawlik TM, Ghiaur G, Ewald AJ, Mezey E, Bader JS, Selaru FM (2019) 
Human primary liver cancer organoids reveal intratumor and interpatient drug response heterogeneity. JCI Insight 4(2), e121490.

50 Associated Press (2017) Lab-made ‘mini organs’ helping doctors treat cystic fibrosis. In the lab. <https://www.statnews.com/2017/08/23/cystic-fibrosis-mini-
organs-lab/> (accessed 11 July 2023).

51 Monash University (2018) Translating colorectal cancer organoids into patient care. Monash Biomedicine Discovery Institute. <https://www.monash.edu/
discovery-institute/news-and-events/news/2018-articles/translating-colorectal-cancer-organoids-into-patient-care> (accessed 11 July 2023).

52 Pistollato F, Madia F, Corvi R, Munn S, Grignard E, Paini A, Worth A, Bal-Price A, Prieto P, Casati S, Berggren E, Bopp SK, Zuang V (2021) Current EU regulatory 
requirements for the assessment of chemicals and cosmetic products: challenges and opportunities for introducing new approach methodologies. Archives 
of Toxicology 95(6), 1867–1897.

53 US Food and Drug Administration (2022a) Advancing New Alternative Methodologies at FDA. FDA, Maryland, United States. <https://www.fda.gov/
media/144891/download> (accessed 10 July 2023).

54 European Medicines Agency (2021) Innovation in medicines, Human regulatory - Research and development. <https://www.ema.europa.eu/en/human-
regulatory/research-development/innovation-medicines> (accessed 10 July 2023); US Food and Drug Administration (2022b) Advancing Alternative Methods 
at FDA. <https://www.fda.gov/science-research/about-science-research-fda/advancing-alternative-methods-fda> (accessed 10 July 2023).

55 Avila A (2021) Microphysiological Systems (MPS): Bridging Human and Animal Research - An FDA/CDER Perspective. <https://view.officeapps.live.com/op/
view.aspx?src=https%3A%2F%2Fwww.fda.gov%2Fmedia%2F147617%2Fdownload&wdOrigin=BROWSELINK> (accessed 10 July 2023).



2.3 Non-animal models 
over the next 15 years

2.3.1 Medical product development 

Over the next 15 years, medical product development 
will see a reduction in animal model use globally. 
This reduction is likely to be accompanied by an increase 
in the use of non-animal models across all stages of 
the medical product development process. The most 
significant growth will likely come from complex in 
vitro models such as organoids and organ-on-chip 
technologies. In silico models are also expected to 
be more widely applied throughout the development 
process; used in conjunction with in vitro models to 
complement and validate findings. Figure 5 summarises 
the anticipated shifts in non-animal model use across the 
stages of the medical product development process.

Figure 5. Expected shifts in the use of models for medical product development


Fundamental research

Organoids and organ-on-chip technologies will likely 
complement and partially replace animal models in the 
exploration of physiological processes in fundamental 
research, particularly for monogenic disease modelling. 
However, implementing these models in fundamental 
research will require overcoming the financial challenges 
of running advanced cell culture techniques (i.e., acquiring 
highly specialised equipment and expensive consumables). 

A transition to using human cells (either primary or derived 
from iPSCs) and more relevant biomaterials is highly 
probable for accurately modelling human diseases in 
fundamental research. More physiologically representative 
inputs can help uncover novel molecular hallmarks 
of disease (i.e., biomarkers), enhance the exploration 
of disease or condition mechanisms, and improve the 
predictivity of model responses to external compounds. 

Bridging the gap between the outputs of in vitro models 
and clinical endpoints observed in an individual will 
likely remain an intense focus. This area will benefit from 
the continuous refinement of in silico approaches like 
quantitative systems pharmacology (QSP) and in vitro 
to in vivo extrapolation (IVIVE), which integrate large 
datasets of experimental data from in vitro models with 
mechanism-oriented mathematical (in silico) models. 
Emerging QSP and IVIVE models could then support the 
prediction of pharmacokinetic and pharmacodynamic 
endpoints, the evaluation of mechanisms of action, 
and dose estimations based on in vitro data.56 

A greater exploration of digital twin approaches 
(an advanced in silico model), paired with personalised 
in vitro models, is also likely to emerge as an area of 
interest for fundamental research before finding more 
specialised uses in discovery and clinical development. 
These approaches seek to anticipate an individual’s 
health outcomes based on key physiological parameters, 
medical data, and experimental responses.57 Advancing 
the use of digital twins in healthcare is the target of 
recent projects like EDITH, which focuses on a multiscale 
virtual human twin, and In Silico World, which aims to 
support in silico models from validation to adoption.58 

Discovery development

Target identification and validation will benefit from in 
silico models that integrate robust ‘omics’ data with outputs 
from complex in vitro models better representing human 
tissues (organoids, organ-on-chip, and tissue explants). 
Iteration between in vitro and in silico models may produce 
more comprehensive representations of human molecular 
pathways and networks in healthy and diseased states.

Compound screening and lead optimisation will 
likely continue transitioning to more complex in 
vitro 3D models like organoids. However, this is 
contingent on achieving technical improvements to 
reproducibility, well‑characterised sizes and shapes, 
precise cellular populations, standardised output 
read-outs, and further cost reductions per data point. 
While organs‑on-chips would be desirable for this 
activity, there is a need for significant increases in the 
throughput of this model type over the next 15 years.

Given the reliance on high-throughput platforms, 
discovery development activities will benefit the 
most from advances in automation, robotics, data 
analytics, and increasingly integrated in silico models. 
Such technologies can leverage complex models’ 
increasing content and information density while 
helping offset their comparatively lower throughputs.

56 Azer K, Kaddi CD, Barrett JS, Bai JPF, McQuade ST, Merrill NJ, Piccoli B, Neves-Zaph S, Marchetti L, Lombardo R, Parolo S, Immanuel SRC, Baliga NS (2021) 
History and Future Perspectives on the Discipline of Quantitative Systems Pharmacology Modeling and Its Applications. Frontiers in Physiology 12, 637999; 
Bell SM, Chang X, Wambaugh JF, Allen DG, Bartels M, Brouwer KLR, Casey WM, Choksi N, Ferguson SS, Fraczkiewicz G, Jarabek AM, Ke A, Lumen A, Lynn SG, 
Paini A, Price PS, Ring C, Simon TW, Sipes NS, Sprankle CS, Strickland J, Troutman J, Wetmore BA, Kleinstreuer NC (2018) In vitro to in vivo extrapolation for 
high throughput prioritization and decision making. Toxicology in Vitro 47, 213–227.

57 EDITH Consortium (2022) European Virtual Human Twin. <https://www.edith-csa.eu/edith/> (accessed 11 July 2023); In Silico World (2023) The In Silico 
World Project. <https://insilico.world/project/> (accessed 11 July 2023); Venkatesh KP, Raza MM, Kvedar JC (2022) Health digital twins as tools for precision 
medicine: Considerations for computation, implementation, and regulation. npj Digital Medicine 5(1), 150.

58 In Silico World (2023) The In Silico World Project. <https://insilico.world/project/> (accessed 11 July 2023).



Preclinical development

For both toxicology and pharmacology, animal models 
are likely to be partially displaced by the increased 
adoption of complex in vitro models that better mimic 
human tissue structures, dynamics, and behaviours 
(i.e., organoids and organ-on-chip). Developing multi‑organ 
models and simulating increasingly complex tissue 
interactions will be technical priorities to enable truly 
systemic toxicology and pharmacokinetic assessments 
in vitro. Modelling these interactions will be facilitated 
by organ-on-chip models where different tissues are 
cultured in independent modules and connected to each 
other. QSP or IVIVE in silico models can complement 
this approach to translate findings to a human scale.

The use of non-animal models in conducting preliminary 
toxicity screens and informing prioritisation strategies 
will enable a reduction in the number of animals used 
during late-phase preclinical development. Animal use 
reductions can stem from the earlier discontinuation of 
candidates with clear hazard signals in vitro or from the 
accumulation of robust toxicology and pharmacology data 
that allows animal testing to serve as a limited confirmatory 
or regulatory step where still required. Non-animal models 
for organ-specific toxicology will rely on developing, 
characterising and validating these models for specific 
contexts of use. These are clearly defined test-output 
combinations where model use is considered valid, with 
resulting data accepted by regulators for decision-making.

Clinical development

Future non-animal model use in support of clinical 
development could occur at four levels: informing 
the decision to advance a medical product to 
clinical trials, planning study populations and 
informing participant stratification in clinical trials, 
investigating adverse events during clinical trials, and 
developing personalised therapeutic regimens.

Even after approval from a regulator based on preclinical 
data, the formal decision to advance a candidate 
into clinical trials carries uncertainty and financial 
risk. Trial pre-screening using models derived from 
prospective participants (organoids, organs-on-chips, 
and tissue explants) could allow candidates to ‘fail 
early’ in clinical development and avoid the significant 
time and resources involved in clinical studies.

If the decision to proceed to clinical trials is made, 
personalised models could also be used to develop 
targeted inclusion, exclusion and stratification criteria 
for clinical trial participants.59 Preliminary testing of 
candidates in models from prospective participants may 
help identify population differences that impact safety 
or efficacy, explore their underlying mechanisms, and 
build study designs that better consider variations due 
to gender, age, genetic background or co-occurring 
diseases. Establishing and storing in vitro models from 
trial participants could facilitate the investigation of 
adverse events during (or after) clinical trials and explore 
adverse outcome pathways for a failed candidate.

In the clinical setting, scalable non-animal models 
for personalised drug screening may be enabled by 
further advancements in iPSC derivation, tissue explant 
culture, integration of genomics and the overall cost of 
personalised approaches. There could be a reduction in 
the time required for iPSC derivation from adult tissue 
to match clinical timelines better and implementation of 
in-depth characterisation of both the starting biological 
material and the cells produced. Similarly, tissue explant 
culture strategies will require improvement to preserve 
native properties over longer periods, while increasing 
interactions with genomics may help link patient-specific 
genetic variants with drug responses observed in vitro. 
Finally, overall reductions in cost will be necessary to make 
this application of non-animal models possible at scale.

59 National Center for Advancing Translational Sciences (2020) Clinical Trials on a Chip, Tissue Chip Initiatives & Projects. <https://ncats.nih.gov/tissuechip/
projects/clinical-trials> (accessed 11 July 2023).



Regulatory approval and compliance

Incorporating non-animal models into routine use for 
regulatory approvals will be incremental. Novel non‑animal 
models will have to be validated by independent 
centres, qualified by large regulators, and accepted for 
specific contexts of use by international harmonisation 
organisations. However, this is a sequential process, 
and in the short term, regulators are likely to prioritise 
the standardisation of the qualification process itself 
over international harmonisation. This prioritisation 
is due to the length of such a process, the impact 
of the international guidance produced, and the 
rapid development of increasingly complex models. 
An initial focus on the qualification process can allow 
the continual definition of specific contexts of use 
for new models, prevent a premature outdating of 
international guidelines for non-animal model use and 
provide sufficient flexibility for the evolving field. 

In silico, in vitro 3D, and organ-on-chip models will 
see faster adoption in simple hazard categorisation 
for product labelling and assessing products posing 
a lower or well-established level of concern for 
regulators (like non-prescription medications or certain 
regulated pharmaceutical ingredients). These regulatory 
applications are the most suitable in the short term 
given their simple assay outputs, the pre‑existence 
of safety data for known substances, and the data 
quality that non-animal models can produce.

In regulatory compliance applications, in silico, in 
vitro 3D, and organ-on-chip models could also be 
used for lower‑concern products or well-known 
pharmaceutical ingredients that require testing before 
batch release. These models would reduce repetitive 
and animal-intensive tests. For example, US agencies 
have identified a potency assay for botulinum toxin 
as a priority area for non-animal model adoption.60 
Similarly, vaccine batch testing via non‑animal 
alternatives (including in silico and in vitro models) 
has been the focus of the VAC2VAC project in the EU.61 

2.3.2 Organ models

Across complex applications, like disease modelling 
in fundamental research or organ toxicity testing in 
preclinical development, the landscape of non‑animal 
model readiness for wider adoption varies by organ system. 
Table 3 summarises several key organ systems, including 
currently used non-animal model types, example 
non‑animal model applications, and an assessment 
of the likelihood that these models will displace the 
status quo within the next 15 years. The organ systems 
for which non-animal models are the most likely to 
disrupt the status quo are the cardiovascular, respiratory, 
digestive (gastrointestinal), and integumentary systems, 
along with the eye and liver as specific organs. 

The future state categorisation is based on currently 
leading non-animal model types, their existing technical 
limitations (see Appendix A.2), and the organ-specific 
challenges to achieving greater physiological similarity. 
Table 3 was informed through literature review and 
stakeholder reviews. Specific modelling applications 
in each organ system vary significantly by model type, 
disease, and medical product of interest. Therefore, 
the readiness of a specific organ-model-application 
combination over the next 15 years could be more advanced 
than the organ system’s overall state presented here. 

60 US Department of Health and Human Services (2023) RFA-TR-22-031: Botulinum Toxin Potency Assay using Tissue Chips (BoT PATCh) (UT1, UT2 Clinical Trail 
Not Allowed). Grants. <https://grants.nih.gov/grants/guide/rfa-files/RFA-TR-22-031.html> (accessed 11 July 2023).

61 European Vaccine Initiative (2023) Vaccine batch to vaccine batch comparison by consistency testing. VAC2VAC EU. <https://europevaccine.wixsite.com/
vac2vac-eu> (accessed 11 July 2023).



Table 3. Potential of emerging organ models to displace the status quo within the next 15 years.

ORGAN SYSTEM

CURRENT ADVANCED 
MODEL TYPES 

EXAMPLE APPLICATION

FUTURE 
STATE 

Cardiovascular system – heart

In silico

Drug-induced arrhythmia testing

 

3D (Organoid) 

Organ-on-chip

Respiratory system – lungs

Tissue explant

Pulmonary fibrosis model

 

Organ-on-chip

Digestive (gastrointestinal) 
system

In silico

Drug absorption and transport modelling

 

3D (Organoid)

Organ-on-chip

Integumentary system – skin

3D (Scaffold)

Wound healing and skin regeneration modelling

 

Tissue explant

Eye

3D (Scaffold)

Retinogenesis and genetic retinal disease modelling (e.g., 
Stargardt's macular degeneration)

 

3D (Organoid)

Organ-on-chip

Metabolic and endocrine – liver

In silico

Non-alcoholic fatty liver disease model

 

Organ-on-chip

Nervous system – central and 
peripheral

In silico

Neuromuscular junction model for ALS mechanism modelling 
and therapeutic response testing

 

3D (Organoid)

Organ-on-chip

Metabolic and endocrine – 
pancreas

3D (Organoid)

Pancreatic cancer and cystic fibrosis drug screening

 

Organ-on-chip

Renal and urogenital – kidney

3D (Organoid)

Kidney polycystic disease modelling or drug-induced 
nephrotoxicity testing

 

Musculoskeletal system

3D (Organoid)

Myogenesis and muscle regeneration model

 

Organ-on-chip

Reproductive system

3D (Organoid)

Reproductive toxicity and endometriosis modelling

 

Tissue explant

Ear

3D (Organoid)

Genetic disease-associated hearing loss modelling and 
corrective gene therapy testing

 

Hemic (blood) system – 
bone marrow

3D (Organoid)

Bone marrow niche and haematopoiesis modelling

 

Organ-on-chip

Inflammatory disorders and the 
immune system

3D (Organoid)

Immune activation and response in a lymph node model for 
influenza vaccine testing

 

Organ-on-chip





FUTURE STATE

Non-animal models demonstrate equivalent, or better, outputs than animal models and are likely to replace animal models 
for this organ system in the next 15 years.

