 Welcome to our panel that we will have today to accompany the Purdue Engineering Distinguished Lecture Series. I'll simply introduce our moderator for our panel, Dr. Bill Clark, who is alumnus of our School of Chemical Engineering, also MD from Indiana University Medical School, and has spent his career in medicine and nephrology and medical devices, including a distinguished career at two medical device companies. So without further ado, let me hand the floor to Bill to introduce the panel. Thank you very much, Dr. Kim, and welcome everyone to the panel discussion after such a stellar lecture by Dr. Janseth will focus the topic now to precision medicine and how biomaterial development kind of fits into that context. The agenda is shown here very briefly. I will give a very short overview of precision medicine, what exactly it is and how biomaterials play into that. We'll have an initial introduction of the panel members and then more formal introductions before each of their short presentations. And then finally we'll have a period of at least 30 minutes, hopefully of questions and answers in general discussion. And by the way, we do have an absolutely hard stop at 5.15 because Dr. Janseth needs to head back to Indianapolis to the airport, so we will have to abide by that very carefully. So this again is just the brief introduction, Dr. Janseth after the very nice introduction by the Dean, I think you know her qualifications. And again, after such a stellar lecture, I think we all have a keen appreciation of how accomplished her academic career has been. And our other panelists, first of all, Dr. Lu and Dr. Solorio will provide faculty research lab perspectives on the role of precision medicine in terms of what they do. Dr. Breitman will provide a bioethics perspective, and finally Dr. Hiles will provide an industry perspective. So first of all, personalized or precision medicine defined. There's no absolutely agreed upon definition, but one take is the combination of clinical practice, at least how it's established by existing clinical data that have been generated from clinical trials involving large number of patients. And taking that information in combination with newer molecular based data, combining that all together and providing more specified and specific therapies to patients. This approach has been already widely used in the pharmaceutical industry, and as you can see, there's a sense that the pharmaceutical industry will even adopt this approach more broadly in the future. The components of precision medicine essentially touch most aspects of the clinical care of patients. Everything, for example, as just one example, risk assessment in a patient who for instance has a family history of some type of genetic disorder. Broadening the scale a little bit, it can be applied to on a larger population detection of a certain disease in high risk population. And then finally, firming up the diagnosis in a specific patient, providing a treatment plan, and then importantly, ensuring and follow up that a certain outcome has been achieved, and that outcome is a durable and robust one. So there was an absolutely fascinating position paper that was written by Dr. Anseth and colleagues, so it's very timely that we're discussing this issue and that she's involved in our panel discussion. So the general idea here is again, is how do we take whether it's a drug or a medical device or a biomaterial that has been clinically qualified in sometimes a clinical trial that involves hundreds or even thousands of patients that obviously have significant biologic variability and heterogeneity. How do you take that particular biomaterial device and customize it so that it's actually very appropriate and specific in a given patient? So that's the challenge that Dr. Anseth and colleagues tried to address in this paper. And essentially, it was kind of a call to action to engineers to start thinking about how precision medical medicine principle should be applied in further development of biomaterials. So the previous figure and the next two figures are drawn directly from the paper. These are just four examples of how this principle of applying precision medicine ideas to development of biomaterials may play out. First of all, material chemistry, and by the way, going from left to right is going from more general to more customized. So the approach on the left for material chemistry is just a relatively nonspecific diffusion of a drug out of some type of polymer matrix all the way to the right. The most customized approach would be a system that reacts to a patient specific enzyme. Another example in terms of fabrication would be a more customized approach would be a 3D printed, for instance, organ. And then even further customized would be allowing for that organ to kind of grow as the patient evolves in his or her clinical disorder. Thirdly, bioactive components in the transplant world and allogeneic transplant, even though it may be, so to speak, well matched, there are still some difficulties that may result once the transplant is done. Autologous transplants of patients' own cells would be an improvement. And then even further and more customized would be an approach in which the cells that are being transplanted actually would be adapted so that they would behave in a certain way in the milieu in which they're going to live in the newly transplanted environment. And finally, we all have heard much about big data and the application of that to medicine. And certainly precision medicine plays a role there. It's anticipated to do so in the future. And again, these are examples of going from fairly general on the left to much more specific and customized on the right, where an in situ approach of some type of data tracking system could provide real time evidence and data in the progression of the disease or maybe the monitoring of some type of intervention in terms of the patient's response. So with that very brief overview, as I mentioned, we'll have two faculty members who are laboratory researchers who