Event ID: 1552515
Event Started: 6/15/2010 7:52:36 AM ET
Please stand by for realtime captions.
8:18 Eastern Time Zone, captioner standing by on hold music, 202-408-6139.
Good morning. Good morning, everyone.
Good morning.
Good morning, Mark.
Good morning.
Good morning. Welcome to the 22nd meeting of the SACGHS. I think that everybody is grateful that this time we're not battling the blizzards.We're in better shape today, just a little rain. I want to thank you everyone who was very accommodating in adjusting the schedule back in February, and also to Alison who bailed many of us out by getting out of here before the great storm.
The public was made aware of this meeting by notices in the Federal Register. I want to welcome the members of the public and viewers who are tuned in via webcast. Thanks for your interest in our work. We will have public comment sessions this morning at 11:15 and again tomorrow at 10:15.We have several topics to cover. We will begin with updates on federal activities. We will hear from NIH's chief of staff.Should be followed by an update on the interim find rule, certification criteria for EHRs by doctor David Hunt. Then we'll hear from Ben Chen on recommendations for good laboratory practices which were developed by CLIAC. After lunch we will spend the rest of the afternoon discussing the implications of the affordable whole genome sequencing. Tomorrow's agenda will include a session on genomickic data sharing. Our final agenda item is to discuss a briefing paper prepared by the committee on inherentable disorders of infant and children.
I have a few updates. Since our last meeting we've finalized our report on direct to consumer testing, and our report on gene patents, which you received a few weeks ago by email. We released our graph report on genetics and training for public comment. For members of the public there is a handout with the URL for the report. I would like to thank Sylvia for her leadership and members of her task force. Sylvia has rotated off of the committee. We hope she listens to the webcast. We appreciate all of the work she did for the committee over several years. The gene patent's report can be found on the SACGHS website. A handout is also available for attendees. In March Jim Evans briefed Dr. Collins. After the meeting Jim and Rochelle sent Dr. Collins more information. The issue of gene patents and whole genome sequencing is also discussed in a "Nature" editorial.
Now that our report is being weighed by the public and second I want to again thank Jim for his dedication and leadership in bringing this report to closure. I also want to thank Rochelle, Bill, Paul, all of whom labored hard after an October 2009 meeting as the report was revised. I would remiss if I didn't thank our staff for their dedication to advancing our work. It was an enormous amount of effort. We will continue to monitor efforts in the area of genetic tests as we move on with our other priorities.
In February Barbara provided a review. The committee approved the draft report and Rexes. The draft report was readvised and shortened. But the recommendation remay as we approved them. It's out for public comment. You should have all received Sarah's email about the report's release for public comment. I Juan to give a special thanks to Barbara for her continuing leadership on this project.
In February we also approved the commentary for submission to a medical journal. We're awaiting more information about its status. A copy of the commentary is in your folders as well. In addition to the presentation and its recommendations for good laboratory practices, CLIC, has been working towards the development of a notice for proposed rule making to update requirements for proficiency testing. During process of developing the rule a working group is engaging with constituents. The proposed rule is expected next year.
I would also like to provide an update on GINA. The EEOC is now engaged in an internal review process. Once approved it will be published in the Federal Register. There have been approximately 08 charges filed since it became effective. The charges, which came from across the country, concern improper requests for family medical history, improper disclosure of genetic information, termination of employment, and denial of health benefits based on genetic information. An article is in your briefing books at tab 10.
Another genetics-related bill may also be on the horizon. It has a number of provisions designed to advance research and personalized medicine. The legislation would establish an office in the department and a national biobank. Copies of the bill have been included in your folders in case you would like to review other provisions of the legislation.
I also want to mention some changes in our committee roster. We would like to thank Dr. Barry [ Indiscernible ] for his services, and welcome Jeff Roach and Jim Rollins as the new CMS ex officios. We also want to thank Robin for her service and dedication. Robin Sue has moved to a new position, Jennifer is taking her place. So welcome back. You've been here before. Good to you have you official.
I call your attention to the lunch order forms in your folder. If you want to preorder them fill them out and give them to Alison at the break. There will be a buffet available in the restaurant for $16.50. For committee members would like to join us tonight for dinner, reservations are at Mio at 6:30. If you want to attend please let Alison know by the end of lunch. We will walk over, it's just a short distance from here.
Before we get into the substantive part of the meeting, Sarah, tell us about ethics.
Yes, thank you, Steve. As you all know, you've been appointed to this committee as a special government employee. You are in a special category, you are subject to the rules of conduct that apply to regular government employees. The rules are in standards of ethical conduct for employees of the executive branch, which you each received when you were appointed to the committee. I will highlight two of the rules. One is about conflicts of interest. Before every meeting you provide us with information about your interests. Information that we use to determine whether you have any real, potential or apparent conflicts of interest that could compromise your ability to be objective. While we waive conflicts of interest for general matters we also rely on you to be attendive during our meetings to the possibility that an issue will arise that could affect your interest in a specific way. We've provided each of you with a list of your financial interests and relationships that would pose a conflict for you if they became a focal point. If this would happen we ask you to leave the room. Also, we're close to the capital today. I want to remind you about lobbying. Government employees are predicted from lobbying -- prohibited from lobbying. If you lobby in your professional capacity or as a citizen it's important that you keep that separate from that of this committee. Keep in mind that we're advisory to the secretary, we don't advise the Congress.
Thank you, Sarah. All right. Let's start off by hearing from our special guest this morning, Dr. Kathy Hudson, who many of you know as chief of staff at the national institute of health. She has worked with this committee on a number of important issues over the years, was really an instrumental part of our task force on the oversight of genetic testing. She's going to tell us about the genetic testing registry and give us an overview of the partnership between NIH and the FDA to advance translational research. Then we will open up the floor for discussion and questions. A copy of the RFI is in your table folders. Kathy, it's really a pleasure to welcome you back. We look forward to what you have to say.
Thank you. It's a little bit like coming home. I realized that I have appeared before this committee many times. I will talk today about the registry and what we have in mind and what we would like to seek from you. But I thought I would talk more broadly about issues effecting personalized medicine before launching into that.
It was probably close to a year something when Francis Collins called me up. He said that he would like me to come back and be his chief of staff. I told him no frickin' way. I clearly lost that arguement, he can be pretty persuasive. I spend a lot of time in this building.Francis spent time think being the horizon and biomedical research and opportunities that are out there. I'm sure that you have seen and heared about his five opportunities for the NIH. These are really guiding the work that we're doing to put together budgets for future years. Those themes include applying high through put technologies to understand fundamental biology, translating into clinical applications, putting science to work for healthcare reform. Already we have funded quite a few initiatives in this arena. Encouraging a greater focus on global health, empowering the biomedical research committee. These are the principles guiding us. We're at a unique time, Francis is an inspiring and ambitious leader, but we also have a very proscience administration with the president and all that they've brought with them into the administration. The president came for a visit in September. I started on September 3, my first duty was to get the president to NIH. I was successful in that endeavor. This is a quote from his appearance. Not only is he committed to science, he's committed to health and healthcare reform and really made, um, healthcare reform a reality in a remarkable period of time. It seems like it was taking forever, but it was really only about 15 months.
