Will Precision Medicine Improve Public Health?

Will Precision Medicine Improve Public Health?

>> This meeting is being recorded. If you have any objections, you may disconnect
at this time. Thank you, and welcome to today's webinar. We ask that if you are not already on mute,
to stay on mute for the duration of today's session, as this will reduce background noise
for the recording. A brief word about logistics, then we'll be
off. Questions are encouraged, and can be submitted
at any time during today's presentation. But all questions will be answered after both
presentations have concluded. Please note that you can submit your questions
using the Q&A feature on the right-hand side of your screen. Please type your question in the provided
field and hit Submit. If you joined us today at NCI, please come
to a microphone to ask your question live in the room. You may join in the conversation on twitter
using hashtag NCI PMPH. And with that it is my pleasure to turn today's
session over to Dr. Debbie Winn. >> Welcome. We're delighted that you are here to listen
to our webinar on Precision Medicine and "Will Precision Medicine Improve Population Health?" This webinar is sponsored by NCI's Precision
Medicine and Population Sciences Interest Group. This is our inaugural webinar, so we hope
you'll stay tuned and listen to learn more in upcoming months about future events. We're extremely delighted to have two fantastic
speakers today. Dr. Galea will be speaking first; he's a physician
and an epidemiologist who is currently the Dean at the Boston University School of Public
Health. He also has been the Chair of the Department
of Epidemiology at Columbia University, and has held other academic and leadership positions
at University of Michigan and New York Academy of Medicine. He's particularly interested in the social
production of health of urban populations, and he's focused on causes of brain disorders,
especially mood/anxiety disorders and substance use. He's particularly interested in issues related
to mass trauma and conflict. Our second speaker is Dr. Muin Khoury; he's
the Director of the CDC Office of Public Health Genomics. He has split his time here with us here at
the National Cancer Institute for many years, divided his — he's been a busy man working
both at CDC and NCI. He's formally the Associate Director of our
Extramural Epidemiology and Genomics Research Program. And currently he's been very actively involved
in promoting a dialogue on precision medicine in public health. He has a PhD in human genetics and genetic
epidemiology, and an MD degree. He has published extensively in genetic epidemiology
and public health. And he's very interested in the translation
of science and research into public health settings. So both speakers have been extensively published
in the scientific literature. And their work is widely read and cited. They're both considered to be very influential
thinkers and speakers, so we're very, very pleased to have them here today. We expect a lively presentation and an exposition
of alternative viewpoints. So to remind you again, our topic today is
"Will Precision Medicine Improve Population Health?" And with that we will just get started. Thank you very much to Dr. Galea for starting
us off today. All yours. >> It's a pleasure to be here and to discuss
this with Muin and with the audience. So I was asked to tackle this question, and
I suppose I come at this from the point of view of, actually I don't know how to advance
slide; can I just say advance slide? Will somebody do that? Can we advance the slide? Hmm. I'm hoping that somebody will be advancing
the slide on the other end. I'm just going to pause until somebody tells
me. >> Yes, we're working on the technical details
of advancing the slides. Just — >> Okay, let me just pause, then. Ah, here we go. So I can just say "Advance slide," will that
work? >> That's fine. >> Okay. So I was asked to take a position as a — what
I'm calling the loyal opposition. I am coming from a point of view of someone
who's going to answer these questions as, I'm not so sure that precision medicine will
improve population health. Next slide. I have been loyal because I'm actually loyal
to the aspirations of science. I actually am excited about the potential
of precision medicine. I think there is a fair bit of mechanistic
insight that will emerge from this exercise. Next slide. But I'm opposed to compelling ideas that I
am not convinced will advance population health, at least in their current formulation. Next slide. So as far as I'm concerned, there's only one
question that matters. Will precision medicine approaches improve
population health? And the answer I'm going to suggest is, next
slide. No, unless. And I'll explain about the "unless" in a second. So let me start with trying to answer my no. And offer three reasons why not. Next slide. So first of all, the first of the reasons
is the challenges of complexity in biology. That a lot of precision medicine efforts are
predicated on the assumption that we shall be able to determine the pathogenesis of disease
so that we can intervene on particular genes, particular molecules which will make a tremendous
difference for individuals. The reality is that the biology is much more
complex than that. Next slide. This is from a paper that just came out recently
showing the population attributable fraction to genes for a wide variety of disorders. And this is really a best estimate summary
I can share. Everybody can see in any number of disorders,
really, the proportion of pathogenesis that is attributable to genes is in the 20 to 30%
range for the most part. That's just one simple illustration of why
genetic factors matter relatively little. Next. Next slide shows a — from the International
Consortium for Blood Pressure a summary of GWAS, genome-wide associational study, showing
that there are a large number of variants that have been identified that are linked
to systolic blood pressure. And to that systolic blood pressure, all of
them showing an effect of about 1 mmHg. Showing an enormous number of variants which
we don't understand how they work, all of which contributing to a fairly minor increase
in blood pressure, which is our target of interest. Next slide. And at the level of trials that have looked
at this — this is from the SHIVA trial, showing that when there was an attempt to use molecular-targeted
agents versus treatment of physician's choice, what you actually see here is the progression-free
survival is roughly the same between molecular-targeted agents and treatment at physician's choice. Which really gets at the heart of the matter
that at core, despite their efforts to molecularly target or genetically target, it's unclear
we're going to have much more gain over the choice of physicians. Next slide. A lot of this has been discussed on the topic
of cancer. And there's no question that for some somatic
cell tumors, precision medicine approaches are probably the most promising, and it's
probably more promising pathology. But even cancers of multi mechanism disease
with a fair bit of [INAUDIBLE] resistance that's predictable. Next slide. And at core, the — a lot of this argument