will provide their perspectives on how precision medicine may apply to what they do. The first laboratory researcher is a colleague of mine from chemical engineering. Dr. Julie Lu is an associate professor of chemical engineering and also biomedical engineering by courtesy. Her laboratory is focused on biomaterials, particularly as they apply to musculoskeletal disorders and surgical applications. So I will hand it over to Dr. Lu. I really just wanted to show you two examples from our laboratory on some problems that we're working on and how the personalized aspect could be applied to this, to these two projects. So the first is Ashley work by Changi Lin, who's in the audience and is defending next week. He's been working on designing these redox responsive hydrogels. And the idea is this is actually a very similar polymer to what Christy talked about. So it's a peg polymer. These have the thyle groups and what he had done is he wanted to make a one-pot reaction where you can have both reducible bonds shown in gold and then the non-reducible bond shown in red. The idea was often if you're using this thyle chemistry, if you get increased mechanical properties, you can't really decouple that from how they degrade. And so he wanted to combine both the non, so basically the permanent and the temporary bonds, and he was able to get gels that had the same stiffness, but you could tune the degradation rate based on how many of the reducible bonds you have. And the gels are degrading in response to reducing conditions. So kind of thinking about how this could be then applied. There's some tissue engineering examples, but the one I have here is maybe a little bit more related to drug delivery. On the left, stiffness might be important in drug delivery if you're thinking about having force actually induce the release of your drugs. So if you have something very stiff it might not compress as much for the deformation which would then release the drug, or in the converse when you're stretching it, depending on how stiff it is, you might have a different release in terms of the molecules that are embedded inside of your hydrogel. On the flip side, so if you're decoupling the stiffness and the degradation, since these are redox responsive, I think it's been well known that tumor cells themselves have a higher intracellular level of glutathione, so a reducing condition. But some recent work also has shown that the extracellar matrix around some of those cells seem to be in a more reducing condition if they're going to be metastatic. And so I think this could be an example of having some sort of depot that could basically, based on how reducing of an environment your tumor is, that's correlated with metastasis, would then deliver some sort of molecule or drug. So that's my first example. The second one kind of ties into this cartilage engineering. So my student Claire Kilmer, who's sitting in the audience as well, she's been working a lot on developing protein-based scaffolds to improve the repair of cartilage that has degenerated due to osteoarthritis. We've worked on these gels in cell culture, so in an incubator, but and looked at how cells respond, but she's also done an animal study where she basically harvested the bone marrow from individual rabbits, so the model here would be the rabbits, the patient, so if you were the patient you would harvest the bone marrow there, culture the cells and then expanded them and then embedded them inside of this gel that was then placed into the defect of the knees that we made into these rabbits. And I'm just showing some histological sections, I wasn't expecting most people to be familiar with histology, but the idea here is that the first column it says collagen call one slash two, that's our kind of improved hydrogel and when you put it in the defect, the defect is marked by the arrowheads. The proteoglycans which is indicative of cartilage matrix shown in pink, you see a lot more of that forming compared to a control of just collagen one which is often used or the empty defect where you don't fill it just to see what would happen, what does the animal do? And so kind of thinking about the personalized aspects of this, certainly we're thinking right now of using autologous cells, so cells that you would gather from yourself, the patient, and in this case we actually preform the gel but I think one of the things we're interested in going towards is maybe thinking about if you have this defect do you want to do bioprinting for cartilage or maybe it's simple enough to just inject it but it'll kind of conform to that patient defect. So there's a little bit of personalization rather than just buying some material that's prefabricated, maybe the surgeon chops it a little bit but here it's just trying to actually conform to the defect size. Those are my two examples. Thank you Dr. Lu and I should mention at this point also that although we didn't want to overly burden Dr. Anseth with another asking her to prepare another presentation for the panel of course we'll welcome any comments or input that she has along the way for sure. So the next panelist to speak is Dr. Luis Solorio from the Weldon School Biomedical Engineering Assistant Professor. His laboratory is very focused on cancer modalities not only understanding of the underlying pathophysiology but also developing platforms for instance that screen for effective therapies. So Dr. Solorio. Thank you for having me. So what I'll talk to you about today is this idea of cancer metastasis and so when we think of cancer I think we're all pretty well versed with this idea that it's pretty terrible disease and it affects lots of people very tragically. What a lot of people aren't aware of is it's not actually the primary tumor that tends to be the cause of patient mortality but it's actually the metastasis that causes that and so what my lab really focused on is kind of trying to understand the steps in the metastatic cascade and where we can use engineering technologies to kind of identify and really find effective treatments