On May 15, the President’s Council of Advisors on Science and Technology (PCAST) received an interim report on the President’s proposed Precision Medicine Initiative (PMI). Opening the meeting, PCAST co-chair and director of the White House Office of Science and Technology Policy (OSTP) John Holdren discussed the ongoing budget process and the challenges associated with the FY 2016 budget. Despite these challenges, Holdren noted, there are a number of exciting initiatives in progress, including the Precision Medicine Initiative.
PCAST co-chair Eric Lander reminded members that President Obama announced his Precision Medicine Initiative on January 30. It is designed to take advantage of the extraordinary opportunities that are available in medicine, including the ability to gather a large amount of data as the result of technology along with the opportunity to turn the health care system into a learning system. Lander noted that the President’s introduction of the Precision Medicine Initiative recognized the scientific opportunities across the health care spectrum with regard to common diseases that affect everyone. He pointed out, however, that there has been a considerable amount of work to follow up on the President’s broad direction.
Tania Simonecelli, OSTP, provided PCAST with an overview of PMI which is defined as “an emerging approach for disease prevention and treatment that takes into account people’s individual variation in genes, environment, and lifestyle.” She acknowledged that while PMI is not a new concept, most therapies available today are designed for an average patient. PMI is about moving away from the “one-size-fits all” approach, Simonecelli stated.
Simonecelli explained that the vision for PMI is to “build a broad research program to encourage creative approaches to precision medicine, test them rigorously, and, ultimately use them to build the evidence base needed to guide clinical practice.” The near-term goal is to apply the tenets of precision medicine to cancer, with a longer-term goal of generating the knowledge base necessary to move precision medicine into all areas of health and disease. She noted the three core components of PMI: (1) cancer, (2) creation of a national research cohort by the National Institutes of Health (NIH) and others, and (3) regulatory modernization.
PMI core activities, Simonecelli reported, include cross-government working groups on participant engagement, stakeholder outreach, data infrastructure, regulatory modernization, and privacy. She also noted the NIH Advisory Committee to the Director (ACD) Working Group which is scheduled to make recommendations on the national research cohort at the September ACD meeting. She highlighted the White House’s commitment to building privacy into the efforts of creating the one million person cohort and the goal to “establish a framework for protecting the privacy and security of participants’ information in PMI.”
Soon after the launch of PMI in January, OSTP organized a series roundtables with bioethicists and consumer and privacy experts to identify the key privacy-related opportunities and concerns that needed consideration in creating a national research cohort. It has since established an interagency working group to build on those preliminary discussions. Efforts include developing a set of privacy principles to guide the development of the PMI cohort, identifying gaps and other issues in current privacy and security protections, and proposing policy changes to address those gaps. Simonecelli concluded by noting that OSTP is appreciative of PCAST’s efforts in a number of areas and found its reports, including its big data and privacy report, helpful in jumpstarting the process.
Richard Lifton, Yale University School of Medicine and co-chair of the NIH’s ACD Precision Medicine Initiative Working Group, provided a glimpse of the working group’s activities. Lifton began by highlighting the lack of biomarkers for diseases such as Alzheimer’s disease, diabetes, and inflammatory diseases. As a consequence, current medical practice requires waiting until people actually develop a disease before intervening and hoping to prevent further progression of the disease. This is one of the opportunities in a national cohort study: to identify new biomarkers with the expectation of being able to identify patients and with a better understanding of the causation and pathways of the disease to intervene earlier.
The idea is to develop a national research cohort of about a million individuals, said Lifton, noting that there is a set of new technologies available today that were not available ten years ago. The advent of electronic health records adds a relatively automatic way to capture the totality of health care exposure and involvement over long periods of time.
Lifton noted that a million person cohort study offers spectacular opportunities, but there are also challenges associated with creating such a cohort, including answering questions like who should be included and how early to start– which dictates how long one is willing to wait. Other questions to consider: Should the pediatric population be included given that we know that adult diseases begin early in life? How do we ensure sufficient representation of the broad swath of the population so that the study allows for measuring the variation across the population, not only by ethnicity but also geographically?
The “how” is also important, said Lifton. Questions that need answering include: Should we start with existing cohorts that were ascertained in the pre-electronic medical record era versus starting anew with a closed system such as the Veteran’s Administration (VA) health system? What to measure in everybody versus what to measure only in a fraction of people? In addition to a longitudinal cohort, should the cohort ascertain some people with incident disease, in hopes of learning something more quickly? There are also questions regarding privacy and how to make the data available to wide body of researchers.
Lifton explained that the process for assembling the national research cohort includes bringing together groups of experts. He reported that the working group has convened a series of meetings that have examined such issues as cohort design, what to measure, how to capture electronic health records (EHRs), and how to make sure that the EHR data is actually usable for the kinds of research that we want. EHRs were built for ensuring billing compliance and not built as a health record, per se, Lifton explained. The goal is for the working group to report out to the NIH ACD in September.
Todd Golub, MIT and Harvard University and chair of National Cancer Institute’s Board of Scientific Advisors, noted that with regards to cancer precision medicine, the emphasis is on optimizing treatment. He cited three significant barriers to the cancer precision medicine vision:
(1) The need to learn from the world’s clinical medicine experiences by aggregating the data across many patients to make the connections. Currently, there is no mechanism to do that, Golub explained. It will require building the infrastructure and working on the culture of sharing patients’ information that respects the patient privacy.
(2) The need to discover all of the cancer dependencies and relate them to the genome. Tumors have vulnerabilities related to the cancer genome; we need to know what those are. There is a need to create a complete dependency map which would allow doctors to infer therapeutic approaches.
(3) The need for a healthy drug discovery drug delivery ecosystem, a therapeutic match of needs and drugs.
Golub concluded that he was optimistic and that we clearly should not wait until all these problems are solved before moving forward.