An Audacious Endeavor: Inside the PMI Research Cohort
The Precision Medicine Initiative (PMI), announced by President Obama during the 2015 State of the Union Address, aims to make it easier to provide each patient with the treatment best suited to his or her specific circumstances. In order to do this, the underlying science requires the study of many, many people. The PMI research cohort has established a target of 1 million voluntary participants to help researchers better understand the factors that affect a person’s unique disease progression and optimal therapeutic options.
The National Institutes of Health (NIH) is convening a series of workshops with scientists, clinicians, government officials, and the public to help design that research program. The latest of these workshops, Digital Health Data in the Million-Person Precision Medicine Initiative Cohort, was held on May 27-28 at Vanderbilt University.
The agenda consisted of nine discussion sessions focusing on existing cohorts managed by healthcare systems and research consortia, the types of data that can/should be collected and the complexities of sharing those data across organizational boundaries, and – significantly – the types of participants that should constitute those million people. These volunteers – both healthy and sick – will provide biological samples for genomic analysis and other testing, and will also consent to have their medical histories, lifestyle data, and environmental exposures tracked over the life of the study (likely many years).
Some of the key issues that were discussed:
- Should the PMI cohort be built de novo or assembled from pre-existing research cohorts that have been built for other purposes? The pros and cons of both were discussed. A new cohort would result in significant start-up costs ($200-250 to enroll a single participant – multiplied by 1 million). However, preexisting cohorts were generally developed to address a specific set of questions or a specific segment of the population. Gluing several together for the PMI would likely result in compromises.
- Regardless of how the cohort is constructed, what is the correct level of diversity (race, ethnicity, socio-economic status, etc.), and how can it be maintained? All agreed that diversity was important, but there was also consensus that participants with low socio-economic status were harder to recruit and retain over the long-term. Special emphasis – and additional resources – would need to be focused on ensuring that these populations would be appropriately represented.
- Should one entity have central control over all the collected data and samples or should that responsibility be distributed throughout a network? Most participants favored a hybrid approach where a core data set would be curated by a central authority, while a network of organizations with more focused research goals would maintain the data most relevant to their missions.
- Should the PMI cohort even be called a “cohort”? Some argued that by strict epidemiological terms, what was being proposed was not a cohort but more of a repository of data and samples that could be used by many researchers to explore a diverse array of questions related to human health. (Very much in the same way that the Human Genome Project serves to power research across the globe.) The principal investigator and chief executive of the UK Biobank, which has similarities to the proposed PMI cohort, describes the biobank as a “resource.”
This project will not be easy, but it can be done. One of the most famous research cohorts, the Framingham Heart Study, started in 1948 and enrolled the third generation of participants in 2002. However, the perils of keeping a large research cohort on course were highlighted when NIH terminated the National Children’s Study in 2014 after 14 years of planning, a pilot study, and $1.2 billion spent. Designing, building, and maintaining an effective PMI research cohort will be a complex undertaking. Everyone at the Vanderbilt workshop was focused on how to maximize the likelihood of long-term success for the research agenda and, ultimately, patient health.
We recognized how important this initiative is in expanding the science of patient input. FasterCures will be working with NIH, patient groups, researchers, and others to insure that PMI launches successfully and has sufficient resources to support its long-term mission.