to prevent this effect. Now when we think about metastasis classically we think about this population of cells that are growing roguely right completely out of control and then they undergo some sort of stress or this could be drugs it could be the environment something is driving them to transition into a new state and then they leave the environment go someplace else and then seed and then ideally they'll switch and they'll grow and it's this process that we think of this accumulation of mutations that starts to change how they respond to drugs. Well it turns out that this classical pathway is is not really correct it's it's kind of old this is the old thought this is and and researched by David Leiden's group Julio Aguiereguiso has really shown that it's actually much more complicated so it's not just a single collection of primary tumor cells but it's it's really a whole physiological system that's working in concert and it's these interactions that change things and not only do we have to think about this time course of events and that metastasis is this in-stage event no it actually happens before there's a poppable tumor so they're actually early disseminated cells that will escape well before there's anything detectable so you have metastasis that occurs throughout a person potentially throughout a person's lifetime they may not ever outgrow but they they could happen and what we're seeing is that this idea of organotropism or tissue specificity develops in response to this so the cancer cells actually release things that create what we call a pre-metastatic niche so this is a small little tissue space that's very hospitable to the cancer cells and allows for metastatic outgrowth and so we we kind of started to focus on this concept of what is this pre-metastatic niche and so what we find is that if we look at a normal healthy lung we see lower levels of fibronectin expression but then during early metastasis we start to see this fibrillar network of fibronectin form and it's this fibrillar network that actually it's this fiber aspect that actually changes these cells and it provides an environment for cells to actually grow and so what we wanted to do was develop a way to kind of mimic and recapitulate that aspect of the tumor space and so we developed a platform that allows us to take fibronectin which is just part of your blood plasma and then transition it from this globular completely inactive state into fibrils that actually drive that and facilitate that metastatic outgrowth and then using the system to create that sort of environment what we were able to show was that if we take cancer cells from a pleural effusion so these are a buildup of lymphatic fluid in a patient's lung it just happens it's it's a course of metastasis and it's very uncomfortable for uncomfortable for the patient it's this giant allure of cells no one really knows completely what's in it it changes constantly so we can take this solution and if we try and grow it on plastic this cells die they don't seed now if we take this synthetic pre-metastatic niche and then seed our cells on top of it we've now selected for that inherent population of cells and then what we find even more exciting is that we're actually able to enrich for the cancer cell population and this gives us a snapshot into a patient-specific cancer so during the time course of all their treatments they're going to have this fluid drain and what we can do is we can sample this fluid put it on our small little chip platform and screen unbiasedly for thousands of compounds and so that's kind of where we're going with respect to the precision medicine aspect is using this approach to really tailor specifically to these cells that that are constantly mutating and different from patient to patient and so it gives us a real strategy really the idea is to give clinicians the tools to narrow down which drugs they potentially would want to give the patient when they have thousands and thousands of compounds to choose from and it just kind of helps to narrow down that selectivity so that hopefully we can actually improve the patient outcome thank you very much the next speaker is Dr Andrew Breitman from the Weldon School of Biomedical Engineering and he has Andrew has a very strong background in among other things in engineering ethics and has led the effort in the School of Biomedical Engineering to make sure that this perspective is incorporated into the curriculum so he will even though precision medicine presents many fantastic opportunities there also are some a lot of issues that have to be raised from an ethical perspective and Dr Breitman will address some of those thank you very much yes I think there's a number of aspects that we could look at I chose to think about personalized bioprinting as one area to consider the ethics Dr Anses work making those materials that could potentially become bioinx and then printing used bioprinting into tissues or organs is a field that is an emerging technology and has a lot of potential in medicine I like that quote that bioprinting 3D organs could have the same revolutionary and democratizing effect as book printing in their applicability to regenerative medicine in industry and I think particularly to personalized medicine there's two aspects that I think are important to consider when you're looking at a bioethical perspective one is the identification of what the ethical issues could be and the second is then what do you do about them or how do you respond to them and in the first aspect I take an approach and encourage this among our biomedical engineers that's built from biomedical ethics using the four principles of biomedical ethics promoted by Beecham and Childress and in their classic work on principles of biomedical ethics we use that as a framework to identify issues the second I want to consider briefly is just a framework of responsible research innovation this idea that two aspects of it one that we if we think about the biomedical ethical issues early in the design and innovation process