okay good afternoon everybody and thank you for joining us for today's uh uh Grand round seminar at the Kuni School of Public Health uh before we start the the seminar uh you are all of you are muted so if you will have questions a Q&A session at the end of the seminar so if you have any questions please post it on chat and then I will ask them on your behalf at the end of the the seminar so um so let's go to the um seminar itself so on behalf of the Kun School of Public Health the Department of
Environmental occupational and Jo special Health Sciences I'm very pleased today to be with all of you this afternoon and have a chance to introduce our guest speaker Dr Robert Wright Dr Wright brings together many skills in basic applied applied and clinical discipline to connect exposomics and phenomics across the lifespan and solve complex problems in environment human health disease and precision medicine the seminar today will tackle these challenges and these topics uh Dr R is the chair o
f the environmental medicine and Public Health at Mount Shai school of medicine and director of the NIH funded Mount siai human health exposure analysis resource targeted labub and Mount Sinai NS P30 core Center he also serves in The Advisory Council for the National Institute of Environmental Health Sciences he also put together one of the first longitudinal Birth Cohort studies back in 2007 based in Mexico City to study the developmental origins of health and disease his funding and Scholar re
cord speaks for itself finally as far as long as I know Bob he just a people's person and we are all for a great treat today without further Ado please join me in welcoming Bob R Bob the floor is yours thank you elas so um this talk is really meant to um bring together medicine and and public health I think of exposomics is really being representative of Public Health and obviously in in the field of Medicine all the talk is about Precision medicine uh and I'm going to try to uh present the reas
ons why we may try to integrate these um based on an old flaner o Conor short story um everything that rises must converge so the outline of the talk is that I'll describe Precision medicine uh which is proposed to be the future uh I think exposomics is also the future and I'll talk about what I mean by exposomics I'll give a working definition uh and how we can apply exposomics to understand health and disease but I think the broader um goal of the talk is really to ask the question can we cond
uct medicine with a public health lens so that's a very different way of actually practicing medicine but I'm going to make the argument that instead of drugs and procedures we need to think about preventing disease progression so think about it within a public health framework so rather than you know can we develop a drug to make Parkinson's disease better or what we do in a person's environment to slow its progression or even reverse in some cases uh disease progression so think of people with
the disease as being the same as a vulnerable population so and now is the time I think to bring together Precision medicine and exposomics so start with exposomics so exposomics is a study of all of the health relevant environmental factors that we experience across our lifespan so it's it's a big task it includes things like allergens it includes Vector born uh infections it includes infections themselves such as um coid and other viruses and bacteria would include the microbiome includes our
social environment our nutritional environment uh which is really tied to behavior our chemical environment and our physical environment all these things are part of our exposome so it's not just chemical so I come from an environmental health background you know my original training was in toxicology and we really just studied uh the effects of chemicals on the body but now I take a much broader approach of what is the environment and another way to define is really everything that's not genet
ics um I don't care for that um definition in part because not everything in our environment really matters that much to our health so I'm trying to what we want to do in exposomic is try to sort of interrogate the environment to try to figure out what are the most important aspects um of the environment that are affecting our health so um I think of all the health relevant environment across our lifespan so what is precision medicine so the national research Council of the National Academy of M
edicine defines it as tailoring medical treatment to Patient characteristics the idea is that um through um understanding primarily genomics um they can understand susceptibility classifications for disease and that can categorize people with regards to their prognosis or response to treatment and if they could do this um it will individualize interventions maximize benefit reduce treatment failure to reduce side effects what it really is about is it means understanding each patient's individual
background and how background influences disease progression and treatment response it operates in a setting where the probability of illness is one that is preventing the illness no longer matters does it mean that preventing the progression of the illness doesn't matter so certainly people who first get diagnosed with diseases like Parkinson's disease or Alzheimer's disease they get worse over time so there is are things that we can do to try to mitigate um the progression of a disease and so
me of those things may be environmental they're not all just treatments and procedures unfortunately for the way that environmental health interacts with uh Precision medicine is precision medicine pretty much ignores environmental health it's really practiced as pharmacogenetics or pharmacogenomics and the idea behind pharmacogenomics is that if we understand genetic variation to treatment and disease we can again Target therapy to the most effective drugs improve efficacy decrease toxicity and
Target behavioral modifications in some cases improve the counseling and decision- making that patients experience improve their outcomes and their satisfaction well that's all very good but it only works if the only source of variation in treatment prognosis and disease progression is genetic if other factors are playing a role and we are absolutely certain that other factors play a role pharmacogen genomics by itself is not going to work as a Precision medicine tool and yet this is what we se
e so if you were to get on the internet and start to Google different Precision medicine programs at major universities across the the country uh you'll basically see variations on a theme where you'll see some um some figure that represents DNA and discussion about genes causing diseases and then targeted drugs that are going to be developed because we're going to sequence everybody who has a particular Gene and it's going to be you know radically changed the way we practice medicine you know t
his has been the argument for about 25 years now since the um Human Genome Project was completed and I can tell you that very little has changed in medicine with respects to genetic causes of diseases in 25 years and how they affect treatment so it's a promise that has been promised for a long time and hasn't delivered and I'm and I believe it's because we're ignoring the fact that genetics is playing only a minor role in most diseases so we know that gentics is only one piece of a much bigger p
uzzle but for most Precision medicine programs there's no mention of environment and this is another example of another website in which instead of being oriented horizont uh vertically it's oriented uh horizontally with DNA a drug stratification and different outcomes um and this is another one that's oriented with the same theme uh vertically but basically all these websites look pretty identical and none of them ever mentioned environment or almost none of them ever mentioned environment but