Non-animal models demonstrate the potential to replicate animal model outputs; further advances could position them to 
replace or complement animal models for this organ system in the next 15 years.

Non-animal models still need to demonstrate equivalent outputs to existing animal models, will require further advances 
to complement animal models, and are unlikely to replace animal models for this organ system in the next 15 years.







3 Non-animal models in Australia

Australia’s medical product development ecosystem is 
active across all organ systems, with each stakeholder 
group (research, clinical trials, and industry) 
demonstrating specific strengths (Figure 6).

CSIRO Futures identified strengths across all 
stakeholder groups for the cardiovascular, respiratory 
and nervous systems. Research strengths not yet 
translated into clinical trials or industry strengths are 
areas to expand Australian commercial offerings. 

Comparative national strengths were assessed via 
publication counts for organ system-model type 
combinations, clinical trial registrations by condition, 
and company counts by therapeutic area. Research 
strength was determined by output to enable a consistent 
examination across areas (where impact metrics may 
be an inadequate measure due to a small number 
of publications) and to support comparison to other 
countries. Organ system-model type combinations 
not identified as comparative strengths in this analysis 
should not be interpreted as an absence of high-quality 
research or a lack of mature models with potential for 
commercial translation. Appendix A.3 outlines further 
information regarding the methodology and results.

Figure 6. Alignment of strengths in Australia’s medical product development ecosystem

Research 
strengths

Eye

2D, 3D (general, scaffold 
and organoids)

Urogenital

2D, 3D (general, scaffold 
and organoids)

Reproductive

2D, 3D (general, scaffold and 
organoids), tissue explants

Ear

Tissue explants

Integumentary

2D

Clinical trial 
strengths

Inflammatory disorders 
and immune system

Industry strengths

Metabolic and endocrine

3D (scaffold and organoids)

Musculoskeletal


Cardiovascular

3D (general, scaffold 
and organoids)

Respiratory

2D

Nervous system

3D (general, scaffold 
and organoids)



3.1 Research strengths

Australia is internationally regarded for its research 
capability in non-animal models for several specific 
organ systems. Between 2018 and 2023, Australia’s 
share of non-animal model publications was above 3% 
and ranked in the top 10 globally for specific organ 
system‑model type combinations, as shown in Table 4.

Table 4. Organ system-model type combinations emerging as comparative research strengths for Australia, based on the share of 
global publication output and corresponding global position between 2018 and 2023.

ORGAN SYSTEM

MODEL TYPE

AU PERCENTAGE OF GLOBAL 

GLOBAL POSITION

Eye

3D (Scaffold and organoid)

3D (General)

2D

7.5%

6.0%

4.7%

5th

6th

6th

Urogenital system

2D

3D (General)

3D (Scaffold and organoid)

4.0%

4.7%

6.2%

6th

7th

8th

Reproductive system

3D (Scaffold and organoid)

2D

3D (General)

Tissue explant

6.1%

4.6%

4.7%

4.6%

7th

7th

8th

10th

Cardiovascular system

3D (Scaffold and organoid)

3D (General)

4.5%

3.7%

7th

10th

Ear

Tissue explant

5.0%

8th

Respiratory system

2D

4.2%

10th

Nervous system

3D (Scaffold and organoid)

3D (General)

4.0%

3.4%

10th

12th

Integumentary system

2D

3.6%

10th

Metabolic and endocrine systems

3D (Scaffold and organoid)

3.4%

11th





Cancer, as a therapeutic area, is not represented in 
the organ system-model type assessment of research 
strengths, given its broad inclusion of multiple organ 
systems simultaneously. However, cancer research was 
noted by consulted stakeholders as an area of strength 
for Australia and ‘childhood cancer’ emerged as a topic of 
comparative strength in a separate bibliometric analysis, 
ranking 5th in the world by publication output.62 

Stakeholders also noted that Australian researchers possess 
world-leading expertise in toxicology, immunology, 
infectious diseases, blood/haematopoietic system research, 
and genetic and rare disease characterisation. While these 
specialties are not presented in terms of non-animal 
model publications and ranking, they represent additional 
areas of capability that could be leveraged to support 
non-animal model offerings domestically and abroad.

Existing research strength in non-animal models can 
unlock economic benefits for Australia, given industry 
interest, expanding applications, and growing markets. 
The global organoid market in 2022 was $1.62 billion 
and is expected to reach $30.91 billion by 2040, with 
a CAGR of 17.4%. This market is estimated to generate 
$1.28 billion in revenue and 4,200 jobs for Australia by 
2040.63 For comparison, the organ-on-chip market is at 
an earlier phase of adoption. It was valued at $110 million 
worldwide in 2022, with a projection of $11.48 billion by 
2040 (29.4% CAGR). The potential share for Australia by 
2040 is $310 million in revenue, with 700 related jobs.64

62 An InCites analysis of the Childhood Cancer Area (Citation Topics – Micro) shows Australia ranks 5th in the world by number of Web of Science articles in the last 
5 complete years (262 in 2018 – 2022), and 9th among the subset of top 15 countries when considering Category Normalized Citation Impact (1.36); US National 
Cancer Institute (2023) Childhood Cancer Model Atlas. Childhood Cancer Data Initiative - Data Catalog. <https://datacatalog.ccdi.cancer.gov/dataset/VPCC-CCMA> 
(accessed 11 July 2023).

63 Based on CSIRO Futures economic analysis of global non-animal model market size data, Australian share of global publications by model, and national 
wage data. See Appendix A.5 for the methodology used.

64 Based on CSIRO Futures economic analysis of global non-animal model market size data, Australian share of global publications by model, and national 
wage data. See Appendix A.5 for the methodology used.



3.2 Clinical strengths

Australia has world-leading clinical trials infrastructure, 
with over 50 trial networks offering Phase I–III trials. 
Australia also provides a range of grants, tax incentives 
and Patent Box schemes, which makes the country an 
attractive destination for conducting clinical trials.65 
Between 2018 and 2023, close to 10,000 clinical trials 
were registered in Australia, with nervous, cardiovascular, 
respiratory, musculoskeletal, and immune systems 
having the highest number throughout the period.66

Australia has registered clinical trials across all organ 
systems identified as areas of research strength. However, 
the distribution of trials per organ system does not yet 
reflect the strengths, highlighting potential gaps and 
unexploited opportunities. For example, Australia’s globally 
recognised expertise in non-animal models developed 
locally for eye, urogenital and reproductive systems have 
not yet translated into increased clinical trial activity. 

3.3 Industry strengths 

Australia’s industry capabilities related to non-animal 
model use in medical product development are highly 
research-led. The market is driven by ad-hoc research 
collaborations between research institutions and 
pharmaceutical companies, and start-ups that have typically 
spun out from research institutions. As such, this report 
takes a broader definition of the industry, to include 
organisations with non-animal model products or services 
that are sufficiently mature to be provided for a fee.

In total, 37 Australian organisations were identified 
to have mature non-animal model-related capabilities 
through stakeholder consultation and desktop 
analysis (Figure 7). Given the existence of ad-hoc 
research collaborations and the ongoing maturation 
of national industry capabilities, this is likely an 
underestimate of the entire commercial ecosystem. 

Australia possesses industry capabilities across all 
non-animal model types. In vitro 2D is the most 
common model type used in service offerings, with 
23 organisations (predominantly research institutions) 
providing this, typically through HTS platforms.

Australia also has market-supporting entities, including 
contract research organisations (CROs), which contribute to 
the commercialisation of non-animal models by providing 
related clinical trial and research support services. 
One such entity is Phenomics Australia, which provides 
infrastructure and supporting services through their 
partner network to enable in vitro genome engineering 
and disease modelling, 3D bioprinting, organoid 
production, human iPSCs derivation, patient-derived 
cell line sourcing, and high-throughput screening.67

Figure 7. Summary of the Australian non‑animal model 
industry landscape

25

Research institutions

9

Companies

3

Contract research organisations

See Appendix A.4 for the full organisation list.





65 Australian Trade and Investment Commission (2022) Insight – Australia: A go-to destination for clinical trials. <https://www.austrade.gov.au/news/insights/
insight-australia-a-go-to-destination-for-clinical-trials> (accessed 11 July 2023).

66 Based on an ANZCTR search of registered clinical trials between 2018 and 2023. See Appendix A.3 for the methodology used and results for the top 5 
organ systems.

67 Phenomics Australia (2023a) In vitro Genome Engineering & Disease Modelling. <https://phenomicsaustralia.org.au/in-vitro-genome-engineering-disease-
modelling/> (accessed 11 July 2023).



3.4 Funding and investment 

Funding of non-animal models within Australia is mainly 
indirect, with no large public or private investment 
schemes specifically dedicated to their development, 
validation, or commercialisation. Stakeholders noted that 
the funding enabling non-animal model development 
primarily comes from National Health and Medical Research 
Council (NHMRC) schemes68 and Medical Research Future 
Fund (MRFF) initiatives.69 For example, non-animal model 
activity benefits from the MRFF Australian Stem Cell 
Therapies Mission funding of $150 million over ten years.70

Funding for NCRIS also supports non-animal model 
development, through access to genome engineering, 
sequencing, functional genomics, histopathology, 
stem cell derivation and disease modelling capabilities; 
activities supported by Phenomics Australia across 
its network.71 Stakeholders noted that NCRIS support 
and funding from Phenomics Australia have been 
instrumental in developing organ modelling strengths 
by individual research groups and institutions.

Given the lack of dedicated funding programs, researchers 
also turn to pharmaceutical company partnerships, 
privately funded fellowships, and philanthropic 
organisations to support the expansion of local 
capabilities and broader model development projects. 
An example of this is the historical backing from the 
Medical Advances Without Animals Trust (MAWA), which 
provides direct contributions and helps steer external 
funding into relevant non‑animal model research.

Funding for the commercial translation of non‑animal 
models, especially from venture capital, is also limited 
in Australia. As a result, while innovative non‑animal 
models are developed with local talent, the vehicles 
to commercialise them are often established 
internationally, forcing novel developments offshore 
(e.g., Dynomics in the US and Mogrify in the UK).72

68 National Health and Medical Research Council (2022) Research funding statistics and data. Funding - Data on Research. <https://www.nhmrc.gov.au/
funding/data-research/research-funding-statistics-and-data> (accessed 11 July 2023).

69 Australian Government Department of Health and Aged Care (2022) All MRFF initiatives. Medical Research Future Fund <https://www.health.gov.au/our-
work/medical-research-future-fund/all-mrff-initiatives> (accessed 11 July 2023).

70 Australian Government Department of Health and Aged Care (2023) Stem Cell Therapies Mission. Our work. <https://www.health.gov.au/our-work/stem-
cell-therapies-mission> (accessed 11 July 2023).

71 Phenomics Australia (2023b) Our expertise. <https://phenomicsaustralia.org.au/expertise/> (accessed 11 July 2023).

72 Dynomics (2022) Dynomics - Therapies to Restore Heart Function. <https://www.dynomics.com/#technology> (accessed 11 July 2023); Mills RJ, Parker BL, 
Quaife-Ryan GA, Voges HK, Needham EJ, Bornot A, Ding M, Andersson H, Polla M, Elliott DA, Drowley L, Clausen M, Plowright AT, Barrett IP, Wang Q-D, 
James DE, Porrello ER, Hudson JE (2019) Drug Screening in Human PSC-Cardiac Organoids Identifies Pro-proliferative Compounds Acting via the Mevalonate 
Pathway. Cell Stem Cell 24(6), 895-907; Mogrify (2023) Mogrify - Reprogramming Health. <https://mogrify.co.uk/> (accessed 11 July 2023).



4 National opportunities 
and recommendations

4.1 National opportunities 

With other nations already leading the way in some 
non-animal model applications and offerings, it is 
important for Australia to focus on leveraging 
research, clinical and industry strengths to pursue 
near- and medium-term opportunities. 

This report defines opportunities as any non-animal 
model product or service offering that could benefit 
the quality of domestic R&D activities or create revenue 
streams. The analysis of national strengths and insights 
from 103 stakeholders around areas of greatest need 
identified four opportunities for Australia (Figure 8).

Figure 8. Four national opportunities

FUNDAMENTAL 
RESEARCH

DISCOVERY 
DEVELOPMENT

PRECLINICAL 
DEVELOPMENT

CLINICAL 
DEVELOPMENT

REGULATORY 
APPROVAL AND 
COMPLIANCE

1. Complex in 
vitro models for 
improving the 
R&D productivity 
of national drug 
discovery.


Adoption of more 
complex and 
biologically relevant 
high-throughput 
in vitro models like 
organoids.

2. Organ-specific 
models for 
preclinical 
development


Research and industry 
collaborations in areas 
of national strength 
to support time and 
animal use reductions 
in preclinical testing.

3. Personalised 
models for trial 
participant and 
clinical treatment 
selection.


Protecting and further 
strengthening the 
competitiveness of 
Australia’s clinical trials 
sector and advancing 
precision medicine 
goals by incorporating 
patient-specific 
models.

4. Onshore production of model components. 


Increasing production capability for critical non-animal model inputs such as human-derived stem cells, 
non-animal derived media, and hydrogels.







4.1.1 Complex in vitro models 
for improving the R&D productivity 
of national drug discovery

The opportunity

Discovery development relies extensively on HTS to source 
novel therapeutic targets, identify compounds exhibiting 
activity against them, and assess preliminary hits. HTS uses 
miniaturised models, concurrent screening, and streamlined 
visualisation or detection of outputs to support large‑scale 
testing within reduced timeframes. The technical 
requirements of HTS have traditionally resulted in in vitro 
2D models being used for most discovery activities, at the 
expense of greater complexity and biological relevance.