we can end up actually with better innovation at the end and part of that thinking about those ethical issues is who gets involved and how to incorporate them and I'll say a little bit right at the end so if we use the four principles of biomedical ethics as a framework they are beneficence or to do as much good as possible and non maleficence which is to do as little harm or no harm as possible the other two on the next slide are justice and respect for autonomy so I just tried to identify a few issues most of these are from the published literature you can probably think of others but that are around the beneficence of or the good that can be done with bioprinting in a precision medicine framework certainly they are allowed to address very specific patient needs you can bioprint an organ that could be very tailored to that unique individual there'd certainly be a reduced need for organ donation if you can create these organs without having to sacrifice anyone else's organs or you would avoid xeno transplantation and all the issues that come from that both cultural and in biological there's a reduced need for immune suppression if you're using the patient's own cells and non-reactive biomaterials certainly providing organs and tissues that could have an adaptability or a growth potential would be great for pediatric applications creating better disease models for testing whether it's new drugs or tissue engineered products and certainly providing alternatives to animal testing would be considered a beneficence issue on the other side there's possible harms or risks of harms there is risk of rejections if you're creating organs bioprinting with using some mixture of animal cells in there there's a possibility that you are running a higher risk of failure these high complexity bioprinted organs if they're going to be functional they're high complexity so they're a lot harder to both develop and then to test certainly there's risks of unknown side effects so longer animal testing and clinical trials might be necessary before they could get into a patient limited irreversibility once bioprinted organs are implanted they get incorporated and then if you've done something wrong they're in the body now and how do you reverse that might be impossible high cost of development and translation with any new emerging technologies you have to consider the cost and the opportunity costs and that could be a potential harm and certainly managing public expectations and hype around a potential emerging technology that might take 10 years before it's ever or longer before it's realized on the on the justice side which is the just distribution of health care and benefits and and risk of harm a few more issues access to advanced technologies may be limited only to those who can afford them or to countries where the infrastructure to develop them and deliver those bioprinted organs legal ownership of the tissues who owns what's been printed do you own your own cells to the company that used them to the hospital that made the tissues for you the rights to market them what if we're designing personalized tissues and then marketing them more broadly if they had broad applicability who gets the the right to market and the funds from that and certainly there's issues around personalized testing and how long and complicated that might be and the respect for autonomy also comes in this right to make decisions so it is a patient get to make decisions about how bioprinted organs made from their tissues are used do you ever have the right to grow your own tissues and then do what you want with them create and that possibility of you know home printing is probably quite a lot longer away than doing it commercially but it's still a possibility creating enhanced organs is another issue that we have to think about what if we want not just disease repair but we want our muscles to have faster twitch or our organs to do specialized metabolism or do we have any rights or regulations to print anything that we want and implant it or use it so the last piece is just a framework so if we have these issues how do we respond to them how do we begin to specify how risky they are or how detrimental they might be and who gets involved in that decision and then how do we regulate that and so the responsible research innovation framework suggests that more than just the technologists the engineers and the clinicians who are involved in this technology you need a public opinion you need social scientists you need ethicists to be involved in the early stages of this kind of innovation to collectively assess the risks mitigate and manage them and be in charge of the overall regulation. Thank you very much and the last presentation is by Dr. Michael Hiles from Cook Biotech Senior Vice President and Chief Scientific Officer there also an adjunct professor in both Weldon School Biomedical Engineering in the College of Veterinary Medicine has a very strong history and relationship with Purdue having gotten both electoral engineering and veterinary medicine degrees and he's been instrumental in that program for for a number of years and finally from an industry side and that's the perspective he's going to provide today is has over 30 years experience in the field that we're we're discussing so Dr. Hiles. Thank you for that kind introduction and for having me this is fun I should say perhaps now for something completely different because I didn't have a lot of pretty pictures are really focused on research instead I was asked to talk about what is it that might be the obstacles or the opportunities to commercializing this stuff so I put together a couple of slides just two about those and so here's a couple of points these processes always lag are all kinds of new research processes always lag behind the technology that actually gets commercialized so there's always a lag time and often a preponderance of safety has to be first established no current paradigms paradigms allow for all outputs of a given process and what I meant by that is is that