we know environment plays a role at least complex diseases um actually have uh risk factors that are not genetic um we've known this for a very long time and so when I say complex diseases what I'm really talking about are the very common diseases that don't have a single cause so mandelian genetic diseases like cystic fibrosis or CLE anemia they have a one primary um cause influenza has one primary cause as particular virus certain bacterial infections obviously have one single cause but the va
st majority of diseases do not have one single cause Alzheimer's disease does not have a cause it has many many causes uh so it is Parkinson's disease so does ADHD so does autism and the list goes on and on and on and so do all cancers so the the real drivers of healthc care costs are actually complex diseases which don't have uniform single causes for their for their ideology and we've known this for well over 20 years in fact this is an issue of science from 2002 excuse me on the puzzle of com
plex diseases and it had an introduction to the special issue that was entitled it's not just the not just the genes and there were articles about Behavior nutrition infections chemicals physical environment culture society and stress and it appeared as though we were on the right path uh back in 2002 but unfortunately we started going down a genetics genomics path and I think that took us away from this idea that complex diseases are truly complex and they're not just complex genetically they'r
e complex on multiple levels and most of those levels are actually fit um a a criteria of being part of the environment so how did we get here so and you'll see some references from different 70s and 80s bands as sprinkled throughout this talk so David burn the Talking Heads so as I said complex diseases are actually the major diseases that are driving health care costs they include ADHD attention deficit hyperactivity disorder obesity asthma COPD renal failure Parkinson's disease the vast major
ity of cancers you know these These are actually what is uh actually causing uh our our health to get worse over time diabetes should also be on that list their ideology is a mix of genetic and environmental risk factors I just showed from that issue of science they're also becoming more common their prevalence and annual incidence is going up and up and up you know one of the arguments that counters the idea that autism is a purely genetic disease is that it's becoming more and more common gene
tics cannot operate on a time scale of even 10 or 15 or 20 years if if something is becoming more common over that time scale it can't be primarily genetic because genetics requires reproduction in multiple generations to pass so you cannot explain increases in these diseases based on genetic risk factors and yet we keep trying to hit this you know um square peg into a round hole all the time with genetics and the causes of these increases in diseases our society is definitely getting sicker and
it's not due to our genes and in fact as I said these genetic main M effects cannot genetic man effects cannot explain Rising rates because a lot more time has to pass in 10 20 or even 30 years the environmental risk factors for all those complex diseases that I showed are largely unidentified we know a few but most of them we don't know because we haven't invested in trying to understand them so we has much less research on them and in fact um I think the language that we use actually in some
ways creates barriers so we tend to think in terms of this old phrase nature versus nurture and um geneticists in particular like to do studies that are called heritability studies where they study Twins and they parse out the percentage of a disease that they say is genetic and a percentage is environmental and there's actually uh these sort of heritability studies that actually say that autism is actually almost entirely genetic and yet autism can't possibly get this much more common over such
amount of time if it's purely Gene genetic and it also ignores the idea that genetics actually confers risk it's not causation there is a difference if you smoke you're at higher risk for lung cancer but not everybody who smokes gets lung cancer so it's a risk factor for lung cancer it is a cause but it's not a onetoone relationship and genetics is not a onetoone relationship with disease either we tend to think that way about environmental factors we we tend to think of genetics as being someh
ow deterministic when it's not and so we need to think of both fact s is being in conferring risk rather than causation and that the levels of risk uh vary for each in fact I'm going to take a step back and I'm going to make the argument that part of the reason why we call a disease a genetic disease or environmental is actually perception and that perception is a function of the prevalence of the risk factors in a population so I'm going to make the argument that if a disease is due to a gene e
nvironment interaction that is complex diseases always have genetic components and always have environmental components if the genetic polymorphism is relatively rare or low in prevalence in a population and the environmental risk factor is common or high in prevalence in the population that disease will appear to be genetic and if the converse is true if the environmental factor is relatively rare or low in prevalence and the genetic polymorphism is very common or high in prevalence the disease
will give an a appearance of being environmental now remember this is a disease that's caused by interactions between genes and environment this is actually a difficult concept to present in humans but it's actually very simple to present in chickens so there actually is a chicken trait called yellow Shanks so yellow Shanks are discoloration of their legs with Artic and they get it this sort of mucousy discharge um it occurs if a chicken has particular genetic variant and it eat eats yellow cor
n as opposed to white corn this is actually an example that Ken Rothman a famous epidemiologist included in one of his textbooks called modern epidemiology the genetic trait follows an autosomal dominant pattern meaning you only need one copy of the Alo to express the disease so imagine two farmers farmer Jones and farmer Smith they live down the road from each other farmer Jones has an inbred chicken flock all of his chickens have the variant for yellow Shanks 100% of them for that reason over
time his family learned to feed them only white corn for generations of farmer Jones's family they fed only white corn one day the store ran out he bought yellow corn all of his chickens got yellow Shanks if you were to ask farmer Jones what is the cause of yellow Shanks he would say it's environmental it comes from eating yellow corn that's his perception farmer Smith on the other hand lives down the road he also has an inbred chicken flock none of them carry the variant for yellow Shanks yello
w Corn's cheaper so his family has always fed their chickens yellow corn one day a chicken gets a new mutation it gets yellow shakes because it has the mutation and it's eating yellow corn and farmer Jones is or farmer Smith is an amateur geneticist and he plots a pedigree evolve The Offspring of that chicken and discovers that half of those chickens get yellow Shanks and the ones who have yellow Shanks half of their progeny get yellow Shanks and ones that don't get yellow Shanks under of their
progyny get it and so he plots a pedigree and he figures out that yellow Shanks is a genetic disease that follows an autosomal dominant inheritance pattern so which is it is it farmer Jones's perspective in which yellow Shanks is purely environmental or is it farmer Smith's perspective that yellow Shanks is purely genetic it's not hard to imagine that similar things are actually happen happening in hum populations and there's this artificial divide that we've created over time through our langua