about the challenge of complexibility rests on the fact that while we are making all this
effort to target molecularly or target genetically, there is this overwhelming role of behavior. This is for men and women. And this looks at the association with cancer
and BMI. And showing this extraordinarily high relative
risk of death with BMI among men for liver cancer. Among women for uterine cancer; also women
around breast cancer. Men, colon-rectal cancer. Showing that much more easily detectable phenotypes
like BMI and behaviors ultimately are directly associated with pathology in a way that we
both understand. And that gives us much more certitude that
we will be able to do something about it. So that's point one. Number one is that we have enormous challenges
and complexity in biology. And I think many of our precision medicine
approaches are making claims that are enliving [phonetic] some of the challenges we are going
to have in getting to a place where these molecular-targeted or genetically-targeted
approaches will make a real difference. Next slide. The next point is that many precision medicine
approaches conflate individual and population. And that observations that have utility at
the population level do not necessarily have utility at the individual level. Next slide. This is a fairly standard and typical dose-response
relationship showing a genotype score. This was show, I think from a group of subjects
here at Boston University. And the likelihood of type 2 diabetes. And I show this slide because it's a fairly
compelling slide that in some respects gets at the very heart of the argument for precision
medicine. If you look at this you'll say, wow, look
at this genotype score. If I had the genotype score, would I have
a higher cumulative incidence of diabetes? That is the type of compelling narrative that
emerges from precision medicine approaches. The problem is that this is true at the population
level. When you get at the individual level — next
slide — What you see is that when you take the exact same data and map it out, genotype
score still on the X axis, and you look at the percent of subjects with and without diabetes,
which are the black and the gray lines, the two curves are fairly indistinguishable. Meaning that having a genotype score tells
you very little about what the likelihood is of you the individual having disease. Which means that a lot of the observations
from population level — which ultimately is what we're doing from large sample studies
that get at the molecular and genetic targeting — have relatively little utility at the individual
level. And in case you're wondering, when do these
findings have utility at the individual level? Next slide. As this slide shows, you need to be at the
bottom right of this figure. This is a mathematical simulation from [Inaudible]
California. Here we need an alteration of about 350 to
have the kind of individual predictive utility that we frequently assume is going to emerge
from precision medicine approaches. Now, here's the good news — or the bad news. The good new or the bad news is that there
is nothing much new in fact about this argument. In that we have much experience with individual
medical approaches that we think make a difference for population health. Remember, the question I'm asking is, will
these approaches improve population health; that is the question I am most interested
in. And that in fact they do not really improve
population health. And to take one very simple example, is the
extraordinary drop in infectious diseases over the course of the 20th Century. And everybody knows that the drop in infectious
diseases over 20th Century is one of the principal reasons for the prolongation of life expectancy. And as this slide shows, you see the drop
in infectious diseases, you see the spike for the influence of pandemic. But you also see where penicillin was introduced. And the introduction of penicillin, as you
can see there, had made barely a dent in the drop in infectious disease mortality. That does not mean that penicillin doesn't
matter. Doesn't mean that antibiotics are not very
important and they're not very important for the clinical care of patients who have infections. What it does mean, however, is that at the
population level there was many other factors that mattered to improve health in the context
of infectious diseases. And one last point on this. Next slide. This is a clinical prediction model, looking
at prediction of diabetes. And what you see here is a variety of lines
on this. The red line is clinical prediction model. And the orange, blue, and purple lines are
clinical prediction models adding multiple genetic loci. And what you see is exactly the same curve
for all along. Suggesting that what we know clinically, what
we can tell phenotypically, looking symptomatically, ultimately gets us results that are as useful
as molecularly-targeted or genetically-targeted approaches. Point three, next slide. Which is that much of the precision medicine
discussion also embeds within it the assumption that if we understand precision medicine,
if we understand precisely what our risk is, that we would behave differently, that populations
would behave differently. Unfortunately, data does not really bear this
out. There are several papers on this, but — next
slide. A recent review that just came out about the
impact of communicating genetic risks on diseases — risks of disease on risk-producing health
behavior. Next slide. Such that expectations communicating DNA based
risk estimates changes behavior is not supported by existing evidence. So those are three reasons why I am actually
skeptical, why I call myself the opposition on what I think is a very compelling idea
that has utility but I'm not so sure has direct utility in the short term for population health. But now, moving onto the next slide — Let
me tell you why I think this matters. I think this discussion is not simply an academic
discussion; it's not an academic discussion between Professor Khoury and myself. I actually think this matters for three important
reasons. Number one — next slide — Number one is
that this kind of discussion, the fact that we are investing a fair bit of our federal
research eggs into this particular basket is distracting us from issues that matter
more. And those issues that matter more are all
issues of population health. Everybody here knows in the United States
health indicators in the United States are nowhere near as good as many of our peer countries. Next slide. What is frequently not known, though — this
is from a National Academy of Medicine report — is that it wasn't always like this. The red dot is the U.S., the gray dots are
other high-income countries. And what you see is that while everybody has
gone up, the U.S. has over the past 35 years drifted inexorably and slowly towards the
very bottom of the pile. This is what's important. And this is what I worry, that distractions
like precision medicine will take away our eye off the prize. And the prize is improved population health. Another illustration is life expectancy, and
there's enormous widening of life expectancy we're seeing in the U.S. Next slide. This looks at life expectancy in the U.S.