over the years we've gotten more and more in vitro diagnostics to the point that the division of devices within the FDA has a whole special sub branch on in vitro diagnostics and I think what we're finding is is that as we're starting to get more and more I'll call it ex vivo personal process developer or personal production that we might need a whole new regulatory framework to regulate these things in other words if you could get a product by process that's a that's a terminology that a lot of patent lawyers use but a product by process and you get an approval for every product that comes out of that process if you can keep it within certain parameters then that might be a regulatory framework that would work for these but we're currently the regulatory climate is not mature enough to determine whether these things are devices drugs or something else and that's why I think maybe there's opportunity for a whole new framework because many of the examples that you've seen today both here at this part and at the original lecture as soon as you start adding something that has a biological function that starts to alter cell behavior it's a drug and to give you an idea at least in the FDA and in the world that we live in a drug takes somewhere north of 10 years and a billion dollars from start to finish whereas an average device you know typically maybe five years and five million dollars a big difference and so if you're just a scaffold without any bioactivity or a device at least right now but as soon as you start tailoring all these things what is it so we need to think about maybe that product by process type of regulatory that might be new but effective and then I just threw in some commercialization questions is the manufacture of end products now subsumed by supplies tools and software so if you think about it if you could bio print right next to the operating room anything that the patient needs then there's no reason for manufacturing the thing that patient needs or you're not manufacturing the endpoint you're simply manufacturing the device to build it and the supplies the inks that it uses and the software that it runs on so you've changed the whole paradigm of what manufacturing is how strict limits will be placed on each component in each process so nothing harmful can be created so that gets this gets back to the product by process and if there's any way to tune it so that you can't make or you can make something toxic or something dangerous then you want to inhibit that I have to tell you a quick antidote and that is that we were at a medical advisory board meeting one time we had a group of doctors there and we were talking about a particular device that we were working on and the doctor got up there and said and no offense to you but he said you got to make this device idiot proof because at the medical schools we'll just build a better idiot finally what tolerances and patient data will cause inappropriate outputs or contraindications for treatment so that's the other thing which is if we're going to do this all personalized medicine and use patients own data to then tailor the device you know what what parts of that data can be used and what parts can't be used and what parts might make the device non-functional or functional so just some ideas because at Cook Biotech we spend a lot of time looking at technologies and what we call a technology readiness level which is goes across a lot of industries but how close is it to commercialization is it is it years away is it months away is it decades away so thanks very much to all the panelists and before we open it up for questions and discussion I'll just check in with Dr. Ansis and see if fish if you have any quick comments or views again before we open it up more broadly well I think the points were very well raised and you know I think it's a challenge for us also to figure out is what is most impactful for us to be doing in this space as well as we go towards more precision and personalization we see all the advances in cancer immunotherapy so work with your T cells and there's car T cell therapy one patient one treatment is $500,000 versus should we be thinking globally and what impact can we do to better understand how different subsets of patients with lower technology can be treated and you know there's no right answer there but I think it's a challenge to think about as well okay well thank you again and we'll open it up for a broad discussion any pointed questions to any of the panelists presentations or your comments a good moderator always has some questions in his back pocket in case they're they're slow to come so I've got a couple so we we kind of touched upon the that how do you clinically validate these these types of new products because as we talked about for drugs you mentioned dr. hiles that it's at least a 10-year process and the phase three clinical trial could be 400 500 million dollars to to to look at two or 3 000 patients so again that's a heterogeneous group of of patients that but then the drug or the technology is only used in a given patient so looking forward I don't think that that paradigm is going to change but what additional clinical validation is going to be required for these for some of these newer approaches to satisfy the FDA and other regulatory bodies well I'll give a stab at that but it's a great question because regulations seem to be getting tighter not looser and the concept of personalizing something I think maybe has to start with all of the components being already approved so that you at least have a basis on which of safety to build and then say okay we're going to take these already approved products and mix them in various ways and then you know as I suggested maybe through a possible new regulatory pathway or some particular pilot program that they might put together specifically for personalized medicine and well and I think we saw one of the examples as well you know I think we need to find ways to fail earlier I think personalized approaches that allow more sophisticated in