In silico studies (digital models/digital twins)

Introduction

The use of digital technology is significantly changing the methods used in health research, in particular by speeding up access to data. In this respect, big data, combined with digital technology and Information Technology (IT), enables the use of Artificial Intelligence (AI) systems[1]. In this sense, the French National Consultative Ethics Committee stated in 2018 that pre-clinical and clinical research is "embarking on a digital revolution"[2]. As a result, there are many applications of IT and digital technology in health research, widely referred to as in silico studies. One of the best example of this revolution are virtual trials, known as in silico trials. This method, which is not intended to virtually replace standard randomised clinical trials, appears to be a fourth pillar (sometimes called 3rd pillar in the literature by the authors who defend it can replace in vivo clinical trials) of health research, complementing in vitro and in vivo research (See Figure 1).

Figure 1: In silico studies as the 4th pillar of health research

In silico studies is therefore a new form of model representation/modelling to support health research. The techniques used make it possible to construct multi-scale models of a disease and to represent the functional relationships between the different biological entities[3].

In extreme cases, in silico methods make it possible to move from modelling a group or cohort of patients to a precise individual[4]. The digital twin technique is used as a patient-specific virtual representation of real-world systems or processes (from organs to the whole body)[5].

These innovative methods have made their mark in the industrial sector (e.g. automotive and aeronautics). However, universities and public research centers have had a significant impact on the use of bioinformatics in healthcare[6]. For example, one of the first ideas for using in silico studies in healthcare was formalised by the IUPS Physiome project in 2001 for a project based on the study of the human genome[7].

Over the last twenty years, in silico studies have been used to optimise the development of medical devices and, more recently, human medicines. Today, the use of in silico methods seems to be more dynamic in industry and, more generally, in the world of private research.

While definitions of in silico studies are emerging, the current lack of harmonisation of the terminology used makes it difficult to identify the methods. In fact, in silico studies or clinical trials can be associated with, for example, Computational Modelling and Simulation (CM&S) or Model-Informed Drug Development (MIDD) methods. Similarly, different methods and models are used: molecular dynamic models, mechanistic models, phenomenological models, pharmacokinetic models, toxicological models, etc.

Main actors

The European Medicines Agency (EMA) is an agency of the European Union (EU) whose goal is to protect and promote human and animal health. The agency is notably responsible for the scientific assessment of advanced therapy medicinal products to be marketed within the EU and the European Economic Area. The EMA can provide support to research stakeholders wishing to use in silico methods, in particular for the qualification of these methods, the acceptance of their use in research and, most importantly, the acceptance of new evidence arising from the use of in silico methods to support the evaluation of clinical and non-clinical data for a marketing authorisation application for a medicinal product under the centralised procedure.

The sponsor is the research organisation, pharmaceutical researcher or industry taking the responsibility for the initiation, the management, and the setting up of the financing of the clinical trial, including the use of in silico methods in research for the development of health products.

The investigator is the researcher (health professional) responsible for the conduct of the research and, where appropriate, the use of in silico method.

Manufacturers of digital tools are those capable of developing the digital tools needed to conduct in silico studies in compliance with applicable regulations. They may be IT start-ups or software publishers that work with sponsors to enable in silico studies.

Definitions

In silico: this term is a neologism referring to silicon, the main material found in computer chips. In this sense, ‘in silico’ refers to computer modelling and simulation techniques.

In silico studies: are all studies (clinical or preclinical) conducted using computer modelling and simulation.

In silico clinical trials: In silico clinical trials are virtual clinical trials using complex technologies based on computer simulation and/or artificial intelligence algorithms in order to support the development or the regulatory evaluation of a medicine, a medical device or a medical intervention.

In silico pe-clinical studies: are studies conducted in the pre-clinical phase using computer modelling and simulation.

Whatever the terminology or model used, the needs are similar. To carry out in silico studies, it is important to have a large amount of quality data (big data), scientific knowledge and high-performance digital and computer systems, or even algorithmic systems. In practice, in silico studies relies on two types of methods, either those based on prior knowledge or those based on new data, which require the automatic development of predictors for the data without necessarily having a causal hypothesis.