for women and men looks roughly the same, dividing people by quintiles. And what you see is that life expectancy has
only increases for women for the richest 20%. In fact, it stayed the same for the middle
60%, and has decreased for the lowest 20%. This is something which most people, when
we first asked them what percent of people have not had an increase in life expectancy,
we tended to not to guess it's actually as high as 80%. And despite the fact that we as a country
have spent an extraordinary amount of time in the past seven years discussing the Affordable
Care Act and focusing on other efforts to improve clinical medical care in this country. Next slide. You see that the percent of adults 65 years
and older who have problems accessing health-care services remains substantially higher than
it is in other peer countries. So point one is that an effort on precision
medicine to the exclusion of all else misses the important questions in population health. Point two, next slide, is that it also makes
for missed resource allocation. And it unfortunately sets us in poor step
in terms of investing in the future. Next slide. I right now happen to live in the State of
Massachusetts. Which considers itself fairly forward-looking
on issues of health. I live in Boston where the health industry
is the largest industry in town. But when you look even at Massachusetts state
spending, you see that healthcare spending — or medical care spending, more accurately
— has gone up in the past 14 years. While spending in everything else — primary/secondary
education, law and public safety, mental health, public health, higher education, early childhood
education, environment and recreation — has consistently gone down. And this is the danger of focusing our attention
too much on efforts that are implicitly medical in their approach about molecular targeting
and genetic targeting, that they distract us from investing in the areas that are ultimately
going to improve the health of populations. And our federal funders nationally are helping,
the most important single health fund in the world, is not immune to this. Because — next slide — If you look at — this
is something you can do yourself from NIH reporter — proportion of NIH funding awarded
to projects with the terms "genetic" or "genetics" in the title, abstract or terms over the past
ten years. The blue line is actual, the red line is just
a fitted line. You see that that's gone up. Next slide. And conversely, proportion of NIGH funding
awarded to projects with the terms "population" or "public" in the title, abstract, or terms
has gone down in the past ten years. So — by the way, I will point out that the
first Y axis was about 30% to 37%, while this Y axis is 4% to about 0.4%, just to give you
sense of the scope. So the second challenge, the second reason
why I think this matters is because of research allocation. Number three, and I think the third reason
why this matters is that we are investing so much of our health capital in the broader
public discussion that the precision medicine approach is leading to hype over hope. And let me just show you a couple of examples
of where hype over hope and how we are repeating mistakes of the past. This slide says, "The time has come in America
when the same kind of concentrated effort that split the atom and took man to the moon
should be turned toward conquering the dread disease. Let us make a total national commitment to
achieve this goal." This was of course President Richard Nixon
declaring the war on cancer in 1970 — well, the National Cancer Act was 1971; he made
the statement in December of 1970. Well, how has the war on cancer done? The next slide looks about and shows that
the war on cancer in and of itself has really made nary a dent on cancer. So what you actually see is the decrease in
cancers largely due to stopping on smoking, having very little to do with the medical
approaches that were advocated. And in much of the approach we're seeing our
precision medicine and many of its ancillary offshoots are doing much the same thing. If you look at the next slide, this is President
Obama saying, "Last year, Vice President Biden said that with a new moonshot, America can
cure cancer. Let's make America the country that cures
cancer once and for all." These are very much repeating the same types
of efforts. But we are investing extraordinary resources
in medical, clinical approaches. While turning a blind eye to what is important,
which is improving population health. And this has consequences. It has consequences for how the health enterprise
is seen in the broader public. Leading us to one of my favorite cartoons,
which is next slide. Which has the news reporter reading today's
random medical news saying that coffee can cause depression in twins. But equally well he could have said that the
computer terminals can cause hypothermia in overweight smokers. And this is the type of roulette that we find
ourselves on as we focus ever more on molecular and genetic targeting at the expense of broader
efforts to improve population health. Next slide. So I said, "No, unless." So let me just talk about "unless," what do
I mean by unless? Well, I mean unless a precision medicine approach
or a precision population health approach — which I will talk about shortly — ultimately
focuses on populations. Unless we take a real hard look about how
this fits in with an agenda that is relentlessly pushing the question of how we improve the
health of many. Next slide. This is a population. Population is messy; it's interconnected;
has spatial interdependencies. People interact with one another and have
behavior changes that are reflected by context and by inter-individual relationships. Next slide. What we are doing with precision medicine
approaches is ultimately taking these populations and focusing ever more on smaller and smaller
subsets. Next slide. What we instead should be doing is we should
be looking at whole populations. We should be saying, these are diseases in
populations. Next slide. And we should be doing whatever we can to
improve the social/economic/cultural context that takes these individuals with pathology
in whole populations. Next slide. And reduces the number of people with pathology. So let me wrap up; I have two minutes left. I realize that in speaking as a loyal position
on the context of precision medicine population health, I run the risk a little bit of painting
