DR. ROBERT GREEN ON THE IMPACT OF INDIVIDUAL GENOMIC DATA

 

Harry’s guest this week, Dr. Robert Green, is a professor of medicine and genetics at Harvard Medical School and director of the Genomes To People research program at Brigham & Women’s Hospital and the Broad Institute of Harvard and MIT. They dig into the individual genome, how individual genomic data is being used, and the impact of individual genomic data on various stakeholders in the healthcare system.

Please rate and review The Harry Glorikian Show on Apple Podcasts! Here’s how to do that from an iPhone, iPad, or iPod touch:

1. Open the Podcasts app on your iPhone, iPad, or Mac.

2. Navigate to The Harry Glorikian Show podcast. You can find it by searching for it or selecting it from your library. Just note that you’ll have to go to the series page which shows all the episodes, not just the page for a single episode.

3. Scroll down to find the subhead titled “Ratings & Reviews.”

4. Under one of the highlighted reviews, select “Write a Review.”

5. Next, select a star rating at the top — you have the option of choosing between one and five stars.

6. Using the text box at the top, write a title for your review. Then, in the lower text box, write your review. Your review can be up to 300 words long.

7. Once you’ve finished, select “Send” or “Save” in the top-right corner.

8. If you’ve never left a podcast review before, enter a nickname. Your nickname will be displayed next to any reviews you leave from here on out.

9. After selecting a nickname, tap OK. Your review may not be immediately visible.

That’s it! Thanks so much.

Transcript:

Harry Glorikian: Welcome to the Moneyball medicine podcast I’m your host Harry Glorikian. This series is all about the data-driven transformation of the healthcare and life sciences landscape. Each episode, we dive deep through one-on-one interviews with lea2ers in the new cost-conscious, value-based healthcare economy. We look at the challenges and opportunities they’re facing and their predictions for the years to come. 

My guest today is Dr. Robert Greene. Dr. Greene is professor of medicine genetics at Harvard Medical School and a physician scientist who directs the genome to people research program at Brigham and Women’s Hospital and the Broad Institute here in Boston. Dr. Greene is internationally recognized for research and policy efforts, accelerating the use of genomics data and precision medicine.

He graduated Amherst College and University of Virginia School of Medicine with mph in epidemiology from Emory University School of Public Health. His specialty training was at Harvard Medical School residences and he has a board certified in both neurology and Medical Genetics. Dr. Green led the first experimental trials disclosing genetic risk for common complex disease, the reveal study and first prospective studies of direct-to-consumer genetic testing PGEN studies.

He currently leads and co-leads the first NIH funded randomized trials of sequencing in adults called the Med seek project, in newborns called the baby seek project. An active duty US military personnel the MIL seek project. He is an associate director of for research of Partners HealthCare personalized medicine and is leading the development of protocols for return of individual genomic data results for All of us research program of the United States Precision medicine initiative and the Google verily project baseline.

Dr. Green also co-founded a company called medic genome medical, a telogenomics technology and services company providing genetic data expertise to patients, providers, employers and health care systems. He also advises a number of companies, many names which you may all recognize. Applied therapeutics, Helix, Ohana Biosciences, Optra health, Prudential, Verily and Veritas. Dr. Green, welcome to the show.

Dr Green: Harry thanks so much for having me.

Harry Glorikian: Dr. Green, I read a lot about your background here for everybody on the show and I’m just curious. You know can you tell everybody on the show what is medical genetics when we think about it?

Dr. Green: Well you know there is this specialty of medical genetics. It’s just like a specialty of surgery or infectious disease or neurology or ob/gyn, it’s just that there’s very few people in it. And it is a two to three-year residency that generally people take after they’ve done pediatrics or internal medicine or some other specialty.

I myself am board-certified in neurology so I had that under my belt and then I came back actually mid-career and retrained in medical genetics. So, I put that on top of my neurology background but I really focus on being a medical geneticist right now.

Harry Glorikian: And so, you know what is a medical geneticist do?