Improvements to the activity of Australian HTS platforms 
and drug discovery programs could be made by adopting 
more biologically relevant but still high‑throughput 
complex in vitro models like organoids. While more 
complex in vitro models are unlikely to fully match 
the throughput offered by in vitro 2D models, wider 
adoption at the earliest stage of development could 
increase the likelihood of identifying relevant targets, 
reduce late-stage attrition of locally developed 
candidates, and increase the productivity of the 
domestic pipeline.73 In instances where in vitro 2D is 
still selected as the primary model type for HTS, using 
organoids for secondary screens can narrow a potentially 
large number of initial hits.74 This complementary use 
could reduce false positives and increase confidence 
in compounds selected for further development.

Why Australia?

Australia has an existing base of discovery platforms with 
HTS capabilities (See Appendix A.4). These platforms 
facilitate the transition from fundamental research 
into preclinical development by identifying hits, 
promoting intellectual property (IP) generation 
and, in some cases, de-risking discovery activities 
through government-subsidised screens.75 

The emergence of Australian facilities capable of producing 
more complex, HTS-compatible in vitro models at a large 
scale, with high reproducibility and minimal cost, can 
help support national discovery platforms. Case study 1 
presents one such facility, with the emergence of others 
necessary to strengthen the local R&D ecosystem. 

Australia also possesses an overarching national 
discovery and development network supported by NCRIS. 
Organ system-specific strengths like those outlined in 
Section 3.1 could complement existing national discovery 
platforms by providing organ- and tissue-specific 
models for specialised HTS during early development. 

CASE STUDY 1: Australian 
Organoid Facility – Large-scale 
production of organoids for HTS

The Australian Organoid Facility (AOF) has invested 
in state-of-the-art automation, high-content 
imaging, and analysis capabilities to produce 
high-quality, reproducible organoids that can be 
used for HTS. The AOF is located at the Australian 
Institute for Bioengineering and Nanotechnology 
(AIBN) at the University of Queensland. 

The AOF provides domestic research and industry 
services across model development, high-throughput 
organoid production, and drug and functional 
screening. The facility specialises in brain, kidney, 
blood vessel and cancer organoids, with a vision of 
expanding to cardiac, intestinal, liver and respiratory 
models. The detailed requirements and specific 
protocols used for organoid production vary across 
tissue types, resulting in different cost ranges. However, 
as an example, the AOF currently provides brain 
organoids for $30 to $60 each, depending on culture 
time and mode of production.76 A focus on automation, 
progressively optimised culture protocols, and scale 
will enable a reduction in production costs over time.

73 Ekert JE, Deakyne J, Pribul-Allen P, Terry R, Schofield C, Jeong CG, Storey J, Mohamet L, Francis J, Naidoo A, Amador A, Klein J-L, Rowan W (2020) Recommended 
Guidelines for Developing, Qualifying, and Implementing Complex In Vitro Models (CIVMs) for Drug Discovery. SLAS Discovery 25(10), 1174–1190.

74 Mills RJ, Parker BL, Quaife-Ryan GA, Voges HK, Needham EJ, Bornot A, Ding M, Andersson H, Polla M, Elliott DA, Drowley L, Clausen M, Plowright AT, Barrett 
IP, Wang Q-D, James DE, Porrello ER, Hudson JE (2019) Drug Screening in Human PSC-Cardiac Organoids Identifies Pro-proliferative Compounds Acting via 
the Mevalonate Pathway. Cell Stem Cell 24(6), 895-907.

75 The Walter and Eliza Hall Institute of Medical Research (2023) Advancing the latest discoveries - National Drug Discovery Centre (NDDC). The National Drug 
Discovery Centre <https://nddc.wehi.edu.au/> (accessed 11 July 2023).

76 Price range and accompanying rationale provided by the AOF.



4.1.2 Organ-specific models 
for preclinical development

The opportunity

Non-animal models that better replicate organ and disease 
physiology can enable more relevant and predictive testing. 
In turn, this can support animal model use rationalisation, 
reduced attrition rates during clinical development, 
and enhanced safety and efficacy profiling of therapeutic 
candidates. Non-animal models can also enable time 
and animal use reductions in preclinical testing. Table 5 
illustrates this, comparing the cost, time, and animal 
use across two validated OECD toxicity tests; one using 
an animal model and the other using an equivalent 
non-animal approach. Additional comparisons and the 
methodology employed can be found in Appendix A.5.

Table 5. Comparison between OECD test guidelines (TG) featuring equivalent animal and non-animal approaches for acute dermal 
corrosion across cost, time, and animal use.77

TYPE OF TOXICITY 

OECD TG

TEST TYPE

STUDY COST 
(AUD 2022)

TIME DURATION 
(DAYS)

NUMBER OF 
ANIMALS USED 

Acute dermal irritation/
corrosion

404

Animal

2,000

15.00

2

Membrane barrier test method 
for skin corrosion

435

In vitro 

4,100

0.2

0

Difference

2,100

–14.8

–2.00





The development, validation, and standardisation 
of improved organ and disease models in 
the Australian ecosystem could potentially 
support three distinct business models: 

1. Commodities: Non-animal models as 
commercialised products for academic researchers 
and pharmaceutical R&D teams, as exemplified 
by the growing organ-on-chip market.78 
2. Services: Non-animal models as validated 
assays for CRO service provision in niche areas, 
sought by pharmaceutical companies to cover 
highly specific in-house development gaps.
3. Partnerships: Non-animal models as versatile platforms 
for developing and testing many different medical 
product candidates. Effectively, means of continuous 
IP generation and enablers of partnerships between 
model developers and pharmaceutical companies for 
larger discovery and preclinical development projects.


77 Adapted from: Marty MS, Andrus AK, Groff KA (2022) Animal metrics: Tracking contributions of new approach methods to reduced animal use. ALTEX 
– Alternatives to animal experimentation 39(1), 95–112; Meigs L., Smirnova L, Rovida C, Leist M, Hartung T (2018) Animal testing and its alternatives – 
the most important omics is economics. ALTEX – Alternatives to animal experimentation 35(3), 275–305; CSIRO Futures calculations.

78 Zhang B, Radisic M (2017) Organ-on-a-chip devices advance to market. Lab on a Chip 17(14), 2395–2420.



Why Australia?

Australia has several characteristics that are key to 
enabling the production of organ- and patient-specific 
models. These include a diverse population, close 
collaboration between clinical and research settings, 
internationally recognised and geographically distributed 
capacity for iPSC generation (important for complex in 
vitro model development),79 and a focus on genomics 
capabilities. As described in Section 3.1, Australia has 
globally competitive modelling and research strengths in 
various areas. These areas could inform initial investment 
priorities that enhance national-scale capabilities in 
organ-specific toxicology and pharmacology testing.

The research-intensive focus of the Australian landscape 
makes the partnership business model a strategic option 
in the short to medium term. Partnerships provide 
pharmaceutical companies with access to both the research 
expertise behind a promising non-animal model and to the 
local capabilities of an institution or biomedical precinct. 
In turn, the institution or precinct benefits from direct 
investment, industry-relevant training, knowledge transfer, 
increased activity for supportive research services and IP 
production that can be out-licensed to attract royalties over 
time. Case Study 2, the reNEW consortium, exemplifies 
the viability and relevance of the partnership model. 

CASE STUDY 2: Pluripotent stem cell 
and in vitro modelling expertise as 
drivers of international partnerships

reNEW is a tripartite collaboration between the 
University of Copenhagen in Denmark (governing hub), 
the Leiden University Medical Center in the 
Netherlands, and the Murdoch Children’s Research 
Institute (MCRI) in Australia. reNEW conducts 
stem cell research in three areas: regenerative 
medicine strategies for damaged tissues, drug 
screening in human models, and genetic disease 
therapies enabled by gene editing. Underpinning 
the collaboration is a EUR 300 million grant from 
the Novo Nordisk Foundation distributed over ten 
years between the partnering institutions.80 

The MCRI was selected as a node due to its 
internationally recognised expertise in pluripotent 
stem cells and connections across the Melbourne 
Biomedical Precinct.81 The MCRI node possesses 
stem cell biology expertise in urogenital (kidney, 
specifically), cardiovascular, blood, musculoskeletal, 
respiratory, and nervous system applications.82 
The node’s expertise in iPSCs underpins strong 
organ and disease modelling capabilities across 
the noted organ systems. This is illustrated by 
organ‑specific modelling projects, such as evaluating 
infectious agents in a stem cell-derived lung model, 
understanding paediatric leukaemia development 
in bone marrow models, or testing novel drugs for 
neurodevelopmental conditions in brain models.83

79 An InCites analysis of the Stem Cell Research Area (Citation Topics – Meso) ranks Australia 15th in the world by number of Web of Science articles in the 
last 5 complete years (764 in 2018 – 2022), and 6th among the subset of top 15 countries when considering Category Normalized Citation Impact (1.37). 
Moreover, iPSCs was consistently noted by consulted stakeholders as an area of recognition and comparative strength.

80 Novo Nordisk Foundation (2022) Novo Nordisk Foundation Center for Stem Cell Medicine (reNEW). Projects. <https://novonordiskfonden.dk/en/projects/
novo-nordisk-foundation-center-for-stem-cell-medicine-renew/> (accessed 11 July 2023).

81 Murdoch Children’s Research Institute (2022a) reNEW - Novo Nordisk Foundation Center for Stem Cell Medicine. About. <https://www.mcri-renew.org.au/
about/> (accessed 11 July 2023).

82 Murdoch Children’s Research Institute (n.d.) Stem Cell Biology. Research areas. <https://www.mcri.edu.au/research/research-areas/stem-cell-biology> 
(accessed 11 July 2023).

83 Murdoch Children’s Research Institute (2022b) Using stem-cell models of the lung to identify cellular therapies for Rhinovirus - reNEW - Novo Nordisk 
Foundation Center for Stem Cell Medicine. Our research. <https://www.mcri-renew.org.au/our-research/using-stem-cell-models-of-the-lung-to-identify-
cellular-therapies-for-rhinovirus/> (accessed 11 July 2023); Murdoch Children’s Research Institute (2022c) Modelling childhood leukaemia to better 
understand the disease – reNEW – Novo Nordisk Foundation Center for Stem Cell Medicine. Our research. <https://www.mcri-renew.org.au/our-research/
generating-models-of-childhood-leukaemia-to-better-understand-the-disease/> (accessed 11 July 2023); Murdoch Children’s Research Institute (2022d) 
Developing a drug screening platform for neurological diseases - reNEW - Novo Nordisk Foundation Center for Stem Cell Medicine. Our research. <https://
www.mcri-renew.org.au/our-research/developing-a-drug-screening-platform-for-neurological-diseases/> (accessed 11 July 2023).



4.1.3 Personalised models for trial 
participant and clinical treatment selection

The opportunity

Personalised non-animal models such as organoids 
and organs-on-chips possess the potential 
to inform clinical trials, improve the safety of 
participants, and support the assessment of the 
clinical efficacy of medical product candidates. 

Personalised models can assist in the screening, 
selection, and stratification of shortlisted participants 
for clinical trials by providing an initial assessment 
of their unique response to a therapeutic candidate. 
This information can increase participant safety and 
reduce development costs by identifying adverse effects 
earlier. This application is best suited for medical product 
development for rare diseases (due to limited population 
sizes); emerging infectious diseases (due to the ethical 
constraints on conducting human trials, especially for 
infectious diseases with high mortality rates); cancers 
(given the ability to collect sufficient tissue samples 
from patients to generate organoids and conduct tests 
that effectively inform treatment decisions); and eye 
diseases (with Australia’s expertise in eye research, 
the anatomical limitations of conducting trials in 
patient eyes, and the prevalence of eye disease).

Personalised non-animal models can also be used in 
precision medicine, benefitting patients by screening safety 
and efficacy responses to shortlisted therapeutics without 
impacting the patient directly. This use supports treatment 
selection and may be primarily suitable for testing cancer 
treatments, due to patients’ sensitivity to treatment and 
the availability of tissue samples.84 Assessing treatments 
of diseases with a strong genetic basis (where treatment 
options are highly susceptible to genetic variability), such as 
cystic fibrosis and rare diseases, is also a viable focus. 
Personalised models also create the opportunity 
for label indication expansion (the identification of 
new uses for already approved therapeutics). 

While few trials have used non-animal models to 
inform therapy selection to date, approximately 180 
planned, active, or completed clinical trials globally 
have utilised organoid models as part of their study.85

Why Australia?

Australia has an established clinical trials sector, 
contributing $1.4 billion to the Australian economy in 2019.86 
With personalised non-animal models in clinical trials 
likely to grow over the coming years, ensuring Australia 
develops this capability will be necessary for maintaining 
and further advancing the sector’s competitiveness.

The clinical application of these models is a form of 
precision medicine, a priority science capability for 
Australia.87 Development of non-animal models for 
clinical applications currently receives funding via MRFF 
initiatives related to precision medicine, including the 
Clinicians Researchers Initiative and Genomic Health 
Futures Mission, which will invest $200 million and $500 
million, respectively.88 Many of the skills and infrastructure 
required for precision medicine are also transferrable to 
this opportunity’s trial participant selection applications. 

While yet to be implemented at scale, research 
groups across Australia are actively investigating 
the use of patient-derived organoids to predict 
colorectal, pancreatic and breast cancer responses 
to treatment, as highlighted in Case Study 3.

84 Chen X, Sifakis EG, Robertson S, Neo SY, Jun S-H, Tong L, Hui Min AT, Lövrot J, Hellgren R, Margolin S, Bergh J, Foukakis T, Lagergren J, Lundqvist A, Ma R, 
Hartman J (2023) Breast cancer patient-derived whole-tumor cell culture model for efficient drug profiling and treatment response prediction. Proceedings 
of the National Academy of Sciences 120(1), e2209856120

.

85 Based on a search of ClinicalTrials.gov using ‘organoids’ as the only keyword in the Other terms field, without time or location restrictions. Excludes 
suspended, terminated or withdrawn studies. <https://clinicaltrials.gov/ct2/results?term=organoids&map=CA > (accessed on 07 July 2023).

86 MTPConnect (2021) Australia’s Clinical Trials Sector: Advancing innovative healthcare and powering economic growth. <https://www.mtpconnect.org.au/
images/MTPConnect_Australia's%20Clinical%20Trials%20Sector%20report%202021.pdf> (accessed 10 July 2023).

87 Australian Government Department of Education (2021) 2021 National Research Infrastructure Roadmap. <https://www.education.gov.au/national-research-
infrastructure/resources/2021-national-research-infrastructure-roadmap> (accessed 11 July 2023).

88 Australian Government Department of Health and Aged Care (2022) All MRFF initiatives. Medical Research Future Fund <https://www.health.gov.au/our-
work/medical-research-future-fund/all-mrff-initiatives> (accessed 11 July 2023).