myself into a corner; as being sort of the guy who doesn't like technology, who doesn't
like innovation, doesn't like advance. And I want to be clear that that's far from
the truth. I actually, I think there is a lot that can
be learned, a lot that can be gained from the type of molecular genetic targeting that
is going to emerge from precision medicine. My argument is not with the utility of these
approaches for mechanistic, biological, and medical work. My argument is with the substitution of an
intellectual and a funding agenda towards the improvement of population health which
ultimately is a medical agenda that will help the few. And, but just to end on a quote and this is
again, the counter to my argument will be well, but look; give this a chance. This is something that we should put our shoulders
to the heavy rock and push it and together. "I know this is a formidable technical task,
one that may not be accomplished before the end of this century. Yet, current technology has attained a level
of sophistication where it is reasonable for us to begin this effort. It will take years, probably decades, of efforts
on many fronts. There will be failures and setbacks just as
there will be successes and breakthroughs. And as we proceed we must remain constant. Isn't it worth every investment necessary? We know it is." And that's the kind of inspirational talk
that I think is reasonable to affix to something that is, you know, a little bit futuristic. Like, surely we should do this, surely we
should invest our money in it. Well — this quote is, by the way, President
Ronald Reagan when he announced the launch of Star Wars as the national defense missile
shield. And I just put it up there because I think
it's got the kind of inspirational language and an untested approach that ultimately ended
up not doing anybody any good. So, just to end. Next slide. I started out by calling myself the loyal
opposition. This is the two kings walking by, one of them
saying, "My local opposition wasn't loyal enough." I do — to go back to my point at the beginning
— I do think that I am loyal to the aspirations of science. I'm excited by the mechanistic and medical
potential of genetic molecular targeting. And much less excited about its potential
to improve population health. And I'll end with a medical — next slide. This is my goldfish — and I'd say it's a
goldfish that I care very much about. I would like the goldfish to be healthier,
to be happier. And I can do two things to make my goldfish
happier and healthier. I can make sure that my goldfish always eats
the little food I put in a measured way and doesn't eat too much, make sure that my goldfish
swims around in his bowl ten times clockwise, ten times counterclockwise on a regular basis. I can make sure that my goldfish has safe
goldfish sex. I can also make sure that I genify my goldfish
and understand fully the mechanisms of its cancer so I can fix it when it develops cancer. Or I could yes, do all that, but I could centrally
and importantly recognize that everything about my goldfish's health is determined by
the water the goldfish swims in. And unless I actually clean its water and
create the social, cultural, and economic environments that promote its health, everything
else I do is futile. Next slide. And that's it; I will stop there. Thank you. >> Thank you very much, Dr. Galea. I'll now turn the mike over to Dr. Muin Khoury. Thank you. >> Thank you very much, Debbie. And Sandro, you're a hard act to follow. My answer to the question, "Will precision
medicine improve population health?" You've answered the "but" in my "yes." But I do not think it's a distraction. There are a lot of challenges you've raised
around the complexity in the biology, the individual versus population focus, the behavioral
change issues, the hype and the hope. And I personally spending relatively a long
time between shuttling between NIH and CDC trying to kind of bridge the worlds of medicine
and public health. I have four themes I'd like to elaborate on. And some of you who are following this discussion
would see a lot of similarities and overlap between Sandro's ideas and mine. So the first theme is that in order to improve
population health, we have always needed medicine and public health. When you get sick, you want to have the best
available drugs and medications and interventions. And when you're not sick, you want to keep,
you want to prevent disease and promote health. And this easy/uneasy relationship between
medicine and public health has been going on for decades. And I know usually medicine gets all the glory. But there is a lot of accomplishments that
public health have made over the years. And as we move from medicine to precision
medicine, I think that same tenet applies. That as we begin to have molecularly targeted
treatments of cancer — and we have a few examples of those; there are some success
stories — we still need public health. So this sort of dichotomy or a distraction
saying that, you know, doing one at the expense of the other is because we always need both. And we can never forget that point. And the thing that I wanted to impress on
the group here is sort of the evolving definition of precision medicine. I'd like to end up with it at the end. Precision medicine is not just about genes. It's a new emerging approach that includes
both prevention and treatment. That also includes environment and lifestyle. So if we subscribe to the idea that precision
medicine is only about genetics, then Sandro's notion is very much fulfilled. So just to highlight the need for both population
and individual approaches, this is a paper from Jonathan Fielding from a few years ago. That highlights the need for both approaches
as you go from well to being sick in bed. And, you know, there are activities you do
at the individual level and the societal level. And all of these things need to be done. And they should not be done at the expense
of one another. A leader at CDC [Inaudible] has put up his
own version of this, sort of the health impact pyramid. The five levels of interventions at the population
level. Of course, if you work at the base of the
pyramid, provide opportunities to wipe out poverty and improve socioeconomic status,
you're going to improve health. Of course, that would have the largest impact
at the population level. At the top of the pyramid, going one-on-one
