 Good morning. It's wonderful to see this growing group of those of us who started from GM one So like to sort of report back the fruits of GM one through three Do I have a point at some point? Okay, so when the cancer group met I think two GMs ago This is such a Talked about anything or decided to work on anything. We chose a couple of things that had pretty good research and I do implement One of them or two of them are relevant to universal Lynch screening one of them from colon cancer and one endometrial and for those of you bring you up to speed that at least in the Western world colon cancer and endometrial are the colon cancers of Lynch syndrome and I'll go into that a little bit and Chose and a tumor where the genetic load is high and that's pheochromocytoma and perigangular Pneuma and a very broad field of somatic genomics and then we acknowledge that there were Crumbs left behind after all we cannot work on it so a report back on universal Lynch screening and a little bit on the field and Something different because we also talked about cross-pollination Amongst the various working groups like to share with you the Cleveland Clinic experience of a prototype cancer family history taking tool So without further ado, let me remind you about Lynch syndrome. It's the most common heritable colon cancer syndrome of the adult onset variety and at least in the Western world Colorectal cancer risk is as high as 40 percent lifetime in women 85 percent in men and in endometrial cancer It's Anywhere between the 25 and 40 percent lifetime risk Importantly there have been very nice population based studies From many centers, especially in Finland and Ohio State when I was there that showed that between three and five percent I'll do say 2.7 percent of all colorectal cancers are due to Lynch syndrome with germ line mutations in the mismatch repair genes Therefore the cellular phenotype of Lynch colorectal cancers inometrial cancers are Micro-satellite instability because of these mismatch repair defects and Using simple in the histochemistry show a lack of protein expression in the mismatch repair proteins Making a diagnosis changes management obviously for the patient and then their family members And just to remind everyone if we did that would come a very long way in meeting one of the two agenda items of our healthy people 2020 the other agenda item is trying to identify every inherited breast cancer syndrome So what we did wasn't as is assessment, which is between 2004 and 2007 So in theory at the time every single colorectal cancer At least on our main campus was put into the pathology workflow to look for MSI and ISD known as for the mismatch proteins if they're not there in the screen negative We say they probably don't have HNPCC We will not talk about the 5FU because this is where it illustrates where the data is good But you shouldn't be using 5FU, but our clinicians find it very hard not to give adjuvant 5FU So we'll just set that aside and if they screen positive it goes in the pathology report And I think most of us would say job well done And then of course the surgeon meeting the pathology report and then referring them to genetics But is it a job well done? So for approach one, which is our SS assessment we had 237 cases and the screen positive was 22% so for those of you who can suggest as you recognize it And when it's a tech high should be about a 10-12% Not capturing enough So of those about half were referred for genetics GC genetic counseling and 12 of those that were originally referred managed to make it for various reasons some them forgot some them didn't think was important and so on and 10 of the 38 or 26% actually pursued Lynch genetic testing and of those 8% were positive So then there was a brief approach to just a very short interval So one year when we said well if the screen positive results We'll also share with the genetic counselor who would remind the colorectal surgeon to call their patients that might work So in the middle column, so of course that you dominate is less. It's only one year Positives were still 20% And then the referral to the genetic counseling is higher But if you look at the bottom line the number of positive genetic tests is still sort of middling. It's not great So family in our third approach, which we have been using since 2008 onwards and I think truly every single colorectal cancer specimen that's been resected at our main campus does undergo this We're trying to reel in the regional practice And so what's the difference we added B ref testing and mlh one of methylation We also so with the agreement of the patients they're given a brochure that says this will happen the results of all the MSI IRC testing are passed through our genetic counselors who would review it and They have been blessed to contact the patient directly and Then referred for counseling. So in that third approach as you can see you'll see that the the screen positive have dropped to 14% So that's pretty good a hundred percent were referred for genetics And if you look at the bottom line, we were able to find 17 out of the 56 with pursuit genetic testing To be to have Lin syndrome So we're a little bit pleased with this. So then the naive side my partner Kate Nathanson and Penn Was going to implement this and I think she's told me that she has challenges and We'll just leave it at that Do you want to say something? Actually, it's I think it's been really interesting because I think one of the problems is the issues of Reimbursement and the issues of what you can do without having an order have really changed over time and since Penn has coming Late to the screen. So for example one accepts that every breast cancer gets ER and PR screening but what's happened is that now because of Compliance issues. In fact, that wouldn't go forward today because there would have to be an order for physicians for things to be done Interestingly enough and so we're actually having to do a whole thing where we're writing a basically a universal order for this to be done and working with the compliance officers to be able to do this and Having a whole process on that this can be done on in patients