ge that actually makes us think that a disease is particularly genetic or particularly environmental when in fact all diseases are environmental and genetic so we really can't say if yellow Shanks is environmental or genetic it's contextual it depends on the prevalence of the genetic variant and the environmental factor in a population and by the same token cancer is not purely environmental or genetic it depends on both the environmental factors that might be causing a mutation the carcinogen a
nd whether or not you have a genetic risk factor for being susceptible to that uh particular carcinogen and even diseases that we think of as being purely genetic pheno kenua is a disease that's thought to be purely genetic it's actually one of the newborn screening diseases if you carry the genetic trait or for feno Ken UA you have two copies of it because it's homozyg recessive and you are exposed to phenol alanine which is a very common amino acid that's found in our diet you'll develop de se
vere developmental delays however through newborn testing you can actually eat very low um pheno alanine diets and you can have a completely normal uh phenotype by avoiding phenol alanine so is it really truly genetic or is it really or is it Pur is it truly environmental or is it a combination of genes and environment so even diseases that we traditionally think of as being purely genetic almost always have an environmental component that we don't think of and usually that environmental compone
nt uh is very very common uh in the population and in fact thinking about heritability where we parse out whether a disease is particularly or what percentage of disease is due to genetics or what percentage of disas is due to environment for far the Jones the heritability of chicken Shanks or yellow Shanks is 0% and for Farmer Smith it's 100% so even heritability is is contextual it depends on the prevalence of the genetic environmental factors in a population because all genes operate in varia
ble environmental backgrounds and there's lots and lots of backgrounds so nutrition is a background so um whether or not you have obesity whether or not you have deficiencies and vitamins and minerals may affect your way that a disease progresses people who are pregnant may make their pregnancy can also often make a disease worse type two diabetes is a very good example type two diabetes often gets out of control uh in women who become pregnant because there are these physiologic changes that oc
cur during pregnancy that can actually exacerbate type two diabetes and then it goes away after the baby is born geography can play a role geography is a background that can affect uh diseases so you're more likely to have a stroke if you live at a very very high altitude and that's because the oxygen concentration at high altitudes is a little bit lower and if it's a little bit lower you tend to make more red blood cells if you make more red blood cells your blood's a little more viscous and so
the probability of forming a clot and having a stroke is a little bit higher and that's been shown that people who live at higher altitudes have slightly higher risk for stroke culture plays a role culture is about behaviors and diet and those things can actually play a very big role uh in whether or not you're going to get a disease or what your response is to a treatment and then environment itself is not a single uh issue so you know very few children with lead poisoning are only exposed to
lead poison they're exposed to many other chemicals we're all exposed to uh combinations of different chemicals or mixtures and we need to start thinking about those mixtures rather than one chemical at a time and then probably the most important background I'm going to argue is having a disease because disease States change metabolism and increase the risk of exposure to multiple drugs so people who have a disease are taking drugs and you're more likely to have a drug drug interaction uh and yo
u're also likely to have a drug environment interaction so there are lots of ways in which environment can play a role in the progression of a person's disease so here's some examples so this is the classic example that you will see on 99% of precision medicine websites so a 53y old woman develops a deep Venus thrombosis which is actually a major clot in their thigh on a trans-pacific flight so one of the risk factors is being relatively immobile for a long period of time so if you're sitting in
coach for 12 hours you're not moving your leg much and you're much more likely to get a blood clot which is what leads to a DVT so after she lands she sees her doctor and she gets given a blood thinner called Warren excuse me and she gets a g bleed so in this particular case she had a genetic variant that actually made her a slow metabolizer warin and because she got a standard dose it was too big for her and she ended up having a bleed because she doesn't metabolize warin uh the same way that
other people do because of genetic variant so in this case genomics would play a role in Precision medicine and improving her health care because she had a genetic variant but for that case you know there there that's a great example but there are many many other situations in which genetics would not play a role so here's an example from a patient I once took care of a 17-year-old child uh with autism has notable increases headbanging and anger outbursts so we tend to think of autism as a genet
ic disease so I can tell you a genetic SC scan found nothing that would explain this also he's 17 years old and all of a sudden he has a change in his behavior he has new head begging and anger outbursts well it turned out he had lead poisoning and it's a little weird for a 17-year-old to get lead poisoning with except 17 year olds with autism so people with Autism have what's called Pica the non-food substances he turned out to be eating newspaper uh newspaper ink has a little bit of lead in it
and he was eating enough newspaper where he was actually lead poisoning himself paradoxically his three-year-old brother who did not have autism did not have lead poisoning either so uh in a general situation a three-year-old is far more likely to get lead poisoning than a 17-year-old and that's because three-year-olds and toddlers tend to have a lot of hand to mouth activity they get dust on their fingers and if there's lead in the environment they'll ingest that dust uh by putting their finge
rs in their mouth 17y olds don't typically do that but this was a 17-year-old with autism and so the background that he had his autism actually made him at risk for lead poisoning we were able to treat him and is actually his headbanging anger up uh improved dramatically once he was chelated so here's another example so a 14-year-old who has gastro Sagal reflux he he's constantly uh burping up fluids um in a history of depression um he breaks his arm and he's treated with morphine for a fracture
and he gets respiratory failure he had a genome scan before they gave a morphine that showed no functional variance in and in genes that could affect morphine metabolism well GE reflux might be treated with over counter medications like Tagamet or renadine which is Al called Zantac and these actually can inhibit the p450 enzymes that metabolize morphine and paxel can do the same same so what was happening in this case is he had a drug environment intera or drug drug interactions which could hav
e been picked up with a exposomic scan of the medications he was taking uh but but because this wasn't part of um Precision medicine wasn't picked up and a genome scan would not actually have picked up anything because it wasn't due uh to um anything in his genetics it had to do with his diet potentially and his medications for example uh grapefruit juice is kind of famous for being a p450 inhibitor as well so maybe he was taking Brave FP juice as well as um over the-counter Tagamet uh and had a
