Event report - DIA Europe 2021: Advancing health priorities

Each year DIA Europe brings together hundreds of life science professionals from all parts of the drug development life-cycle. 2020 may have changed the setting of the conference, however it did not change the focus on the health of patients world-wide.

I attended this year’s virtual conference and primarily focussed on attending the sessions around pharmacovigilance. As well as COVID-19, Intelligent Automation (IA), Machine Learning (ML), Robotic Process Automation (RPA), and Artificial Intelligence (AI) were prominent players in the pharmacovigilance discussion topics.

Although we missed out on the in-person interaction, I have to credit DIA for creating a very user-friendly and interactive virtual conference setting with jam-packed content.

Below is my takeaway of some of the insights shared during the following sessions:

• Vaccination: Exploring risk communication needs.
• Globalisation of pharmacovigilance and Impact of the COVID-19 pandemic.
• Poster Session – Safety & pharmacovigilance.
• Changes in PV and the impact of EU legislation on patient safety.
• Advances in evidence-based quality and risk management for pharmacovigilance decisions making.
• Strategies to implement routine Intelligent Automation technologies within pharmacovigilance for the ICSR process.
• A digital channel pharmacovigilance requirements database with integrated Robotic Process Automation (RPA).
• The use of Artificial Intelligence/ Machine Learning in pharmacovigilance.

The Impact of COVID-19 on pharmacovigilance

We are currently in novel territory as safety signals are scrutinised on a daily basis by the media and Regulatory Authorities. In March, we saw 13 EU countries suspend the AstraZeneca vaccine over blood clot fears even after UK’s Medicines and Healthcare products Regulatory Agency (MHRA), European Medicines Agency (EMA) and the World Health Organization (WHO) assured no link between the doses and clotting had been established. A reported adverse event does not mean that the adverse event is attributed to the vaccine. Causality assessments need to be further reviewed before we can attribute blood clots to the vaccine.

Effective and tailor-made risk communication of vaccines is needed especially in the post COVID-19 era. Transparent, user-friendly and clear communication in simple language must be provided to the wider public. Social media has been widely used and we have seen global regulatory authorities create a reach with the general public through videos and social media content.

There has also been public scepticism around the speed of bringing the vaccine to market. It is important to understand that Pharma, Biotechs, CROs, and Regulators have a very good understanding of the safety profile despite overlapping clinical trial steps across Phases 1 - 3. It is also important to understand that in the current climate, we cannot have trials on-going for 2 years or more, and that it is in the public's interest to have vaccines available as soon as possible. The critical review process of the safety of medicinal products and vaccines, to ensure continual monitoring of safety of vaccine post-approval, is very robust and has been in place for a long time.

The future of pharmacovigilance

The pandemic has both strained the pharmacovigilance system and forced innovation. Changes to pharmacovigilance include electronic reporting of ICSRs and other safety documents (PSURs/ RMPs etc.), virtual audits and inspections and global collaboration in safety monitoring and assessment.

Over the past few decades, we have seen the evolution of formal and detailed pharmacovigilance requirements such as the the introduction of the PV legislation in the EU in 2012 to strengthen the handling of drug safety by the by industry, authorities, and health care professionals.

To continue strengthening the world of pharmacovigilance, we should look to elevate the pharmacovigilance profession as there is currently limited availability of talent and this is a major setback for future development. Opportunity areas include investing in continuous training, education and awareness in pharmacovigilance across healthcare professionals, healthcare scientists and students, those working in the pharmaceutical industry and also the general public. We also need to look at enhanced partnerships across health systems and the scientific community, and a possible global roll-out of a scheme similar to the successful MHRA UK yellow card scheme.

Intelligent automation is also a positive way forward for pharmacovigilance

Rule-based static systems are in place across pharmacovigilance and we are now seeing Intelligent Automation-based static systems emerging. Intelligent Automation-based dynamic systems are a likely eventuality and these will probably be Intelligent Automation-based static systems which are continuously updated based on new source data.

Machine Learning is here, has been for a while and is a developing area, both for industry and regulators. The industry needs to trust in outputs: Robust, reliability and reasoning; and should also have realistic expectations of machine-learning outputs. Suggestions around co-creation with regulators and alignment between regulators is the best path forward.

Intelligent automation solutions have many areas of application in pharmacovigilance and require a combination of factors to be successful. There is an importance of these tools fitting into existing workflows and culture.

Examples include:

• Auto-translation systems of source documents.
• Entry of individual adverse event reports in to the safety database - Model based on machine learning for causality assessment of ICSRs.
• Identification of Safety Signals.
• A digital channel pharmacovigilance requirements database with integrated Robotic Process Automation (RPA). This is a solution designed by Roche to screen websites for potential ADR case reports.
• As per ICH E2D, MAHs should routinely screen websites under their responsibility or management for potential ADR case reports.

Dominic Nguyen-Van-Tam (Medicines and Healthcare Products Regulatory Agency, United Kingdom) spoke about Intelligent Automation (AI) in pharmacovigilance case processing – An MHRA pharmacovigilance inspector’s perspective. As part of the AI Focus Group Outreach Program, the MHRA held discussions with the majority of global pharmaceuticals companies and large AI vendors. The exciting news is that the vast majority of AI activity seems to be taking place in the area of pharmacovigilance. The MHRA expects to encounter AI systems in pharmacovigilance inspections in the context of case processing and possibly signal detection. Pharmacovigilance inspections will evolve to encompass inspection of AI systems. The MHRA has produced a 2nd draft for GPvP AI in post-marketing which they are aiming to finalise within the next few months.

All in all, these DIA Europe 2021 sessions provided a great insight into various on-going and planned global pharmacovigilance related activities. It is mind-blowing at how far pharmacovigilance has come in the past few decades. With Intelligent Automation, I look forward to where this industry takes us over the next 20 years.