that we as part of the DRG that we don't have to wait Until people get out of the DRG so that it doesn't it can be costed against the inpatient costs It's actually been a very long complicated process, but we've just submitted a letter That we've all agreed on to try to get it adapted. It's been a very difficult process Yeah, and I think just it's before we move on to the peregrine Leoma certainly the experience that we had at Intermountain Emphasize the idea that it unless you take a systematic approach to this and develop a care process You will really have less than optimal results and for those of you who are including Kate We're in the process of implementing it. We have two publications that Have come out related to this one is the business case and decision modeling that we use to Look at these issues specifically the issues relating to a pay or mix and the implications for of a hospital related to the additional cost under a DRG or a fee for service arrangement that modeling is Something that most of your business folks should be familiar with and might help to Short-cut the process the other question that comes up on the clinical side Frequently is the impact of whether or not we should be imposing an age cutoff So rather than screening everybody should we only be screening people under the age of 50 with colorectal cancer or 60 under the And so in the journal of oncology practice There should be an article appearing shortly that Where again we applied our decision model and looked at the hum the impact on sensitivity and specificity that would be Realized if you impose an age cutoff based on the data that we have and the use of the Ohio State data Right and so in real-time for all endometrial cancers or GYN Ong's in fact wanted to impose a 50-year-old age cutoff And so empirically we're probably proving what you're showing with the models that we actually miss quite a few of general mutation positives, so there you go I Think it is important I think one of the things that the cancer work group can really do is again aggregating these types of data because you know the modeling is good But you know the joke is what's the difference between the statistician and an economist? statisticians actually need data so You know the models are only as good as the data that underlies the assumptions and so the more data that we can pull together that Actually looks at the outcomes of systematic Screening approaches we can refine those models so that they really are much more reflective of real-world And that's a very nice tangible next steps for the lynch part for our working group So we welcome more people who fancy that sort of work So off to pheochromocytomas So pheochromocytomas and the related perigangliomas are neuroendocrine tumors and these are things that derive from the neural crests They can be malignant or not usually they're not Even if they're not malignant they can be in very inconvenient sites They're also inconvenient when they're hormonally active So they secrete adrenaline Nodrenaline and various other things and this can result in sudden death at worse and hypertension stroke, etc interestingly unlike most other cancers where the genetic component is 5 to 15 percent in Pheos and perigangliomas 30 to 40 percent of all comas irrespective of whatever have a germline mutation in one of 10 genes that we know of There are gene-specific risks of various cancers of malignancies of other cancers and gene-specific management therefore There's a genotype clinical outcome association So we felt that was rather important and it's actionable and of course like everything else that we've spoken about There are no practice guidelines So out of GM 2 and 3 came we gathered ourselves and we talked and we talked and we gathered for health systems That seem to be from Wisconsin eastwards They reflect a bit of diversity, but because they're all anchored by a central university based Main campus and regional practice But you've also seen that we've gathered the PI's who have very very different backgrounds, but the necessary multidisciplines What we wanted to do is to develop a very systematic approach for ascertaining every single field chromosome periganglioma patients almost like the colorectal cancer model and Lin syndrome and But we said I think in the modern era can the EMR help us and so we were going to look into that and evaluate perigangliomas are In the ganglia, but they're considered extra adrenal. There's several definitions of cost Perigangliomas are very easy because they are limited by a few diagnostic codes and it just pulls it out things in the adrenal a little hot Exploring whether the physician or physician assistant can help that once that thing alerts them Whether is it just an adrenal incident in Loma, which is going to be most of the cases versus in the adrenal medulla itself Then we wanted to see what was the most impactful way of testing a patient the our European American field chromosome study group Actually has a population base and we now have about a thousand Population-based registrants that based in Germany and central Poland and with that we were able to actually come up with an algorithm Based on data of if you see certain features, how should we prioritize testing? So of course at the Cleveland Clinic being one of the consulting sites We do follow that we also acknowledge that other sites like pennette and so on might actually do either full-bundled testing Rarely all 10 genes but bundled testing now for our approach if we go step-by-step There are downsides as you can imagine it takes time and after a while that's patients get what we call testing fatigue What's of all even though I said that 30 to 40 percent of these patients have a general mutation half of the ones that don't have it Still have the clinical red flags that say it has to be genetic so too young multifocal disease You know family history where it's not in one of these 10 genes. So we were going to compare the efficacy to Exome sequencing I almost said whole but I took it out and Finally, we were going to look at the psychosocial impacts of one of us with the patient say who I like it Just you know drawn in one