Challenges

There are many challenges currently facing the development and use of in silico studies. The main challenge for the use of in silico studies is the lack of expertise in Europe for the evaluation and validation of these methods and the resulting data[8]. The skills required to run in silico studies are scarce and expensive. In addition, in silico requires a mastery of computer modelling and simulation, which means that pharmaceutical R&D needs to open up to other scientific practices. The use of in silico also relies on widespread access to high quality, standardised data.

These innovations appear to break with 'traditional' research methods and therefore require new ways of generating scientific evidence, which may be met with some reluctance in practice.

Using in silico studies in the context of health research is a challenge in itself. The limitation of in silico use could be holding back potential improvements in clinical research and thus the potential to accelerate pharmaceutical discovery or the development of healthcare products to improve human health[9].

The lack of a regulatory framework or support for the use of in silico studies and the acceptance of the resulting digital evidence is hampering potential advances in research. This lack of guidance is currently justified by a lack of harmonisation of terminology and practices, a lack of communication between stakeholders and regulators, and a lack of experts in the field at European level.

Certain industries and research organisations turn to the US, and in particular to the FDA, for assistance in the oversight of their in silico methods, thus creating disparities in research at the European level[10].

The use of in silico studies is in the ethical and societal interest for public health, as it promises to speed up and reduce the cost of research and thus of healthcare products, particularly in areas under pressure such as rare diseases. However, to ensure the safe and acceptable use of these methods, it is important that the principles of transparency and trust are reaffirmed. Equally, equitable access to these methods must be ensured so as not to create an unbalanced environment that undermines the principle of freedom of research.

Legal and ethical requirements need to evolve in line with the risks that the use of in silico studies accentuates or creates.

There is currently a real need for dialogue with regulators, in particular to clarify and coordinate the existing frameworks that overlap and that are applicable to in silico methods (see figure 2).

Figure 2: Overlap of regulations and standards applicable to in silico studies

Advantages

The use of computer modelling and simulation can improve randomised controlled trials by providing new forms of evidence. The contribution of in silico methods can be seen in the testing of subgroups, by raising issues that have not yet been studied, by completing the research or regulatory protocol, and by limiting the presence of approximations in research protocols. In this sense, in silico methods can be presented as a tool to anticipate deviations from the protocol, making it possible to propose the best way to conduct research. This is made possible by providing greater precision, which makes it possible to identify and refine the target populations most relevant to the research, and also by the possibility of predicting efficacy data, by providing information (proof of concept)[11] as well as interpretations before the start of the research[12]. These advantages guarantee real savings in terms of resources, time and money[13]. Computer modelling and simulation allow multi-dimensional analyses to be carried out to identify the probabilities of success of the research, whereas in conventional trials a large number of molecules and combinations have to be tested to identify the most appropriate ones[14]. The trial's chances of success are then increased because the trials conducted are based on consolidated hypotheses[15]. For all research organisations, whether public or private, and especially for the pharmaceutical industry, in silico studies are a means of increasing productivity and promoting innovation in the field[16].

In silico methods add a new dimension to research, making it possible to extrapolate the kinetic behaviour of molecules between species, between different routes of exposure, at different doses, and to anticipate interactions between molecules[17]. The use of computer modelling and simulation involves comparing two identical research protocols, but with certain differences, such as dosage. This is almost impossible to reproduce in real-life conditions due to various factors, such as laboratory or clinical conditions or the choice of participants, which affect the results of the research[18]. In this sense, the use of in silico allows a series of different hypotheses and conditions to be crossed, thus overcoming these biases. It is easier to isolate and identify the effects of variation that may be causing undesirable effects. Thus, in silico studies bring an aspect of knowledge and understanding that changes the way we look at the trial[19].

