Lack of advanced Models
Predictive models are in many cases not advanced and robust enough to be used in the reduction, refinement and/or replacement of the clinical trials required for the certification of new medical products.
Better models: develop more advanced models for the evaluation of safety and efficacy for both medicinal and medical device products that address specific Contexts of Use (CoUs), but at the same time enable a completely new class of models of disease with a similar pathogenesis.
Lack of independent validation collections
We need widely available collections of curated medical data specifically designed for the development and the validation of In Silico Trials technologies.
Validation collections: define requirements for data used as inputs of In Silico Trials models; analyse contexts of use and define requirements for validation collections; organise collections of existing data; support the formation of prospective collections.
Poorly informated stakeholders
It is crucial to give clear and correct information to all stakeholders (policy makers, patients, doctors, regulators, healthcare payers, Contract Research Organisations (CRO), technology providers, executives of biomedical companies, etc.) on the opportunities and the risks associated with the use of In Silico Trials technologies. Each stakeholder will thus benefit of a rigorous risk-benefit evaluation framework to assess the adoption of In Silico Trials technologies.
Responsible Research & Innovation: policies-watch; risk-benefit and cost-benefit analyses for healthcare professionals; monitor perception of IST among senior management of medical industries; promote citizen trust in IST.
No clear regulatory pathways
The credibility assessment proposed in the ASME VV-40 technical standard is limited to models used to assess medical devices. Pharmaceutical products, Advanced Therapy Medicinal Products, and Combinatory products do not have yet an equivalent standard. Also, the VV-40 is not harmonized in the CEN standardization framework, which makes it unclear on whether it can be used in relation to CE marking submissions. “First in kind” qualifications of In Silico Trials methods for the evaluation of drugs with EMA are proceeding quite slowly. Also, we need to explore new regulatory pathways for class III medical devices and pharmaceutical products that allow a significant reduction of human experimentation with In Silico Trials technologies, and the adoption of a rigorous post-marketing surveillance that monitors the validity of the in silico predictions as the new product is adopted.
Technical standards and regulatory barriers: develop good regulatory practices for the qualification of In Silico Trial models to assess medical devices and medicinal products; pursue first in kind qualification of existing solutions.
Poor scalability and efficiency
At the moment, most In Silico Trials models have insufficient scalability, as long as the lack of efficient computing of large-scale population models in easy-to-use, safe, and trusted environments.
Scalability and efficient computing: improve efficiency (cost and duration) of IST simulations; improve scalability of existing IST solutions, to handle large-scale simulations (from sensitivity analysis to populations); from big data to big simulations: user interfaces to supervise thousands of simulation results; implications of the General Data Protection Regulation 2016/679 (GDPR) for large-scale IST simulations.
Lack of trained workforce
It is rare to find professionals that have the necessary technical skills on In Silico Trials to work in industry, regulatory agencies, research hospitals, etc., because of the lack of opportunities for training and re-training.
Re-training and education: re-training of technical and non-technical professionals on the use, opportunities and threats IST can provide; revision of the curricula of both technical and non-technical biomedical education to include IST elements.