This is a cartoon that I had when I was at John Hopkins on the coffee board, some people will remember it. This does reflect the new spirit that the scientist community has. There was tension and some animosity between NIH and the administration in the last eight years. Now it feels like everybody is dressed up in the same uniforms and playing for the same team. Just a word about healthcare reform, what it means for us at NIH and the biomedical research enterprise. I do want to say a few words about the [ Indiscernible ] acceleration network, which was included in healthcare reform. What the network is seeking to do is to bridge the valley of death and increase the repiedty to translate research into the medical cabinets. This article in "Newsweek" talked about this problem about being able to bridge the valley of death. The pipeline is familiar to many of you. What we do at NIH is focused on the left end of this pipeline in disease identification, target identification.We have made new investments in this pathway over the last year, and will continue to make new investments through the NIH directer's common fund. There's about $5,440,000,000 -- $540 million at his disposal.
In order to bridge the valley of death we have to partner with pharmaceutical companies, biotech and our academic-funded centers. We have over the last nine months put together some new and exciting partnerships with the FDA, that partnership is going to be critical. In February we announced a new NIH/FDA joint leadership council.That joint effort will make available a funding opportunity in regulatory science. We have those applications in, they're about to be reviewed. Together the agencies made available nearly $7 million for that. A couple weeks ago we had a public meeting where we listened to public input on the high priority issues that FDA and NIH should tackle as we begin this partnership together. We are planning a meeting for the joint leadership council hopefully in June or July.
So back to the cures acceleration network. We have efforts already ongoing. It will provide additional funds. Specifically will provide us new ways of doing business so we can be more nimble. This is the goal from the legislation that was included in healthcare reform.The legislation is focused on high-need cures. Of course, if you or your family are affected by a disease or a disorder that is by definition a high-need cure. We will have to spend a lot of time on defining that criteria. It was authorized for $500 million in FY2010. That's not an appropriation. We don't have a check. The network would be situated in the office of the directer, it would provide some nifty funding and other authorities that would allow us to have more flexibility in how we fund research. Currently from the time that we have an idea of something which want to fund until we put money in hands is a fairly long process. And this would provide other transaction authority.There's been a lot of support for the cures acceleration network. This was a letter sent to the Senate in April. There's been a similar letter sent to the house by a remarkable group. It is a large number of diverse advocates from across medicine and patient advocacy who are supporting this.
So a good cure, of course, requires a good diagnosis.The development of new diagnostics could be a high need cure, as well. I want talk about what we have been doing lately with genetic testing and where we are with the registry. It has increased in number, it has changed a lot, it's increasing in complexity. As we read about virtually every day. It's changed, it's offered more frequently. It's offered through the internet. It's available to you if you want to be an undergraduate at the University of California-Berkeley. Genetic testing has increased in clinical relevance. You could get the information about your future, but do little about it in the past. Now you have refined information about the risk for future disease and having the ability to do something about that increasingly. My favorite example is plaivics. Nearly a third of the population does not respond to this drug because of variants in the pathway. Because it's relevant, I want to review the drug relabeling that went on. It's an interesting challenge that we face today. If a drug is already out on the market and we find a subpopulation that responds adversely, or with whom it's not efficacious the FDA and the sponsor and the diagnostic developer have to work in a complicated way to get that information into a drug label.
Clough pied grel was relabeled in 2009 to add information, it didn't say much, except that genes were involved in the metabolism of this drug. In November of 2009 the precaution was upgraded to a warning, in March there was a block box warning added to the label of play vix -- Placix.
I want to switch gears and talk about -- I know that you have your report out and it's been transmitted to the secretary. Certainly the case from March comes to bear in terms of how the development and availability of diagnostic tests. I'm sure you are family with the sweet decision that was brought by the plaintiff. They challenged the validity of claims on BRCA 1 and 2. If you haven't read the case I commend it to you, it is fascinating reading. In the decision the judge refers not at all to any of the previous cases that have been argued over the patentability, argues that DNA is not patentable subject matter, period. The two findings were first that isolated DNA containing sequence are unpatentable, and the claims for comparing the sequences to [ Indiscernible ] type sequences is an abstract mental process. These are two very sweeping rulings by the judge. To my knowledge this court decision has not yet been appealed.
I think this case raises some important questions in terms of its implications for the development of DNA diagnostics. It could be interpreted broadly to apply to nucleic acids used their piewtically.
I will say -- get to the issue of the oversight of genetic testing. Which I have had a long-standing interest in. Today in Dr. Collins and Dr. Hamburg have published a perspective. Look at that, it's 9:00, it's now online. Copies are available over at that table. It's interesting, it is a shared vision. The NIH directer and the FDA commissioner on what they feel are the key steps along the pathway to realizing personalized medicine.I think that's a pretty clear statement of a shared vision that these two leaders have. And now we will move forward in implementing that vision in a coordinated way.
Of course this comes at an interesting time. The Congress has been paying quite a bit of attention to genetic testing oversight of late. Congressman Waxman and his investigations in oversight have been looking deeply into this issue and launched an investigation into genetic testing, focused on the DTC tests. They sent letters to the manufacturers asking them for a fairly substantial amount of information by June 23, that should be resulting in some sort of analysis by the committee soon. Dr. Collins is testifying before this committee this afternoon, I anticipate that questions about come up. Subsequent to the energy and commercial investigation, FDA sent letters to a number of these companies indicating that they thought that these products may be divisive. They invited the companies to come in and talk with them. It's not the kind of invitation that you ordinarily want to get. I expect that those companies are working rapidly to come meet with FDA. We're not a regulatory agency. We've been working with FDA, have lots of friends and colleagues at FDA who we have been interacting with. On our end what we are able to do is to build this genetic testing registry. The goals of this registry, which were inspired by you in your oversight report, is to improve research and public health flew increasing transparency, increasing consumer, physician, and researcher access to nfght. We have published a request for information in the NIH guide and in the Federal Register. I think it was published last Friday. Is that right, Sarah. We posed a question of questions. I will run through some in just a minute.
The team of people who have been working with this have involved the office of the directer and also the inform mattics guys at the National Library of Medicine. The exciting thing to me is it will be gee grated into these other resources in the library so you can easily go from one resource to another resource to another resource. We are seeking input, hope to gather information from you all over the course of the next 30 days. Then have some public meetings and private meetings. If you have information that you want to share with us our doors are open. These are some of the issues that we are seeking information on. For some reason the word "test" is bigger. I don't know why. I won't read through this because you are all lit rate.I will stop there and say I'm excited and looking forward to getting your input and develop a beta database and get the thing up and running in the next calendar year. We have big challenges ahead in all areas of genetic medicine and building this database I think will help us get from. I will end with that, be happy to take your questions. Or not.
Great. Let's start with Andrea and then Paul.
Thank you very much for a wonderful presentation, Kathy. It's good to see one of the recommendations actually is -- moving forward, you were a part of that, too. I understand these are still an evolving process. I wanted to find out if you have in mind to have a single registry? Or having one for healthcare providers? Another one for consumers? The information might have been related in a different way for different groups.
We would certainly be open to comments on that very topic. I think that, um, at least in my own head I was imagining a single database that would be searchable, it would have information that is fairly technical because of the nature of the information that we're trying to contain. Whether it could link over to other more lay resources is a possibility.
A follow-up question, are you envisioning [ Speaker/Audio Faint or Unclear ] to ensure the accuracy of the information?
