Precision Medicine and Public Health: Improving Health Now While Generating New Knowledge for the Future

In a previous post, I commented on the importance of a public health perspective to ensure the success of the proposed precision medicine large national research cohort. Here I offer additional thoughts on the need to balance short term public health gains with long term knowledge generation from this effort.

There is no doubt that “the time is right” for a large precision medicine cohort of 1 million people or more due to advances in scientific knowledge, technology and computing, empowerment of patients and the general public, as well as existing resources of millions of people who already participate in ongoing cohort studies. While insights from this cohort would take years, if not decades to materialize, what are the opportunities for short term public health impact? At a recent NIH workshop, Dr. Francis Collins used examples of early success such as pharmacogenomics tests, new therapeutic targets for common diseases, insights into resilience and healthy aging, and new ways to evaluate m-Health technologies for chronic disease management. He also described a hypothetical 50-year-old woman with type 2 Diabetes with suboptimal glucose control in spite of using a common prescription drug. Suppose within the next two years, she becomes part of the national cohort. A sample of her DNA is analyzed and she agrees to track her glucose levels via a tiny implantable chip that sends wireless signals to her watch and the researchers’ computers. Using these data, she changes diet and her medicine dose schedule. Five years later, her doctor switches her to new drug based on improved molecular understanding of diabetes. Of course, in order to fulfill the vision of better diabetes control based on precision approaches, researchers would not only have to make new discoveries from this large cohort but would also need to conduct follow up studies including randomized trials to compare the relative clinical utility of interventions in diabetes control. At the current pace of research translation, it may be more realistic to expect a switch to improved diabetes therapy to occur much more slowly, perhaps over decades, a reminder of the translation bottlenecks that occur beyond “bench to bedside”.

I believe the proposed large national cohort would provide a unique opportunity for near term impact, by assessing and enhancing implementation of already proven interventions, especially if the cohort can adequately represent various population groups, including minorities and underserved populations. Moreover, since most members of the cohort will be healthy at the time of enrollment, we can combine discovery of new precision tools with implementation of what we already know now to prevent disease and save lives. Using the example of type 2 diabetes, we know that finding and enrolling people with pre-diabetes in a diabetes prevention program can help prevent the onset of disease. Millions of people with pre-diabetes in the United States can benefit from such interventions but do not know they have pre-diabetes. The proposed cohort can assess how to identify thousands of people with pre-diabetes and connect them with healthcare and lifestyle interventions. There are dozens of evidence-based prevention recommendations that we know work in clinical or community-based settings but are not optimally implemented in the general population. Follow up of a cohort of a million plus people would allow us to assess how to enhance the delivery and impact of proven prevention guidelines on population health.

Of course, a unique feature of this proposed large cohort study would be the whole genome sequencing of participants. While this would lead to numerous discoveries of new genome-based associations and possible interventions, it would take time to yield dividends. In the meantime, we have a real opportunity for near term genomic health impact by focusing on conditions for which evidence-based applications are available. We have created a three-tier classification schema of genomic applications based on the methods of evidence-based medicine. A growing list is available on the CDC website. Similarly, Berg et al proposed binning the human genome into buckets based on clinical validity and utility of genomic variants.  Tier 1 (bin 1) genes and their variants are those with sufficient evidence for clinical validity and clinical utility to provide meaningful and actionable information to consumers and providers. Tier 2 (bin 2) genes/variants are those with established evidence of validity but insufficient evidence of utility to support a recommendation for use. Tier 3 (bin 3) genes/variants are those with either sufficient evidence for a lack of utility or presence of clear risk for harms, or those with insufficient evidence for both validity and utility.

Using this approach, we can evaluate how to obtain immediate benefits for participants of the cohort and their families. Examples of tier 1 conditions for which genetic testing is recommended in healthy people at high risk (for example to due to family health history) include hereditary breast and ovarian cancer syndrome (BRCA mutations), ;, Lynch syndrome, associated with increased risk of colorectal cancer; and familial hypercholesterolemia, associated with increased risk of premature heart disease. An estimated 2 million people in the US have one of these conditions and most are not aware of their risk. Once identified, there are evidence-based interventions that can significantly reduce their risk. The challenge lies in how to identify these individuals in the population. This cohort of million or more people would be expected to include thousands of undiagnosed, unrecognized patients with these 3 disorders alone, who can be identified through genomic sequencing and clinical data. These individuals and their relatives can take advantages of interventions to reduce their risk—an immediate health impact from this study.

In addition to these 3 conditions, other potential targets are the genes recommended by the American College of Medical Genetics and Genomics to be included in reporting of results of clinical sequencing performed for other reasons (incidental findings) because of clinical actionability. Moreover, a growing number of pharmacogenomic traits with varying levels of evidence is available for dozens of currently used drugs. And not to mention the possibility of expansion of carrier testing for a wide variety of genetic disorders [PDF 1.31 MB] that could be used in making reproductive decisions. Thus, the proposed cohort provides an immediate opportunity to evaluate the use of genome information in disease prevention and health care services, and to conduct applied research involving communication and behavioral sciences, and outcome research, on patients, families, health care systems and communities.

In conclusion, a large precision medicine national cohort should allow us to learn in short order how best to implement evidence-based genomic and non-genomic applications in real world settings, and how to use technologies and enhanced connectivity to achieve early gains. A continued dialogue among the stakeholders will ensure balancing near term successes with the long term goal to “generate the knowledge base necessary to move precision medicine into virtually all areas of health and disease”.