fell soup. I don't have keep coming back. You keep telling me It's negative. It's negative. Let's go to the next gene Finally, and I think this is probably the most ambitious, but it also touches great practicality So here we say alright, it's well and good that you can say ma'am You have a mutation in axia this you must do this we can track compliance with the screening recommendations and We can record the incident new new plages or size growth and so on so this is a surrogate for are we catching things in time? And this is very acceptable in clinical cancer genetics in the fact in a even rarer syndrome leaf from any syndrome This has been done and published Finally because there are no cost-effective modeling in fear and periglium and testing and management We were going to do that with my catan who is expert in this and from all the data that we give it to slowly implement the evidence to Inform guidelines and Kate is in charge of those guidelines and We submitted this to the pilot demonstration projects, and yes, we know it's a cow We should also say so a little not to the somatic genomics, but also the interaction of work groups and other NHGRI Initiatives Kate is co-chair of the TCGA for fuel and periglium is which did not come out yet at the last meeting But she's there and with give it a multinational group to participate So now this is a little bit off can the various groups interact with other groups and there's a family history group here as well so this is a Pro this is a report back on free experience of a prototype tool Which our institution forces to do but without giving us resources. So anyway, so scotch duct tape cancer family history taking tool So we implemented it piloted it in Cleveland Clinic oncology focused clinical settings the scheduling of the qualifying appointment is by Appointment type so when a patient gets a specific code it triggers the patient via email to complete my family health history at a secure portal and We viewed this as a quality improvement initiative Because we know how terrible good our clinicians are at taking family histories We also use epic and not the fancy epic that Penn has It's terrible And so it doesn't get done So we analyze this for uptake so in other words do will we capture the population of people just it's too much trouble So we looked at uptake personal diagnosis of neoplasm sex age and social economic class So you can imagine the hypothesis perhaps people who've already had cancer might be more motivated People under 65 we like to play with our computers Perhaps uptake would be higher there too and of course a very famous one of course with higher socio-economic status The uptake is better. So we collected we have 877 or 76% of the cohort had a prevalent history of neoplasia already Most were female which is interesting 87% were under 65 And the socio-economic status was estimated by medium family income by zip code To all pleasant surprise there was no difference in completion rate of my family health history By diagnosis prevalent diagnosis of neoplasm by sex all by SES, which is very good news We weren't surprised of course that there was a decrease odds of completing for folks over 65 So this is all univariate analysis. So when we put in the multivariate model age still came out in the wash So then we thought well, this is rather important lots of people over 65 and Lots of people over 65 are considered the family Family the keeper of the family history So we thought we would go into a focus group and survey for barriers And then there was a suggestion that it might not be age because of course It's not a hundred percent uptake for all the other groups, but are there shared domains or shared? People who are not completing so we'll have to do that. I'm a little daunted because it's a lot of work I mean I have a wonderful and large team, but it's still a lot of work So with the prototype speaking about Institution what they want or need to actually release resources. So after this prototype the Cleveland Clinic released resources For us to build a scalable family history Tool and so again, it had me web-based patient entered and Of course, we need a clinical decision support at the point of care So the theme that we have heard yesterday and before and yes, it would be EMR compatible we built in hl7 language and We built a platform with modules for automatic risk assessment in our prototype We were the automatic risk assessment So that's not very convenient if you're going to deploy it in large health systems So the first few modules were made we tried to Take the most common heritable diseases We also Did focus groups of our end users the non-gen X clinicians whether they thought this would be Good ones to begin with so after Labor Day in a rolling model We have started beta testing my family in different settings and the most important settings are all Primary care providers out in the regional practice So far so good of a few hiccups We're going to take feedback after we formally analyze it and continue to tweak our tool and to build modules That doesn't end And the reason why we build it in modules is as already alluded here Guide lines will change because the evidence will change It will be easy to revise them whole module versus taking down the entire platform and revising it and We also heard the theme that not every hospital or healthcare system or place Our clinic or private practice is going to be like the Cleveland Clinic And so we have approached all places that are not like us have approached us for beta testing Oh Yes, you might be interested to hear you're probably thinking now. How did they make those algorithms? So we Used the IOM guidelines to make the algorithms. We started with the existing professional society guidelines We gave it a group several groups per module that had representation from genetic counseling from the content expertise of the non-genetics clinician and the geneticists and behind the content experts of the vest of the surgeon for example, he or she will also have Their support behind them. So the stakeholders the people who are actually going to use this and so that's That's a process that was really