Great question. We will not be double checking the information that is submitted by a laboratory or a manufacturer. We can't really annotate it. I think the best example of a database that operates under a similar premise is clinicaltrials.gov. Sponsors of clinical trials enter information about the trials into a database that is run by the National Library of Medicine and the library does not attest to, and really can't attest to, the accuracy of the submitted information. That said, because the library has the publications and other genetic databases affiliated with it, I think there be a natural way by which people will be able to assess whether or not there's something amiss in the information. But we will not be curating --
There's a big difference between the clinicaltrial.gov and where people get information about current trials that are ongoing. People will start ordering the tests and act on that, the genetic tests. It's important to have in mind that people -- might have the public -- on that particular issue.
Right. There is today a registry that has a more limited set of genetics tests in it. I don't believe that the information is -- that's run out of the National Library of Medicine, gene test. It does not double check with the laboratories about the valid dy of the submitted data.
You put the name of the test that you do there, there's no other information. When you start putting in other information --
Yep, yep -- I think the analogy of clinical trials is a good one. We are relying on people to submit their results, and do it in a way that reflects straight forwardness. I imagine there is some high incentive to submit accurate information and not mislead the government. I'm sort of an optimist, I will look on the bright side. Because all tests that are entered into the registry will have an identifier things will be linked to that. If three people publish data showing that Test A doesn't detect what it supposed to that will become affiliated with it. We don't have -- even though I published an article arguing we did have the clear legal authority to require, to mandate the submission of information and the ability to then go and bop people over the heads if if he don't provide correct information -- that's not what we do at the NIH.
The issues of the data that you be requiring, it's important to seek public comment on that. One of the areas of concern is clinical utility. How you define it, the evidence needed to determine it, and information is not available, you know, we might provide more harm to the patient. That information might be misused. I will ask you to look at what information you will be requesting and what will be the consequences of having that information.
I will ask you to tell us what you thing would be useful information to include. And where certain kinds of information might be eerlgt burdennensome, or where it's not useful information to the users of the database. We're looking to you to tell us to where information would be different to produce and not valuable.
Are you asking the committee to do that?
I'm asking the public, of which you are a part, to do that.
Paul, and then I have Mara, Sam, Jim, and Mark.
Thanks for a great talk. I have learned there's students wanting to have their sequences looked at. The VA has started their million veteran project. I'm curious how the programs will communicate and learn from each other.
That's a great question. I have just been put on the advisory committee that advises the VA on the million vet program. That's one mechanism of interaction, at the highest levels we've been keeping in touch with and coordinating with. I think the more important coordination is in the conduct of their study. And also learn from them, because we're doing other large cohort studies as well. We want to be able to learn from each other.
Great.
Mara?
Again, Kathy, thank you. I appreciate the context of the broader picture. One of the issues that this committee and the industry has discussed is genetic exceptionalism, and how that the perception of that has changed as genetic tests, um, not even into the issue of definition, because we'll be here all day. How do you think about that? How did you think about that in putting this registry together?
Good question. You know, if you look at the total number of tests out there, coming up with an accurate number is a little difficult. But maybe in the orders of 5000 to 6000 distinct tests, I'm making that up. And what proportion are genetic? It's an interesting question to pose of whether it would be useful to axe panned this -- expand this registry to be a clinical test registry, rather than just a genetic test registry. In fact, you know, comments on that subject would be welcome. I think there's a little bit of a notion right now there's -- we don't want to bite off more than we chew. So if we can start with defined set that might be an easier way to go, and maybe expand later if there is interest in doing that.
Sam?
Kathy, thank you for this comprehensive review. Is an exciting time. We see that enthusiasm on your part. The question that I have is that genetic testing, as it continues to expand may be somewhat different than most diagnostic tests that are ordered by professionals. And to build on Andrea's point, do you believe there needs to be a special sort of registry provided to consumers? Because so much of this may be marketed directly to consumers going forward. And the reason I say that is because with this information, as we will talk about over the next two days, it may guide people in directions that even clinicians won't have clarity. Although clarity will be identified in the future I think that consumers may benefit if they have a trusted site to go to, where they can have this information layed out this them. Both the promise of genetic testing and personalized knowledge of medicines and what we do know today.
Yeah, I think that's a good point. It will be a challenge as we move forward to see whether or not we can meet that need as a part of this registry, or if something else is needed. I think that, um -- understanding what consumers who are purchasing direct to consumer testifies understand about -- tests understand about the tests that they're ordering, and what actions they're taking based on the results is an important issue. I was involved in designing such a study at John Hopkins before I left there. Hopefully they will be report on that before too long. I think that will help to inform us as we move forward.
Jim?
Great. Yeah, I just wanted to amplify something that Andrea said, I think it's important going forward with an uncare rated database where you are relying on people to fill in fields to make sure that those fields don't try to capture more than they can. What I'm getting at there is it's one thing to ask for analytical validity, that seems verifiable by the mechanisms that you mention. Clinical utility, of course, means as it should, different things to different people. In the end it's an iterative process. So I would just advocate being careful about trying to create a field that is really undoable, unfillable, right. So I just want to -- in the spirit of not biting off more than we can chew, or consume, I think that trying to think hard about the real limits is really important to making it work.
Yeah. I think after we have a basic structure layed out it will be critical to partner with laboratories with a subset of well defined tests and try it out.
And clinicians. I can't emphasize enough that much of where the rubber hits the road in the utility of tests, be they genetic or what have you, I think that clinicians need to be involved because they are the most direct representation of patients.
A word about that, the National Library of Medicine has an advisory group that includes clinicians, medical geneticists and whatnot. They will provide insight.We will pull together experts to advise us internally.
I wanted to say to Jim that there are groups that would view these. One could include those things on a website.
Yeah, that would be a mechanism. It runs the risk when you try to capture very broad things in discreet fields in a database, being so all over the map that it becomes pointless. I'm just saying to think hard about what fields make sense.
Utility that people really need to know --
Exactly. That doesn't mean that it's easy, right.
No, no -- that's why we have people like Kathy work on it.
Mark?
I wanted to comment on the last bullet of the last content slide. I'm curious on your thoughts of the effective system to manage conflicts.
I want to delete that bullet --
I noticed it, unfortunately for you. I think is a very important issue because there are a lot of stakeholders with a lot of perspectives. I'm curious if you have some ideas about who needs to be at the table, what sort of forum would be needed. I think it's an interesting thing to articulate and put up there.
I think the issue of managing conflicts, um, you know, sort of surrounds the whole enterprise. If we're going to engage in move things we will have to work closely with industry and how to do that and yet not lose the trust of people supporting the science. You may know we've put out proposed new rules for governing [ Speaker/Audio Faint or Unclear ] funded investigated. We haven't yet gotten in the comments in, it's out for public comment now. In those proposed rules we lower the dollar amount that is subject to disclosure, which will be probably inconvenient for many. I know when I came into government and had to disclose everything under the sun to join the government I found it very inconvenient, you certainly have made those disclosures in being a part of this committee. Sunshine is a good dis disineffectant. -- disinfectant. We've had some pretty e graingeious cases over the last several years. Investigators have failed to disclose enormous sums of money, that has not served us well, we need to clean up our act. We thing we're on the way to doing that with these new conflict of interest rules.
Well, great. Any other questions for Kathy? Thanks so much. We're really delighted to see all of this work going forward. You were an important part of this --
As you were.
It's terrific to see all of this going forward. Clearly a dynamic environment. Thank you for all of that important work.
Thank you, thank you.