vitro assays are important but I also think on the flip side too there was a New York Times article about how personalized and precision medicine was failing the average patient or person because it wasn't really benefiting them so you know the average response and the average treatment can actually be harmful to many many people so how do we navigate that that space I have one colleague just as one example you know a woman is not a small man is the talk she gives yet we dose drugs based upon weights you know and and so I think there's at some level the economies of what what subsets can we look at and the markets can bear financial development for for different sub populations is is a really important thing and I think engineers have to play a role in that well I think one thing we're starting to do now though is that we because we're collecting so much data on patients that if there's a particular ailment like they have a hernia and they need a hernia repair and there's a dozen different types of devices that they might use they can go back and look at the patient and there might be some reason why they might be contraindicated for a particular therapy so now it's not just the physician deciding that oh I like this product better so I'll use it in everybody they are starting to look at the demographic of the patient and tailor the solution if it's already on the market to the patient but it's not being made next door so I would also say that while clinical trials were designed for testing both safety and efficacy protecting the patient safety is a high value we we do as the emerging technology need to think about new paradigms for clinical testing and part of that are engineers coming and clinicians as well working together to come up with these new forms of testing we we really need a testing science a revised testing science for clinical testing when it's a personalized medicine or otherwise we're going to have safety issues all the way down the road because you you can't generalize those kinds of technologies Dr. Ramakrsta. I'm curious about to what extent does FDA get involved in using personalized strategies for example let's say you base it on mathematical modeling and point out to the clinician that each individual patient there is a way to learn about the patient's special response to a given drug and if you were to come up with strategies for dosage will you could take the attitude maybe that could be used as a suggestion to the clinician and it would be up to the clinician's discretion as to whether use it or not or would the FDA get involved that even the clinician can't use it until we approve it it's a good question and I think it's a hard question but first of all physicians can't prescribe anything that hasn't been approved already with very few exceptions there are some emergency uses and things like that but very few exception but the FDA also very carefully says we do not regulate the practice of medicine and what they say by that is is that we will not stand between the doctor and the patient and so if the doctor prescribes a drug that is supposed to have this regiment and instead he prescribes that regiment that's between him and the patient it's FDA doesn't step in but the drug couldn't have been there at least unless it was approved for a particular thing which gets to a whole other discussion about off-label use of medical products whether it's drugs or devices and a lot of people want to outlay off-label use but if that's true then for diabetics and for and for pediatrics or children or whatever a lot of things would go away because none of them were made for that in the first place and I can follow up with that on some level because when a clinician so say a patient is prescribed with a HER2 positive cancer and they want to get they want to give them the Herceptin the drug to treat it based off of the receptor status and it fails well once that standard of care therapy fails the clinician is you know they're kind of going blind at that point and so a lot of the times already they have to rely on just their past experiences to guide and dictate what that patient has and there's no evidence of it's not evidence-based and so I think with respect to the FDA approval it's I mean the standard would already be there and let the doctors you use the information as they see fit and I think that would probably be a wise strategy and maybe just to add a slightly different point too is that as we have more data and certain areas of medicine modeling artificial intelligence machine learning can sometimes detect things better than the doctor for treatment you know for for example certain macular degeneration and it's been shown but as a regulatory aspect we have tort law in our country as well and even though some of this technology exists most medical practices are not yet willing to trust the outcome of a image diagnosis over the face-to-face of what a doctor recommends to a patient so kind of echoing these doctor-patient kind of relations so we aren't there yet so it's interesting I have a question about privacy um so currently when we collect data about patients bioinformatics we have ways to anonymize the data if it's just bits and bytes right if we start harvesting tissue from patients in order to regenerate organs for them and they consent to donate the residual tissue to the company for further research are there any ways for us to anonymize tissue so that the company can keep working with it without any sort of privacy concerns from the patient there are and and biobanking ethics deals with this quite often we already collect a lot of tissues and samples what I think you're getting at that which is even highlighted in personalized medicine is where when that tissue is donated a lot of screening is going to be done that will point to that as a unique piece of tissue that data will be stored and protected um lots of other information may be collected about that particular patient their background and other aspects genetics and as that tissue and that data is being used