CASE STUDY 3: Translating colorectal cancer organoids into patient care

The Monash Biomedicine Discovery Institute, 
in partnership with Cabrini Health, the University 
of Melbourne, the Peter MacCallum Cancer Centre, 
the South Australian Health and Medical Research 
Institute (SAHMRI) and The Walter and Eliza Hall 
Institute of Medical Research (WEHI), has developed 
a colorectal cancer organoid platform for patient 
care and has demonstrated the potential use 
of organoids for treatment selection.

The initiative, funded by Cancer Australia’s Priority‑driven 
Collaborative Cancer Research Scheme (PdCCRS), has 
found that patient-derived colorectal cancer organoids 
retain key characteristics of the tissue from which they 
are derived. The colorectal organoid platform is derived 
from 50 patient tumours, normal adjacent tissue and, 
in some cases, matched metastatic tumours from the 
same patients. While work is underway to ensure this 
translates into the clinic, this research is a strong step 
towards utilising organoids to assess individual patients’ 
drug sensitivities and guide therapy selection.89

89 Cabrini Health (2022) New research puts personalised treatment for colorectal cancer on the cards. Research. <https://www.cabrini.com.au/organoids/> 
(accessed 11 July 2023); Monash University (2018) Translating colorectal cancer organoids into patient care. Monash Biomedicine Discovery Institute. 
<https://www.monash.edu/discovery-institute/news-and-events/news/2018-articles/translating-colorectal-cancer-organoids-into-patient-care> 
(accessed 11 July 2023).



4.1.4 Onshore production 
of model components

The opportunity

Many components used in non-animal models face 
challenges arising from their poor characterisation, 
batch variability, and limited traceability. Additionally, 
with limited large-scale production of non-animal 
model components in Asia‑Pacific and components 
produced within Australia typically prototyped on 
an ad-hoc basis, Australian organisations typically 
source components from international suppliers.

Onshore component production can provide benefits for 
sovereign capability. Increased domestic production of 
non-animal model components could alleviate susceptibility 
to international supply and price fluctuations, streamline 
Australia’s R&D supply chain, and enable higher throughput 
in R&D for non-animal model capabilities and broader 
biomedical research. Onshore production can also present 
export opportunities after establishing sufficient scale. 
International stakeholders noted the benefits of proximity 
between organisations using models that require tissue 
samples and established component manufacturers to 
reduce transit risks such as preservation of tissue quality. 

Three components were most identified by stakeholders 
as being valuable for onshore production:

• Human-derived stem cells: Human-derived stem cells 
(adult stem cells and iPSCs), are sought-after components 
for complex in vitro models such as organoids and 
organs-on-chips. However, those from genetically 
and phenotypically well-characterised donor pools 
are difficult to source. The variability of currently 
available donor cell lines increases development 
costs, as customers must test and characterise 
batches individually before using them in models.
• Non-animal-derived media: Currently available 
animal-derived media components face poor 
characterisation and batch variability challenges, 
particularly foetal bovine serum (FBS), a 
widespread input into cell culture media. 
• Hydrogels: Hydrogels are a synthetic animal-free 
alternative to Matrigel (a common animal-derived 
cell culture matrix) with reduced batch variability.90


Producing all three component types onshore would 
create an opportunity to sell them in combination 
as model development kits. This opportunity 
would represent a unique, well-differentiated 
international and domestic market offering. 

90 Curvello R, Alves D, Abud HE, Garnier G (2021) A thermo-responsive collagen-nanocellulose hydrogel for the growth of intestinal organoids. Materials 
science & engineering. C, Materials for biological applications 124, 112051; Curvello R, Kerr G, Micati DJ, Chan WH, Raghuwanshi VS, Rosenbluh J, Abud HE, 
Garnier G (2021) Engineered Plant-Based Nanocellulose Hydrogel for Small Intestinal Organoid Growth. Advanced Science 8(1), 2002135.



Why Australia?

Australia’s strength in stem cells provides a strong 
foundation and competitive advantage for developing 
well-characterised adult stem cell-derived and iPSC‑derived 
models. Stem cell technologies are advancing quickly 
within Australia, with service facilities such as the 
Stem Cell and Organoid Facility providing iPSC-derived 
organoids for pharmacology testing and evaluating 
advanced technologies such as gene therapies.91

Australia is a notable global producer of FBS. 
As non‑animal models become increasingly commonplace 
globally, producing and exporting non-animal-derived 
media will reduce demand for FBS. Developing an 
alternative media production capability will help 
prepare Australia for the market’s likely transition 
away from animal-derived inputs such as FBS.

Cell culture matrix material is not easily accessible in 
Australia. Matrigel, for instance, is not produced at scale 
domestically, causing a dependence on international 
Matrigel supply. While small-scale production of hydrogels 
is present in Australia,92 increasing domestic production will 
reduce reliance on offshore Matrigel supply, prepare for 
the transition away from animal-derived inputs, and 
address issues associated with batch variability.

Finally, Australian companies have successfully 
established scaled production of important 
non-animal model components in Australia for 
domestic and export markets. One such example 
is Schott Minifab, described in Case Study 4.

CASE STUDY 4: Local in vitro 
component manufacturing

Schott Minifab is headquartered in Victoria and 
manufactures and exports millions of high-standard 
(ISO9001 and ISO13485 certified) components 
for in vitro use per year. Founded as Minifab in 
Melbourne in 2002, the organisation was acquired 
by Schott AG in 2019 due to the two organisations’ 
complementary capabilities in polymers and glass 
components, leading to a broadened product 
range for their overlapping customer bases.

Schott Minifab partners with clients to customise 
and develop point-of-care cartridges, lab-on-a‑chip 
solutions (including organ-on-chip) and other medical 
device components for in vitro use. Their capabilities 
include microfluidic design (fluids at the nanolitre 
scale), polymer-based material selection (for the 
precise movement of these fluids), prototyping, 
and advanced manufacturing processes for high 
volume production.93 As of 2022, Schott Minifab is 
one of the world’s largest contract manufacturers of 
microfluidic devices, with 12.6% of the market.94

91 Children’s Medical Research Institute (2023) Stem Cell & Organoid Facility. <https://www.cmrijeansforgenes.org.au/research/research-facilities/scof> 
(accessed 11 July 2023).

92 Fatimi A (2021) Hydrogel-Based Bioinks for Three-Dimensional Bioprinting: Patent Analysis. Materials Proceedings 7(1), 3.

93 MTPConnect (2023) MiniFAB (Aust) Pty Ltd. <https://www.mtpconnect.org.au/Company?Action=Profile&Company_id=99> (accessed 11 July 2023); SCHOTT 
MINIFAB (2018) Diagnostics & life science product development & manufacture. <https://schott-minifab.com/what-we-do> (accessed 11 July 2023); SCHOTT 
MINIFAB (2019) SCHOTT AG to acquire Australian microfluidic expert MINIFAB. <https://schott-minifab.com/item/43-schott-ag-to-acquire-australian-
microfluidic-expert-minifab> (accessed 11 July 2023).

94 Yole Intelligence (2022) Status of the Microfluidics Industry 2022. <https://yolegroup.com/articles/> (accessed 11 July 2023).



4.2 Recommendations 

This section outlines ten recommendations that aim 
to provide Australia with the foundation for pursuing 
the four identified opportunities. Recommendations 
were co-developed with industry, research, and 
government stakeholders. While the opportunities 
were developed by considering what could be possible 
within a 15-year time horizon, setting Australia on 
a path towards these opportunities would require 
actioning all recommendations within five years.

Within these five years, recommendations can be 
prioritised and ordered by themes, with those aimed 
at coordinating and updating existing processes 
considered the most important first steps by those 
consulted (Figure 9). These activities would set a strong 
foundation for the remaining recommendations, which 
aim to integrate local capabilities into medical product 
development before strengthening production and 
commercialisation. Recommendations for non-animal 
model validation data will provide the evidence base 
to generate momentum across the other themes.

Consideration and implementation of the proposed 
recommendations would benefit from national 
coordination. The Australian Government would 
likely lead initial decision-making in these areas. 
However, many of the recommendations will require 
dedicated support and implementation from other 
levels of industry, research, and government.

While the design of these recommendations is to 
mature Australia’s medical product development 
capabilities, implementation of these recommendations 
could also benefit applications in other fields such 
as veterinary and agricultural medicines, industrial 
chemicals, cosmetic testing, and eco-toxicology. 

Figure 9. Recommendations for strategically maturing Australia’s non-animal model capabilities 

Coordinate and update 
existing processes

1. Establish a national 
consortium that coordinates 
and promotes Australia’s 
non‑animal model capabilities 
2. Develop national data 
collection standards on the 
use of animals in scientific 
research, teaching and testing
3. Align TGA processes and 
industry guidance with new 
FDA procedures for accepting 
non-animal model data


Integrate local 
capabilities

4. Develop a national biobanking 
and tissue collection network 
5. Integrate outputs from 
NCRIS platforms into a 
coordinated pipeline for 
non-animal models


Strengthen production 
and commercialisation

6. Facilitate IP management and 
material access for research 
and industry collaborations
7. Enhance commercial 
skillsets across the 
non‑animal model sector
8. Update biomedical R&D 
infrastructure to support 
non‑animal model capabilities


Validation data

9. Conduct retrospective studies that compare animal and non-animal model predictivity
10. Conduct systematic reviews of locally and internationally developed non-animal models




RECOMMENDATION 1 
Establish a national consortium 
that coordinates and promotes 
Australia’s non‑animal 
model capabilities 

The Australian non-animal model landscape has a breadth 
of expertise, primarily operating in siloes. Forming a 
consortium centred on Australian non-animal model 
capabilities and 3Rs advancement would strengthen existing 
integration initiatives, increase the chances of productive 
collaboration, and facilitate advocacy for local capabilities.

Consulted stakeholders identified existing entities that 
share some of these responsibilities in the broader 
Australian biomedical sector and could form the basis 
of the national consortium. These include the Australian 
& New Zealand Council for the Care of Animals in 
Research and Teaching (ANZCCART), CSIRO, MTPConnect, 
Phenomics Australia, the Therapeutic Goods Administration 
(TGA), and Therapeutic Innovation Australia. 

Regardless of where specific responsibilities sit, strong 
collaboration with these existing organisations will be 
critical for the consortium’s success. Adequate resourcing, 
both in terms of funding and dedicated personnel, will also 
be essential for the consortium to succeed. Exploring and 
identifying mechanisms that can support its establishment 
and sustainability over time should be initial priorities. 

Stakeholders highlighted several activities that would 
benefit from being governed by the consortium:

• Supporting the establishment of networks of local 
developers, input producers, and prototyping groups 
to reduce manufacturing costs and encourage 
the formation of small-scale value chains.
• Fostering communication and collaboration 
between research, industry, government, and 
regulators, both domestically and internationally.
• Identifying, promoting, and facilitating partnerships 
of local institutions with multinational companies by 
matching their organ or disease modelling interests 
with locally developed models and expertise.
• Tracking key national non-animal model 
R&D and industry capabilities and promoting 
these domestically and abroad. 
• Representing Australian interests at global 
non‑animal model forums and existing international 
initiatives around global harmonisation, regulation, 
standards, and privacy. For example, the Putting 
Science into Standards projects with CEN-CENELEC 
or the Microphysiological Systems Affiliate of the 
International Consortium for Innovation and Quality 
in Pharmaceutical Development (IQ MPS).95
• Developing, maintaining, and promoting a 
publicly available resource that collates emerging 
funding opportunities (public and industry-led 
initiatives), both national and international, relevant 
to non-animal models and 3Rs advances. 
• Building on reports such as this one, develop a 
national strategy that includes investment priorities, 
the coordination of efforts across Australian entities, 
and the broad analysis of global unmet needs. 
The EU’s Smart Specialisation initiative represents a 
successful model to consider for national coordination, 
leveraging localised strengths and differences.96 
• Facilitating the collaborative development 
of frameworks for locally developed models, 
including reporting requirements for research 
involving non-animal models, performance 
standards, validation requirements, qualification 
pathways, and steps for regulatory acceptance.
• Developing manufacturing standards, standardised 
protocols, and validated guidelines for model 
inputs and complete models to minimise the 
need for repetitive validation by end-users.


95 CEN-CENELEC (2023) Putting-Science-Into-Standards (PSIS). PSIS. <https://www.cencenelec.eu/get-involved/research-and-innovation/cen-and-cenelec-
activities/putting-science-into-standards/> (accessed 11 July 2023); IQ Microphysiological Systems Affiliate (2019) About Us. IQ MPS. <https://www.iqmps.
org/about-us> (accessed 11 July 2023).

96 European Commission – Joint Research Centre (2022) Smart Specialisation. EU Science Hub. <https://joint-research-centre.ec.europa.eu/scientific-activities-z/
smart-specialisation_en> (accessed 11 July 2023).



RECOMMENDATION 2 
Develop national data collection 
standards on the use of 
animals in scientific research, 
teaching, and testing

Data regarding the use of animals for scientific purposes 
is collected and shared in different forms by State and 
Territory governments and animal ethics committees 
in Australia. Most of this data lacks granularity around 
the specific research or activities, making it challenging 
to understand where and how animals are being used. 
Funding bodies, such as the NHMRC, have no role in 
the standardised collection or use of this data.

National data collection standards on the use of animals 
for scientific purposes, aligned to established international 
standards, will help to inform funding decisions better to 
support the 3Rs, track the impact of 3Rs initiatives, guide 
infrastructure development, rationalise animal use, and 
inform priority areas for non-animal model development.

The EU’s ALURES database is one example of 
an international standard Australia may look 
to align with. Specific animal use information 
collected under this system includes:97

• The number and species of animals, 
their origins, and dates of use.
• Type of research (basic, translational, or 
regulatory) and relevant sub-categories.
• Purpose of research (biomedical or 
ecological/environmental purposes).
• Diseases, therapeutic areas, or organ 
systems being researched.


This activity would require a coordinating entity, which 
would be responsible for consulting with key research, 
industry, government, regulatory groups, and animal 
protection organisations across all jurisdictions. This entity 
would also need to develop standards, and data collection 
and reporting systems. Stakeholders suggested several 
existing organisations could be appropriate for this 
responsibility, including the Australian & New Zealand 
Council for the Care of Animals in Research and Teaching 
(ANZCCART), CSIRO or the Federal Government.

RECOMMENDATION 3 
Align TGA processes and 
industry guidance with new 
FDA procedures for accepting 
non-animal model data

Stakeholders cited a lack of clarity on the acceptability 
of non-animal model data to fulfil regulatory 
requirements as a barrier to the greater adoption of 
non-animal models in medical product development. 
Leveraging new biological models and associated 
data to support medical product development will 
require more explicit guidance and formal regulatory 
precedents of non-animal model-assessed products.