history of depression was taking pone all these things uh interfered with his ability to to metabolize morphine and that's why he got respiratory failure but again a genome scan would not have picked that up so if we're going to really practice Precision medicine we can't just isolate genetics and think that we're actually going to change much we have to include the environment we have to include what I what I call exposomics all aspects of the environment because I think the most important bac
kground um in Precision medicine is the disease itself and the treatment for that disease and if we understand the way that environment interacts with the disease and the potential treatment for the disease we'll be able to to prescribe more precise medic medical interventions many of which come from the public health field of just avoiding certain environmental exposures or promoting nutrition so but to start to get Physicians to think this way we have to do the studies that patients you know t
hat Physicians need in order to treat patients so right now we're not really doing that there are very few examples of Environmental Studies that take place in people with a disease so and there's a reason for that because the public health perspective and the medical perspective are very very different so this is the medical perspective so as a doctor I want to be able to diagnose the patient so I want the research that's done to help me diagnose the patients so sometimes EP exposomics and epid
emiology can help so risk factors are kind of weighted variable I'm 59 years old if I had chest pain and presented to an emergency room uh particularly if I was a smoker um you would probably assume based on the research that's been done that I'm at very high risk for for having a heart attack and you would assume that that might the cause of my um cause of my chest pain was a heart attack so help me treat the patient well there it breaks down a little bit as a physician the fact that you know I
'm a 59y old man who smoked has nothing to do with my treatment it's not going to change anything it's Not Gon to change the medications I'm G to get it's not going to change whether I'm gonna get surgery or catheterization it doesn't help at all in fact I may even ignore the results of such studies because it doesn't help me treat patients I'm no longer trying to prevent the disease from occurring in this scenario I'm trying to treat the disease so the risk factors that caused it don't really h
elp very much so in order for us to bring environment into medicine we have to ask different questions because there are no right answers to the wrong questions so what are the roadblocks during environment uh into medicine well medicine operates at the individual level you know Public Health operates at the population level um when people present to a doctor something's going on the doctor may not know what it is is they may not have diagnosed them correctly but there is a problem that exists i
n public health people start out healthy um and they may develop a problem later on but at the moment that they typically get enrolled in a study they actually don't have the disease yet and then it's all about treatment and whether not there's side effects and what treatment will get the best outcome so in public health as I said instead of population individual level it operates at the population level we think about risk factors if you're exposed will you get sick if you smoke will you get lu
ng cancer yes you may get lung cancer but not everybody will so it's really a probability it's not a one to one relationship um you know I had a my father-in-law was a two to three pack a day smoker for about 60 years and never got um a heart disease or or lung cancer before he passed away so it doesn't always fit one to one although definitely there's a higher risk of getting L lung cancer or or heart disease if you smoke that's absolutely clear CLE but it's not a one toone relationship because
it's really about in public health about prevention at the population level that's why it doesn't work as well at the individual level because it's just harder to get people to change their behavior if they're not actually sick um Public Health thinks about susceptible populations children pregnant women the elderly tend to be the classic susceptible populations but it's a different perspective and part of it is just time so medicine gets involved after the person is sick Public Health gets inv
olved before the person is sick because public health is trying to prevent that person from getting sick which is a very Noble goal but there are already millions of people who have a disease so what can Public Health teach medicine about how to help them because outside of primary care medicine is very little interested in why you are sick they want to understand how to treat the disease but perhaps there are environmental factors that can actually actually help treat the disease that's all abo
ut the research that needs to be done so in public health as I said uh the research tends to be before a disease is present so on this graph um to the right a disease or phenotype is present to the left it's not the x-axis is time and then the different types of of study designs that we do we can do a prospective cohort study where we follow a group of healthy people over time and some of them develop the disease and some of them do not and we see which risk factors had a higher probability of p
redicting the disease or we can do what's called a case control study where we take a group of people with a disease and a control group and we see which risk factors they have and we try to go back in time to see whether or not that risk factor preceded uh the onset of the disease and in genetics they typically do uh some variation of case control family based Association because your DNA sequence never changes across your lifespan so you almost never see a cohort study in genetics because it's
not very very um efficient because your DNA sequence never changes so geneticists tend to think only in terms of case control or family based Association studies Physicians tend to think in terms of um randomized clinical trials because they're um trying to see whether a particular treatment is better than a no treatment or better than a different treatment or they're doing a clinical observation study where they take a group of patients with a disease to try to see whether or not in many cases
a genetic risk factor predicts a a higher or faster progression of the disease but medicine tends to think in terms of what what to do after the disease occurs and public health tends to think of what to do before to prevent the disease from ever occurring so it's a two different types of mindsets Public Health thinks of causation risk factors medicine thinks of diagnosis and treatment so can we link these and if we're going to we have to start studying what happens after someone gets a disease
and in public health we very very rarely do that so what I'm arguing for is that patients with a disease should be thought of as a vulnerable population and one of my Arguments for this is I'm not against prevention we still have to do prevention but there are millions of people with type two diabetes and obesity there are millions of people with Alzheimer's disease we have to start thinking about what we can do for them and there are environmental modifications that can help them so why now so