talking to people and clinical interventions will achieve lesser of an impact. So what am I saying here, am I agreeing with
Sandro? Yes, of course I'm agreeing with Sandro. But we cannot have this discussion and say
precision medicine versus public health. To me it's a false dichotomy — we need both. We need both in an era where medicine can
become more precise, as well as public health can do its work. And I think my ending argument is at the end
that public health will become more precise, as medicine will become more precise. So the second theme of my talk is that there
is a lot we can do now to implement what we already know. This is not a pie-in-the-sky. There is course a lot of hype. But there is a lot of hope. And I reflected that on the paper I wrote
last year with Jim Evans about balancing long-term knowledge generation, which is now geared
up with a one-million person cohort. With early health benefit. If you take a look at a cohort of a million
people, there are thousands of people that can benefit from interventions at the molecular
or other level that they may not currently be benefiting from. So melding intervention and implementation
science with discovery science is very important at this point. The CDC has been maintaining this simple tiered
classification of genomic tests — and let's stick with genomics for a little while. The tier ones are those that are applications
that are ready to be put in play. A few of the cancers, newborn screening which
affects all newborns in the United States. And all the way ranging to tier three, which
is not ready for prime time, as Sandro mentioned about personal genomic tests and direct-to-consumer. But there is a whole shades in between of
the tier two. And if you think about some of the tier one,
there are already many examples and millions of people that can be affected by these things. Two examples, hereditary breast and ovarian
cancer and Lynch syndrome. Yes, do affect a small fraction of all cancers. But that translates to thousands of preventable
cancers every year in the United States. And there are actions that public health can
do now to prevent certain cancers associated with these hereditary conditions. And a public health/healthcare partnership
will be needed. Because healthcare alone does not seem to
be ascertaining those people who need the services, genetic counseling, and some of
the interventions that currently work. And there are disparities. And of course if you are in public health,
we have to worry about disparities. Not only in the [Inaudible] research, but
implementation. And this is data from a very large data set
of 15 million insured people that shows the disparities of utilization of BRCA testing
in young women with breast cancer by race and ethnicity. So the third theme what I want to continue
on this journey is that public health is a partner in the development of precision medicine. And what we call public health sciences — I'll
tell you a few of them as we go along — are essential in ensuring the success of precision
medicine. Let me elaborate on that. So a few years ago, Francis Collins put out
this vision for the future of medicine in which he portrayed this 23-year-old man named
John who goes for his primary healthcare checkup. And he gets this fictitious printout of genes,
and then a personalized intervention strategy to mitigate the risk of these various diseases. But let me ask the question here, where do
you get these numbers to begin with? These numbers can be only gotten from large-scale
population studies. And as Sandro showed us, some of these current
kits are not that predictive. And I'll elaborate on that a little bit more. But the question is, what do you do with these
numbers when you get them? Because there is a whole lot of things that
can be done. So a case in point is this discussion about
mammography, which is sort of a population-wide screening effort that's been discussed at
the forefront. And all these changing strategies in how often
do we need them, at what age. And the various recommendations by different
groups and different countries. And if you think about these sort of not very
predictive genetic factors that Sandro showed us a bit earlier, they don't have to be too
predictive. But you can think about whether or not they
can be used in crossing a certain threshold for utilization. This is data from last year from JNCI that
shows on the basis of 80 snips, and now probably there's about 200 of them, the ten-year absolute
risk of developing breast cancer for women with and without family history by this polygenic
risks score that was mentioned earlier. And depending on your cutoff and screening,
you might decide to use the risk score. Or at least in the informed shared decision-making. Because if you see a certain cutoff, sometimes
people cross that threshold at a younger age group if they have a higher genetic load. And it doesn't have to be as predictive of
the phenotype that we have in mind. So the question is, the public health sciences
are eminently needed to translate this activity of what's going on with precision medicine
discoveries. And Sandro showed us a slide that showed that
we don't have enough of these population sciences here in the division of cancer control and
population sciences, about 30% of our portfolio is in population sciences and precision medicine. Much of it is in epidemiology, but some of
the other areas are growing. The literature actually shows a deficit of
this public health genomic science beyond bench to bedside. About 1% of published genomic research is
beyond the initial discoveries. And half of it obviously is in cancer because
it's a hot area of discovery right now. So I want to close with Theme number four,
which if you remember how we defined precision medicine at the beginning. I want to posit to the group here and have
some maybe dialog and discussion afterwards, as medicine moves into a precision medicine
territory, I submit to you that public health is moving in that same space towards an era
of precision public health. And let me explain that a bit more. Yes, we know that there is complexity in biology. But there is more than biology involved. Health and health outcomes are determined
by multiple levels, ranging from the molecules all the way to environmental and socioeconomic
factors, the exosome in big ways, and all this technology is going to allow us to measure
these things a little bit more than we've had before. So it's not just about genes. I've showed you examples of a few single gene