Aligning Australia’s national regulatory processes and 
guidance material with leading international jurisdictions 
with more advanced non-animal model ecosystems will 
provide researchers and industry with greater confidence 
to plan for the use of non-animal models. The FDA 
and the EMA are at the forefront of global efforts to 
integrate non-animal models into regulatory practice 
and represent important institutions for continuous 
engagement by the TGA. Aligning TGA processes with 
the FDA is proposed due to the US regulator’s proactive 
role in collaborative programs that are continuously 
producing non-animal model improvements and guidance, 
as well as the extensive public funding provided to the 
sector in the US. Both are factors likely to spur faster 
regulatory innovation that the TGA can build upon. 

Beyond the FDA and EMA, a joint effort between the TGA 
and other national regulators for producing harmonised 
guidance on non-animal model data acceptability could 
further facilitate the growth of non-animal model 
applications. Global harmonisation is a longer and more 
complex process than aligning to a single regulator, but 
it will encourage greater market access for Australian 
collaborations and commercial offerings in the long term. 

While harmonisation activity is being explored, several 
short-term initiatives that clarify regulatory requirements 
could help prepare Australian stakeholders for the 
challenges and opportunities of greater regulatory 
alignment on non-animal models. These initiatives could 
include conducting reviews of non-animal model data 

97 European Commission – Directorate-General for Environment (n.d.) Statistics and non-technical project summaries. Animals in science. <https://environment.
ec.europa.eu/topics/chemicals/animals-science/statistics-and-non-technical-project-summaries_en> (accessed 11 July 2023).

98 Baran SW, Brown PC, Baudy AR, Fitzpatrick SC, Frantz C, Fullerton A, Gan J, Hardwick RN, Hillgren KM, Kopec AK, Liras JL, Mendrick DL, Nagao R, Proctor 
WR, Ramsden D, Ribeiro AJS, Stresser D, Sung KE, Sura R, Tetsuka K, Tomlinson L, Van Vleet T, Wagoner MP, Wang Q, Arslan SY, Yoder G, Ekert JE (2022) 
Perspectives on the evaluation and adoption of complex in vitro models in drug development: Workshop with the FDA and the pharmaceutical industry 
(IQ MPS Affiliate). ALTEX – Alternatives to animal experimentation 39(2), 297–314.



acceptability across countries and actively participating in 
forums where the requirements for advanced non-animal 
models and the suitability of their data for regulatory 
purposes are being discussed. This includes attending the 
joint IQ MPS-FDA workshops with industry representatives 
and regulatory agencies and engaging the regulatory 
boards of dedicated organisations like EUROoCS.98

RECOMMENDATION 4 
Develop a national biobanking 
and tissue collection network 

Patient-derived cells and tissues are essential for 
the development of more biologically relevant 
models of disease that improve patient outcomes. 
A national biobanking and tissue collection network 
can provide researchers and industry access to 
well‑characterised cells and tissues that support 
preclinical testing of therapeutic candidates and 
personalised models for clinical treatment selection.99

A national biobanking and tissue collection network could 
also facilitate a greater representation of the Australian 
population in medical product and non‑animal model 
development through more extensive, geographically 
and ethnically diverse patient cohorts. Greater population 
representation will enable the earlier identification of 
potential adverse events and better-suited treatments 
for a broader range of patients. Individual nodes 
could exist across States and Territories that exchange 
and share methodologies, clinical practices, standard 
operating procedures, expertise, and data.

While Australia has the foundation to enable a national 
biobanking and tissue collection network – including 
existing biobanks – there is still a requirement for planning 
and coordinating contributing facilities and programs 
at the national level. A set of national patient registries, 
organised by therapeutic area, could offer an initial basis 
to build a national biobanking and tissue collection 
network. Registries provide a standardised format 
and integrate patient characteristics with therapeutic 
decisions, collected tissue data, and clinical outcomes.

Stakeholders noted that the network’s set-up should 
consider the biobanking and tissue collection needs 
and integrate multiple discrete steps and participants 
into a streamlined pipeline. The needs include:

• Patient recruitment, support, retention, 
and follow-up, with assistance from nursing 
staff experienced in clinical trials.
• Sample collection, in coordination between clinicians 
and researchers for patient and tissue selection.
• Sample processing, derivation into stable in vitro 
models for longer‑term storage, and standardised 
characterisation with strict data quality controls.
• Logistics for patient tissue delivery and access 
pathways to stored in vitro models for researchers.
• Large-scale digital infrastructure that allows 
secure data storage and sharing, monitoring 
tissue activity (e.g., transport, provision, use), and 
integrating de-identified clinical data to support 
future precision medicine applications.
• Research partnerships to enable access and 
uses by industry, which involves testing novel 
therapeutics on a range of patient-derived models 
and delivering the data to partnering companies.
• Coordinated ethics frameworks considering broad 
subsequent uses for the collected tissues from the 
start to support R&D activities and data generation.


99 Roth A (2021) Novel human cell models in drug development: How 3D, Organoids & Organs on Chips can improve and renew current paths – and our 
vision for the future. Danish 3R-symposium. <https://en.3rcenter.dk/fileadmin/user_upload/Editor/images/Om_Danmarks_3R-Center/Symposium/
Symposium_2021/Adrian_Roth-Danish_3R-symposium_Nov_2021-_FINAL.pdf> (accessed 11 July 2023).



RECOMMENDATION 5 
Integrate outputs from NCRIS 
platforms into a coordinated 
pipeline for non-animal models

High-quality biological model research outputs 
(e.g., biological materials, novel model types, 
human-relevant data, and hardware) are produced 
by multiple NCRIS platforms. However, a sequential 
process that supports the streamlined production 
of non-animal models is yet to connect them. 

Increasing service coordination at the national level 
is supported by the formation of the NCRIS Health 
Group, comprising Bioplatforms Australia, the National 
Imaging Facility, Phenomics Australia, the Population 
Health Research Network, and Therapeutic Innovation 
Australia.100 Their capabilities, in combination with those 
of other relevant NCRIS platforms like the Australian 
National Fabrication Facility (ANFF), could be explicitly 
integrated for non-animal model development, 
production and use. This approach would mirror the 
pipeline already established for novel therapeutics.

A holistic pipeline for non-animal model activity would 
use the outputs of one NCRIS platform as inputs for 
another, encouraging model standardisation, user-guided 
improvement, and increased uptake. A national pipeline 
builds upon the foundation set by the NCRIS Health Group, 
which could lead this initiative. The following discrete steps 
may help inform and advance the integration process: 

• Identify all outputs relevant to non-animal 
models currently produced across the various 
NCRIS Health Group platforms and other 
platforms not yet aligned with NCRIS. 
• Examine existing barriers to sharing and using platform 
outputs in an integrated, sequential process. These may 
include challenges or gaps in governance, funding, 
communication, ethics and privacy, or logistics. 
• Propose specific outputs, models, and mature 
applications to support a small-scale pipeline. 
For instance, organ-on-chip production for trial 
pre-screening using ANFF-produced hardware 
and patient-specific iPSCs derived by members 
of the Phenomics Australia network.


RECOMMENDATION 6 
Facilitate IP management and 
material access for research 
and industry collaborations

Variable IP management can represent a significant 
challenge for collaborative and fee‑for‑service 
work involving non-animal models.

A dedicated working group could co-design a streamlined, 
nationally coordinated framework that facilitates material 
transfer agreements between organisations, supports IP 
protections for outputs from non-animal model work in 
domestic and international markets, and allows sufficient 
clarity and flexibility for industry collaborations. 
The working group could include representatives 
from IP Australia, universities, medical research 
institutes, pathology labs and hospitals, industry 
organisations, funding bodies (e.g., NHMRC and 
ARC), and relevant professional associations like 
the Australasian Research Management Society.

100 Australian Government Department of Health and Aged Care (2023b) National Critical Research Infrastructure initiative. Our work. <https://www.health.gov.
au/our-work/national-critical-research-infrastructure-initiative> (accessed 11 July 2023).



RECOMMENDATION 7 
Enhance commercial skillsets 
across the non‑animal model sector

Stakeholders described the limited commercial expertise 
of non-animal model researchers and developers, and a 
need for non-animal model expertise among commercial 
professionals nested within non-animal model developing 
organisations, as significant barriers to commercialisation 
in Australia. Skills and expertise necessary to improve 
commercialisation include business case development, 
economic modelling, capital raising, IP and sales, 
bench‑to-bedside product translation, and biotechnology 
commercialisation leadership. Several initiatives could 
assist local organisations in addressing these skills gaps:

• Establishing knowledge transfer opportunities 
(e.g., secondments) for commercial professionals 
from local non-animal model developing 
organisations with leading international 
commercial non-animal model organisations.
• Recruiting experienced international 
biotechnology commercialisation specialists 
into key translational institutions.
• Providing commercialisation training to 
non‑animal model researchers and developers. 
Such initiatives can draw from and build upon 
programs like Cicada Innovation’s NSW Health 
Commercialisation Training Program.101 


RECOMMENDATION 8 
Update biomedical R&D 
infrastructure to support 
non‑animal model capabilities

Enabling the broader use of non-animal models across 
the Australian medical product development process 
requires addressing infrastructure needs across 
fundamental research, translational research services, 
clinical applications, and manufacturing capabilities. 

For instance, ensuring that equipment assessments 
at universities and research institutes are periodic 
and have a forward-looking approach that considers 
compatibility with complex in vitro models can further 
adoption in fundamental research. These assessments 
can help minimise long-term overreliance on animal 
and in vitro 2D models due to a lack of alternatives.

To fully enable the clinical applications of non-animal 
models, it will be essential to develop infrastructure 
near clinical settings for processing and using 
patient‑derived tissues in personalised complex in 
vitro models. This proximity can help guarantee tissue 
viability and reduce testing timelines, making the process 
more compatible with clinical practice and needs.

Additionally, increasing the number of Good Laboratory 
Practice (GLP) and Good Manufacturing Practice (GMP) 
accredited facilities can benefit the broader non-animal 
model ecosystem. Identifying high-quality settings 
within Australia and supporting their accreditation or 
re-accreditation can be an enabler of greater activity, 
particularly for organisations in the translational 
research services and manufacturing segments. 

101 Cicada Innovations (2023) NSW Health Commercialisation Training Program. <https://www.cicadainnovations.com/nswhealthctp-updates> (accessed 11 July 2023).



RECOMMENDATION 9 
Conduct retrospective studies 
that compare animal and 
non‑animal model predictivity

The lack of direct performance comparisons regarding the 
predictivity of clinical outcomes has limited the uptake 
of non-animal models in regulatory toxicology testing.

Retrospectively applying non-animal models can 
provide valuable insights on several fronts. Assessing 
withdrawn medical products of known clinical toxicity 
can help benchmark non-animal to animal model 
predictivity performance (including species-specific 
differences). Similarly, a retrospective assessment of 
medical products that exhibited preclinical toxicity in 
animal models and did not advance to clinical trials 
can explore the loss of potentially useful medical 
products due to animal-specific toxicity that would not 
occur in humans. Both comparisons can encourage 
the adoption of non-animal models, help improve 
performance and usability, strengthen a knowledge 
base for regulators’ future interpretation of data, and 
support a broader economic case linked to their use. 

Moreover, conducting retrospective studies involving 
medical products of known toxicity could serve as the 
last evaluative step in the qualification and regulatory 
acceptance process of novel non-animal models.

RECOMMENDATION 10 
Conduct systematic reviews 
of locally and internationally 
developed non-animal models

The many non-animal models already developed, locally 
and abroad, across different organ systems and for distinct 
applications represents both a significant challenge 
and an opportunity. Essential data that can inform the 
greater use of non-animal models throughout the medical 
product development process are available but require 
an objective, comprehensive and technical-oriented 
assessment. Conducting thorough systematic reviews 
of non-animal models and their documented uses in 
medical product development has been highlighted as a 
need by consulted stakeholders. Such reviews should:

• Assess models’ capabilities, limitations, use aspects, 
and predictive performances. Comprehensive 
assessments should focus on the extent to 
which models replicate well-defined pathways 
or endpoints of human toxicity and disease. 
• Cover different therapeutic areas and specific diseases 
with their respective clinically relevant endpoints. 
• Where possible, support or integrate dedicated 
prospective or retrospective analyses, directly 
comparing predicted toxicity with clinical trial results. 


Reviews are regarded as a cost-effective approach to 
assessing the field’s current state, informing changes to 
local regulatory practice, identifying areas of unmet need, 
and further building support for the use of non‑animal 
models in Australian research and industry. The EU 
Reference Laboratory for alternatives to animal testing 
(EURL ECVAM) has led comparable initiatives previously 
for seven disease areas.102 An Australian project would 
benefit significantly from leveraging the work already 
done and engaging closely with EURL ECVAM. Systematic 
reviews could also identify suitable areas for Australian 
projects exploring the deliberate integration of non-animal 
models in medical product development, similar to the 
EU’s overarching project on ‘Innovations to accelerate 
vaccine development and manufacture (Inno4Vac)’.103 

102 Disease areas include respiratory tract diseases, breast cancer, neurodegenerative disorders, immuno-oncology, immunogenicity testing for advanced 
medicinal therapy products, cardiovascular diseases, and autoimmune diseases. European Commission - Joint Research Centre (2022) Review of advanced 
non-animal models in biomedical research. Biomedical research. <https://joint-research-centre.ec.europa.eu/eu-reference-laboratory-alternatives-animal-
testing-eurl-ecvam/biomedical-research_en> (accessed 12 July 2023).

103 European Vaccine Initiative (2023) Inno4Vac project. <https://www.inno4vac.eu> (accessed 11 July 2023); Innovative Medicines Initiative (IMI) (2021) 
Innovations to accelerate vaccine development and manufacture (Inno4Vac). Projects & results <http://www.imi.europa.eu/projects-results/project-
factsheets/inno4vac> (accessed 11 July 2023).



4.3 R&D priorities

The recommendations outlined in the previous section 
will need to be supported by the research community’s 
ongoing maturation and validation of non-animal models. 
For researchers seeking to contribute to these efforts, the 
priorities for advanced model development most frequently 
suggested by consulted stakeholders are presented below. 

The specifics of these R&D activities will ultimately vary 
by model and tissue type. These priorities also represent 
areas of opportunity for novel commercial offerings or 
collaborations, given their status as unmet needs globally. 

Support the economic case for non-animal 
models in medical product development

• Perform comprehensive economics studies 
that quantitatively assess non-animal models’ 
direct and indirect economic benefits, including 
time and cost impacts across the entire product 
development process and broader impacts 
on patient safety and health outcomes.