Advantages of in silico preclinical studies:

At the stage of fundamental research and preclinical studies[20], in silico methods allow to analyse toxicity, structure/activity relationships or to provide proof of concept. Modelling and simulation can be used to target the properties of a compound and predict its development in different contexts. It helps to identify the most suitable compounds, which can then be tested in clinical trials under real-life conditions[21]. It is also possible to identify biomarkers to optimise subsequent research.

Advantages of in silico clinical trials:

In silico can be used in phase I of the clinical trial to provide information on the most appropriate dose or the most relevant sampling, for example, in preparation for phases II and III. Finally, its use in phases II and III can optimise regulatory submissions[22] and decision analysis[23]. In particular, decision-making involves a 'go/no-go' approach to the question of whether or not to conduct a clinical trial, in particular by identifying elements (in particular biomarkers) that can clarify and facilitate decision-making[24].

In silico clinical trials can be of benefit to patients by reducing, or at least anticipating, risk[25] (by identifying the causes of potential side effects or unacceptable adverse reactions)[26], suggesting a more targeted trial (e.g. by defining an optimal dose) and optimising[27] or even reducing the selection of trial participants[28]. In addition, by using in silico methods, it is possible to test exposures between individuals (inter-individual variability) and for the same individual (intra-individual variability). This makes it easier to extrapolate results between different population groups, taking into account age, ethnicity, genetic variations and physical conditions (pregnancy, obesity, multiple pathologies, etc.). In silico studies could thus make it possible to tailor a substance to the precise characteristics of a patient and even to adapt it on an intra-individual basis by monitoring changes over the course of a lifetime (health status, anatomy, physiology, etc.)[29].

In silico clinical trials complement traditional clinical trials, making it possible to take better account of trials carried out on vulnerable people, who are generally under-represented[30]. In silico clinical trials are also useful in situations where it is not possible to create parallel groups because the number of participants is too small[31]. This is the case, for example, in paediatrics, geriatrics[32] or rare diseases[33]. As a result, the use of in silico clinical trials provides a better representation of the participants[34].

Advantages of In silico studies for other uses:

In silico can also play a role in phase IV. Indeed, going further, we note that they are also tools that can be used for monitoring beyond the traditional clinical trial, i.e. during post-marketing studies[35]. The use of in silico methods can enable population biomonitoring and very long-term monitoring of the evolution of molecules in organisms.

More specifically, computer simulations and predictions make it possible to target biomarkers or genes that may be of particular interest in the treatment of a disease[36]. These numerical predictions are then a real asset in providing certain answers for optimising the payload, designing vectors, assessing genotoxicity or even identifying promising experimental research protocols[37]. In silico methods can also be used to optimise and complete the databases available for the development of particular therapies, such as gene therapies. Finally, various artificial intelligence tools such as machine learning, deep learning or even generative AI can be used, which are particularly relevant for identifying genes that may be of therapeutic interest or helping to select donors for allogeneic cell therapies.

The contribution of in silico studies to the development of gene therapies overlaps with the promise of personalised medicine, which is particularly highlighted by the EU-STANDS4PM consortium[38]. Through collaboration between stakeholders, EU-STANDS4PM aims to raise awareness of the need to integrate health data (in particular genetic data) in order to optimise the development of in silico models based on data of particular interest for improving the development of gene therapies[39]. This work highlights the first obstacle to the use of in silico methods for gene therapy: access to relevant and controlled data.

Figure 3: Uses and benefits of in silico at different stages of research

Opportunities and incentives

The EU encourages research organisations to use in silico studies (more especially in silico pre-clinical studies) as an alternative to animal testing[40]. Therefore, these innovations are particularly identified as methods that comply with the European 3Rs (Refining, Reducing, and Replacing animal testing) pursued by Europe.

At the same time, the EMA is providing incentives by supporting the qualification of digital technology-based methodologies to support approval of medicinal products[41].