disorders earlier that account for about 5 to 10% of the human disease. But our complexities which are major killers
— heart disease, cancer, and diabetes, that account for the majority of human diseases
— hopefully can be approached with this melding of all the exposures that come together. So it's not about our susceptibility any more,
it's about various types of exposures and the genomes that we come across, including
the microbiome that there is more of them than us in our own body, if you will. There is that complex interactions that we're
just beginning to skim the surface. And there are a few success stories, but tremendous
challenges. A promising area is epigenetics. Which is more than just about the gene sequence. It really, it's the ultimate gene-environment
interaction. Because it uses the life course approach to
look at the impact of environmental factors on gene expression across multiple generations. And the promise for that is yet to be explored. But cannot be over — sort of overruled, in
a way. So I wanted just to end with a couple of these
slides. Because we get stuck on words. And we have been using a lot of words. And sometimes the words have a purpose of
their own. Precision medicine has had its roots in the
words "personalized medicine." Which means that we apply things at the individual
level. If you look at this Google term search, the
blue shows how the trends and utilization of "precision medicine" has taken off after
the announcement of a precision medicine initiative in 2015. But there is more to that. Because the 2011 Precision Medicine National
Research Council actually outlined a case for precision medicine beyond utilization
of these technologies at the individual or personal level. We know that there are lots of individual
information that are very imprecise, like some of the data that Sandro showed us. That the active genetic testing movement is
a very imprecise movement. And while it's really highly personal, it
doesn't advance the cause of public health. Whereas sometimes you have precision medicine
discoveries that combine biological and environmental insight, where the applications are not necessarily
for the people with a rare disease condition. For instance, the use of statins. It was discovered because of a rare genetic
disease of cholesterol, familial hypercholesterolemia. But a lot of people are benefiting from statins
at the population level regardless or not whether they have familial hypercholesterolemia. Our challenge ahead, really, is using and
measuring all determinants of health. From the macro to the micro. To develop analytical approaches to population
health. It's a question. And I think the technology is moving in that
direction. So last year I published this paper about
big data meeting public health. And asked the question, can our public health
functions become more precise? We do a lot of surveillance, where we practice
these at the population level. We do develop policies, and then we deliver
interventions. Can we use the tools of analytical data science
and big data to do public health better? And as a matter of fact there are three upcoming
areas that we have initial success in using these tools at the population level — modernizing
surveillance, pathogen genomics, and then the targeting. Pathogen genomics is really the early frontrunner
of precision medicine for public health, whose applications are not necessarily at the patient
care level. But more at the population health and outbreak
detection and response, effective antibiotic use guidelines, and reducing the burden of
these conditions. And this is just one example from data from
CDC showing that after the use of whole genome sequencing in surveillance for Listeria outbreaks,
we are now more able to link more of these outbreaks because of the sequence with the
source of the food from which the outbreak is derived as a result of whole genome sequencing
of Listeria samples. Ultimately at the population level, we want
to use tracking in order to solve public health problems. So this quote from WGS to GPS is reminding
us, all of us in public health know about John Snow. And the Broad Street pump. And this is sort of a quote from Harvard that
essentially posits that if John Snow was alive today he would have solved the outbreak much
more precisely and more quickly than he had at his disposal in 1859, or whenever he had
to painstakingly try to map where people drank the water that was infected with Vibrio, and
correlating it with the outcomes of interest. And that's what at least I would like to mention
Sue Desmond-Hellman, CEO of the Gates Foundation, has been thinking along the same lines. Of trying to come up with a new era of precision
public health. For those of you who don't know, Sue Desmond-Hellman
was at UCSF. She was one of the architects of the National
Resource Council Report of 2011. And is now leading the Gates Foundation global
health effort to think about how we can bring the analytical tools of data science and others
to do tracking and understanding disease at the population level. So that we can improve health not only for
rich people, but across the globe. So in conclusion, the four themes that I have
presented to you are as follows. That as we move into these new tools and technologies,
we still need both medicine and public health to improve population health. That was true before precision approaches,
will always be true even after precision approaches. Partnerships are needed to implement what
we already know. And what we will know even ten years from
now. Because there are opportunities to save lives
now that we are not capitalizing on in that precision medicine space. Then public health sciences are needed to
generate and implement the new knowledge that's coming from the bench. And finally, we are entering a new era of
precision public health that's not just about genes, drugs, and diseases. And with that, I conclude my talk. And turn it back to Debbie. Thank you very much. >> Thank you to Sandro and Muin for their
presentations. We will now open the floor for questions. As mentioned, questions can be submitted using
the Q&A feature on the right-hand side of your screen via WebEx. Please type your question in the provided
field and hit Submit. For those in the room at NCI, we ask that
you approach and activate a microphone before asking your question. With that, I'll turn it over to NCI's Dr.