there have really been exponential advances in exposure science in the last 10 years exposomics is really about measuring as much of the environment as you can um there are what are called untargeted chemical assays so our lab at mons can do these these can measure tens of thousands of chemicals in a sample of blood or urine uh our we we can also measure targeted chemical panels we can measure phalates we can measure past the forever chemicals in the news a lot uh we can measure most of the che
micals that happened in that East palest Palestine uh train derailment and um we can we can do that um which M what hundreds and hundreds of chemicals now at a time so we're getting much better at the chemistry at the same time uh because of Big Data uh we're able to measure a lot of other things I know there's folks in the geography Department that probably do satellite remote sensing to estimate um exposure to air pollution temperature Green Space light at night among other different types of
of variables public databases can give information about uh the geospatial distribution of diseases such as coid but also things like sces um uh historical records that show how redlining has occurred and the effects of racism on on health all these things are things that we can do now with public database mining that we couldn't do 10 years ago we can mine electronic medical records and then almost everybody has either a smartwatch or at least a cell phone so there all of us are wearing we devi
ces or many of us are wearing wearable devices that actually give us information about our physiology our activity our heart rate um all kinds of things even like things like EKG strips can now be actually derived from weo devices and the Big Data computational infrastructure to analyze all this data simultaneously now exists that didn't exist 10 to 15 years ago so the time to do your exposomics really is now there have been so many advances what we really need is some larger umbrella to bring i
t all together because all of these fields are somewhat disperate the expertise that uh is needed to do satellite remote sensing is very different than the expertise needed to do untargeted chemical assays and so the field is a bit disperate because of that because different groups of people are working in different aspects of the exposive so as I said public health is interested in studying vulnerable populations but is largely ignore the role of environment in response to treatment and the eff
ects of environment on people with the disease and the disease severity or its progression as I said the largest vulnerable population to environment may actually be people with a chronic disease and we need to study their environments so I'm going to bring some ideas together about ways we can use some of the exposomic methods that I just uh described uh in in studies so untargeted chemical acids this is what's called a chromatogram so these are features of a chromatogram so this is a mass spec
device so you put Ur you process urine you enter it into the the instrument it um it tells you based on the molecular weight and the charge a distribution of chromatograms and these are what are called features and the features represent individual chemicals so untargeted assays may add May Aid in Diagnostics so if you have a PE people with a disease and people without a disease maybe there will be different signals that are seen in people with a disease or not so a great example might be say E
ating Disorders like bulimia people with bulimia are going to have different kinds of metabolites in their bloodstream or in their urine that actually may be unique enough so that you can use it uh to actually diagnose the disease and in fact what if we focus only on cases maybe uh if we had patients with a particular disease and we measured the untargeted signals over time we could see whether or not there's a correlation with their signal patterns and the in the progression of the disease or i
t's severity so um we need to actually start looking at these cases of different metabolites and chemicals and see whether or not these signals can actually form um what are sometimes called exposomic risk factors people talk about polygenic risk factors we have how many um genetic variance do you have and how does that increase your risk for a disease why not think about exposomic risk factors and how many uh clusters of different metabolites you have and how does that uh correlate with your ri
sk for disease progression so other ways that we can operationalize the exposome and precision medicine uh using untargeted assay so we need to run untargeted assays and cohorts of patients longitudinally so that we can predict the risk of complications so maybe on randomized control trials uh maybe we can start getting some of these assays longitudinally for from each visit and we can actually start to predict the response to treatment maybe the people with particular signal set of signals at t
he start of a trial are more likely actually respond to the treatment than people who won't so we can actually identify clusters of people uh who may actually benefit from the treatment and those who may not so and then there's wearable devices so um also the internet things so a lot of the devices in our lives are connected to the inter internet phone apps cars appliances sensors Etc these devices exchange data with the internet and with other connected devices um we can download GPS data in ph
ys logic data directly and there's even a very low Tech exposome um wearable device so it's actually a silicone bracelet um there's actually a company called my exposome um that you can actually Google uh you can order the bracelet they'll send it to you you wear it for a week you send it back and they'll do an assay on on the on the bracelet then tell you what environmental chemicals uh you were exposed to um in the last week so obviously most of this is going to be stuff that fell on it from t
he air but also things that may may have excreted in your sweat um or from your clothing that may be touching it um so it's not quantitative but it is a pretty decent qualitative uh estimate of what you've been exposed to while you were wearing it so this is becoming more and more popular and it's now its own company so we don't know a lot about these weable devices uh in patients with diseases I think there is some work being done right now um but you know we need to start tracking people who h
ave say Parkinson's and Alzheimer's disease for their physical activity maybe um Apple watch can actually help tell us whether somebody with alzheimer's disease is actually starting to progress to the point where they're not able to take care of themselves Based on data and patterns in that data in in the Apple watch so maybe it actually tells us something about chemotherapy as well patients who are less active may be having more severe reactions to chemotherapy so using these wearable devices i
n patients actually a diagnosis and and treatment and understand progression we need to start thinking about um other ways that um the environment affects uh patients so maybe air pollution which we know is a risk factor for heart disease actually may exacerbate uh patient symptoms and complications after patients had coronary artery bypass one of my favorite recent studies uh was a study on a un University of Washington a few years ago which showed that patients who had had a lung transplant ha
d a higher five-year mortality rate if they lived in an area of higher air pollution than an area of lower air pollution so there are definitely effects of air pollution on our lungs and on our heart we I think that's been pretty well established but what we really haven't studied is what about people who have diseases of their lungs and their heart what is the air pollution doing to them and are there things that we can advise them on such as HEPA filters or even quite frankly moving to an area