Moreover, the EMA is responding to the growing interest in the emergence of new digital techniques in the use of medicines, in particular by including a reference to these methods in revised guidance[42]. It also clarifies its intentions in the EMA's Strategic Action Plan 2025, which places great emphasis on the development of computer modelling and simulation[43] and the qualification of new methodologies[44]. The document reveals that by 2030, members of the working group hope to have developed a catalogue of in silico studies[45].

Finally, various players and stakeholders are currently joining forces to promote the use of in silico methods, notably under the aegis of consortia supported by the European Commission, such as the Avicenna Alliance or VHP Institute, or national initiatives.

Interaction with regulators

Interaction between the sponsor and the EMA or national agencies:

In the context of in silico applications, interactions with regulatory authorities are mostly early interactions in advance of the study or trial. For example, sponsors of a clinical trial using in silico methods can request support from the EMA. In addition, interactions with the EMA or national agencies should be considered so that they can recognise the reliability of the numerical evidence generated by in silico methods and thus the validity of the product developed.

At the EMA level, the CHMP provides scientific advice on protocols and methods designed to develop innovative approaches, with the goal of progressing toward formal qualification of novel methodologies for medicine development. This advice is based on an assessment of the scientific justification and any preliminary data submitted to the EMA as well as on recommendations from the EMA's Scientific Advice Working Party. Since 2013, the CHMP and the Scientific Advice Working Party are supported by an operational group of experts on modelling and simulation[46].

If the novel methodology under review shows promises but is not yet ready for full qualification, the EMA may offer a letter of support as an option. These letters aim to encourage data-sharing and facilitate further research toward eventual qualification of the methodology.

A letter of support typically includes:

A high-level overview of the novel methodology,
Its intended context of use,
Available data, and
Details of ongoing and planned investigations.

More information on the EMA website here.

Other stakeholders interactions:

Associations of industrials and academics involved in the use of in silico methods to improve health are proving to be key players in raising awareness of the challenges of in silico methods and in trying to steer European and international policies towards greater adoption of these innovations for the benefit of human health[47].

European Union Guidance

In 2016, the EMA developed guidelines[48] to support researchers in the use of a physiological pharmacokinetic modelling and simulation methods (which are just one example of possible models among many others, that are not regulated). Through these guidelines, the EMA emphasises the direct link it draws between verification and validation of methods and the need to qualify them.

The EMA also supports stakeholders through a guidance document on the qualification of new methods for the development of medicinal products[49]. As mentioned above, this voluntary approach is based on a CHMP scientific advice on protocols and methods designed to develop innovative approaches.

At the EMA, the Paediatric Committee is particularly involved in these considerations, as in silico methods are used in a number of areas, including paediatric pathology, rare diseases and cancer.

In addition, in July 2023, the EMA launched a call for consultation on a discussion paper on the use of Artificial Intelligence in the life cycle of medicinal products for human and veterinary use[50]. This document has been developed in collaboration between the EMA's Committee for Medicinal Products for Human Use (CHMP) and its Committee for Veterinary Medicinal Products (CVMP). This work is part of the joint initiatives of the HMA-EMA Big Data Steering Group (BDSG), which aims to develop the capacity of the European medicines regulatory network in the area of data-driven regulation. The public consultation process reflects a genuine desire to open a dialogue with the developers of algorithmic systems developers, academics and all regulators. The final reflection paper has been adopted in September 2024: EMA, Reflection paper on the use of Artificial Intelligence (AI) in the medicinal product lifecycle, EMA/CHMP/CVMP/83833/2023, 9 September 2024[51].

Between November 2024 and February 2025, the EMA launched a public consultation on guidelines relating to medicine development based on modelling and simulation (MIDD). The first draft, developed within the ICH framework, was published on 11 February 2026. It proposes a harmonised regulatory framework for these methods and associated evidence[52]. Since November 2024, the FDA has also been developing a similar project, reflecting a shared commitment to harmonisation between Europe and the United States. The guidelines set out the criteria for recognising evidence derived from digital methods, placing particular emphasis on the context of use, the impact of models on decision-making processes, and the analysis of risks and errors. They also provide practical tools, such as summary tables, and require the implementation of an analysis plan and a structured report of the results.