Thanks for coming. [ Applause ]
Um, I will turn at this moment to Mark Williams. In February we had some discussions about clinical utility and where we need to go. We've progressed a bit since then. Mark, I think if you are prepared to fill us in on what has happened since then, and what we should be thinking about as we move forward.
Thanks. I'm semi-prepared. I thought this might come tomorrow. I think we can probably fumble our way through this. As Steve mentioned, there was a request at the previous meeting to kind of do a little more laying back on our work on comparative effectiveness to get a sense for the activities going forward. Behind Tab 10 you see a report on obligations, expenditures, et cetera. As I reviewed that I thought it was quite useful in terms of getting a sense for where things are. In the presentation that I gave in February I gave some ideas about where NIH was expending some funds related to Jen om nicks and personalized medicine and where there are some infrastructure things being funded through AHRQ that could potential benefit genomics and personalized medicine. At that time we were waiting for more formal announcement about the discretionary funds that were given to the secretary's office. I want to walk briefly through the report to highlight a couple of different things.
[ Captioner Transition ]
Please hang up the telephone so the new captioner can dial
In. Thanks!
Further along and it is out of the $400 million bid and has already obligated to hundred million dollars with a another 200 million unobligated. However, with AHRQ, basically they are through all of the announcements and are basically in the final stages of making the awards.
On page four of that can doubt there is an outline of how they are doing and how they're spending -- there is an outline of how they're doing and how they're spending the money. Of the 58 million that they have, they are going to be, they are going to be spending $10 million on methodology Development effectiveness Research which I think would be of some what of interest to this group and I also would wanted to highlight on page five that at the presentation that I gave in February I indicated that I believe there were seven awards and cancer original mix that had come out of that ARRA -- cancer genomics that had come out of that ARRA funds and maybe could prevail on doing -- if you are still around, oh, there you are. To comment a bit on that when I'm done talking. So that has been the most tangible evidence of commitment to some moneys going into the comparative effectiveness of genomic and personalized medicine.
And last are the discretionary funds to the office of the Secretary which is again a total of $400 million the secretaries report which begins on page six the outlines of I think any military high level what the secretary is planning to do with those funds but at the present time there is not a lot of detail related to what is the plans of there and if you look at how the moneys are allocated, the secretary has actually extended only about $10 million of the $400 million to this point. Part of that is related to some review that you have to go through other groups at the administrative level before they were able to release funds but has is noted in a number of these, shockingly I think to most of us is much harder for the government to spend money than we sometimes think it should be war think it is -- Or think it is.
The secretaries still have an unobligated balance of 380 million so nearly all of the funds were unobligated and 246 million of this have actually been put out into funding opportunity announcements both grants and contracts and those are closed in the review stage and that is what I was asking more information about because I was not actually aware of or somehow missed those announcements so I am very interested to know what those were and what they were specifically looking for. There are $53 million worth of funds that are in announcements that are currently open and I guess I am probably more interested in knowing where those are because I might be interested in applying for them. And lastly I think of most relevance to our committee is that there is $85 million that has not yet been posted and so as I was thinking about the role of the prize three committee in this effort, it seems to me that we could potentially make some recommendations to the office of the Secretary about how at least some of those remaining unannounced funds could be allocated to work in the genomics and personalize medicine space and as we were sort of brainstorming ideas in anticipation of the meeting there would be used that emerged that I think we could bring into discussion. Things like how to determine evidentiary standards, how did the standards. How to reset the bars for information that needs to be accrued relating to genetics and genomics subgroup analyses some warm-up illogic kind of things in another area that would beat of potential value of is again a discussion that we have a lot of lost time and will be continuing to some degree in the next session is about the informatics infrastructure and moneys that are being devoted to the High Tech Act and the ability to capture information about genomics and personalized medicine might be of some importance on how we look at the impact of those types of infrastructure decisions to do comparative effectiveness in this space and I am sure that others around the table have a number of other ideas. I am looking to Sarah and Steve. Did I miss anything that we talked about about possible opportunities?
No, I think it is important that the money that you are talking about has to be expended by September 30th so there is a very narrow windows.
[ Indiscernible ]
Obligated.
This is not the end of this whole issue because we will have Research at the institute that is going to be created and it will need to carry on this way. And there is a lot of work going on and we will have some others talk about a meeting at Beckley this fall on the interface and personalized medicine and comparative effectiveness. Although we have a specific opportunity right now the window is. Narrow and the question is if we do something we will need to do something tomorrow. We need something really drastic and I am not sure exactly what that would say. Does anybody have a specific proposal? Because I think we will be taking of the whole issue of clinical utility and comparative effectiveness probably in October. Because we defer to that at the moment but before we decide that to do something I will let you say something.
I said no but I think there are things.
[ laughter ]
I did not really mean no [ laughter ]
I am thinking about the fact that drafting something would be basically me drafting something and neither which are feasible however I do think there would be opportunities to take suggestions that we have made from previous reports relating to some of the things from our oversight reports that would directly impacts so I think we could potentially be purpose some of our existing recommendations into something that would be semi coherent that could potentially influence the announcements related to that last $85 million that might be actionable by this group by tomorrow.
Yeah.
I worked with them some time ago when they were working on comparative effectiveness as well.
We weighed in on this and we can probably pull together something that is reasonable.
It is my understanding that the Institute of Medicine was charged with privatizing CER and they came up with a list of 100 that we know and I am wondering if we can crosswalk that list of top priorities to our thinking and secondly if we have or if we have not there may be a list of number 101 to 500 that could accelerate us taking work that we have done in a very broad context.
Let me respond to that. We did in fact crosswalk both the IOM report and the of FCPER reports which were meant to be considered by the secretary and she looked at allocation of funds. There was less from the IOM report other than the oncology realm that could be directly purposed towards compared to the effectiveness and the secretary has essentially said that the moneys that are under her discretionary power will be less at these articulate their projects for specific disease entities which is what the IOM report essentially we've been at and more at issues of infrastructure and methodology which were covered in the FPCER -- I may have left a "C" out there. So we did a look at those and those are issues that we did look for and that is why I am suggesting this infrastructure may be the way to go.
So much to talk about this topic but Mike mentioned some of the NCI projects and I spent a lot of time at the NIH and we were able to use ARRA to fund seven groups in the genomic and personalized medicine. They have a very active agenda right now including a pathogen project that was mentioned is setting the groundwork for sort of a -- sort of evaluating methods Development as well as trying to figure out sorted using modeling and other approach is working with the working group, ways to exhilarate devaluation of clinical utility. So in this context -- waste to accelerate the evaluation of clinical utility so in this context with emphasis on translation and personalize medicine it looks like CER is one mechanism because not all genomics applications will be subjected to randomized clinical trials and many of them are already in practice or will be in practice so this will be one way to get the information in a usable fashion. Just one comment on the genetics that made that comment when Kathy was here. I think as people have been saying, the information that is part of the registry will have to be looked at using independent bodies such as kind of like EGAP and others especially from a clinical facility perspective and having this -- the evidence of chemical and utility and validity as well as some the size and the evidence. So any mail as part of our initiative I wanted to share with our group what the real -- recent offering has been.
The way that I envision a day would go hand-in-hand with the genetic registries. So you can go on the Web and look at some of these.
All right, so Mark. We consider me remind the Secretary about King's we have already done -- about things and have already done. Is that something that we should be doing at this stage?