Dr. Green: A medical geneticist really conventionally looks at rare diseases, looks at children often children, sometimes adults. Who appear to have a disease that is genetic in nature so they have a family history of abnormal heart or a family history of an abnormal skin condition or hereditary hearing loss or any of the myriad rare conditions, that we know are subsumed by usually single gene disorders, monogenic diseases. And that’s traditionally been the place where medical genetics has been so it’s been this little specialty tucked away often kind of in the background of Pediatrics.

But now I think there’s a certain new brand of medical geneticist and it’s also other specialties as well, that’s really talking about the ways in which genetics and individual genomic data can influence everyday medicine, can influence common complex diseases, can influence reproduction on a large scale, can influence all of the medications that we take from day to day. So, I think there’s a new specialty it’s not really formally named this but I think of it as a genomicist, really someone who’s interested in applying the large-scale benefits of the genome to human health.

Harry Glorikian: Yeah so, when you’re talking to someone who’s not from the healthcare industry, how do you describe individual genomic data in medicine? And let’s leave you know the oncology and all the stuff we hear about all the time, when people are really desperate for say genomic data but the average person and how this data would be used to help them, I don’t know live a healthier life understand how to eat, relative to what their genome might say. How to prevent disease?

Dr. Green: Yeah, I often start by saying you know how long we’ve been talking about genetics and DNA in in our society but when’s the last time your doctor said, hey let’s check out your genome and see what diseases you’re at risk for. And almost no one has had that kind of clinical experience and that’s by way of sort of bringing home the point that we’re not using medical DNA testing in our day-to-day healthcare right now.

We simply aren’t, there’s a number of factors we can talk about but you know what most of them are is – it’s still relatively high cost. It’s not reimbursed by insurance, there isn’t a sufficient evidence base that is in place to convince the academic recommenders of medical care that this is a valuable component. And yet Harry, everyone sort of anticipates that this is going to happen. So, there’s a strange dichotomy right now, where nobody’s really using it for day-to-day medicine and everybody believes it’s going to be used for day-to-day medicine. And a lot of my research is how do we get from this point we are in now to that point that we all expect to happen? 

Harry Glorikian: Well no but we have companies like 23andme and you know if you remember deCODE genetics back in the day and so forth. And they seem to be playing on one end of the spectrum and then clinical practice, actual using it in the daily life of an individual seems to be on another. Now there seems to be a lot of new companies coming up for let me do your genome and you tell you eat better and so on and so forth. What’s your take on this this continuum that that you see there?

Dr. Green: Well the early direct-to-consumer genetic testing companies have democratized the notion of sending away your spit and getting something back and as you know those are really focusing on genetic markers, so called genotyping which is the low-cost genetic information, we can get right now. But I think we’re moving to a place where you’re going to be able to do much more extensive sequencing and other omics of your DNA. And you’re going to get a lot richer more personalized information out of it and we’re also moving to a place where that information gets fully integrated into the day-to-day practice by your doctor and what I mean by that is when you’re a little baby, this predicts whether or not you’re at risk for a childhood disease. When you are a teenager and young adult, this gives you insight into what recessive carrier variance you’re carrying. 

So that you can plan your own family without having a devastated child. When you’re in middle age it’s giving you insight into common complex diseases like type 2 diabetes risk, and heart disease risk so that you can really hone in on lifestyle choices, diet and exercise and preventive medications that can keep you safer.

And when it’s – and throughout the course of your life when you do encounter medications, it’s helping you understand whether you start at a higher dose or a lower dose and which of those medications is likely to have adverse effects for you. One of the reasons it’s so hard to collect data to justify the value of sequencing and other omics is because their value amortizes over decades, even a lifetime.

And I think that this is really something that’s very hard for our medical care evidence gathering system to contend with. How do you – if I tell you that the greatest value in your genome might be to explore it when you’re a newborn baby and then to have it available for every doctor visit and every medical interaction for the rest of your life. How do you actually prove to the world that this is beneficial and cost-effective?

You almost have to believe in it rather than be able to prove some sort of data gathering over 20 30, 40 or 50 years and so that’s part of the fascinating intellectual struggle that my research is dealing with is – how do we get proxy markers of value and by value of course we mean effectiveness, that’s clinical. Limited number of adverse effects and cost-effectiveness, how do we how do we prove that value to the world and accelerate that use of individual genomic data to help human health?

