OR WAIT 15 SECS
Is the much anticipated Europe action plan on personalized medicines, unveiled recently by an EU-sponsored international consortium, worth paying a close eye on?
You’ve waited long enough for this … the European Union (EU)-led attempt to move decisively ahead at the international level on personalized medicine has just published its action plan. Now decide if it was worth the wait.
The wait has lasted for more than a decade. The action plan, released in late March by the ICPerMed consortium, notes “previous initiatives in personalized medicine at EU and national level” going back to before 2007, and even including a 2015 agenda calling for “a different approach” to research. That emphasized the need for collaboration “across the entire healthcare value chain,” and listed some of the desiderata: research cooperation with regulators, healthcare providers, policymakers, ethical, legal and social experts and patients; extending research funding to communication, training, outreach and non-research cross-sectoral projects; and new transnational networking structures.
What this EU-sponsored international consortium presents us with this spring is an action plan-but not yet any action. And in line with the plan’s own acknowledgement that plenty of work has already been done in identifying the challenges of personalized medicine, what we get is a lot of confirmation of what was already agreed in 2015-and, indeed, what has been also been repeatedly pointed out by other organizations working in this field over recent years. “In order to develop and use the full potential of personalized medicine, truly interdisciplinary research and inter-sectoral collaboration is needed,” says the 2017 action plan. “Knowledge, research activities and best practices need to be shared and where appropriate, aligned.”
The ICPerMed plan goes on to offer the now-familiar lists of areas requiring attention: generation of best available evidence, attention to ethical aspects, information for citizens and patients, healthcare professionals and providers, payers and researchers. It calls for comprehensive, validated, accessible and interoperable datasets, harmonized and with quality data that also takes account of data protection and ownership. And it genuflects to the need for “patient-centred, strategic approaches” that are nonetheless cost-effective and capable of “ensuring the sustainability of national health systems.” It also sets out the requirements for an effective context for personalized medicine: infrastructures, resources and frameworks for education and training, and empowering citizens to manage their own datasets.
So what is there new in all of this?
In concrete terms, there is an attempt to set some specific goals, and some indicative timelines for achieving them. Data features prominently in the most urgent needs. Two immediate objectives-defined as needing action over the next two to four years-include research projects at the European level to ensure the quality, completeness, validity and analysis of datasets, and studies on data integration and interpretation of multifactorial diseases-work which should be shared at European, regional and national levels.
Up to seven years is the medium-term timeframe set for other actions, such as support for research on enabling the extraction of structured data from unstructured sources, pilot projects to assess the impact of sharing data for researchers and others, research projects to optimize data security, privacy and ownership within personalized medicine approaches, and research projects to develop innovative decision support tools for healthcare providers. It isn’t all about data, either. The same medium-term timeframe is indicated for development and implementation of high-throughput preclinical models, and translational programs with shared access to genetically defined patient populations.
Seven years is seen as a reasonable window for integrated research into clinical validation of pharmacogenomics, research into "adequate regulatory structures and pathways in personalized medicine" and clinical trials methodology, as well as longitudinal cohort studies of disease outcomes. A similarly soft deadline applies to action to engage patients, to conduct research into postmarketing surveillance methodologies for accessing patient outcomes, developing tools to enhance digital health literacy, and sharing best practices in conducting and evaluating the impact of public and patient engagement.
Work should start immediately on some longer-term infrastructure preparations, such as "development and maintenance of biobanks and population/disease cohorts," but less urgency applies to "supporting interdisciplinary research in challenges and drivers that influence bringing PM innovation to the market, from ethical, legal, and societal perspectives."
Another area where ICPerMed claims to be innovative is in ordering the often fragmented approaches to personalized medicine. It is, it boasts, “the first time that such an impressive number of research funders and policymakers have identified and published a consensus view on the R&D needed in order to further advance personalized medicine.” It certainly meets its claim on numbers, because it does bring together 30 European and global research and health ministries and funding agencies, including such big hitters as the UK’s Medical Research Council, France’s Agence Nationale de la Recherche, and Germany’s Federal Ministry of Education and Research.
The plan also offers what it calls a chance for "enhanced coordination and alignment of research activities," with its "work program for ICPerMed member organizations," and engagement in a European research area network, which it aims to have up and running by the end of 2017. The plan talks of having created "a blueprint for a coordinated approach for personalized medicine research and reasonable implementation."
Its "blueprint" does contain some view of how detailed interlocking of many distinct strands could promote progress. For instance, data harmonization requires a prior definition of minimal datasets for clinical and general population databases, since "only then can all available and relevant datasets such as from medical/clinical, lifestyle and other sources be utilized." Similarly, "it would be of great benefit if existing information, for example, within text documents, could be edited in a way that would ensure that these sources could also be utilized." Or it proposes a three-stage approach to integrating research for clinical validation of pharmacogenomics: "promoting clinical trials of new diagnostic systems; innovative technological platforms for clinical validation and standardization, and dissemination in order to facilitate the implementation in ordinary clinical practice."
The plan also proposes a three-level process for standardizing clinical trial designs to demonstrate the feasibility of personalized medicine: new clinical trial design methodology and guidelines for approval and reimbursement, clinical and cohort pilot studies and new stratification strategies based on biomarkers, and combination trials on companion diagnostics and therapeutics.
Just another blueprint?
It would be churlish to dismiss the work that ICPerMed has put into developing this action plan-and I declare a marginal interest, in that I moderated the closing session of the Personalized Medicine Conference in 2016 that contributed to the process. ICPerMed makes much of the collaboration this has entailed. "Reaching this shared understanding was a worthwhile process in itself," says the action plan. But credibility for this exercise will depend on a lot more than "worthwhile" collaboration. To be taken seriously, the effort will need to show that it is delivering on some of its goals in a way that has not been done before. Otherwise, the risk is that it will come to be seen as just one more in a long series of European talking shops that have talked and talked, but done nothing.
Peter O'Donnell is a freelance journalist who specializes in European health affairs and is based in Brussels, Belgium