Presumably.
But what she did not have that she does now is $85 million that she could actually devote to them should she choose to do so. So II think -- as I think about the process and the fact that this is a delivered to a group to understate this case dramatically, to come up with something Tinoco by tomorrow that the group could actually agree to it is unrealistic but if we take things that the group has previously agreed to and say this is something that could be purchased within the context of comparative effectiveness research I think that could be valuable to try and frame the discussions about the remaining announcement.
So it seems to me -- because this has to be done pretty quickly if we're going to have any effect. We could write a short cover letter that would take a document that you talked about, the overset report and the letter to the IOM and basically transmit them to her with a date of important recommendations that could be, could inform her decision making.
I think the other document that we could probably sell the portions of would be the comments relating to meaningful use because with in those documents there is specific verbage about the use of genomics and personalize Medicine Research.
Let me hear from the group, do you think that is something that would all be useful? I think we can do that quickly and something we can pretty much do offline. Anybody think that is not a useful thing to do? We will take no comment as a yes.
Any further thoughts on this?
All right Bidwell, thanks, Marc. Sorry to put you on the spot and I think we will be hearing more about clinical utility and where the field is going and some of the things you may want to react to in the fall.
While Kathy is speaking I saw David Hunt coming in. He is still over there, he moved. I think with your permission is just as well if you are willing to be before the break rather than after words. It is great to have you here. We had the pleasure of hearing from you before I found it was at our last meeting and we responded to some of the things that is really important but we have actually invited David back to give us the update on the interim final rule related to electronic health records and from the deidentification of the safe harbor identification process and to insure private security.
He is the chief medical officer for how the affirmation adoption at ONCHIT and it is always great. We were very eliminated by your last presentation and look forward to the same. Thank you very much for coming here and we appreciate it.
Thank you very much are very grateful to provide an update of our activities. When we were last together we met at the start of a quite remarkable streak of winter weather here in D.C. and that time really tested the city. It really wore out the statement that John Kennedy made about our city before that D.C. is notably a city of northern charm and southern efficiency.
[ laughter ]
While I do not know if we are more charming or more efficient we have been doing our best to be a responsible stewards. But before I go to a tape of are highlights if you will of what we have been doing since he last met let me give a quick recap of what we discussed before.
I pointed out the tremendous responsibility and trust was placed in ONC with the HITECH Act would choose a section of the American recovery and reinvestment act. It seems like another one are too small pieces of health care legislation may have passed in the 481 days since this came to conclusion. That is the HITECH Act and to them and our hope that HITECH was a prelude to significant health care reform has really been answered and while we have a few important responsibilities within the new health care reform law, we are still pretty center during the tremendous amount of work that we have been given with this legislation in 2009. Still I will be the first to admit that this form is to the pool bit more confusing so I am always thankful that Dr. Blumenthal could articulate our priorities.
And HITECH as we discussed before they are to define what the meaningful use of electronic help record is. We want to support the medical community in meeting that definition and we plan to establish a public sector which is so very important in a health care system that actually ever it is this to the full extent, information technology and finally we want to make sure that all of our work helps to foster greater innovation in this field. And to that end this is eight very high level view of our operations -- this is a very high level view of our operation and adoption of electronic health records as well as exchange of information all feed into the meaningful use of electronic help records. And the final goal as you can see is to approve individual and population health outcomes to increase transparency and efficiency. And to improve our ability to study and improve health care delivery. And finally at the bottom, under all of this B plan to continue to foster innovation and promote research that actually improves our ability to use this technology as well as the technology itself.
To that end, I will pick up in terms of the highlights I am not sure we were able to share much about this at the last meeting. This has been a piece -- or a lack of the better term, the R&D world that we are trying to catalyze in terms of enhancing our knowledge base and how we are able to leverage this technology. This is called the Strategic help IT Advanced Research Project and we're trying to get some play on the DARPA acronym from the floor.
That investment of $60 million was the entire size of our 2008 budget so we have come quite a long way. The strategic held at IBM's research project team we have a number of institutions that have -- health advanced a research team, with security of health IT which remains a fundamental importance to all of our work. Calmative support. How do we help leverage information technology to be a is a rebuttal assist device and delivery of care -- cerebral assist device and Libya. We have to optimize health and finally we have to help enhance our ability to use the information that we have in more than one form or for more than one purpose. These are just the legal organizations, the leading institutions as well as the lead researchers investigators for this work.
I should point out that each of these four areas are actually working with about a dozen other institutions and researchers so this really represents a good cross-section of the United States Health Services Research and Medical informatics community.
This work has two major deliverables. The first is the short-term deliverables. It is a bit quixotic and, please bear with us when we talk about research and short-term deliverables most would say that is five for 15 years but in most cases it is about one year and we want some significant milestones met after one year and one -- most of those are dealing with the known, specific problems.
And long-term deliverables, the second large set of deliverables we are thinking on the order of about four years or so that they will be able to produce some results and inform us of how to actually improve the health IT system. You can keep a close eye on this work and all others on our website and I will give the URL for that in just a minute towards the end of my presentation but I should also note that this group will be regularly updating the health IT Council which is one of our two mean advisory committees that meet on a monthly basis. So look out for the results of this work as we are continually updating with that.
Having said that, I looked at the research portion and our ability to innovate. I have to get back to sort of the nuts and bolts of what we have been doing. To that and one of the biggest pieces of work were centered and focused around the meaningful use as has already been mentioned of electronic health records. I purposely for this presentation have a very little information specifically about the final rule regarding meaningful use speed I mentioned earlier that that is going to be released in the late spring and you are here.
[ laughter ]
You still have a few days left of spring and we are actively working. But we expect to meet our timeline, plus or minus one or two weeks let's say.
For the final rule with regard to meaningful use and just to remind everyone also that will will speak primarily to what the expectations of around the meaningful use for electronic health records for the 2011 period. We all recognize that that has got to be a bit of a special time given the short period of time we have to ramp up all of these programs and expectations have to be appropriately manage with that a guard the rule may be to some of the expectations for the second and third periods of meaningful use which are 2013 and 2015 respectively and we will be able to outline or three about is what some of the trajectory will be for that end.
One of the important pieces, you know this might previous slide highlighted the fact that the exchange and some of its components have to be a huge support for that work and we have released an interim final rule. We set an initial set of standards back in December shortly before we met that discussed basic functionality, and interoperability and the expectation is that we will have a final rule Commines in very close proximity perhaps before the actual meaningful use rule is released in that substandard and certification will really be what is used to help support all of this work for this initial period.
You will also notice that in March we were able to release a process for how the certification of these records will proceed. The HITECH Act is complex but incredibly elegant in some of its goals and that is it has given ONC a dual responsibility. One is to create a set of standards that will serve as a base for all electronic health records as well as we have taken back the control of HHS, the overall role of overseeing the process of certification and we will look to accredit a number of entities not just the wine that is in the form -- not just the one that is in the form of CCHIT paid the expectation is that we will have a number of entities to provide capacity as well as some specification as the number of and demand different groups have indicated that we need to have testified that these records actually meet their requirements.
So you will note that the final rule, the interim final rule was passed and the final rule will be coming out in close proximity to the final rule for meaningful use. One thing that I like to point out and this is incredibly important particularly for a number of communities that have asked about what the electronic health record will mean for those perhaps who are disabled or in some way have been disturbed. We like to point out that with in the interim final rule we did highlight that nothing is required. In the expectation would be that that also holds for the final rule to be construed and other legal requirements under federal law so all important regulations that are meant to assist compliance and access to the health care system will still be in place as a result of our actions.