Amy Kennedy to moderate the Q&A. Amy? >> Okay, as we wait for questions coming in,
is anyone in the room here at NCI just speak up, come up to a microphone, or — >> Hi, Sandro; it's Ann Geiger. How are you? >> Hi, Ann. How are you? Good to hear from you. >> You too. So I'm going to pose to both of you this question. Which is, it seems to me that there's an important
matter of economics — [ Silence ] >> Yes, my light isn't lighting up. But anyway — >> Ann, all I heard was "economics," then
you cut out. >> Okay, we're going to move microphones;
here we go. So for me, this, you know, Sandro, these issues
of funding basic science over population science have existed for decades. You know, they're a legacy I think we're all
aware of. I think my question for both of you is, we
have limited budget. And so, you know, I wonder if the two of you
have a different perspective on where is bang for the buck? Because I think I heard implicit, difference
in the two of you. Sandro, you referred to things like antibiotics
and I would say public works. That are actually pretty cost-effective for
what we got from them. And for me the question is, is precision medicine
screening public health going to be similarly cost effective? So I'm curious to know what you two think. >> Well, I can take the first crack. I actually think economics is at the very
heart of my argument. That if we have limitless resources, I actually
probably would take a lot of my argument off the table. Because then I would say, God bless, let us
invest in molecular genic targeting, and let's also invest in efforts that improve the health
of populations. So we can actually create a better world. But the challenge, of course, is as everybody
know, that we are in this country — just focusing on the U.S. for a second — we spend
roughly 6% more in GDP in health than any other country. We have the worst indicators of any of our
peer high-income countries. And a lot of that is ultimately misallocation
of resources. So my — at the heart of my argument is a
concern with resource allocation. Now I'm well aware of the fact that the resources
come from different buckets and different pools. And it's not a simple matter of saying, well
let's take the money that's being spent on precision medicine and shift it into more
population-based approaches. It doesn't work that way. But at the end of the day it's the broader
national conversation that determines where we put our — which basket we put our eggs
in. And my worry is that we have a momentum behind
an individualistic approach that has reached its apotheosis in a precision medicine approach. and my worry if would crowd [Inaudible]. Again, just to be very clear about my argument,
I think there is a lot of utility at the level of science that can emerge from these precision
medicine approaches. My worry is that it crowds out everything
else. So really it comes down to politics and resource
allocation. >> This is Muin here. So I do not disagree with you, Sandro. And I think the question is well posed. And the economics is always an issue. And as someone who has spent most of his career
in a public health agency, I can attest to that. The question is right now is whether or not
investments in both the population sciences and basic science — I mean, the case that
I made is that a more holistic investment is first likely to make precision medicine
success. And second, it would lead to a new era of
precision public health. So that's the sort of the argument I'm putting
forward. I'm not trying to pit one agency versus another. I'm saying that, you know, that the investment
— the return on investment if we are to take on precision medicine moving forward would
be much better and much faster if we, as we learn new knowledge we implement what we already
know. And we use the same technology at the population
level to track and measure population health problems. And then target the interventions where they
are needed the most. Based on any number of factors. And they don't have to be genetics; they could
be based on geography, on the resource allocation. And so economics is always at the heart of
it. But I'm hoping that moving forward this would
not be a competition but a collaboration between the various sectors of public health, medicine,
and the basic sciences. [ Silence ] >> Okay, this is Debbie Winn. Question for both of you. Sometimes fundamentally I think, well, precision
medicine really all that means is risk stratification. Or prognosis stratification. Or stratification on the basis of treatment. And I'm trying to reconcile that with equity
access. To some extent if we really are going to achieve
some level of equity and reduced disparities, that there has to be some level of detailed
and precise stratification of groups of people into subgroups. And that by doing that we could perhaps in
appropriately targeting those subgroups, we could have better health than we could at
some measure that might be more cost-effective and reach a larger community, but maybe not
have quite the impact. Do you want to talk about that issue of equity
and access disparities? Thanks. Muin will go first. >> I mean, equity, access and disparities
are at the heart of the public health, I guess the public health mandate. And if — just to repeat some of the same
arguments that I just elaborated on, that for precision medicine to succeed and succeed
widely across the population it's not enough to do the discovery. Even if the discovery includes a wide cross
section of the population. Because I know the current precision medicine
initiative is trying very hard to be inclusive across the population. But also dealing with the equities of implementation
of the new discoveries. And as I said earlier — I mean, I showed
some data that shows even with the things that we know what to do, like with any new
technologies, they're not well implemented across the population. So could we use — if we put genetics aside
for a minute, some of the new analytics and data science and visualization, could we be
able to show more robustly and more persuasively some of these disparities that are rampant
across the population for which resource allocation needs to occur? So I think I'm not too far from Sandro on
that point. Because it's at the heart of the argument
that there is disparities that need to be dealt with. >> Dr. Galea? >> Yes, I don't disagree. I think the only thing I would add is that