where there's less air pollution if you can afford it uh that may actually prolong their lives and there's lots going on in what I'll call the external exposome in geom medicine uh we can measure air pollution in climate exposure to Green Space um we can we can measure temperatures traffic patterns and noise access to healthy foods and the built environment and uh even like one of my favorite studies is one of our faculty actually you you can actually get GPS coordinates off Twitter uh who actu
ally published a study on the geospatial distribution mean tweets uh in the Boston metropolitan area so there's lots that can be done almost everything can be mapped and in fact I think even those untargeted assays that I were describing with the chromatograms if you know somebody's address you can map it so I think we can start to map all kinds of aspects of the exposome the social exposome the Infectious exposome the chemical exposome as well as more more traditional measures like weather vari
ables and air pollution uh noise and light all those things can be mapped and if we start to Overlay those Maps we can start to see the patterns of how the exposome um interrelates with uh health so how might we operationalize uh exposomics in medicine for those of you who like the Ramones um uh wearable devices so one day I Envision a day where you'll take a wearable device uh that connected to your um cell phone that tells you where you are um Google and apple probably know where you are right
now they can even tell whether you're indoors based on the the signal strength of the of the uh internet or the cellular um signal and um you'll get information highly accurate information if it's overlaid with maps of air pollution temperature and other physiologic data that links that physiologic data to weather and air pollution as well as activity and that could be downloaded U by your doctor into um a Mainframe computer that might analyze it using machine learning to understand what risks
you are at going forward uh if we have your address history we can actually link these Maps back in time to where you used to live if we have your occupational history we can probably estimate different types of exposures that are probably more likely to occur with certain occupations and job activities so simple questionnaires um job job Matrix uh job matricies may actually help us understand the exposomics long-term term if this would become part of your standard medical record so one day you'
ll give your cell phone uh at the front desk when you walk in the doctor's office and the data will be um with your permission obviously you have given form can set the data will be downloaded so that we can estimate uh your exposure to all these different environmental variables you'll probably um take an untargeted screen of your blood or urine um and that what chemicals you're exposed to will be linked to your genome scan that um all the different uh precision medicine um uh systems want want
to use and they'll integrate that data with the environment and the genomics to get a risk score for How likely you are for a disease to progress or How likely you are to get develop a disease that and maybe even some uh simple recommendations for what you can do to reduce your risk so I think that is the goal one day for exposomics in Precision medicine and right now there's a lot that can be done if you think about apply environmental health or exposomics to diseases and this is just off the
top of my head um thades are thought to be a risk factor for lung disease and they're thought to be a risk factor for diabetes Well people with cystic fibrosis a melan genetic disease have higher rates of diabetes and pancreatitis and they also have lung disease so what are thads doing to them so we think that their disease is almost purely due to the gy trait and I'm not disputing that but thid exposure on top of that might be actually making their disease phenotype more severe and that's an ar
ea that needs to be studied we think that environmental obesogens like pasas forever chemicals are a cause of diabetes Well type 2 diabetes is an extremely common disease in the United States there's tens of millions of people with type two diabetes they're all exposed to pasas is it making their diabetes worse is it actually uh predicting whether or not they need a higher or lower dose of insulin that's a study that hasn't been done that's a study that probably should be done if we think that t
hese that these environmental exposures actually cause diabetes we need to ask the question what's it doing to a patient who already has diabetes and then we think a lot of about metals that are neurotoxic such as lead or Mercury or arsenic well regardless of whether you think they a cause of Alzheimer's disease and you don't even have to think they're a cause of Alzheimer's disease but if somebody has Alzheimer disease and they're exposed to lead what's it doing to their Alzheimer's disease and
that's another era of research that needs to be done maybe the air quality in surgical icus is actually leading to a slightly increased um uh mortality and morbidity rate so we're not routinely measuring that and untarget chemical assays might actually as I said earlier be a risk factor for eating disorders such as anorexia or bmia so unlike you know traditional Public Health causation research if you flip the question and say well rather than thinking about whether or not their likes cause dia
betes uh and say what is it doing to people with diabetes if you just do that simple exercise in your head the field is so open that almost all the diseases are low hanging frood so I'm getting close to the end um exposomics I think I've shown is a very complicated field uh it requires the integration of multiple types of data some of which are going to be questionnaires I think questionnaires will always have a role uh particularly for social stressors or job occupation uh we can't go back in t
ime to measure everything so we're going to have to just deal with the fact that everybody's memory is imperfect so we might get imperfect data but it doesn't mean that it's not valuable uh we'll be using GIS based models satellite remote sensing uh sensing um actually also mining public databases I think photographs are going to be used more and more often in research uh I think if we took a picture of that cabinet underneath your sink so we could see all your planing products we could tell a l
ot about the uh types of VC's that are actually likely be in the environment of your apartment or your house and then obviously we going to do biomarkers and bloody urine um there's going to be a role for mobile devices like smartphones and accelerometers to actually tell us what's going on both where you are but also what you're exposed to uh in your local environment um and what your physiologic reaction is that is to that and then this data I think can be linked with other types of omx when p
eople think about genomics or epigenomics or proteomics I think all those things can be linked together I think the missing piece of the puzzle though is this exposomic piece that's actually at the top of this graph so I think in the future exposomics in Precision medicine are going to enable better diagnostic tests so we can identify subpopulations that have different prognosis of disease uh we can understand the relationship between a patient's uh chemical nutritional physical and social envir
onment and their prognosis and we can also identify optimal treatments based on someone's exposome so that we can reduce variability um to a treatment response by uh modifying their environment reduce side effects and increase compliance so that Precision medicine programs instead of being purely genetic will look like this where people are stratified based not just on their Geno but on different characteristics of their environment that actually I think are going to be more likely to predict a