Dr. Green: Well some of that is actually I mean if we think about it it’s the end right? It’s the number of genomes that we have and if I if I go around the world everybody’s got their – I have five, I have ten, I have a hundred, I have ten thousand. They’re not necessarily aggregated in one place, one of the other problems is how did you do the work and how did you get to the information that you got to – that’s a whole other issue that I think that we need to standardize.

But so how do you see individual genomic data in medicine affecting different areas of healthcare, sort of across the different specialties and how do you see a changing medicine?

Harry Glorikian: Well I think you’re absolutely right there’s we must be approaching millions of people who’ve been sequenced for example in some way or other, but very few of them have had a comprehensive analysis of mendelian, that’s at monogenic risk factors, polygenic risk factors, pharmacogenomic risk factors. Everybody’s had little pieces of this and very few of them have had it fed back to them personally or almost no one has had it integrated into their day-to-day practice of medicine. 

So, one of the things I think your question illuminates is how fragmented we are in the practice of medicine. You know you’ve got your OBGYNs who are dealing with the pre pregnant and pregnant woman but not really thinking about cancer predisposition. You got your oncologists who are thinking about cancer predisposition but they don’t want to have anything to do with the pharmacogenomics of an anti-epilepsy drug.

So, you’ve got your entire-we forget how much our field of medicine is fragmented and therefore like the proverbial elephant they’re all looking at the genome and at the other omics from their tiny little corner of intellectual space and trying to again – trying to communicate and advocate for the value of the entire genome over time, over decades is something our medical care research system and our medical care clinical system is just not well suited to pursue.

Harry Glorikian: But it’s interesting that you say that and I think to myself okay well if I – the data is the data. And so what if there was a whole series of apps that looked at the same piece of data and but were designed to give you a different answer in one system right? So, you know when I’m driving my car there’s one engine but there’s a number of gauges to tell me what the different pieces of the engine are doing but it’s all integrated into one system per se.

Dr. Green: And funny we should use that analogy because many people talk about a dashboard right for their health care or for their individual genomic data. So, that’s a great analogy.

Harry Glorikian: Yeah so, I mean even if we aggregated different pieces of information from different you know medical specialties. You could give that to a you know informatition that was you know steeped in this and design something that might get them to the next answer so –

Dr. Green: I think that’s right but think about that I mean that’s almost a retooling of the entire orientation of our healthcare system. Which as you know well is really oriented towards sick care not health care it’s really oriented toward responding to people with mid or late stage disease rather than preventing illness, we’ve just started a preventive genomic data clinic at Brigham Women’s Hospital. 

And the idea is not that individual genomic data is the be-all and end-all of prevention and there’s lots of other things involved but genomic data is an underrated component of prevention because there are probably 10 to 15% of people out there walking around who are carrying a single gene risk variant, that may imply implicate their health in the next couple of decades. There is – everyone’s carrying recessive carrier traits that are relevant to their reproductive health and almost everyone is carrying some kind of atypical pharmacogenomic variant that could influence drugs, that they’re going to get at some point in their life.

So, we’re seeking to try to learn how to arm people with this preventive information now you know for you Harry and for other futurists people who are really looking ahead, this is like what why hasn’t this already happened? Why aren’t people doing this already? And all I can tell you is that the health care system is relatively resistant to change.

We’re a fairly you know we’re a fairly slow-moving cultural behemoth of a system that just doesn’t turn on a dime.

Harry Glorikian: Well I was you know I was thinking that you almost are forced to have an external party come in and we’ve heard a lot about you know Apple and Amazon and so forth. You know Apple making an EMR now portable on your phone. You know they almost now become the conduit to could you actually get a you know piece of your genome or your genome in your phone, if it was small enough and could you transfer that to another piece of software, that would then do the analysis for you and I think I see that coming. But how do you see this changing drug discovery or Diagnostics or treatment of patients? I mean I I’ve always believed that once you – the datafication of an industry takes place.

That there’s no one business model that stays the same that it is totally up for manipulation and I don’t mean in a bad way but and that – the monetization of it doesn’t have to be CPT codes and reimbursement and everything else, the way that we think about it. So, how do you see the individual genomic data changing these areas and does the revenue model change?