particularly if there's modeling going on or other kinds of databases are involved or that data needs to be shared with other kinds of researchers or even clinicians we run into the potential for de-identifying a particular patient because those algorithms even if they're encrypted or as you're sharing them there's more opportunities for those to be matched or mapped back to the original patient so we have to have processes to that are heightened in terms of policies of protection for the patients I think that's also another interesting point so also what what is patient data and you know I think in a medical context we have all kinds of protections and rules and regulation but now a lot of this technology even just gets out into the general public going directly towards getting their genome sequence themselves and loading it up to ancestry.com it's public data and it's available and how is that used I think apple and our phones and the watches we wear can collect data on our health and that can be a very good thing it can lead to interventions or monitoring but that data is not subjected to this same type of privacy issues and I think a lot of people don't know that and so it's it's an interesting time with not just how can we lead to improvements in medical care but what are implications health insurance you know life insurance so there's there's both sides to it which I think you raised as well. I just wanted to add I'm sure many of you who work with cells have heard of HeLa cells and there's a very famous well there's a book that's been published from the last five years talking about Henrietta Lax right so she was the donor and I think we also had be cognizant about she didn't really realize what kinds of purposes her cells will be put to and I think a lot of her children were actually quite upset to know that there were pieces of her living out there and I think part of your question yes we have to de-anonymize or how do we do that when we have so much specific information but how do we also educate the person so they really understand what's going to be done with that tissue and what kind of information will people know about them. I would also say bioethicists are struggling with this idea of what does privacy mean in both in the medical and human research because we've come into a an era it seems where it's not a binary it's not either private or public there are grades of what privacy is which means our policies need to reflect that shift in our current status of how much we're willing to give up and for what value or return but understanding the the uses of what that data is going to be critical to that gradation of what privacy means. Since we have representatives, was there a question? Since we have representatives from academics and industry here maybe I'll float a question to everybody does generally speaking the field of precision medicine does it offer a lot of opportunity in terms of industry academic collaboration or what what you're feeling about about the future especially given the pharmaceutical industry's apparent kind of latching on to it. So my personal feel interacting with other companies is a sense of cautious optimism and that you know we've gone down these roads before thinking you know this is going to be the next big thing and then a lot of money's been invested in it and it's kind of fizzled out and I think you know there's definitely excitement and but there's also let's see what you can do from a research side and let's see how real it's a little bit more I think from industry perspectives it seems that they're a little bit more cautious than they might have been you know 20 years ago if we would have been at the same state just because there have been failures in the past they're really advanced things at the rate that you know were promised from academia. I think there's a great amount of opportunity I mean I don't know whether you have to basically narrow it and narrow it and say as a company we're going to come talk to these researchers about a very particular disease and a very particular disease state in a very particular population maybe that's where it'll start but I think there's a lot of opportunity. Any other questions or is there another okay yes a question from the audience. Back to the topic on FDA approvals for particular devices so as we know on the topic of off-label use it can be a physician's discretion to try to treat these patients when there are no other solutions and if enough data presents itself that can be published in case studies and what was once off label can become on label through FDA approvals so my question is can personalized medicine keep pace with some of these innovative discoveries that are found somewhat organically or are we risking an opportunity where innovation may be stifled in ways that are yet unknown how do we encourage those types of innovations through academics to be sure that we're not missing out on opportunities to treat a variety of disease states that the original intent may not have been there. Any takers? Well my first thought is better and better modeling if we can prove that models are validable so that we can actually guess in silico or whatever that this is really going to work and those models get better and better then we might be able to improve that in that way. It is I mean to your point it seems like a little bit different situation though for instance off label use of a device or a drug is typically just as you know is just taking that technology and using it in a patient population that it wasn't not indicated for based on the regulatory process but here I think we're talking about maybe even if they involve kind of off the shelf cleared or approved products it still is almost by by definition kind of mixing things together to form a package whether it's a diagnostic with a drug or a sensor with a medical device or or whatever so it's maybe a little bit different pathway and maybe that's what makes it a little more complicated. Yeah I mean I think the pessimist so I've been in in some discussions