This initiative represents a significant step forward in regulating the use of simulations in research, an area which was previously lacking in operational standards. It remains to be seen whether these guidelines, due to come into force in July 2026, will meet stakeholders’ expectations and be effectively implemented.

References

[1] SEBBAN E. and BABINET G., Santé connectée: demain, tous médecins? Une révolution avec les professionnels de santé, pour les patients, ed. Hermann, Paris, 2015, 216p.

[2] CCNE, Report by the National Consultative Ethics Committee, "Numérique et santé: quels enjeux éthiques pour quelles régulations", 19/11/2018, p.6.

[3] F. ROUSSEAU C., CROZATIER C., A. DUVERLE D., BUYSE M., E. CARTER S., M. VOISIN E., BOISSEL F.-H., " A regulatory landscape shift for in silico clinical trials ", Regulatory rapporteur, Vol. 16, N°11, november 2019, pp. 18- 21.

[4] BAJARD A., CHABAUD S., CORNU C. and all, "An in silico approach helped to identify the best experimental design, population, and outcome for future randomized clinical trials", Journal of Clinical Epidemiology, published online 2015.

[5] EDITH, “Building the European Virtual Human Twin”, Deliverable 3.2 First draft of the VHT roadmap, Report, 31 July 2023. (https://zenodo.org/records/8200955).

[6] Avicenna Alliance, "Avicenna Alliance members case studies on in silico medicine", April 2020, p.8: Example of the Extrapolis project. [Available online: https://avicenna-alliance.com/files/user_upload/PDF/Avicenna-Alliance_Members_Case-Studies_04-20.pdf] (Consulted on 07/05/2020).

[7] HUNTER P. J., “The IUPS Physiome Project: a framework for computational physiology”, Progress in Biophysis and Molecular Biology, Vol. 85, Issues 2-3, July 2004, pp.551-569.

[8] GIRARD P., CUCHERAT M., "Simulation des essais cliniques dans le développement des médicaments", Thérapie, May-June 2004, n°3, pp.294.

[9] Leem, "Essais cliniques 2030", survey of French and international experts by IQVIA, 10 March 2022, 97 p.

[10] T. MUSUAMBA F., et al, "Verifying and validating Quantitative Systems Pharmacology and In Silico Models in Drug Development: Current Needs, Gaps, and Challenges", CPT Pharmacometrics & Systems Pharmacology, 2020, 3p.

[11] GIRARD P., CUCHERAT M., "Simulation des essais cliniques dans le développement des médicaments", Op. Cit.

[12] MORRISON Tina, "How simulation can transform regulatory pathways", FDA, 09/08/18, [Available online: https://www.fda.gov/science-research/about-science-research-fda/how-simulation-can-transform-regulatory-pathways].

[13] E-SANTE EUROBIOMED, "Les essais cliniques in silico : le numérique et la modélisation au service du médicament", Rubrique actualité, 29/03/2018, [Available online: https://www.eurobiomed.org/actualite/detail/News/les-essais-cliniques-in-silico-le-numerique-et-la-modelisation-au-service-du-medicament/], (Accessed on 11/04/19).

[14] MANOLIS E., T. MUSUAMBA F., E. KARLSSON K., "Regulatory Considerations for Building an In Silico Clinical Pharmacology Backbone by 2030", Op. Cit.

[15] AFFOLTER C., "Novadiscovery aims for more efficient R&D", AGEFI, 13/03/2020, p.3.

[16] Ibid.

[17] A. SENI D., " L'expérimentation in silico et les nouvelles technologies moléculaires en biopharmaceutique stratégies de R&D des firmes de l'industrie de la région de Montréal ", Gestion, 2003, vol.28, p. 88.

[18] BRIVES C., " L'individu dans un essai thérapeutique. Sur quelques aspects du devenir objet dans les expérimentations scientifiques", Revue d'anthropologie des connaissances, 2013, vol.6, n°3, pp. 653-675.