quite good and it also generated a lot of support as you can imagine So thank you and I'll be happy to take questions Thank you. Here's a comment and a question Going back to the colorectal cancer. I just wanted to reflect for the group as a whole You know, we've talked a lot about The idea that how to get things represented in some of the important guiding Documents like USP STF or other things of that nature. I think one of the things about the Lynch syndrome Project that's really critically important is that that is in fact represented at least hereditary colorectal cancer is represented as a goal and healthy people 2020 So we can link it to that and also that has been one of three genetic use cases that has been flagged by the CDC's Office of Public Health Genomics is being achievable as a public health Genomic goal, so I think that this Effort has a lot of opportunity for visibility in relation to those things and that may be the one of the first within the scope of what we've been talking about to to do that so Again more reasons to be involved and engaged in that the question I had cares as you did reference in the last part of your talk the family history work group of the genomic medicine Group and so I was just curious. I didn't see in the presentation The specific if any consultations that you'd had or a cross representation between The family history group that Jeff leads and your group and so if you could just expand it It's just it's just been informal so up until we finished a prototype. I mean frankly I and Cleveland Clinic felt too embarrassed to be part of your group. So it's true and with this though Would like to ask whether we can Fully join a group and I'll send someone else in since obviously I cannot be in every single group Embarrassment is a good thing I don't need to be embarrassed for sure, but I think this is great And I the idea that the working group as it takes on its next iteration of its life to include Other initiatives like the one that you're doing one that Marin is doing and others to really Be a cohesive force around family history for all the reasons you articulated Would be a terrific thing for us to be doing. Thank you Brad So to Mark's comment, how do you engage the families in the Lent syndrome? Project and and what's the uptake by the families? Yeah, so the patients themselves are given a brochure and Before surgery and the family they're encouraged to share the family members So it's the family member part that we don't know about and of course like all nice tertiary referral hospitals A lot of the family members are not in Cleveland So that would be a lovely next next step into our cancer working group, especially for the Lynch group It's always the family Yeah, it is a challenge And essentially what you end up doing is whoever has the initial contact with the Person who's identified as part of the screening program You're reliant on their ability to kind of capture that data on Whether or not family members have been contacted whether they're seeking help and so that is it's very manual meaning It's very resource-intensive. There's no really automated way to do that very easy easily But it clearly very important because as we know from the e-gap working group report And some of the subsequent cost effectiveness analyses have been done using that data The true cost effectiveness of the Lynch syndrome screening relates to the identification of unaffected relatives and instituting preventive services or early Identification services to avoid or at least identify cancer at an early stage So it is something that we need to You can imagine that if we can somehow link it in front of the family history group stuff Maybe having them Fill in the family history or help the pro band fill in the family history would be engaging enough and draw them in So as we are Beta testing these people we actually go out to do tape interviews with the patients With the care providers just to make sure that a the patients are not disrupted and be the care The flow the clinical workflow is not disrupted because it's like oh heaven forbid and and all the patients have said How did you pick me for this one points that they were randomly pick? I'm so honest. It's so fun You know, I call my family members and the family members are very engaged And I think there's also an opportunity I know that some people have talked about it, but there hasn't been a lot of work But one of the ideas is to use social media approaches to family history that could you create a Patient family history where there could also allow connections with other family members to contribute additional family history data It's something that's been discussed conceptually, but again because of some concerns about Again privacy and other things the idea of doing it within a health care system versus outside a health care system Is is a bit of a challenge, but but I think it's a it's an interesting way to think about, you know taking a You know a genealogy, you know type approach, but have it with the medical family history So Joan Yeah, I wonder if there's an opportunity for a synergistic of efforts around developing Some of the clinical decision support tools that are being done, you know We've gone through an extensive evaluation just before even developing the clinical decision support for the prenatal family history tool Of what should we be screening for and what are the practice guidelines and what are the recommendations? And we're currently doing as you know the same thing with the pediatra for a pediatric Family history tool and even that decision about what do you what do you in? Questions what's what are you screening for is in of itself a huge effort and then you have the development of the actual Clinical decision support and what are the messaging? Etc. So there and I know there are other groups who are involved in the same and the same efforts So it seems to me there's an opportunity to share some of that Of that information and experience so that we aren't all reinventing the wheel and not only do we share that I have to point Up that Emily Adelman who was office genetic counselor project manager went to work