ret treatment response uh and failures so we instead of um looking at medicine from a purely medical perspective look at medicine from a public health perspective and think of patients as vulnerable populations and then um one of the programs that um I'm part of is called here the human health exposure analysis resource um we will do a wide range of assays um that can be done used in um different Precision medicine and issues or traditional epidemiologic and public health studies untargeted chem
ical assays metabolomics which actually come along with the untargeted assays untargeted assays include both the metabolome and um exogenous chemicals one of our faculty Manish rora is a dentist and a chemist and he specializes in assays that can actually measure chemicals in teeth teeth grow like trees uh so you can actually go back in time based on the ring pattern uh in a tooth uh and actually measure rings that are were laid down different life stages and what chemicals got laid into those R
ings is something you can do very analogous to what happens in geology with trees we can measure heavy metals uh and all that data could be uh actually linked to different omic datas although I think the most important omic obviously and perhaps I'm a bit bias is the exposit which I think is going to complete the complex disease puzzle um by asking the right questions that benefit patients so and that is my last slide so thank you all thank you very much uh I mean one of the problems with the on
line is we cannot really uh congratulate for this excellent presentation so I have um there's a couple of questions for you um yeah um man uh Paulo can you um uh unmute the dean okay thank you Professor Riot this is I'm Mendes I'm the dean of the school and uh really enjoyed your presentation very much uh H having spent a very large part of my career in medicine and then moving now to Public Health I'm intrigued by the theory that you presented in its breadth and and I have to ask you perhaps a
naive question I I'll I'll give you two diseases that like farmer Smith I presumed were genetic diseases uh let us say duen muscle distrophy or sickle cell disease as to known genetic diseases is it because I'm limited in my knowledge or is it because we have not discovered the environmental component that uh allows this disease to manifest or can can you help me uh well I've taken care of a lot of patient with the sickle so so I'll focus on that one because I'm not I'm less familiar with u musc
ular disy other than it's exlink recessive and I don't know what the what the gene is actually involved in but for a sickle cell if you think about the Paradigm of it it whether the disease appears is genetic or environmental depends on the prevalence of the genetic Factor environmental Factor so the genetic Factor um is relatively rare and the environmental factor is oxygen which is very very common and so cyel anemia appears to be genetic because the enviral factor is pretty much unavoidable s
ame thing is true for Cystic Fibrosis which is sodium potassium pump it's completely impossible to avoid sodium potassium so it's going to come across as a purely genetic disease CLE cell is really interesting from environmental perspective because you know there are things you can do uh in your environment that can help um obviously high altitudes can sometimes trigger CLE cell um crisis um patients who have infections may have may get crisis and there's all kinds of uh environmental factors th
at can cause inflammation so certainly any uh chemical that was a risk factor for um pulmonary toxicity is probably something someone CLE cell needs to avoid and so there may be ways that in the future we might screen patients with CLE cell anemia for what's in their local environment to try to help them avoid things that might be toxic to the lung so obviously when when I was young and I was I was a dishwasher in a restaurant I foolishly mixed ammonium and bleach and I made chloramine gas and I
was coughing for about a month if I had SI cell uh that would be something that probably could have killed me um and so there are probably household items that people with sickle cell may want to avoid so that I see that as something that may be part of Medical Practice in fact I think one day um there'll be the equivalent of genetic counseling for exposomic so there'll be people who learn how to counsel based on um exposomic U measurements in patients uh how they can actually uh improve improv
e their health so hopefully that will one day be actual profession so to complete the cycle my understanding and the CLE cell uh situation is you're really talking about the disease symptoma rather than the disease in its very purest pathology in in the uh construct of the hemoglobin or the appearance of the red belt cell you're talking about the symptomatology of the disease being triggered by environmental circumstance yeah um yeah the well the the the um the variant will sickle and low oxygen
will make it worse so oxygen is the Environmental factor for the gene for the particular Gene variant and then all the other things like crises or Strokes they're Downstream complications of the disease got it thank you great question okay Andrew do you want to pause the question or uh can you unmute [Music] Dr hey EXC me thanks so much Elis Dr R thanks for for the talk it was really interesting um my question is not so technical it's more um uh practical I guess so some of the studies uh that
you're you're suggesting more people should do require a really broad um range of skills and backgrounds and data sets uh so forth so on so like in your experience what's the best way to actually convene those groups since it's rare you're G to have one person that knows all those things what do you think like an efficient and effective way to convene this group of very different people that often speak functionally different languages yeah well um the folks on this call have the skills that are
very very different than the medical folks so very few Physicians understand how to measure the environment probably the majority of them are probably not even aware that air pollution is a cause of heart attacks um it's just not taught in medical schools so some of it is um bringing people together the Physicians will not have the epidemiology skills they will not have the environmental health skills but they will have the patients and they'll understand the biology of the patients and so the
public health professionals and the exposure scientists I mean to me it's like a perfect team because everybody brings something complimentary to the table um the key quite honestly is yeah I'm trying to say this in the nicest possible way because you know genetics is the 800 pound gorilla and so right now geneticists are not interested in the environment I have sensed a change in the last few years I think there's starting to be some frustration probably from Congress that Precision medicine ha
sn't done anything yet and so I think from my read is that NIH is starting to look for other things other than genomics because they know that genomics by itself isn't going to change anything so I do think in the next few years the potential for those sorts of Partnerships to develop is going to be greater and greater and people with public health particularly people with exposure science skills are going to be able to form those Partnerships much more easily I would argue in the next five to 1
0 years because there's going to be a big demand for them so I I think it's coming um exposure science is going to become much much bigger because it makes sense there's just if you look at you know almost every example of genomic causes of almost any complex disease you can name whether even autism or Alzheimer's or asthma or cancer aners it explains at most about 25 to 30% of the disease the vast vast majority is actually due to environment and I think that's starting to be accepted finally so
