An undeclared civil war is breaking out in biomedicine. On one side is precision medicine, with its emphasis on tailoring treatments to ever-narrower groups of patients. On the other side is population health, which emphasizes predominantly preventive interventions that have broad applications across populations.
Which vision will provide the most durable and efficient path to improved health for all?
Precision medicine is a merger of molecular genetics, the dominant vision in biology, and genomics, the expression of that vision in human health. Disregarding the “breakthrough” announcements that appear on a regular basis, the question of whether precision medicine will lead to better health for all remains an open one.
Advances in public health such as cleaner water, safer and more nutritious food, prevention of infectious disease, and the prevention and cessation of smoking had much to do with adding 20 years to the lifespan of the average American between 1900 and 1950. Advances in biomedicine, such as better diagnosis and treatment of cardiovascular disease and cancer, added another 10 years since then.
Enter molecular genetics and genomics. The launching pad for this new era was the Human Genome Project, led primarily by the National Institutes of Health at a projected cost of $3 billion. This massive project encouraged people in the genomics community to see themselves as transformational actors in all forms of medical research and to promise equally transformative benefits for health.
As President Bill Clinton said in a ceremony celebrating the completion of the Human Genome Project in 2000, it offered the hope that “genome science will revolutionize … the diagnosis, prevention, and treatment of most, if not all, human diseases.”
The development of genomics since 2000 has been accompanied by an enormous growth in resources devoted to research, training, technology development, and implementation. This new emphasis has required a substantial reallocation of resources that were once directed toward traditional biomedical disciplines. The NIH, with its nearly $42 billion budget for 2020, currently invests approximately half its resources in genomics-related research, but less and less on research into prevention and public health.
That the NIH is bullish on genomics and precision medicine is clear from a 2018 tweet by its director, Francis S. Collins: “We would expect to see more effective prevention of many diseases, fewer diagnoses of serious illness, and an extension in health span.”
Great question! We would expect to see more effective prevention of many diseases, fewer diagnoses of serious illness, and an extension in health span. #NIH #JoinAllofUs https://t.co/pDbnju6ZTN
— Francis S. Collins (@NIHDirector) May 8, 2018
Offering personalized medicine as the foundation for population-based medicine of the future, as well as the ongoing and proposed investments in it, means the time has come for open debate to achieve balance and consensus on the fundamental priorities that will drive further gains in healthy lifespan in the United States and other countries.
We believe that genomics and precision medicine have ridden a wave of hype without substance for far too long. Unless they are able to go well beyond their thin record of empirical success and demonstrate their effectiveness in meeting the actual health needs of populations, they will be marginal players with regard to any lasting impact on the health of the public.
Fortunately, it appears that the tide is beginning to turn toward population health, especially as a more balanced perspective of the value of polygenic risk scores — one of the most widely advocated innovations of the precision medicine movement — is beginning to emerge.
Initial studies using genetic markers to predict the future risk of coronary heart disease received widespread attention and the marketed algorithms have been used by numerous medical centers. However, back to back studies in JAMA — conducted using improved statistical comparisons with standard risk factor profiles — demonstrate little or no value for these scores. An accompanying editorial provides further context for premature acceptance of polygenic risk score tools and reaffirms the need for population approaches to the prevention of cardiovascular disease, which have proven to be so successful over the last five decades.
Genomics has clearly led to major advances in many areas of biomedical research, and has myriad clinical applications, such as improved screening for familial diseases, enhanced diagnostic capabilities for infectious diseases, targeted therapy of immune-related conditions, and the capacity to provide guidance for greatly improved therapy for a subset of cancers.
But the “grand vision” accompanying projects like the NIH’s All of Us Research Program, which is currently enrolling 1 million volunteers with what Collins promised would be more “effective prevention of many diseases, fewer diagnoses of serious illness, and an extension in health span,” far exceeds what can be supported based on current evidence. Genomics, as Nobelist Harold Varmus once pointed out, “is a way to do science, not medicine.”
The genetic variants we inherit in our DNA have little impact on common diseases, and we need to draw a sharp dividing line between the majority of illnesses that influence population health — which are driven overwhelmingly by lifetime exposure to harmful air and water, poor nutrition, infectious agents, physical trauma, and psychological insults — and the minority that are driven by harmful genetic variants.
As we argued recently in the journal Issues in Science and Technology, choosing which road to follow to achieve significant improvements in health — like the 75% decline in cardiovascular mortality since 1968 and the 30% fall in cancer death rates since 1990 — is of enormous importance to public health. Although genomics has arrived as a new and exciting frontier of science, the opportunity costs for biomedical research if we continue down that road are enormous.
With a substantial chunk of the NIH’s budget devoted to genomics-related research, landmark trials like SPRINT, which demonstrated that lower treatment thresholds for blood pressure could reduce cardiovascular deaths by 110,000 per year in the U.S. alone, can no longer be funded because of the diversion of funds to genome sequencing projects like TOPMed.