I think I did highlight some of this before but again, so much of our federal procurement rules handcuff me an being able to see things before they are actually final. And I know that when we spoke before the concept of Beacon communities was on the cusp of.
Released and as you can see on the cusp in February turned into May and on May 4th the Secretary and vice-president announced 15 the Beacon communities that have been given resources to actually affect what we like to think of as the full flower of what health IT can mean to the medical and health care communities. These communities just to recap and their applications have to demonstrate that they are in a little bit more enhanced in terms of some of the infrastructure and kaysix around deleveraging health IT.
So they have an infrastructure in place but most important the they have a culture that actually increases and is looking to leverage health IT to a greater sense. To that and these committees have already set up or made significant progress in having all of the imported stakeholders' at the table -- important stake holders at the table to speak of how they can most effectively use Information Technology.
With all of this elegant and efficiency there are significant numbers -- or there are a number of defects in the HITECH Act bid one of which is it is that speak to the entire continuum of the health-care community. Many specialized groups did not see themselves as having a piece of the people use or some of the incentive programs to that end and some in the mental Kolff community in that aspect. There are a number of different constituencies that really were not as fully fledged out as state primary care in this traditional path of medical specialty in the HITECH Act is.
The Beacon community program is expected to be mediate many of those defects in that these resources are targeting committees that were in a little bit further advanced. I like to think of them as our A-students. The expectation is that they have everybody at the table and long-term care. And we were asking them, what could you do with additional resources to really leverage all that can be done with health IT? We are very pleased and went to see some of the communities that we have a I think he will be pleased also with some of the responses that we had paid most of my slides as a ploy to are taken directly from my website and again the URL will appear at the end of my presentation.
Virtually everything that we do will be able to be found in greater detail on our web site. These are the communities that had been awarded the Beacon community resources. You can see the art a wide range geographically as well -- you can see that they are a wide range geographically as well and holds a Oklahoma and -- Tulsa, Oklahoma and Brewer, Maine.
I would encourage any of you to go to our web site. There are descriptions of all of the work that each of the communities are on our web page. And I think you'll be able to see that these represent really a. Exciting group and we expect a great deal from them.
I should note a couple interesting points. The first is that we have realized that we have resources to fund two additional Beacon communities. So while we have 15 the expectation is that we will be able to find two more so that process is in the works so we will have two more coming out. As old as the fact then, it sort of wrote a little bit under the radar when we first announced the concept.
That is that the VA and DoD are working together around the same concept of Beacon communities. The Nuance facets of procurement in the VA/doD system really meant that we had to go and all parallels rather than integrating all these programs but the leader expectation is that we will be able to leverage the information from those Beacon communities in that believe they will be funding a total of four and leverage the information and the results from them just as well as we do from our own HHS funded Beacon communities.
Back to some of the nuts and bolts pieces. Almost as exciting, let's say. And that is our regional extension Center program and that actually -- you have to think of as a centerpiece to all of our work and I would say that is. The regional extension Center program is a group of now 60 centers throughout the country who would be providing the technical assistance, boats on the ground if you will assistance, to practices and providers to obtain a meaningful use of electronic health records. These groups will contain an eclectic staff of understanding the new ones of security and health IT networks as well as the software pre as well as the most important link, the steps that are needed to take a practice to the total transformation that is the result of really having an electronic health record in place so to that and they have individuals who are skilled in the new ones of getting practices and have been patients blow three York office now that you have a completely electronic of setup. So they're providing again boots on the grounds technical assistance to providers.
The goal is to have over 100,000 providers through these programs of teen the meaningful use of electronic health records -- up teen the meaningful use when we last spoke we announced 32 of those grantees. And since that time, we have been able to announce another 28. The total funding for this program is on the order of about $600 million. We have a great deal of high expectations for the Centers and being able to be the centerpiece of resources in their medical community of how to actually obtain the new use of health IT.
So they will be the local experts. This program was somewhat derivative of the agricultural extension program that many of you may know something about. We have actually local experts that are able to help guide individuals with their agricultural needs.
One thing that I like to point up about our regional extension centers. They really have a charge to obviously provide resources and expertise the rub the medical community's. But they are specifically charged -- up about the medical community's and they are specifically charged to work without providers that are least likely to do this on their own. Small practices with ten or fewer physicians or conditions. Practices in rural environments, hard to reach locations and serving the underserved communities. Those are basically with a waiting room full of patients and very little time to go through a 12 step program on how to change and transform their practice.
And those are the individuals that these centers are specifically going to be working with. I should also note that in addition, since our last announcement, while we did have resources to have these extension centers work. Obviously with individual practices and also hospitals. Since we last spoke we have been very pleased to announce that we have been able to add some additional resources that had these extension centers work a little bit more closely with critical access hospice. I am not sure how many of you are familiar with those but they are, as the name implies, small hospitals usually about ineluctably remote areas. Traditionally buried very small but they provide a critical link to health care service to all this country. They are the place to go in many places in the more and rural areas. And they are sort of, if you will, an island in the sea cadet so many of these communities depend on.
We recognize that the above all others really, really need to be able to leverage held information technology to its fullest. So we are very happy that we have been able to add some resources to help the regional extension centers work in the little bit more closely with them -- a little bit more closely with them.
That is a roundup of but we have been doing and I am hoping we can have a bit of discussion, questions and answers as far as some of the other details that we have been doing. I would like to leave with a highlight. And this is my highlight. We have been doing an awful lot and in some cases some folks would assume that we are in a little bit lost given the magnitude and scope of we are doing but we are not lost at all but we will admit to have been confused for a few weeks [ laughter ]
So with that let me stop and see if there are any questions and again, this is the URL and I read the might encourage everybody to go to this web page and you can sign up for our list serve, you can sign up for all of them or a subset of that and you will be the first to find out the actual minute that the meaningful use was published in the standards and certification as well as keeping up-to-date on so many of our other programs so I highly encourage you to go and see that.
Thank you so much for the update. One of the things that we were hoping that you could speak it is some of the issues regarding the identification and privacy. We have a whole group working on the genomics data sharing and the mechanism as well as a challenge on this court. So is wondering before we turn it over to General discussion and all aspects of this, if you could share a little bit about where you are with her the deidentification security issues.
Getting back to the original HITECH legislation one of the things that you notice, the first of our responsibilities is to make sure that this information be made secure and available to help support institutions of public health. And as part of that rule, the act actually provided for the first ever key privacy officer at HHS. We have been very pleased that we have been able to bring on board Joy Fritz to be the chief privacy officer working with our group. She is the direct report to the Secretary. And with her leadership, we also have been able to began on a series of projects to help inform our policy-making with regard to security and privacy and particularly the deidentification of information.
On our web site you will notice that we came up with a white paper in the middle of March which actually is a framework for how we will be candid to discuss and began -- begin to discuss a and I will refer everyone to that white paper and the interim final rule a round meaningful use will also speak to some of the aspects around security and privacy that are there.
The most important piece around security and privacy is obviously when we are discussing the exchange of information across and outside of a particular practice or a particular provider's perview and the NHIN, the National Health Information Network has released a set of standards and protocols called NHIN Direct. Which is a baseline to begin to start to exchange and use some of disinformation and in some of these protocols you will see some rudimentary and fundamentally important requirement of around how data will be encrypted pit how it will be on itemized and how it will be actually shared.