the question underlines a tension that I don't think we have been fully honest in grappling
with about the society. Which is a tension between advancing efficient,
targeted efforts that improve the health of a few and the inevitable trade-off that results
in terms of widening health equity gaps. And will high-end medical therapies based
on genetic and molecular targeting result in potentially better clinical care for a
small subset of people who get really sick? Probably. Will that also widen health gaps? Almost certainly. Is that a price we're willing to pay? I don't know that we have discussed that as
a society. And I also don't think — the discussion playing
field is not level. And for obvious reasons. And that there are many special interests
that push such discussions in a very particular way. So it strikes me as almost canonical that
these approaches, as with any other high-end technological approaches, will result in widening
health gaps in the short- and medium-term. Whether or not they have success in improving
health in the long term and eventually in narrowing of health gaps remains a subject
of debate. >> Okay, taking a question from online. Would you say that familial hypercholesterolemia,
a common genetic disorder affecting one in 250 people would be a good example of what
precision health could mean for heart disease prevention at the population level? >> Can I take that one, Sandro? Because — >> All yours. >> Yes, our office at CDC has been — has
promoted and put familial hypercholesterolemia as one of these tier one green applications. It didn't show up on my slides today because
I'm at the NCI and the focus is mostly on cancer. So let's take FH. FH is an interesting story by itself. It's the genetic disease that's not rare;
it's actually the most common genetic disease. We have data that shows from [Inaudible] that
it's about one in 200, 250 people in the U.S. have it. Very high, early premature heart disease risk. But that accounts for less than 5% of all
heart attacks. So in terms of the magnitude of the burden,
it's not that big in terms of a population attributable fraction. But there are thousands of patients who have
FH. And the relatives, who could benefit from
early detection and treatment with statins that are not being cared for right now. And I think using these precision tools to
find them and, you know, registering them. And there are different studies that have
been shown recently to be able to track FH at the population level is, yes, an example
of a precision approach at the population level. Similarly to what you can call newborn screening
as a program. Newborn screening, I mean, we screen four
million babies every year to find 10,000 or so who have one or more genetic conditions
like PKU. That's truly a population-wide effort that
benefits the few that otherwise would not have benefited from health care by themselves. Or they could have had mental retardation. So it is an example of precision public health. [ Silence ] >> Another question from online. Muin, I think it's directed towards you. "Can you elaborate on the potential of epigenetics
to bridge the approaches of social, environmental, behavioral, and genetic factors' influence
in population health?" >> Well, I mean like with all technologies,
there is both hope and hype about epigenetics. And I'm here at NCI where there are many more
epigenetics experts than I am but I don't see anyone here. But I want to refer people to a blog I wrote
a couple of years ago after I gave a talk at a public health, the Association for Tape
and Editorial Health Officials annual meeting on why epigenetics is so appealing to public
health. You know, when you talk about genes, you can't
change your genes and their effects are small. But epigenetics is one of those areas where
truly the environmental, social determinants of health really come face-to-face with the
biology. And, you know, both from the experimental
studies and even some of the human studies, we know that genes can be turned on and off,
especially at the right time in growth and development. Now, are the applications here? I mean, there are a few applications. It's mostly the promise of using epigenetics
and epigenetic markers as a union between nature and nurture. Well, I'll stop there and if people want to
know more info, I'm happy to correspond to them. >> Sandro, I don't know if you have anything
else to add? >> No, that's fine. [ Silence ] >> I think in the absence of any additional
questions, we should give both of our speakers a few moments for a couple of last words. >> This was a good discussion today, Sandro. I appreciated your willingness to engage in
this discussion. I'm hoping that this dialogue will continue
and some of the challenges that you raised will be looked at very seriously. Both from a funding perspective as well as
societal perspectives and so on and so forth. Because at the end of the day, all the new
tools and technologies with all their promises are not going to lead to improved population
health unless we use them and we use them judiciously at the right time. And do not forget about some of the other
things that we have to do as well. Thank you. >> Yes, thank you for having me; I enjoyed
the discussion. My feeling about this is that there are a
lot of very good scientists. And health thinkers who are engaged in this. And perhaps this is why I started my presentation
by casting myself in [Inaudible] position. I'm deeply committed to the goals and aspirations
of science towards improving health. And my commitment and beliefs is that we need
to have this kind of conversation to try to not just end where we should be, and figure
out what that space is. So I've enjoyed this conversation. Thank you. >> In these last moments, I'd like to take
time to thank Drs. Winn, Galea, Khoury, and Kennedy for their time. Thank you to all of us who joined today in
person via the web. This session has concluded. You may disconnect at this time. [ Applause ]

Related Posts

1 thought on “Will Precision Medicine Improve Public Health?

  1. This webinar series is a great initiative. With whatever I am seeing and learning, Precision Medicine will surely improve Public Health. But it won't happen overnight. We are talking about complex diseases like cancer, diabetes, etc. It will take several decades before we can expect significant impact from genomics on public health.

Leave a Reply

Your email address will not be published. Required fields are marked *