[19] InSilc, project initiated by the European Commission, official website: https://insilc.eu/, 28 November 2019. European Commission website, https://ec.europa.eu/digital-single-market/en/news/eu-funded-insilc-project-promises-reduce-size-and-duration-human-clinical-trials, 28 November 2019.

[20] M. STOCKES J., YANG K., SWANSON K., S. JAAKKOLA T., BARZILAY R., J. COLLINS J., "A Deep Learning Approach to Antibiotic Discovery", Op. Cit.

[21] BOUZOM F. et al, "Physiologically based pharmacokinetic (PBPK) modelling tools: how to ﬁt with our needs?", Biopharmaceutics & Drug Disposition, 33(2), 2012, p. 55-71

[22] KASTRISSIOS H, GIRARD P., "Protocol deviations and execution models. In: Ko HC", Duffull SB, editors. Simulation for designing clinical trials. New York, USA: Marcel Dekker, 2002

[23] BAJARD A., CHABAUD S., CORNU C. and all, "An in silico approach helped to identify the best experimental design, population, and outcome for future randomized clinical trials", Journal of Clinical Epidemiology, published online 2015

[24] MELLERIN M., " La complexité croissante des études cliniques : Analyse et adaptation des laboratoires pharmaceutiques ", Dir. TABOULET Florence, Thèse pour le diplôme d'état de docteur en pharmacie, Université Toulouse III Paul Sabatier, 8 December 2017, p.80

[25] BOUZOM F. et al, "Physiologically based pharmacokinetic (PBPK) modelling tools: how to ﬁt with our needs?", Op. Cit.

[26] F. ROUSSEAU C., CROZATIER C., A. DUVERLE D., BUYSE M., E. CARTER S., M. VOISIN E., BOISSEL F.-H., "A regulatory landscape shift for in silico clinical trials", Op. Cit.

[27] VEYRAT-FOLLET C., BRUNO R., OLIVARES R. et al. "Clinical trial simulation of docetaxel in patients with cancer as a tool for dosage optimization", Clinical Pharmacology and Therapeutics, n°68, 2000, p.677-687.

[28] HAS, "Evaluation of connected medical devices. Spécificités méthodologiques d'évaluation clinique d'un dispositif médical connecté (DMC)", Service évaluation des dispositifs, 29/01/2019.

[29] https://insilc.eu/about/

[30] NATHAN N., ABOU TAAM R., DELACOURT C., DESCHILDRE A., REIX P. et al, "A national internet-linked based database for pediatric interstitial lung diseases: the French network. Orphanet Journal of Rare Diseases, n°40, 15/06/12.

[31] BAJARD A., CHABAUD S., CORNU C. and all, "An in silico approach helped to identify the best experimental design, population, and outcome for future randomized clinical trials", Op. cit.

[32] EMA, ICH Topic E7, "Studies in support of special populations: Geriatrics. Questions and answers", EMA/CHMP/ICH/604661/2009, European Medicines Agency, London, 2010.

[33] Avicenna Alliance, "Avicenna Alliance members case studies on in silico medicine", April 2020, p.8: Example of the Extrapolis project. [Available online: https://avicenna-alliance.com/files/user_upload/PDF/Avicenna-Alliance_Members_Case-Studies_04-20.pdf] (Consulted on 07/05/2020).

[34] Regulation (EU) No 536/2014 of the European Parliament and of the Council of 16 April 2014 on clinical trials on medicinal products for human use and repealing Directive 2001/20/EC, O.J., L 158, 27 May 2014, Article 10.

[35] MADABUSHI R, BENJAMIN JM, GREWAL R et al, "The US Food and Drug Administration's model-informed drug development paired meeting pilot program: early experience and impact", Clinical Pharmacology & Therapeutics, n°106, 2019, pp. 74-78.