where people think the pace of innovation is slowing because of our some of our regulations and systems and what we focus on but I think there's many reasons to be optimistic as well. You know unfortunately you might know the statistics better but medical devices and some of their approval you know I think 80 to 90 percent get approved through what's called a 510 k where it's based upon something that exists and you want to make what you're doing sound as close to something that's already a product and in some ways that suppresses or hides some of the innovation and what's going forward and I think there's movement a field right now of what's going to happen maybe in the future with some of medical devices but you know I think also when you think globally I think there's also a huge opportunity for innovation and and what we know now is based really even a lot on smaller sub-populations European and US and I think there are different populations and groups in in in Africa where there can be lots of tremendous innovation applied in different ways depending on how you define it. I think a risk averse society as well as cost containment and health care both work against the innovation I'm not saying it's killing it it's not an indictment of anything other than it just makes more challenges because something new that you've got to bring you've got to bring to the table and realize that it may not it's not going to be perfect so this gets me at least sometimes back into this better modeling approach and I think you raised an issue where in just very recently the FDA paying more attention to real-world data from use whether it's off label or other in the ability to approve I think it opens a possibility if people want to do modeling and they want to look at that real-world data and go back and say oh maybe this is only working for this sub-population or maybe this works best in this population we might be able to get some innovations out of that real-world data that will inform how the research progresses. Another question here whether it's actually germane to this discussion or not it's an expression of frustration that comes about from the difference between the culture of physicians and engineers. One of the issues with modeling that has been hard to get across to physicians is the fact that equations can or base or the model equations are based on some view of what's actually happening and individuals are represented by the parameters that are in there and it's been very hard to get that point across to physicians that you can write the same equation and be able to describe the response to a drug of two entirely different people different patients in the sense that perhaps some of the processes that are occurring in one patient are a little bit different quantitatively quantitatively speaking this is an issue that has not been easy to get across to physicians I don't know what others feel about this but I've tried to get across for example the idea that water and air are two very different fluids they're both Newtonian as long as you don't worry about compressibility viscosity and density are the two things that you use the equations are the same but there's nothing similar about water and air this is an issue that I dearly want to get across to physicians and Mike you it goes back to your point of better modeling and yeah well at the risk of sounding rumsfeld-esque there's the known knowns the unknown unknowns and and everything in between and and I think it's hard to convince the doctors that your model is good when they have patients that are sitting in front of them that they could easily enumerate you know 150 different variables and did you take all those into account or not or did you decide that most of them were irrelevant and so it becomes at least in I think in a physician's mind it comes a very complex process that that can be simplified if you're an engineer and take through it can be simplified but communicating that is as hard you're the only physician my specialty is very quantitative nephrology is but I would say in fairness most physicians are relatively non-quantitative I would I would agree and they balk at the idea of being able to reduce a complex uh human down into elements that can be modeled that I would I would say that's a very general statement but nevertheless probably pretty accurate I would maybe say two things right so we are seeing examples of medical schools now starting that have an engineering-centric curriculum so I I think that will be interesting to see what that generation of physicians and how they think and I do think you call to a point that right now we're in an era of quantitating and measuring lots and lots of things right so the whole genome the whole proteomics and the approach has largely been to try to understand every single aspect of the changes that are occurring but what I where I think what you mentioned and what modeling and engineering brings to the table is we understand a system we reduce complexity we understand governing principles and somehow I think we've got to play a role in that that how simple becomes complex enough to model or provide guidance to treat a patient or a person one other thing I'm not sure if I can elaborate or you know state this properly but basically if you're treating a patient and you can find the parameters of a particular variable that you're trying to tweak that don't cause harm so you've got this range in which you're pretty sure you're not going to cause any harm but yet you have no effect there are a lot of people out there that say that if you administer a therapy and it has no effect that that's harm because you've set up an expectation that you didn't meet so that makes the definition of harm even fuzzy at time so I think we're kind of winding down time-wise so if there aren't any more questions or comments from the audience I'd like to say thank you very much the panelists and and thanks once again to Dr. Anseth for her stellar presentation and participating in the panel and thanks for joining us