Improve analytics for increasingly complex 
in vitro models 

• Strengthen 3D model analysis capabilities by linking 
semi-automated testing, high-content imaging and 
read-out capture, and data processing pipelines.
• Implement detailed phenotyping technologies 
(e.g., single-cell RNA sequencing and live imaging) 
to characterise the temporal and spatial dynamics 
of cellular populations within models to assess 
the accuracy of tissue and disease modelling 
and the detailed effects of therapeutics.
• Embed real-time physicochemical and metabolite 
sensing capabilities in locally developed 
culture hardware to characterise culture 
microenvironments and assess cellular responses.
• Refine, validate, and implement data analyses powered 
by artificial intelligence approaches to leverage 
increasingly complex biological model outputs and 
clinical data for safety and efficacy assessments.


Advance the quality of model inputs 
and hardware

• Improve the design and composition of biomaterials 
used in in vitro 3D models to ensure maximum 
relevance to human physiology through appropriate 
signalling and biomechanical properties.
• Implement epithelial or endothelial barriers that use 
disease-relevant cell types and spatial arrangements 
instead of synthetic and non-physiological proxies.104
• Transition towards hardware materials that, 
unlike those currently used for prototyping, are 
optically clear, chemically inert, non-absorbent, 
and better-suited to large scale manufacture.
• Improve model designs (across hardware, biological 
inputs, and use protocols) to allow greater ease 
of use, manufacturing under strict QA and QC 
conditions, and off-the-shelf applications. 
• Improve the design of the microfluidic chips used 
in organ-on-chip models to reach simultaneous 
culture, reproducibility, and output read-out levels 
compatible with high-throughput applications.
• Improve and standardise perfusion strategies, including 
vascular precursors or bioprinting, for in vitro 3D 
models that do not rely on external microfluidic 
systems (i.e., organoids and tissue explants). 
Enhanced perfusion is critical to increasing model size 
for better reflecting native tissue size and function. 


Extend the capabilities of in vitro models 
for a closer recreation of human physiology 
across healthy and diseased states

• Optimise multi-organ modular culture approaches to 
enable physiologically relevant Absorption, Distribution, 
Metabolism and Excretion (ADME) studies, systemic 
toxicity, and pharmacokinetics/pharmacodynamics. 
• Incorporate immune system interactions into in vitro 
models through optimised simultaneous culturing of 
immune cells, systemic distribution modelling, and 
replication of migration and tissue infiltration.
• Where relevant to the disease or condition 
of interest, extend the effective culture time 
of complex models to the multi-month range 
to reflect disease onset and progression and 
chronic and cumulative exposure effects.
• Optimise co-differentiation protocols to reduce 
variability in differentiation efficiency across batches, 
minimise unwanted cell phenotypes, and closely 
control the identity of cell populations in a model.
• Incorporate tissue-relevant biomechanical cues in 
complex models to support more comprehensive 
differentiation and maturation strategies.
• Deliberately integrate in silico modelling expertise 
at the outset of data-intensive biomedical research 
projects, such that the iteration between in silico and 
in vitro models leads to improvements in both.


104 Irrechukwu O, Yeager R, David R, Ekert J, Saravanakumar A, Choi CK (2023) Applications of microphysiological systems to disease models in the 
biopharmaceutical industry: Opportunities and challenges. ALTEX – Alternatives to animal experimentation 40(3), 485–518.



5 Appendices

A.1 Consulted stakeholders

CSIRO would like to thank the following organisations for their contributions to the project through interviews and 
reviews. The insights expressed throughout this report were developed by considering the collective views obtained 
alongside independent economic and qualitative research. They may not always align with the specific views of one of the 
consulted individuals or organisations. This list is not to be interpreted as an endorsement or promotion of this report.

23 Strands Pty Ltd

CSL Seqirus

Phenomics Australia

360biolabs

The Defence Science and 
Technology Group

QIMR Berghofer Medical Research 
Institute

ACT Health 

European Commission Joint Research 
Centre (JRC)

Queensland Department of Agriculture 
and Fisheries 

ARC Centre for Personalised Therapeutics 
Technologies (CPTT)

European Organ-on-Chip Society 
(EUROoCS)

Queensland Health

AusBiotech

Flinders University

Schott Minifab Pty Ltd

Australian & New Zealand Council for the 
Care of Animals in Research and Teaching 
(ANZCCART)

Gelomics Pty Ltd

Tessara Therapeutics Pty Ltd

Australian Clinical Trials Alliance

Griffith Institute for Drug Discovery 
(GRIDD)

The Australian National University

Australian Ethical

Harry Perkins Institute of Medical 
Research

The Medical Advances Without Animals 
Trust (MAWA)

Australian Government Department of 
Education

Humane Research Australia

The Peter Doherty Institute for Infection 
and Immunity

Australian Government Department of 
Health and Aged Care

Inventia Life Science Pty Ltd

The University of Adelaide

Australian Government Department of 
Industry, Science and Resources

Leiden University Medical Center

The University of Melbourne

Australian Government Office of the Chief 
Scientist

Medicines Development for Global Health

The University of Newcastle

Australian Organoid Facility

Melbourne Centre for Nanofabrication

The University of Sydney 

Avicenna Alliance

Moderna

Therapeutic Goods Administration (TGA)

Bico

Monash Biomedicine Discovery Institute

Therapeutic Innovation Australia

BioMelbourne

Murdoch Children's Research Institute 
(MCRI)

Transport Canberra and City Services 
Directorate

Center for Alternatives to Animal Testing 
(CAAT)

National Health and Medical Research 
Council (NHMRC)

UK National Centre for the Replacement 
Refinement & Reduction of Animals in 
Research (NC3Rs)

Children's Cancer Institute

New South Wales Department 
of Primary Industries

University of South Australia

Children's Medical Research Institute

New South Wales Ministry of Health

US National Institutes of Health (NIH) 
– National Center for Advancing 
Translational Sciences (NCATS)

Cortical Labs Pte Ltd

OminiWell Pty Ltd

Victor Chang Cardiac Research Institute

Crux Biolabs Pty Ltd

Opthea

WA Health

Commonwealth Scientific and Industrial 
Research Organisation (CSIRO)

Peter MacCallum Cancer Centre

Walter and Eliza Hall Institute of Medical 
Research (WEHI)







A.2 Biological model comparisons 

Table 6. Biological model comparisons105

IN SILICO

IN VITRO

IN VIVO 

QUALITATIVE CRITERIA

2D

3D

TISSUE EXPLANT

ORGAN-ON-CHIP (OoC)

ANIMAL MODELS

Biological resemblance

How close is the resemblance to the native 
process or tissue being modelled?

Low

Mathematical and probabilistic 
models focus on specific 
outcome prediction rather than 
the comprehensive replication of 
tissues and processes. 

Low

Conventional 2D cell cultures 
do not replicate tissue 
microenvironment architecture, 
composition, biomechanics, or 
the cellular dynamics they enable 
(e.g., migration and improved 
signalling). 

Low-Medium

Conventional 3D models are static, 
rely on matrices with limited human 
relevance, and face challenges in 
vascularisation and reproducibility 
(e.g., size, shape, and cell types 
present).

High

Carefully sourced and cultured 
explants preserve relevant 
characteristics of tissues in vivo: 
diverse cellular populations, spatial 
arrangements, architecture, and 
complex interactions.

Medium-High

Organ-on-chip devices enable 
continuous flow, controllable 
biomechanical cues, and 
multi‑organ interaction but still 
require improvements for complex 
3D architectures and reducing the 
use of extraneous materials.

Medium-High

Animals are compatible with 
genetic modifications and transitory 
phenotype inductions capable 
of replicating human diseases. 
However, genetic, physiological, 
and behavioural differences persist, 
limiting model accuracy and result 
reproducibility. 

Technical expertise

How much training is required 
for effective model use?

Medium-High

In silico models rely on knowledge 
and skills across chemistry, 
pharmacology, engineering, 
mathematics, cell biology and 
human physiology.

Low

Cell cultures in a conventional 2D 
setting require limited training, 
core laboratory skills, and basic 
knowledge of cell biology. 

Low-Medium

In addition to the requirements for 
2D cell culture, 3D models require 
more extensive knowledge of 
human physiology and materials 
engineering, plus training in the 
imaging and analysis of 3D cultures. 

High

Sourcing, preparing, and culturing 
human tissue explants requires skills 
(or interdisciplinary collaborations) 
across surgery, histology, anatomical 
pathology, and organotypic culture 
techniques.

High

Organ-on-chip models leverage 
the skills described for conventional 
3D models and require additional 
training in microfluidics and 
biomechanics. 

High

Using animal models requires 
specialised skills in animal 
husbandry, surgery, histology, 
and cell culture, and significant 
knowledge of pharmacology and 
a particular species’ physiology.

Technological maturity

How close is the model to formal and widespread 
adoption of its evidence in regulatory submissions?

Medium-High

In silico models are well 
characterised and amenable to 
validation, with some variants 
(like quantitative structure‑activity 
relationship models) already in 
regulatory use. However, more 
extensive human datasets could 
further optimise them. 

High

2D models are widely accepted for 
specific regulatory applications 
(particularly chemical assessment) 
and are well-characterised and 
optimised. 

Medium-High

3D models have well‑established 
production methods, hardware 
compatible with existing equipment 
(e.g., plates) and are increasingly 
used for internal decision‑making. 
However, better characterisation 
of cellular populations and inputs 
(e.g., matrices) is necessary to 
achieve the standardisation required 
for regulatory uses.

Low

The high variability of primary 
tissues, the relative difficulty in 
sourcing them, and the lack of 
guidelines for their use in medical 
product development limit this 
model’s maturity.

Low-Medium

The technological maturity of 
organ-on-chip models is limited 
by the use of materials unsuitable 
for large-scale manufacture, the 
lack of minimum standards for 
commercially available platforms, 
and the still-unresolved formal 
validation and qualification of this 
model type for an initial context of 
use. 

High

The technological maturity of 
animal models is supported by 
extensive historical use, detailed 
characterisation, and prominence 
in regulatory applications.

Direct and indirect costs

When considering acquisition, associated 
costs and operational expenses, how 
comparatively expensive is the model?

Low

Once developed and trained 
(which can carry significant costs), 
model use is relatively inexpensive, 
limited to computational resource 
cost and result analysis. 

Low-Medium

Well-established practices, 
mass‑produced consumables, 
existing economies of scale, and the 
limited expertise required minimise 
costs. First-time acquisition of 
the equipment can significantly 
increase the overall cost. 

Medium-High

3D models maintain the same 
direct and indirect cost categories 
from 2D culture, with increases due 
to consumables (e.g., specialised 
plates, matrices, and growth factors) 
and more complex characterisation, 
imaging, and analysis.

High

High expenses are associated with 
organotypic culture and with the 
use of primary tissue, which is likely 
to be surgically extracted, scarce 
and costly.

High

Besides retaining the increased 
costs of 3D models, organ‑on‑chip 
devices require extensive supporting 
equipment (e.g., pumps), specialised 
consumables (e.g., microfluidic chips), 
and additional training for 
reliable use.

High

Animal models involve direct 
expenses on husbandry, histology, 
and analysis, and indirect costs from 
extensive planning, preparation, 
and significant space demands.

Throughput

What is the model's capacity to test many 
experimental conditions and generate 
endpoint‑relevant results in a short period of time?

High

Modelling an interaction or 
predicting an outcome is only 
limited by computational resources.

High

2D models are compatible with 
quick culture times, basic readouts, 
assay miniaturisation, and simple 
handling. This compatibility makes 
them amenable to standardisation, 
automation, and large-scale parallel 
culture, which are vital for high 
throughput.

Medium-High

Due to their increased complexity, 
3D models require longer 
production or culturing times 
(e.g., organoids) and specialised 
characterisation, output detection 
and analysis. However, they remain 
compatible with large‑scale 
production, automation, and 
parallel testing.

Low

The lower availability of primary 
tissues, short culture longevity and 
technical complexity of organotypic 
culture (often dependent on manual 
interventions or custom steps) limit 
the use of explants for large-scale, 
parallel testing.

Low-Medium

Available organ-on-chip devices still 
need to improve in their throughput 
due to the extensive equipment 
required to operate individual 
chips (which precludes scalability) 
and the prevalent use of few wells/
systems per chip (which limits 
parallelisation). 

Low

While one animal may be used to 
assess multiple distinct endpoints, 
it remains a single experimental 
unit. Animal models also rely 
on extensive manual labour and 
require longer overall timeframes 
to yield results (e.g., acclimation, 
testing, processing). 





Notes: (i) ‘in vitro 3D’ includes scaffolds, spheroids, and organoids. (ii) There are significant differences between animal models (e.g., rodents vs primates).

QUALITATIVE SCALE

Capacity to perform comprehensively in a criterion, or requirements needed for use

Low

Limited capacity or limited requirements

Low-Medium

Minimal capacity or minimal requirements

Medium-High

Intermediate capacity or intermediate requirements

High

Significant capacity or significant requirements





105 Ekert JE, Deakyne J, Pribul-Allen P, Terry R, Schofield C, Jeong CG, Storey J, Mohamet L, Francis J, Naidoo A, Amador A, Klein J-L, Rowan W (2020) 
Recommended Guidelines for Developing, Qualifying, and Implementing Complex In Vitro Models (CIVMs) for Drug Discovery. SLAS Discovery 25(10), 
1174–1190; Irrechukwu O, Yeager R, David R, Ekert J, Saravanakumar A, Choi CK (2023) Applications of microphysiological systems to disease models in the 
biopharmaceutical industry: Opportunities and challenges. ALTEX – Alternatives to animal experimentation 40(3), 485–518; Low LA, Mummery C, Berridge 
BR, Austin CP, Tagle DA (2021) Organs-on-chips: into the next decade. Nature Reviews Drug Discovery 20(5), 345–361.



A.3 National strengths analysis

Methodology

Table 7. Search methodology

RESEARCH 
QUESTION

DATA SOURCE

METHODS

ANALYSIS 

Research strengths

What non-animal 
models does 
Australian research 
have comparative 
global strengths in?

Web of Science 
(WoS) Core 
Collection

Bibliometric analysis 
via MeSH defined 
search term lists per 
non‑animal model 
type and organ 
system (19/02/2018 – 
19/02/2023).

Metric: Publication counts per organ system/non-animal 
model type combination (e.g., 3D-Nervous system), 
Australian percentage of global publications, and global 
ranking for Australia.

Based on these metrics, combinations were deemed to 
be research strengths (Australian percentage of global 
publications > 3%).

Clinical strengths

What condition 
categories does 
Australia conduct the 
most clinical trials in?

Australia and 
New Zealand 
Clinical Trials 
Registry 
(ANZCTR)

Search of condition 
categories that can 
be directly matched 
to an organ system 
(19/02/2018 – 
19/02/2023)

Metric: Number of registered clinical trials, with recruitment 
in Australia, per condition category (e.g., neurological).