This is going to be a learning process beyond all else and we expect that over the course of prop open the fall, Joy will be able to release a number of policy and policy statements in regards to data sharing insecurity with the expectation that we have a kernel that a sufficient to help support meaningful use by the start of 2011. So you will see more and more coming out on this whole section and privacy and security policy moving forward.
That is an area we looked forward to hearing a whole lot more about, helping people getting clinical utility of this information be I saw Mark's hand.
Yes, I appreciate the Boone quote and a Japanese philosophers said had been no location you are never lost. And I want to go back to your discussion on the Beacon communities and as you talk about communities that may be feeling somewhat disenfranchised from some of the meaningful use process and we represent one of those committees as well. I wanted to clarify something that you said about the purpose of the Beacon committees in be a little bit more focused and not my question is sounds like what you said that the beacon community is really being constituted to pull in more broadly a lot of these different areas to have discussions that are more holistic if you will and given that they are two available opportunities, are in need of the beacon committees focused around one specific area? And if so or if not would there be receptiveness from the office of the National corn interrelated to constituting these communities around the issues of genomics, personalize medicine and other informatics concerns in this realm?
I am glad that you asked that question actually. Each of the committees themselves were asked to speak to a broader range of inclusion. And as you mentioned, provide a more holistic approach. Each of them is also looking at a specific and specialists said the various. And I think to speak to the genomics and Genetics peace and I should have that -- I believe the group in Rochester at the Mayo clinic may have a little bit more of a look at that and some of the committees are making sure that the interest-rate mental health to a greater extent. Others are working in the area of disparities. So each of the committees well providing broad the Beacon of how we can actually leverage all of these technologies to have a somewhat specialized area of focus. And if you go into our web site and click on any of the links associated with these various committees, you will see able to description of what their proposal entailed. And why they were chosen to actually do this work in the generic sense and specialist sense of what they will be able to bring to the table in a specialized area.
Some are working to focus on, as I mentioned, the underserved. What is actually focusing on the integration of long-term care within the community they have a very interesting and diverse group of specialized interest so some of those to seek to more specialized interest and the final two that we will fund additionally also have that same aspect. So all of them, the one defining characteristic among all of them is that they have a. Large table where virtually everyone, but the committee has a seat.
The one thing that we know is that the delivery of health care and the offense of medical science is not complete until you have everyone at the table. Not just those specifically involved in the practice of medicine but those involved in working with some of the social this earnings if you will of healthcare. And these Beacon committees I think he will be pleased to see that they speak to these issues within a very clear fashion.
All right. I am going to ask Jennifer to talk about some of the privacy issues.
Thank you very much for a comprehensive presentation. Two questions, one is on interoperability and that is one of the key issues that has been and Apple people continued to be critical in terms of not just individual committees -- that has been and continues to be critical in terms of not just individual communities. What do you think about that and what specific initiatives do you have which was specified as one of the key issues of health care reform and I have a follow-up question.
Actually if we do nothing else at ONC, we must be able to support interoperability. This is all and only about the exchange of information and exchange of data. Which may actually be to the exchange of polish. That is a concept and not sure if we are going to achieve that but we have a number of different programs and everything that we do is trying to leave for that and and it has been pointed out that if we with our current or rather haphazard system, if we did not affect interoperability, all we have done is placed the system that we have in individual electric silos that still will not affect the full flower of health care. Still on our office of interoperability and standards actually, they are releasing the protocols and the standards and the expectation and among those standards you'll see expectations of around the ability to achieve Information.
And the earliest levels will be rather rudimentary and to that end, all certified and at checkout records will be able to exchange information regarding the patient's medication. They will be able to exchange information of dirty a problem with an olive tree as well as cyclical summary. affirmation of an allergy or a problem with a clinical summaries.
Moving beyond that we will have interoperability standards that will continue to come up and the best place to see those will be in the NHIN world.
So for instance with in the Beacon communities might you have parlayed -- parlance for them with level two interoperability?
Yes, again being the A-student, the expectation is having everyone at the table every one would be able to exchange and share and use this information so we expect to see robust information exchange and many of the Beacon Communities and I am looking towards Providence Rhode Island and the community in Indianapolis, the Indianapolis helped Exchange and the community in eastern, Maine. They actually are held -- health information exchanges so everyone expects that within the Beacon Community there will be robust exchange.
If there is not exchange than you do not have anything.
So lately one of the key pieces of that are the service providers, the vendors, not for profit and how do you impact with that committee at all, and how did they understand your priorities and focus and any pro-active or prescriptive programs with the vendor and outside provider committee's meeting it back providers that health care providers. How does that work?
In the lead up to the meaningful use rule we held a series of meetings with different groups and one of the groups that we specifically wanted to speak to was the vendor community that -- and there were actual building the systems and we listened very carefully to what they felt they were able to provide and what type of support and assistance they might need to be able to do this work. And we took their suggestions under advisement. And the Office of interoperability and standards to a large extent is there crafting the fools around the standards and requirements -- the rules are all the standards and requirements. When we spoke to the certification. When we learned more from the certification group and CCHIT was the first in the gate and they spoke to as what we could expect in the first step of interoperability and standards. It will be a continuing dialogue. It has got to beat eighth. -- be a very careful dance because we have to be responsible stewards of the resources that have been given to us. We have to be prudent in the expectations that we have.
And when we look at the comments actually that came in from the vendor community for the meaningful use rule, we were incurred by the number and breath of comments from all segments and. While we have a number of aspirational goals it appears as though we will be able to do this. We will definitely be able to meet our goals. Particularly for the 2011 and as you point out the Beacon Community, the expectation is that many of those communities have a smaller group of vendors that they're working with and I am thinking of Mayo and the Tyson Group and Utah.
Once that are in common.
Exactly.
And we will see how far they continued to advance this breed and also again I come back to the NHIN work, that core set of standards that will be used to share and send information back-and-forth freed and we will continue to see that grow. We will have more information on NHIN Direct coming out over the course of the summer and I do not want to steal that thunder but I think most will say we have a good start of been able to share and exchange information. And we have taken the advice and the words of the vendor can be these two cards in terms of what we can expect them to do put on the same token it is the important that -- many thought that some of our current circumstance maybe because we either have not been as focused as we could be in challenging the vendor community and as focused as we could be in challenging the vendor community.
So this is going to be a short cut, this is going to be challenging and a number of the pieces that you will see in the press have been speaking to that challenge paid to the greatest extent I think we believe that the vendor community will be able to step up. That is fundamental. Through the HITECH Act, they provide incentives for their customers to purchase and use their services and products. So to that extent, Steve high standard will be expected of them -- a very high standard will be expected of them.
And the rest of the Technology Center, those standards and expectations have been met and in many cases exceeded I hold in my pocket right now a phone that is the equivalent of my desktop computer ten years ago and I think they will be able to step up and do it. Also I should say it has to be integrated and that is also some of what we expect from the SHARP researchers. Many of them working particularly around the Harvard group platform architecture how to exchange information and they will continually to hopefully guide and give us advice.
A great creative Sam and Jennifer and then we will wind up this session.