[36] E. MANOLIS, F. T. MUSUAMBA, K. E. KARLSSON, "Regulatory Considerations for Building an In Silico Clinical Pharmacology Backbone by 2030", Clinical Pharmacology & therapeutics, 2020, vol. 107, n°4, pp. 746-748 (747).

[37] https:/whitelabgx.com/

[38] https://www.eu-stands4pm.eu/

[39] BRUNAK S., BJERRE C., CATHERINE E., Ó CATHAOIR K., GOLEBIEWSKI M., KIRSCHNER M., KOCKUM I., MOSER H. AND WALTEMATH D., "Towards standardisation guidelines for in silico approaches in personalised medicine" Journal of Integrative Bioinformatics , vol. 17, no. 2-3, 2020.

[40] https://www.eu-stands4pm.eu/

[41] EMA, « Questions and answers: qualification of digital technology-based methodologies to support approval of medicinal products », Human Medicines Division, EMA/219860/2020, 1/06/2020: https://www.ema.europa.eu/en/documents/other/questions-and-answers-qualification-digital-technology-based-methodologies-support-approval-medicinal-products_en.pdf

[42] EMA, ‘Guideline on quality, non-clinical and clinical requirements for investigational advanced therapy medicinal products in clinical trials’, EMA/CAT/22473/2025, 20 January 2025, p.52.

[43] EMA, ‘EMA Regulatory Science to 2025. Strategic reflection’, December 2018, 60p: https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/ema-regulatory-science-2025-strategic-reflection_en.pdf

[44] EMA, ‘Future-proofing Qualification of Novel Methodologies (QoNM)’, Action plan, September 2024: https://www.ema.europa.eu/en/documents/other/future-proofing-qualification-novel-methodologies-qonm-action-plan_en.pdf

[45] MANOLIS E, MUSUAMBA FT, KARLSSON KE, ‘Regulatory Considerations for Building an In Silico Clinical Pharmacology Backbone by 2030’, Op. cit, pp. 746-748.

[46] EMA website, https://www.ema.europa.eu/en/committees/working-parties-other-groups/pdco-working-parties-other-groups 21 October 2019.

[47] MARTIS B. S., ROUSSEAU F. C. and all., “The potential of in silico approaches to streamline drug development”, Avicenna Alliance Position Paper, Avril 2023 ; GERIS L. and all., The role of artificial intelligence within in silico medicine, Avicenna Alliance and VPH Institute white paper, August 2022.

[48] EMA, "Guideline on the reporting of physiologically based pharmacokinetic (PBPK) modelling and simulation", EMA/CHMP/458101/2016, 13 December 2018, 16p. [Available online: https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-reporting-physiologically-based-pharmacokinetic-pbpk-modelling-simulation_en.pdf].

[49] EMA, Qualification of novel methodologies for drug development: guidance to applicants’, EMA/CHMP/SAWP/72894/2008, November 2014, 15.p: https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/qualification-novel-methodologies-drug-development-guidance-applicants_en.pdf

[50] EMA, "Reflection paper on the use of Artificial Intelligence (AI) in the medicinal product lifecycle", EMA/CHMP/CVMP/83833/2023, July 2023, 17p.

[51] EMA, “Reflection paper on the use of Artificial Intelligence (AI) in the medicinal product lifecycle”, EMA/CHMP/CVMP/83833/2023, September 2024.

[52] EMA, “ICH M15 Guideline on general principles for model-informed drug development. Step 5”, EMA/CHMP/ICH/4966426/2024, 9 February 2026.

Acknowledgements

Published: 27/04/2026

Authors:

Noémie Dubruel, PhD student in health law, Maurice Hauriou Institute, University of Toulouse Capitole & CERPOP, BIOETHICS team (Inserm UMR 1295) and University of Toulouse

Aurélie Mahalatchimy, EuroGCT WP4 Convenor, UMR 7318 DICE CERIC, Aix-Marseille University, CNRS, Aix-en-Provence, France

Help and support

FAQ's

Glossário

Escolha o seu idioma

Select language