Individual condition categories were sorted based on the 
absolute number of clinical trials for the period, with priority 
assigned to the top 5.

Industry strengths

What therapeutic 
areas does Australian 
industry participate 
the most in?

AusBiotech 
Australian 
Life Sciences 
Innovation 
Directory

Search of primary 
therapeutic areas that 
can be directly matched 
to an organ system 
(as of 19 February 2023)

Metric: Number of companies operating in relevant primary 
therapeutic areas (e.g., Diseases of the nervous system/
neurology)

The therapeutic areas AusBiotech-listed companies operate 
in were sorted based on the absolute number of companies, 
with priority assigned to the top 5.

National strengths: Areas of alignment that emerged after matching organ systems (research) to their equivalent condition categories 
(clinical) and therapeutic areas (industry).





Model assumptions and limitations

• The research strengths component focuses on 
publication output by organ system, non-animal model 
type, and the combination of the two. No impact 
metrics were applied (e.g., category normalised citation 
impact) because of the small number of articles in 
some combinations. The focus on output supports a 
consistent approach across the different combinations 
and the comparison to other countries. However, it 
does not capture non-public R&D outputs: industry 
platforms, commercial-in-confidence work by research 
groups, and in-progress research. Commercially oriented 
data sources like patents were also out of scope. 
• Absence of representation in the analysis results 
should not be interpreted as a lack of high-quality 
research occurring in an area or mature models 
that could evolve into commercial offerings. 
• Despite the use of structured lists in the research 
strengths analysis, the output of each search 
contains a percentage of mismatched articles 
due to language overlap between non-animal 
model categories. Publication counts should be 
considered ‘likely hits’ instead of exact figures. 
• Some of the examined organ system categories 
are not mutually exclusive between themselves 
(e.g., urogenital, and reproductive systems). They 
are analysed and presented individually despite 
their overlap to allow comparison to ANZCTR 
Condition Categories and Australian Life Sciences 
Innovation Directory Therapeutic Areas.


The analysis does not consider certain condition 
categories (clinical trial strengths) and primary 
therapeutic areas (industry strengths) as they cannot 
be directly matched to an organ system. Such is the 
case with cancer/neoplasms/oncology, as it represents 
a broad category encompassing all organ systems.



 Summary of results 

Table 8. Australian share of global publications by organ system and non-animal model type for combinations where national activity 
accounts for ≥3% of global output.

ORGAN SYSTEM

NON-ANIMAL MODEL TYPE

AU PERCENTAGE 
OF GLOBAL 
PUBLICATIONS

GLOBAL 
POSITION

GLOBAL 
TOTAL

AU TOTAL

Eye

3D (Scaffold and organoid)

7.5%

5

372

28

Urogenital system

3D (Scaffold and organoid)

6.2%

8

921

57

Reproductive system

3D (Scaffold and organoid)

6.1%

7

904

55

Eye

3D (General)

6.0%

6

599

36

Ear

Tissue explant

5.0%

8

40

2

Eye

2D

4.7%

6

610

29

Reproductive system

3D (General)

4.7%

8

2380

113

Urogenital system

3D (General)

4.7%

7

1779

83

Reproductive system

2D

4.6%

7

2073

96

Reproductive system

Tissue explant

4.6%

10

130

6

Cardiovascular system

3D (Scaffold and organoid)

4.5%

7

617

28

Respiratory system

2D

4.2%

10

1839

77

Urogenital system

2D

4.0%

6

1738

70

Nervous system

3D (Scaffold and organoid)

4.0%

10

1557

62

Cardiovascular system

3D (General)

3.7%

10

1401

52

Integumentary system

2D

3.6%

10

1030

37

Metabolic and endocrine 
systems

3D (Scaffold and organoid)

3.4%

11

1104

38

Nervous system

3D (General)

3.4%

12

2801

96







Table 9. Organ system-model type combinations not emerging as current globally comparative strengths in the publication analysis

ORGAN SYSTEM

NON-ANIMAL MODEL TYPE

Blood

2D, 3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Cardiovascular

2D, Tissue explant, Organ-on-chip

Digestive system

2D, 3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Ear

2D, 3D (General, scaffold and organoid), Organ-on-chip

Eye

Tissue explant, Organ-on-chip

Inflammatory disorders and immune system

2D, 3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Integumentary

3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Metabolic and endocrine

2D, 3D (General), Tissue explant, Organ-on-chip

Musculoskeletal

2D, 3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Nervous system

2D, Tissue explant, Organ-on-chip

Reproductive health and childbirth

Organ-on-chip

Respiratory

3D (General, scaffold and organoid), Tissue explant, Organ-on-chip

Urogenital

Tissue explant, Organ-on-chip





Table 10. Number of registered clinical trials per condition category (Top 5), with recruitment in Australia (2018–2023, 
as of 19/02/2023)

ANZCTR CONDITION CATEGORY

NUMBER OF REGISTERED TRIALS

PERCENTAGE OF TOTAL

Total

9633

100.0%

Neurological

1108

11.5%

Cardiovascular

893

9.3%

Respiratory

840

8.7%

Musculoskeletal

831

8.6%

Inflammatory disorders and immune system (immune)

695

7.2%





Table 11. Number of companies active in the top 5 therapeutic areas, operating in Australia in 2023 (as of 19/02/2023). 

AUSBIOTECH PRIMARY THERAPEUTIC AREA

NUMBER OF COMPANIES

PERCENTAGE OF TOTAL

Total106

359

100.0%

Diseases of the nervous system / neurology

101

28.1%

Cardiovascular / cardiology

70

19.5%

Respiratory / pulmonology

66

18.4%

Endocrine, nutritional, and metabolic diseases / endocrinology

61

17.0%

Musculoskeletal system and connective tissue

57

15.9%





106 As some companies present in the Life Sciences Directory operate in more than one therapeutic area, percentages for individual areas do not add up to 100%.



A.4 Australian industry capability

The organisations and capabilities listed below provide an overview of the Australian non-animal model industry. 
Table 12 includes organisations offering mature non-animal model products or services within Australia. 
Table 13 includes organisations offering supporting or adjacent services (including CROs with non-animal 
model capabilities, HTS providers and component producers). These organisations were identified through 
online searches and consultations. As such, this may not be an exhaustive list of all relevant industry capability.

Table 12. Organisations with mature non-animal model product or service offerings

ORGANISATION NAME
Click for more information

ORGANISATION 
TYPE

PRIMARY MEDICAL 
PRODUCT 
DEVELOPMENT STAGE

NON-ANIMAL 
MODEL 

APPLICATION 
(ORGAN OR DISEASE)

23 Strands Pty Ltd

NSW

Company

Clinical application

In silico

In vitro 2D

In vitro 3D

Cancer, heart, 
reproductive

CellBank Australia (Children's 
Medical Research Institute)
NSW

Research 
institution

Various

In vitro 2D – 
iPSC, Cell lines

Various

Centenary Institute
NSW

Research 
institution 

Discovery development

In vitro 2D – 
iPSC and others

In vitro 3D

Cancer, heart, 
respiratory, skeletal

Cortical Labs Pte Ltd
VICCompanyVariousIn silicoIn vitro 2DBrainGriffith University, Griffith Institute 
for Drug Discovery

 #
QLD

Research 
institution

Fundamental research, 
Discovery development 
(HTS), Preclinical 
development

In vitro 2D

Various

Harry Perkins Institute of 
Medical Research, Translational Cancer 
Research Program in Oncology ⁰
WA

Research 
institution

Preclinical development

In vitro 3D – 
Organoid

Cancer

Hunter Medical Research Institute

NSW

Research 
institution

Various

In vitro 2D

In vitro 3D – 
Organoid

Brain, cancer, immune 
system, reproductive, 
respiratory

Inventia Life Science Pty Ltd #
NSW

Company

Fundamental research, 
Screening and lead 
optimisation (HTS), 
Preclinical development

In vitro 3D – 
Organoid

Brain, cancer, liver, 
lung, ovaries, 
prostate, skin, stem 
cell, others

Monash University, Monash Organoid 
Program & Monash Genome 
Modification Platform ⁰
VIC

Research 
institution

Screening and 
lead optimisation, 
Preclinical development

In vitro 3D – 
Organoid

Breast, colorectal, 
epithelial tissue, 
gastrointestinal

Murdoch Children's Research 
Institute, iPSC Derivation & 
Gene Editing Facility ⁰
VIC

Research 
institution

Discovery development

In vitro 2D – 
iPSC‑derived

Various

OminiWell Pty Ltd
SACompanyTarget identification, 
ToxicologyOrgan-on-chipBrain, breast, gut, 
blood vesselsQIMR Berghofer Medical 
Research Institute


QLD

Research 
institution

Target identification, 
Toxicology

In vitro 2D

In vitro 3D – 
Organoid

Brain, heart







ORGANISATION NAME
Click for more information

ORGANISATION 
TYPE

PRIMARY MEDICAL 
PRODUCT 
DEVELOPMENT STAGE

NON-ANIMAL 
MODEL 

APPLICATION 
(ORGAN OR DISEASE)

RewiredBio Pty Ltd

NSW

Company

Target identification

In silico

Various

Stem Cell and Organoid Facility 
(Children's Medical Research Institute)
NSW

Research 
institution

Preclinical 
development, Clinical 
development

In vitro 2D – 
iPSC‑derived

In vitro 3D – 
Organoid

Various

Tessara Therapeutics Pty Ltd
VICCompanyTarget identification, 
Screening and 
lead optimisation, 
Preclinical developmentIn vitro 3DBrainThe University of Melbourne, 
Faculty of Medicine, Dentistry 
and Health Sciences

 ⁰
VIC

Research 
institution

Target identification, 
Preclinical development

In vitro 2D

In vitro 3D – 
Organoid

Tissue explant

Cancer, central 
nervous system, ear, 
eye, gastrointestinal, 
respiratory, stem cell, 
others

The University of New South Wales
NSW

Research 
institution

Preclinical 
development,

Functional assays

In vitro 3D – 
Organoid, adult stem 
cell derived

Gut, lung, others

The University of Newcastle

NSW

Research 
institution

Preclinical development

In vitro 3D

Tissue explant

Cancer, heart, 
intestine, respiratory, 
reproductive, others

The University of Queensland, 
Australian Organoid Facility #
QLD

Research 
institution

Screening and lead 
optimisation (HTS), 
Preclinical development

In vitro 3D – 
Organoid, adult 
stem cell- and 
iPSC‑derived

Brain, blood, kidney, 
cancer, others

The University of Queensland, In vitro 
Genome Engineering and Disease 
Modelling Service ⁰
QLD

Research 
institution

Target identification, 
Preclinical development

In vitro 2D

In vitro 3D – 
Organoid

Various

The University of South Australia, 
Centre for Cancer Biology

SA

Research 
institution

Target identification, 
Screening and 
lead optimisation, 
Preclinical development

In vitro 2D – Cell 
lines

Tissue explants

Brain cancer

The University of Sydney
NSW

Research 
institution

Toxicology, 
Pharmacology 

Organ-on-chip

Lung

Victor Chang Cardiac 
Research Institute #⁰
NSW

Research 
institution

Screening and lead 
optimisation (HTS), 
Functional assays

In vitro 2D – 
iPSC‑derived

In vitro 3D – 
Organoid

Heart

Westmead Institute 
for Medical Research 

NSW

Research 
institution

Target identification, 
Screening and 
lead optimisation, 
Preclinical development

In vitro 2D – 
Cell lines

In vitro 3D – 
Cell lines, Organoid

Tissue explant

Breast, cancer, 
immune system, skin







Table 13. Supporting service providers for non-animal model production

ORGANISATION NAME
Click for more information

ORGANISATION 
TYPE

SUPPORTING 
SERVICE

NON-ANIMAL 
MODEL 

APPLICATION 
(ORGAN OR DISEASE)

360biolabs (Burnet Institute)
VIC

CRO

Functional assays, 
Pharmacodynamics, 
Pharmacokinetics

In vitro 2D

Cancer, infectious 
disease, others

ANU Centre for Therapeutic Discovery ⁰
ACT

Research 
institution

HTS

In vitro 2D

In vitro 3D

Various

Children's Cancer Institute Drug 
Discovery Centre
NSW

Research 
institution

HTS

In vitro 2D

In vitro 3D

Various

Codex Research

NSW

Company

Component 
production

Components for in 
vitro 2D and 3D

Various

Crux Biolabs Pty Ltd
VICCROFunctional assays, 
Pharmacodynamics, 
PharmacokineticsIn vitro 2DImmunology, 
inflammation, othersFlinders University, Cell Screen SA


SA

Research 
institution

HTS

In vitro 2D

Various

Gelomics Pty Ltd

QLD

Company

Component 
production

Components for in 
vitro 3D

Cancer, endothelial, 
mesenchymal, hepatic, 
others

Griffith University, Griffith Institute 
for Drug Discovery #
QLD

Research 
institution

HTS

In vitro 2D

Various

Inventia Life Science Pty Ltd #
NSW

Company

HTS

In vitro 3D – Organoid

Brain, cancer, liver, 
lung, ovaries, prostate, 
skin, stem cell, others

Monash University, Monash Institute 
of Pharmaceutical Sciences, Centre for 
Drug Candidate Optimisation
VIC

Research 
institution

HTS, 
Biopharmaceutics

In vitro 2D

Cancer, central nervous 
system disorders,
cardiovascular disease, 
infectious disease, 
metabolic disease

National Drug Discovery Centre 
(Walter and Eliza Hall Institute 
of Medical Research)
VIC

Research 
institution

HTS

In vitro 2D

Cancer, cardiovascular 
diseases, degenerative 
disorders

Schott Minifab Pty Ltd

VIC

Company

Component 
production

Components for 
in vitro 2D, 3D and 
organ‑on‑chip

Various

The University of Queensland, 
Australian Organoid Facility #
QLD

Research 
institution

HTS

In vitro 3D – 
Organoid, adult stem 
cell- and iPSC-derived

Brain, blood, kidney, 
cancer, others

The University of Wollongong, 
Translational Research Initiative for 
Cellular Engineering and Printing
NSW

Research 
institution

Bioprinting

In vitro 3D

Blood, ear, eye

Victor Chang Cardiac 
Research Institute #⁰
NSW

Research 
institution

HTS

In vitro 2D – 
iPSC‑derived

In vitro 3D – Organoid

Heart

Victorian Centre for Functional 
Genomics (Peter MacCallum 
Cancer Centre) ⁰
VIC

Research 
institution

HTS

In vitro 2D

In vitro 3D

Tissue explant

Anal, breast, colorectal, 
gastrointestinal, heart, 
ovarian, pancreas, 
prostate, stem cell

vivoPharm Pty Ltd

VIC

CRO

Toxicology

In vitro 2D

Various





Notes: (i) ‘Various’ refers to technologies that can be applied across a broad range of organs and diseases or medical 
product development stages. (ii) ‘Research institution’ includes charities, not-for-profit organisations and universities. 
(iii) # = Provider of both non-animal model and supporting service. (iv) ⁰ = Phenomics Australia supported.