Dr. Green: Yeah, I think you’re right I mean we’re in this fascinating period where companies are launching, if you like experiments. So, we’ve had the direct-to-consumer genetic testing experiments which you know millions of people have signed up for. We’ve had other companies that are not truly direct-to-consumer but are what I’m calling consumer directed physician mediated companies.

So, you know they advertise to the consumer the consumer interacts with them over the Internet and then they either hire a company like ours, Geno medical to be your advocate for ordering the test and interpreting it or they have access to their own physicians, that they have ordered the tests for you. Then you’ve got systems that are trying to be more forward-looking. So, you’ve got Geisinger and you’ve got Northshore in Chicago and you’ve got Sanford in South Dakota and you’ve got these systems, that are trying to be leaders usually with some philanthropic support or in the case of Geisinger, a big shot of funding from industry.

They’re trying to really integrate individual genomic data on the front end of your care and so those are sort of more conventional experiments. Then you’ve got the precision medicine initiative, the All of us research program. Which is the ambitious attempt by the United States to create a 1-million-person biobank for the entire country and to genome sequence every single person, after having collected a hugely diverse collection of Americans. 

And we’re as you said in your introduction, I’m helping the group design a return of – a basic return of results strategy for the individual genomic data component of that. So that people have access to some of their own information and all these things are driving just as you said new models, new ways of thinking about it. And Apple the whole Apple rollout with their enormous and loyal customer base, new ways of thinking about the electronic health record, new ways of digitizing your moment-to-moment healthcare information.

That’s really exciting part of it, so whether we’re talking you know how many steps you’ve walked or your Apple watch collecting cardiac rhythms or genome sequencing and integrating that into day to day healthcare. I really think you’re right, new data approaches are driving new ways of thinking about healthcare. Now I tell you Harry where this excites me and where it worries me it excites me for the reasons you suggested because I think it’s going to force a revolution of our healthcare system it’s going to force a reconsideration of our reimbursement model. but it worries me in that so much of it is commercially driven.

I’m very committed to the all of us project because it’s an NIH project that’s really trying to do this without any kind of commercial pressure but so many of these other initiatives are commercially driven. Which is great, they’re efficient they’re thoughtful, they’re creative and they’re fast-moving and nimble but they also have at their root, profit motive.

And there’s a it’s very easy to blur the line when you’re when you’re a data is a advocate it’s very easy to blur the line between what is really scientifically valid and what allows you to sell more product. And this is never more salient than in the wellness space, the space around diet and exercise and we’re seeing all sorts of sort of data-driven initiatives around diet and exercise some of which struggle to have a real scientific basis and quite legitimate and others are really rather sadly approaching a kind of fraudulent space.

So that’s a long-winded answer but I think that the area is tremendously exciting but we’re gonna have to be vigilant about putting too much of our future in the hands of commercial entities. Who, well of course they want to do good they also have a very strong motivation to be financially solvent and financially profitable and I think sometimes those motivations are gonna be in tension.

Harry Glorikian: Oh no I totally agree and I’ve always believed that and and fee-for-service, I think is what does the almost the opposite to a certain degree right? Is you’re not always at the push to change in and adapt and be on that next frame, whereas a fully competitive environment is Apples coming up with a new chip, you know Nvidia is gonna come up with a new chip. Everybody’s trying to one-up the other right and push the envelope because that’s what makes them competitive. Competitiveness is not what we necessarily want per se in health care but if you had a value-based model where you’re paying on outcomes, now  you’re sort of – the system gets pushed along to do the best and and measure everything, and compete accordingly.

So, I’m hoping that as we move in that direction for me you know US perspective that we don’t miss the opportunity to change our system to a value-based model. I’m not saying it’s going to be easy it’s going to be incredibly –hard

Dr. Green: Oh, it’s going to be so painful. 

Harry Glorikian: But I think that in certain areas we are moving in the right direction and Geisinger wanting to incorporate individual genomic data, their self-insured systems.