Perhaps more important, an entire generation of young scientists who are being trained in genome-related laboratory and statistical disciplines are coming to believe that this form of biomedical investigation is uniquely important, to the detriment of other forms of research.
Although advocates of a more public health focused approach to biomedical research are beginning to make their voices heard, and the failures of the predictions of genomic enthusiasts are becoming increasingly evident, the biomedical research establishment continues to dig in its heels.
A course correction is urgently needed.
Richard Cooper, M.D., is a cardiovascular epidemiologist and chair of public health sciences at Loyola University Medical School. Nigel Paneth, M.D., is a professor of epidemiology and pediatrics at Michigan State University.
People are terribly ignorant of their bodies & what it is that maintains homeostasis. Teach someone to ride a horse and they’ll enjoy the ride; else, have trouble. Teach people to avoid disease, and they’ll avoid trouble. Education is highly lacking but very cost-effective.
Insightful article. Really like how you highlight that there are hyped expectations for genomics. However, I don’t think that means we should turn away from mechanism-based approaches to managing disease. It is very clearly that most of our definitions of disease are arbitrary and genes are one component of human biology. Molecular profiling with other ‘omics’ holds for the potential to identify aberrent pathways that are common to multiple diseases. What is most interesting is that advances in molecular profiling are making it easier to combine population health research with mechanism grounded science.
>>Advances in public health such as cleaner water, safer and more nutritious food, prevention of infectious disease, and the prevention and cessation of smoking had much to do with adding 20 years to the lifespan of the average American between 1900 and 1950.
And since 1950? What has been the effect of the USA’s policy on subsidized corn syrup? Allowing widespread use of pesticides that are banned in Europe for decades, looser rules on toxic VOC’s in workplace and homes, toxic plasticizers, the prohibition of healthy intoxicants like cannabis while alcohol is widely promoted. Water and air pollution standards much looser than Europe since 1990. Slow-walking regs on PCB’s, PFC’s, lead. What are the effects of those policy choices compared to Europe since WWII?
It seems we’re all subject to the very powerful groupthink of American exceptionalism. We all want to quote successes and omit the spectacular failures from mention because we’re Americans. The focus on precision medicine is of course due to our health care system’s profit-driven focus, where personal greed overwhelms concern for public health.
Most of the USA population is completely unaware that this country is in the grip of a cancer epidemic that is not present in rest of the world. We’d get better decision from policy-makers on investment priorities if we truthfully, and thoroughly, portray scientific reality to the public.
@Robert Where can I read more about the evidence for this epidemic and its causes?
@Richard Cooper
I am a young scientist- what do you think we should work on? Precision medicine is a natural consequence of the low-hanging fruit of mass applicability being plucked. If I could cure the most prevalent diseases of our time by screening soil samples as in the 1950s, I would do it. If not, I do the next best thing and work on precision tweaks. We know we can find something useful by applying the recipe of genomics. It seems you think we can find something better by doing something else. What else is there?
Thanks for the thoughtful question Alex. There is of course no short, simple answer. You might want to look at our piece in Issues in Science and Technology which provides a framework. The basic concern is whether we are concentrating to intensely on areas that will lead to niche advances (ie, driven by molecular technology) and neglecting those areas that have an impact on population health and medical care advances with broad applicability. Any individual scientist of course must combine the intellectual problem that interest him/her the most with the opportunities available to you, which are unfortunately often going to be the local academic environment where you find yourself. The challenge, as I see it however, is to also become adequately sophisticated in the structure of scientific enquiry and the fundamental lessons about how biomedicine advances that we don’t get side-tracked into a narrow technology driven agenda. I can speak from experience, because I have been down both roads. It is a profound challenge.
Thanks for the perspective Richard, I’ll give it a read.
The question of resource allocation is important but so is the science of molecular manipulation. When we speak of efficacy of cost to populations we shouldn’t constrain ourselves to current living beings. We need to think about the ramifications for future generations of both fauna and flora as well as human. A corrected or perfected gene can yield health results for all ensuing issue. These numbers of prevented diseased beings would, in short order, dwarf the numbers that exist today. Hopefully, we are not talking about a binary choice that has to be made. We need to serve both current and future need.
I agree – these are not “all or none” binary choices. It’s about balance. If the results from SPRINT were implemented widely we estimate there would be >100k deaths postponed in the US every year, and now estimates show prolongation of life expectance. (Circulation. 2017; 135:1617–1628. doi: 10.1161/CIRCULATIONAHA.116.025322). Precision medicine is primarily important for rare diseases and the magnitude of impact will be smaller, by definition. Of course there are many other applications of molecular technology (tracking cornoavirus . . ) but not massive cohort studies like All of Us that consume federal budgets.