I I think it's going to change and I think your skill set is going to be complimentary to the Physicians um as a physician I can tell you that the average physicians's understanding of environmental health is next to nothing they just have to start to be convinced that it's important I think that's going to come in the next few years uh before I give it I have a one question on the chat um David M is asking uh whether the rise on the disease uh we've seen it is because we are more aware of abou
t the diseases right now and is diagnosed more frequently than used to be that that's the argument that the geneticists make about autism because um it's been pointed out to them that it's impossible for genetics to explain a rise in autism I think it used to be something like one in 1500 children and now I think the most recent is like one in 40 children um it's just grown exponentially I think it's gotten to the point where people are are are starting to question whether it's just there in the
early days there was probably more screening and probably more detection I think it's gotten to the point where it's a little hard to believe that that percentage of children should have autism that something hasn't changed also for other diseases um such as obesity and Asthma those diseases are clearly becoming more common is no there's no question about them and those are complex diseases alimer disease is clearly becoming more common although we're getting older uh as a population but um it
it's prettyy clear that for the vast majority of diseases they're just more common okay get Dr Pao can you unmute hi you would need to unmute your mic yeah Dr Solan and then Dr yusf can you hear me now yes all right well thank you so much for the presentation it was very informative um you you focused more on the external expoo um I work on the internal exposome I usually do untargeted metabolomics I started first by targeted metabolomics a while ago long long time ago because people were accusi
ng uh with the fishing Expedition so now I'm doing the untargeted metabolomic high resolution targeted metabolomics in cells so I look at ayoy cells or beta cells and I I I differentiate between different metabolisms and then I look uh using the um a predictive uh Pathways in um statistical software to try to find the relation between the uh metabolite versus the pathways versus the uh the proteins do you envision this type of strategy would work on the long run with uh comparing disease State v
ersus non- disease state and instead of just using objective models using also experimental genomics data plugging it in the model to get more onto the signaling and the pathways and the affected genes well um I think untargeted metabolomics and untargeted chemical assays um have a clear role I do think we have to recognize you you you talked about fishing Expedition that it's Discovery research when you do an untargeted assay because you're you you don't have an opiary hypothesis about what you
're going to find so you know you're you're more likely to make a type two error and that's okay because um you know that's where replication comes in so we have to replicate what we find um when we do Discovery research whether it's from the you know external exposome you can do Discovery research by measuring a 100 different external environmental factors or you can do Discovery research uh using untargeted assays and just just have to be we just have to recognize the role at chance so you kno
w we're trying to discover what's going on and then replicate it genetics got into trouble you know 20 years ago when you know they were doing you know gws and Publishing their gws Discovery analysis as if it was you know something real when it's it is technically a fishing I hate to use the term fishing Expedition but that that is the term that gets used and it has to be replicated first and we we don't want to fall into that trap we have to replicate our findings too if we start with a discove
ry approach which I would argue untargeted chemical assays are we do have to consider a multi-stage design where we also replicate yeah so I'm not totally blind in terms of uh Discovery because I have a hypothesis about a pathway nutrient sensing pathway that is mtor pathway so what I do I just use the Prototype Romy which only Blocks part of the functions of mtor and then I use another uh drug which inhibits both mtor complexes and I compare between them and the control with the hypothesis that
mtor controls proliferation that will lead that's reled to cancer but it also controls anabol it's an anabolic uh kise so it affects uh diabetes and from that perspective I'm comparing whether this pathway is inhibited versus this pathway is intact by using the drugs to inhibit it so it's it's not just looking at the metabolomics from on on fishing Expedition I'm just having a hypothesis about the specific role of uh kyes sure I mean there um I'm aware of people that we've worked with that have
an operary of a component of the screen but not but still analyze the whole screen so yeah one component can be if it has an OP priority that that's perfectly that's that's a hypothesis test but the other things are not so we just have to be cognizant of that so thank you um P oh thank you very much such a fascinating presentation I wonder um um this is a very selfish question but uh it's related to my fi of research which is Hepatitis C among people who inject drugs um and we often wonder two
aspects of hepatitis C one is the exposure and of course among people who use drugs the environment where the drugs are being injected and the nutrition the the the nutrition of the person over time um might have an effect on the actual sport and um perhaps equally important on the natural clearance so I wonder under the your espic framework what are your thoughts about that interplay between environment virus natural clearance and nutrition over time um well as a as a toxicologist when I think
of you know liver toxicity hepatitis to see um being a hepatic virus you know I would think of chemicals that affect the liver um manganese and cadmium that bioaccumulate the the one that's in the news all the time is pasas that bioaccumulates in the liver uh and is associated with a high risk of fatty liver disease so I would wonder what those chemicals are actually doing in patients with hepatitus C maybe it's uh increasing the progression of their disease and um and from nutritional perspecti
ve I mean obviously I'm sure you're you're thinking about alcohol but you know there are other uh aspects of the diet that actually like iron for example uh gets stored in the liver that may actually play a role in modifying the progression of the disease so yeah I think I think the sort of mental game I kind of played when I said this hypothesis should be tested this hypothesis should be tested if I was thinking of hepatitis yeah I think of certain liver toxins and instead of thinking of them a
s the cause of the liver disease obviously it's not the cause in most cases it's the hepatitis C but it may still be making the hepatitis C worse if you're if you have hepatitis C and you're exposed to excess cadmium or you're exposed to excess pasas so measuring those chemicals in in patients with hepatitis C may actually give some insights as to why some patients are worse than other patients thank you I think that concludes our session for today i' would like to thank Bob W for uh joining us
today it was very informative presentation and Mount Sinai is just down the road so there is plenty of opportunities I know some faculty in the school already work with you guys and uh and as well as part of the as asrc initiative so uh so again thank you very much and uh thank you everybody for attending the event and um making time today thank you thank you allate
Comments