Dr. Green: That’s right

Harry Glorikian: So their incentive is keeping their population healthy, the healthier they are, the more profitable they are. And-

Dr. Green: Exactly and we’re finding this in a number of arenas like that for example you take very large companies, who have essentially supporting the healthcare cost of their entire employee base. They also recognize that anything they can do to convert themselves from a reactive to a preventive posture on health, is going to be down to their bottom line and to the health and happiness of their employees. And so I think we’re Geno medical for example is really talking to these employers about making individual genomic data services available to their employees.

And we’re not talking necessarily futuristic or experimental Omic services, we’re talking their employees who by existing standards say have a family history of cancer and should be tested for carrying a hereditary cancer susceptibility mutation, and have never even known that they meet those criteria. So, we’re talking about standards that are in place today, we’re not even talking about sort of futuristic sequencing for downstream disease.

Harry Glorikian: Well let’s – you know let’s talk about you know what I think is is gonna move the field forward even faster. How do you see you know the AI, machine learning moving things forward? You know I, we talked a little bit earlier before we started the podcast about Joel’s Dudley’s paper and how herpes simplex six you know could be a major driver of Alzheimer’s, where you would have never you know dawned on anybody that that could have been the culprit, and so you treat for herpes and you know – not cure somebody but are able to actively treat somebody for Alzheimer’s. Is a huge step and that was all driven by you know data correlations and sticking the data in and wasn’t a human brain necessarily that came up with an answer.

How do you see this new technology driving that change based on the rapid evolution of the software?

Dr. Green: Well of course I think it’s incredibly exciting and I think what people should keep in mind is that it is most salient for hypothesis generation, and what I mean by that is we have known for a long time that correlation does not equal causation. Right now if you put a big data engine together and tried to correlate alcohol intake and lung cancer, you’d find a tremendous correlation. Why? Not because alcohol has anything to do with lung cancer per se but because alcohol is related to smoking which has to do with lung cancer. So, if you have smoking in that equation you can tease that out, if you don’t have smoking in that equation you actually end up with the wrong conclusion.

And your AI confidently predicts that those things go together. So, I think we need to respect the ability of large data algorithms and artificial intelligence, active and deep learning paradigms to generate questions but I don’t think we can abandon the proactive, the randomized clinical trial as so many people have sort of advocated for. 

They’ve predicted that the ear of this sort of slow painstaking methodological data collection is over. I mean I do think we’re going to have to be very selective about what we invest in for those traditional methodologies, but we also have to be cautious about the ability of algorithms to draw conclusions that are not necessarily true. So, Joel’s paper I think is a tour de force and fascinating sort of resurgence data-based resurgence of hypotheses, that have been around for a while that viral load particularly herpes virus might be associated with either the development of Alzheimer’s disease or the acceleration of symptoms.

And I think it’s well worth looking into so but we shouldn’t we shouldn’t conflate that with the notion that it is established truth, because the algorithm has found it. So, I think if we keep in mind that artificial intelligence and big data is a fantastic hypothesis generating tool, we should be in great shape.

Harry Glorikian: So, do you see any systems or platforms that that you think are driving you know that the bleeding edge at this point?

Dr. Green: I think well obviously big data writ large is driving this, the standardization and phenotyping of electronic health records into ontologies, which can take the crazy subjective nature of a doctor’s note and either by computer algorithm natural language processing or other situations, can sort of create structured data files from that. The collection of phenotype data, both genomic data, activity level, biochemical.I’m looking to get one of those little glucose monitoring things that you can actually stick into your skin and it just sits there on your skin with a little probe just under your skin, it doesn’t hurt and it tells you when you eat something how your insulin is spiking and how your blood sugar is changing.

And apparently there – this is not my field but apparently there are genuine differences in how people – insulin and blood glucose spiked to the exactly the same carbohydrate load. So, you can see if you’re one of those people who’s who is more likely to have those kinds of spikes. At the same time that we collect all this data we’ve got to be a little bit patient as it gets collated, integrated, analyzed and proposed for interventions.

We can’t – there’s gonna be lots and lots of people out there showboating and saying oh this is the answer. This is gonna make you healthy this you know this one thing or this one channel or this one product or this one device. So, I think it’s going to be an iterative process. I think medicine is going to be stubbornly slow to change. Which is bad because we need to change more quickly but it’s only good in the sense that I hope we keep bringing the lens of evidence and outcomes to the table.