A.5 Economic analysis methodology

Market sizing approach 

The 2040 global opportunity for organoids and organs-
on-chips was modelled based on existing market reports. 
Australia’s potential global market share in 2040 was 
calculated using estimates of Australia’s current market 
share percentage. Potential headcount employment 
was also calculated. Insufficient data prevented the 
modelling of other non-animal model types.

Current estimate of global organoid and 
organ-on-chip market size and CAGR

Current estimates of the global market for organoids 
and organs-on-chips were based on revenue averages 
provided by market research reports published between 
2019 and 2022. All revenue figures were converted and 
adjusted for inflation.107 The global organoids market 
in 2022 is estimated at $1.62 billion,108 while the global 
organ‑on‑chip market in 2022 is estimated at $0.11 billion.109

Forecasts of the global market size for organoids and 
organs-on-chips were also collected. The year 2026, within 
the forecast range of several market research reports, 
was chosen as the starting year for projections out to 
2040. The global organoids market in 2026 is estimated 
to be $3.26 billion, while the global organ‑on‑chip 
market in 2026 is estimated to be $0.31 billion.110

Non-animal model market report sources estimate 
revenue to grow by an average of 17.4% in the organoids 
market and 29.4% in the organ-on-chip market.111 
The 2026 revenue estimates were projected to 2040 
using the respective average growth rate, relying 
on the assumption that the average CAGR for these 
models are constant over the years up to 2040.

Market share captured by Australia by 2040

Australia’s market share for these model types was 
estimated using the number of published non‑animal 
model research articles related to each model type 
as a proxy.112 The average percentage of global 
research articles published over ten years, between 
2013 and 2022, for organoids and organs-on-chips 
involving Australian researchers or institutions 
were 4.1% and 2.7%, respectively. These market 
shares were assumed to be static over the years. 

Headcount employment for these 
models in Australia by 2040 

The ratio of wages to revenue for Australian scientific 
research services was used as a proxy to estimate the 
relationship between wages and revenue for both 
non-animal models.113 Figures taken from an Australian 
industry report include forecasts out to 2028.114 A ten-year 
average of the wages to revenue ratio from 2019 to 2028 
was estimated to be 39.1%. Similarly, the average wage 
per worker in Australian scientific research services was 
used as a proxy for the average wages in the non‑animal 
model sector. Average wages in the scientific research 
services are currently $101,300,115 and the ten-year 
average growth rate of wages from 2019 to 2028 (0.9%) 
was calculated from the same industry report and used 
to forecast wages out to 2040. The wage/revenue ratio 
and average wage were then applied to the Australian 
2040 market share to derive headcount employment.

107 ATO End of financial year rates: USD 1 = $0.69 for year ending Dec 2022; USD 1 = $0.75 for year ending Dec 2021; USD 1 = $0.69 for year ending Dec 2020; 
USD 1 = $0.72 for year ending Dec 2019; Figures were adjusted for inflation using ABS CPI: All groups, Index Numbers and Percentage Changes for March 
Quarter 2023.

108 Technavio (2022), Human Organoids Market by End-user and Geography - Forecast and Analysis 2022-2026; bccResearch (2022), Laboratory Animal Models, 
3D Cultures and Organoids: Global Markets; Allied Market Research (2022), Organoids and Spheroids Market by Type, by Method, by End User: Global 
Opportunity Analysis and Industry Forecast, 2021-2031 Market; The Insight Partners (2021), Organoids Market Forecast to 2027 - COVID-19 Impact and Global 
Analysis By Type, Application, Source, and Geography; Markets and Markets (2020), Human Organoids Market by Product, Usability, Application, End-user - 
Global Forecast to 2025.

109 Allied Market Research (2022), Organ-on-Chip Market by Type: Global Opportunity Analysis and Industry Forecast, 2020-2030; Data Bridge Market Research 
(2022), Global Organ-On-Chip Market – Industry Trends and Forecast to 2029; bccResearch (2022), Organ-on-a-Chip: Global Markets; GMD Research (2021), 
Global Organ-on-Chip Market 2020-2030; Technavio (2022), Organs on Chips Market 2022-2026 by End-user and Geography – Forecast and Analysis 2022-
2026; Yole Group (2019), Organs-On-Chips Market and Technology Landscape.

110 The average global forecasted market size for 2026 was calculated by extrapolating the closest market size figures to the year 2026.

111 Based on an average of the growth estimates provided in the cited market research reports.

112 Based on web of science search results using keywords ‘organoid’ and ‘organoids’ for number of published organoid articles and ‘microphysiological systems’, 
‘microphysiological system’, ‘organ-on-a-chip devices’, ‘organ on a chip devices’, ‘organ-on-a-chip device’, ‘organ chips’, ‘organ chip’, ‘organ-on-a-chip’, ‘organ 
on a chip’, ‘organ-on-a-chips’, ‘organoids-on-a-chip’, ‘organoids on a chip’, and ‘organoids-on-a-chips’ for number of published organ-on-chip articles.

113 While the development, adoption and commercialisation of non-animal models requires a diversified workforce (from research scientists to industrial 
developers and manufacturers to IP experts) the proportion of labour demand for research scientists would likely be greater, at least in the short term.

114 IBISWorld (2022) M6910 Scientific Research Services in Australia Industry Report.

115 IBISWorld (2022) M6910 Scientific Research Services in Australia Industry Report.



Summary of reported results

Table 14 summarises the results of the market sizing 
approach. The global organoids market in 2022 was 
$1.62 billion and is expected to reach $30.91 billion by 
2040, with a CAGR of 17.4%. Under current research 
output trends, estimates of this market result in 
4,200 jobs and $1.28 billion in revenue for Australia 
by 2040. By comparison, the organ-on-chip market is 
at an earlier phase of adoption. This has been valued at 
$110 million worldwide in 2022, projected at $11.48 billion 
by 2040 (29.4% CAGR). The potential share for Australia 
by 2040 is $310 million in revenue, with 1,000 jobs. 

Table 14. Summary of economic analysis results by 
non‑animal model. 

 

ORGANOIDS

OOC

Potential global revenue 
by 2040

$30.91B

$11.48B

Potential Australian 
revenue by 2040

$1.28B

$0.31B

Potential Australian 
headcount employment 
by 2040

4,200 jobs

1,000 jobs





Case study approach

This case study aimed to get a sense of cost, time 
and animal-use differences between animal and 
non-animal models in preclinical testing. 

Animal use differences

Marty et al. (2022) estimated the potential percentage 
reduction in the number of animals if an equivalent 
non-animal method is used instead of, or in complement 
with, an animal model for several OECD-approved 
toxicity tests.116 The potential percentage reduction in 
the number of animals used depends on factors such 
as regulatory acceptance and the adequacy of the 
information provided by a non-animal method versus 
an animal model. Table 15 incorporates this data. 

Cost differences 

Using information from Marty et al. (2022), equivalent 
animal and in vitro tests were grouped and compared 
based on their prediction of the same toxicity category.117 
The cost data for these tests were then extracted 
from Meigs et al. (2018).118 Only one in vitro and one 
animal test were then chosen for each category based 
on the lowest cost of each test. It was assumed that 
laboratory decision-makers could choose the most 
cost-effective option for the given category. 

The costs, in Euros and assumed to be in the base year 2018, 
were converted to 2022.119 There was no consideration 
of additional costs that might arise from repeating the 
experiment or introducing new endpoints or tested doses, 
as the magnitude of such costs can vary significantly and 
is dependent on the outcome of the primary experiment.

116 Marty MS, Andrus AK and Groff KA (2022) Animal metrics: Tracking contributions of new approach methods to reduced animal use. ALTEX – Alternatives to 
animal experimentation, 39(1), 95–112.

117 OECD TG 474 and OECD TG 475 were assumed to be equivalent during grouping.

118 Meigs L., Smirnova L, Rovida C, Leist, M and Hartung T (2018) Animal testing and its alternatives – the most important omics is economics. ALTEX – 
Alternatives to animal experimentation, 35(3), 275–305.

119 ATO End of financial year rates ending 30 June 2019: EUR 1 = $0.66 for year ending December 2018. Figures were adjusted for inflation using ABS CPI: 
All groups, Index Numbers and Percentage Changes for March Quarter 2023.



Time differences

Data on the duration of equivalent animal and non‑animal 
studies were collected to determine possible time 
savings. Technical information provided by OECD Test 
Guidelines (TG) documents was used for this purpose.120 
To accurately compare the duration of both animal and 
non-animal models, the time taken to prepare for the 
tests,121 the observation period122 and any post-exposure 
procedure123 up until data analysis and reporting were 
all considered. The maximum duration of the test within 
the range provided was assumed to be the duration of a 
test, in line with a conservative approach. For instance, 
in OECD Test no. 487, the treatment time is said to 
last 3 to 6 hours and 6 hours was assumed to be the 
duration of the treatment time. It is important to note 
that the estimated time duration data does not capture 
confirmatory tests in the case of animal models and 
repetition of the experiment in the case of non‑animal 
models, as the decision to conduct them is uncertain and 
contingent on the results of the primary experiment. 

Summary of reported results

Table 15 summarises the results of estimating cost, time and 
animal use differences of equivalent animal and non‑animal 
models. Table 15 is an extended version of the table used 
in section 4.1.2 and provides additional case studies. 

For instance, when comparing non-animal testing to 
animal testing for eye irritation in the table below, it is 
observed that the former incurs an additional cost of 
$1,000. However, this can be offset by significant time 
savings of around 21.8 days and a reduction 2.7 in animal 
use. In the case of mutagenicity, using a non‑animal test 
will only partially replace the animal model. It cannot 
be said with certainty how this will impact the cost and 
duration of the animal test used.124 In this case, animal 
reduction is estimated to be between 0 to 6.4. 

120 OECD (2021) Test No. 405: In Vivo Eye Irritation/Serious Eye Damage; OECD (2020) Test No. 437: Bovine Corneal Opacity and Permeability Test Method; 
OECD (2015) Test No. 404: Acute Dermal Irritation/Corrosion; OECD (2015) Test No. 435: In Vitro Membrane Barrier Test Method for Skin Corrosion; OECD 
(2010) Test No. 429: Skin Sensitisation; OECD (2022a) Test No. 442C: In Chemico Skin Sensitisation; OECD (2022b) Test No. 442D: In Vitro Skin Sensitisation; 
OECD (2022c) Test No. 442E: In Vitro Skin Sensitisation; OECD (2016a) Test No. 475: Mammalian Bone Marrow Chromosomal Aberration Test; OECD (2016b) 
Test No. 487: In Vitro Mammalian Cell Micronucleus Test.

121 The preparation period for animal models involves examination of the animal or animal parts for quality control purposes, the acclimation period, and 
any additional procedures such as de-skinning the animal. For non-animal models, the preparation period refers to the pre-culturing of cells, equilibration 
period or pre-incubation period.

122 The observation period for animal studies refers to the time elapsed after administration of the test substance during which animals are observed for signs 
of toxicity. In non-animal models, this time period is often referred to as the incubation period, treatment time or exposure to the test chemical.

123 Post-exposure procedures are more common in non-animal models and refer to any steps taken after the observation period such as ‘post-immersion’, 
‘post‑incubation period’, ‘HPLC analysis time’ or ‘centrifugation and staining of cells’.

124 As a result, we cannot talk about cost and time savings in this case. The same applies for skin sensitisation.



Table 15. Comparison between OECD tests featuring equivalent animal and non-animal models for various toxicity endpoints, 
across cost, time, and animal use.125 

TYPE OF TOXICITY 

OECD TG

TEST TYPE

STUDY COST 
(AUD 2022)

TIME 
DURATION 
(DAYS)

POTENTIAL % 
REDUCTION 
IN ANIMALS 
USED BY USING 
EQUIVALENT 
NON-ANIMAL 
MODELS

NUMBER 
OF 
ANIMALS 
USED126

POTENTIAL 
NUMBER OF 
ANIMALS SAVED 
BY USING 
EQUIVALENT 
NON-ANIMAL 
MODELS127

ACUTE ENDPOINTS 

Eye irritation/corrosion 

Acute eye irritation

405

in vivo 

2,000

22.0

N/A

2.7

N/A

Bovine corneal opacity 
and permeability test 
(BCOP)

437

in vitro 

3,000

0.2

100%

N/A

2.7

Skin irritation/corrosion

Acute dermal irritation/
corrosion

404

in vivo 

2,000

15.0

N/A

2

N/A

Membrane barrier 
test method for skin 
corrosion

435

in vitro 

4,100

0.2

100%

N/A

2

Skin sensitisation128

Local lymph node assay

429

in vivo 

7,100

11.0

N/A

16

N/A

Direct peptide 
reactivity assay (DPRA)

442C

in vitro 

6,400

2.3

33% – 50%

N/A

5.28 – 8

ARE-Nrf2 luciferase 
LuSens test 

442D

in vitro 

6,300

3.1

33% – 50%

N/A

5.28 – 8

Human cell line 
activation test (h-CLAT)

442E

in vitro 

11,700

4.1

33% – 50%

N/A

5.28 – 8

GENOTOXICITY ENDPOINTS

Mutagenicity 

Mammalian bone 
marrow chromosomal 
aberration test

475

in vivo 

105,600

19.0

N/A

64

N/A

Mammalian cell 
micronucleus test 

487

in vitro 

14,400

2.3

0 – 10%

N/A

0 – 6.4





125 Adapted from Marty et al. (2022); Meigs et al. (2018); CSIRO Futures calculations.

126 Data on the number of animals used was derived from Meigs et al. (2018).

127 The figures provided are only average estimates and are not rounded to a full number of potential animals saved.

128 In the case of skin sensitisation, the three non-animal model tests must be done in complement to benefit from 33% – 50% reduction in animal use. As a 
result, the total cost of using non-animal model approaches would be $24,400 and the duration of those tests will be around 9.5 days assuming that they 
are not done concurrently. It cannot be said with certainty how this will impact the cost and duration of the animal model test. 



For further information

CSIRO Futures

Greg Williams
+61 3 9545 2138
greg.williams@csiro.au
csiro.au/futures

As Australia’s national science agency, 
CSIRO is solving the greatest 
challenges through innovative 
science and technology.

CSIRO. Unlocking a better future 
for everyone.

Contact us

1300 363 400
+61 3 9545 2176
csiro.au/contact
csiro.au