Background

A pan-European Expert Forum to tackle the complexity of big data integration for in silico methodologies in personalised medicine

On January 1^st 2019 the Horizon2020 Coordinating and Support Action “EU-STANDS4PM - A European standardization framework for data integration and data-driven in silico models for personalised medicine” launched activities with support by the European Commission Directorate-General for Research and Innovation. During the next three years EU-STANDS4PM will initiate an EU-wide mapping process to assess and evaluate strategies for data-driven in silico modelling approaches. A central goal is to develop harmonised transnational standards, recommendations and guidelines that allow a broad application of predictive in silico methodologies in personalised medicine across Europe.

Contribution to the sustainability of health research by using the power of health data

It has already been demonstrated that the analysis and interpretation of Big Data holds great potential for a deeper understanding of the mechanisms that lead to a transition from wellness to disease (Butte, 2017). EU-STANDS4PM will thus promote a broad and safe exploitation of health data in collaborative research and relevant stakeholder communities to enable better public health benefits through predictive in silico approaches in personalised medicine.

Harmonisation of health data integration strategies and data-driven in silico models in Europe

During the last decade in the related field of systems biology many community based standards have been developed to describe biocomplexity (of diseases) with the help of mathematical and computer-based in silico approaches (Stanford et al., 2015). This led to a high degree of fragmentation and many specialized community standards that are very often not compatible. EU-STANDS4PM will counterbalance the high degree of fragmentation of the many specialized communities that now seek transition into more applied fields —such as systems and personalised medicine— through a clear focus on reproducibility, interoperability and harmonisation on a European level.

EU-STANDS4PM is an open network and will lead to an increased flow of information about hurdles, bottlenecks, and expectations but also about best practices, optimisation strategies and success stories across disciplines on a broader European level. In the long term, this will have a positive impact on the actual clinical implementation of in silico models as well as data-driven health analysis strategies and lead to a more pro-active/predictive/personalised medicine in Europe.

In silico methodologies applied in personalised medicine

Despite the ever progressing technological advances in producing data, the exploitation of Big Data information to generate new knowledge for medical benefits, while guaranteeing data privacy and security (OECD-Council, 2017), is lacking behind its full potential. A reason for this obstacle is the inherent heterogeneity of Big Data and the lack of broadly accepted standards that allow interoperable integration of heterogeneous health data to perform analysis and interpretation for predictive in silico modelling approaches in health research such as personalised medicine. Further obstacles are legal and ethical issues surrounding the use of personal data.

To overcome these obstacles, EU-STANDS4PM will establish a pan-European expert forum with two main objectives:

1. To assess and evaluate national standardization strategies for interoperable health data integration as well as data-driven in silico modelling approaches.
2. To harmonise and develop universal (cross-border) standards as well as recommendations for in silico methodologies applied in personalised medicine approaches. 

By harmonizing and standardizing data-driven approaches across multiple disciplines and stakeholders, EU-STANDS4PM will ease an accelerated use of health data in clinical research and practice and will contribute to unfold the potential of predictive in silico models in personalised medicine.

Legal and ethical issues surrunding the use of Big Data

Big Data which will form the basis for in silico modelling are potentially vast in scope and both complex and heterogeneous in nature. There will be a significant variety in the content and the geographical and institutional origin of these data, as well as in the manner in which they will be processed. Consequently, the legal and ethical issues which might arise in the use of these data will be correspondingly diverse. While it may not be possible to provide quidance for every possible scenario, a focus can be put on the likely types of information contained within the data, the sources of the data, and the ways in which the data will need to be used. This will in turn allow identification of the main applicable laws, and ethical concerns for the various following data sources and data protection issues as well as issues related to research ethics regulation and patients’ rights. EU-STANDS4PM will to provide guidance on these legal, ethical and policy considerations arising from the use of in silico modelling for personalised medicine.

Key barriers relating to data access, dissemination and governance

The current diversity in administrative, ethical and legal requirements has made it unnecessarily complex and, in some cases, impossible to carry out integrative analysis across identifiable data, such as sequencing-based data, which is essential for personalised medicine. EU-STANDS4PM will address this challenge through  an innovative framework for data access and governance. Through an assessment of Dynamic Consent, harmonised Data Access Agreements, registered and cloud-based access (Budin-Ljosne et al., 2017; Dyke et al., 2016; Onankunju, 2013) EU-STANDS4PM will create a consciousness among the consortium and its extended European stakeholder network to develop codes of conduct for conditions under which not only non-personal and reliable data can be made securely available across borders but also conditions under which personal data (properly protected or with low sensitivity, low risk and non-stigmatizing health-related) can be used to facilitate data discovery.