So, that there is this balance between innovation and commercialism on the one hand that’s surging forward, tradition evidence-based a bit of conservatism in the traditional medical structure that’s kind of anchoring it but hopefully moving forward with it. And I hope this is a healthy tension that moves us forward into a new era of preventive medicine and preventive health.

Harry Glorikian: I’m totally with you I I do believe that industry if nothing else as well as a shift to value-based medicine. Assuming people down in DC don’t accidentally blow up the whole system is gonna move us in the right direction and I think will be an example for the rest of the world to follow. So, I see I mean a potential huge shift there – which is I think going to affect every piece of the healthcare industry. Everything from basic research all the way through to drug discovery and managing patients and everything. 

Dr. Green: And you asked about drug discovery before and I didn’t say much about it, but this is again not really my area but I think it’s intuitively clear that when you collect a lot of this medication of this medication information, when you collect a lot of this genomic data, when you collect outcomes information and you are able to cross-pollinate this. That you’re number one, able to use current drugs more effectively and figure out which situations they work in, and which situations they don’t.

But secondly your genomic data has the opportunity in particular to tell us about extreme cases which then teach us about new drugs we can use for garden-variety cases. So, the case in point that you know well is the PCSK9 story, where there’s this rare set of mutations that causes people to have extremely low LDL levels. LDL cholesterol levels, and by analyzing that pathway and those few outliers, there has been the development of some new anti-lipid medications so that’s an example of ways in which we hope we’re going to be able to better treat diabetes, heart disease, cancer all sorts of common diseases.

Harry Glorikian: So, I’m sure you’ve seen you know there was a bunch of articles recently about how precision medicine is not achieving its goal. And there’s a gentleman I forget his name now who is this sort of a gadfly of disputing that this does that and do you do you see examples where that’s just we’re seeing the opposite, we’re seeing the forward movement of this technology.

Dr.Green: Yeah you might be talking about Sandro Galea at who is the Dean of the School of Public Health at Boston University, a colleague and friend actually who I’ve invited to debate this a number of times and he writes very eloquently and cogently about the divide in socioeconomic status, the potential improvements that we could garner in health with some less sexy interventions around obesity, smoking cessation and things like – basic things like blood pressure control. And I guess my position is that that he is correct but that both are true that we should be investing in those basic health interventions and we should be pursuing new paradigms for our entire healthcare system, driven in part by a cutting-edge scientific discovery. 

I think he and others do point out that if these new techniques are only available to people who are able to pay out of pocket for them. If our health care system goes into bankruptcy at the same time it’s costing more and more for you to buy fancy gadgets and get your human genome sequenced to get your individual genomic data, and you know, have day-to-day monitoring of your blood glucose.

If those two things are going in the opposite direction and regular people can’t even afford regular health care, we’ve got a problem that I think echoes general societal problem of income distribution. We’ve got the haves and the have-nots, not only in terms of wealth but in terms of health care. And I think that’s something we really want to try to avoid and I’m not sure what the answer is but I don’t think the answer is reining in precision medicine or somehow turning down the enthusiasm on genomic data and revolutionizing healthcare. I think the answer is bring these costs down accelerate our use of data, make sure that we’re getting good outcomes and then have the will to implement novel interventions across Medicare, across Medicaid and across Obamacare safety nets throughout the country. Now that’s a big order, some of these right now are very expensive but if we if we put our mind to it – I think we could get there and the example is cell phones.

You know cell phones, smartphones have come down to the point where they are affordable to almost everyone in society at this point and it’s something that almost everybody carries around with them and wouldn’t really dream of living without, and I think a evolved health care system could eventually be the same.

Harry Glorikian: Well on that note I want to thank you for your time, it’s been wonderful as always. I wish you well in all of your journeys and the success in in all these projects that you’re doing. I know that you’re trying to drive forward answers and bring things to help patients live a healthier happier life.

Dr. Green: Thank you very much Harry it’s been a pleasure speaking with you.

Harry Glorikian: Thank you and that’s it for this episode hope you enjoyed the insights and discussion, for more information please feel free to go www.glorikian.com. hope you join us next time until then, farewell.