for niche pack We even share people it's wonderful and while institutions actually not touching The prenatal and the pediatric because why reinvent the wheel? Yeah, and I think if you think about this again in the more modular perspective you can the underlying family structure issues are Universal I mean those are cross cutting across all family history tools You have to be able to represent the pro band the parents the siblings etc. Etc. So that type of structure Can be done and once it's done then what you overlay in terms of the information can vary depending on the specific use case So I think a lot of it relates to the you know What specific things are you trying to collect and when so it's not only the guidelines that are out there But also when do you clinically when do you present a module to a patient in the course of their clinical care? Which is another interesting question. So again, I think there are a lot of these types of questions that Can be addressed and it's I think it's really exciting to think about The idea that a family history is something a document that gets richer and richer and richer as people progress through their lifespan But also through their clinical encounters Jeff I think one of the points that John was making is that maybe there's an opportunity here to create a common toolbox for various platforms that family history is using in other words guidelines are obviously The standard and as Karris pointed out it takes a lot of effort to bring the guidelines into a clinical decision support engine And through open CDS. I think we could Think think through how to make these guideline driven engines available to a number of different Types of platforms that may be used in different health arenas for capturing family history and delivering Guideline informed decision-making. I think that's an excellent point. Thanks, Jeff. Katie So I wanted to just mention that in that in the spirit of this being a meeting that is focused on education I wanted to point out that the AMA and niche peg collaborated recently on educational resource for physicians and other health care providers on hereditary colorectal cancer syndromes and it goes over risk assessment Identifying red flags collaboration and referral to other providers and it's based on Guidelines that are well established the eGap guidelines and CCN guidelines Guidelines from the gastroenterologists and one unique feature that we we really put a lot of work into was In addition to being able to take this as a traditional Online CME. There is also a performance improvement Version of this so a physician or other health care provider can actually assess his or her Skill in Recognizing red flags and and referring and ordering genetic testing when appropriate both before and after Taking the the course and so we're hopeful we're going to be Analyzing some of the data from the groups that have taken both the online Traditional format and then the format that involves a performance improvement arm We're going to be hopefully comparing those and finding out whether that performance improvement activity this included Actually really does produce sustained Improvement in the way that physicians and other health care providers recognize These families and go about caring for them and and referring to to specialists And that can be found on the AMA's website or a niche pegs website And it's freely open to any physician or health care provider. Great. Thank you Erwin last question Carries this this is very significant effort that you've undertaken in particular in the context of working with epic Which is certainly covering a lot of the HR market across the nation and I want to congratulate you on that and for your beta tested version And you mentioned this. Oh my god the clinical workflow. I mean, that's obviously the holy grail in our operations so Where does the information from my family then actually reach the physician and at what point is it on opening the chart? I mean or how for those that have enrolled in my family Give us the flavor of how it then looks in the doctor's office. Yes. So this is very true So the two parts one of it is alluded. So in other words, if there's a family history that has been flagged as high hereditary risk Is the physician liable? So we were concerned about that clinicians were so we got legal involved and The solution was because if the patient doesn't show do you still have that legal? Responsibility and and the answer is no if you don't open it So the this my family is actually locked until the the the patient comes in and registers and then it opens up And so it's so the famous left-hand comic epic. You can you actually click family history And and it opens up. So, you know, we geneticists like pedigree So because there's a pedigree drawing part most primary care physicians Forgetting they don't like pedigrees. So it actually shows up immediately as a sort of a descriptor and then The decision support is ranked in order of red to green So the most heritable stuff all the way to general population risk It also has a little button right now that says although we were told the color was wrong So there's the first feedback they can find it will change it it will say accept and Refer button refers to genetics or if they say something looks funny Just doesn't seem right the risk looks too high to low and they can hit review So our staff will review it and they can they can review it and ask for referral or review it and hold and then We'll get back to them. So that that's how part of the feedback is taken Very good. Thank you We are running ahead of schedule Which is a good thing We will be more ahead of schedule because as best as we can tell Andrew futrell is not in the building So if you've been hiding or underneath the table or something, we'll let you talk but otherwise one of the things that we've Encouraged over the different genomic medicine meetings is that when we have new groups new participants that are joining we'd like to hear a brief summary of the activities that are going on in the implementation space and so I'm pleased to Have John Harley present on behalf of Cincinnati Children's Hospital