David, thank you for providing the process that you're making be the question I have you look at the HITECH Act and other investments is some more like $40 billion in health IT so this is our rescue people we are all claiming that this will all improve the quality and affordability of health care. The two questions are a number one a look at the Beacon Community you are struck by these are committees that have made impressive and -- advancements and most if not all are sorted in the secondary or tertiary committees and markets. The question is how do you get to the New York, the Los Angeless, where people are living basically Edward care is not as high liquidated when you had Eastern Maine health care system.
The reason I ask that is because is that the Holy Grail and how did we get drug information, health and for reaching and release decision support particularly in areas that are advancing property --rapid like molecular genetics, how to get back to the American people?
That is a great question and I would point again to our regional extension center program. Our Beacon Communities, there are very few end of is the date cannot be that supportive although I will speak to a couple of the groups and mainly the Delta Health Alliance and Mississippi. They really are working and if you read their proposal, they really are speaking to making sure this is where people are.
The Regional extension Center is actually probably what will meet the needs, to answer your questions more than any. They are the boots on the ground that will actually be working hands-on technical assistance in this community with providers to teach the first hard to adopt, how to implement and an exchange paid within all of this and I am sorry that I gave a too short discussion. You will notice that we have a whole section on health information exchange. We have a parallel program for state based health information exchanges that are going to be working again closely with the extension centers. And the expectation is that in each of the state's providers will be able to through the extension centers where how to adopt and implement this technology and then learn how to join and plugging to lack of a better term their state based health information achieved and begin to exchange business information. Are absolutely right, many of the Beacon communities, they have actually done this to some extent and they are dominated by one or two or the are dominated by some significant forces each within their own community. I will not see each common many of them.
Again I will point to the Delta Health alliance. The committee services also in Tulsa, Oklahoma as well as the western New York clinical Information Exchange. I think that we will see even with and that the Beacon Community action were the people live. Specifically the expectation is that the regional extensions of the program working in close coordination with our health information exchange will be providing the technical assistance and the services to the majority -- well over 90% of the population centers of the United States appeared to be able to get this and for the health care providers to be able to do this.
Still with all of that is important to note that the office of the national coordinator was given about $2 billion of resources and we are leveraging about 20 to $40 billion of incentive programs to the Centers for Medicare and Medicaid services. While that is significant, that pales in comparison to the goals that we're trying to accomplish. The actual goals, the resources required are much, much higher and we're hoping that will be able to use the principles as a lever to advance even further. The notion that we're trying to overcome is much greater.
Jennifer and I worked obviously a lot with the privacy Rules and I wonder if you could sort of see some of your thoughts about this.
Sure, absolutely. I can address your question about deidentification. We did hold aid to the workshop earlier this year and the sea on the topic of the -- in DC this year on the topic of deidentification and a variety of topics. The safe harbor method and the experts statistical methods and other topics including reidentification issues. The workshop is linked on our web site so if you go to our privacy website there is a link where you can see the web cast, the workshop as well as presentation materials that were there.
We will be issuing guidance on deidentification at some point in the near future and there will be opportunities to submit comments to the new guidance. There was a link on our website and will be again if it is not currently there to submit comments.
Do you have anything specific to genomics as part of the DA did the -- as part of the deidentification process?
That topic did come up in the workshop and as you are we're is not listed currently has one of the 18th identifiers with NBC -- within the safe harbor method and that is something that we're happy to receive comments from the community on.
Thank you.
And if I could dovetail in that, that was wonderful. It ain't all of the information you can by keeping it in balance I need to learn how to do that.
[ laughter ]
HITECH also speaks to this important point and we do not like to highlight this for -- this very much but it also speaks to the enforcement issue of privacy and security. And the hope is that is in addition to the policy the increase resources for the enforcement of privacy and security will go a long way.
Thank you so much for joining us. His office the critical for the well-being of our health-care system -- it is of is the critical so thank you so much and we look forward to having you back and I am sure we will hear more and thank you Jennifer for your additional comment.
[ applause ]
Before we break for lunch you notice that we have the second edition, Version 2.0 of the menu. We had a little, what do we say -- exchange of information. You have a larger sheet so those of you who want to order please to it at the beginning of the break because they need to get it ready. Otherwise let's go ahead and take a ten minute break and we will be back here at five-oh.
NIH SACGHS Meeting on a ten minute break and will reconvene at 10:50 a.m. Eastern time.
As I mentioned earlier we are going to move on to the next section on CLIAC and Bin Chen will talk about the recommendations for biochemical testing and I know she has some questions about us to respond to and this certainly is a major focus of our oversight report on genetic testing so welcome, Dr. Chen and we look for to your presentation.
Thank you for the introduction my name is Bin Chen and I am from the CDC's new office of laboratory services and I can say it is great to be here but two of the youth on the CDC's efforts on the guidelines of genetic testing.
So for my talk ion going to highlight the recommendations provided by the Conoco -- I'm going to have a the recommendations provided by the advisory committee highlighting the practices in genetic testing is particularly the recent recommendations for biochemical genetic testing and newborn screening and I will be discussing the development of the CDC guidelines and also our upcoming new guidelines for biochemical genetic testing and newborn screening and then at the end as Dr. Teutsch pointed out we have some issues and questions for input.
Just a brief overview for the current landscape of genetic testing the CLIAC regulations apply to all patients performed on the U.S. a patient specimens and correctly because most genetic tests are considered high complexity and testing so laboratories providing these tests are subject to the general requirements of for non testing and a --in the personal requirements and there are also specific requirements and the qualification requirements for the Technical Supervisors. There are no known specialty requirements for pilots, co genetic testing because these are not considered a -- biochemical and besides regulations, FDA has regulations for products and the oversight is also enforce such as the MRA.
Some state programs such as New York State and Washington state and other organizations maim have more specific standards for genetic testing laboratories. Also they are the -- There are the professional guidelines but in addition there are expert practices implemented by individual laboratories to comply with dilatory requirements and also add here with voluntary standards and this is the realm that we tried to help with, with the guidelines.
So the guidelines that I am reporting today are based on the recommendations of the clinical laboratory and advisory committee which is eight federal advisory committee dislike SACGHS and CLIAC was established back in 1982 and provides technical [ indiscernible ] to the government including clinical laboratory standards including CLIA regulations and their impact on medical practice and a CLIAC also hopes regarding the need for modification to existing CLIA requirements to accommodate advancing technologies. CLIAC reports to the HHS secretary and the assistant secretary for help and the leading officials at the CDC, CMS and FDA end is managed by the Laboratory of science and standards which is where I work in the CDC.
So they need for improving quality assurance practices and oversight was recognized back in the late 1990's and in 1997, they started to work with advisory committees including CLIAC and other stakeholders to improve the quality of genetic tests and to enhance the oversight for genetic testing by providing guidance and training to CMS by developing educational materials for testing laboratory in regarding CLIA compliance by collecting data on quality issues the.
In 2008 we published the report costs of genetic testing which supported these approaches and in the temper 2008, CLIAC provided recommendations for practices and molecular genetic testing and a document to address biochemical genetics testing.
In 2009 date published the report publication that incorporated the CLIAC recommendations for molecular genetic testing. So this was the guideline for molecular genetic testing and this is a comprehensive document that includes discussion on the oversight issues and the Quality assurance recognized in the molecular genetic testing as teleprocessing on how this guideline was developed in the main body of the document provides recommended practices to address the issues that had been recognized as specifications for laboratories to comply with the existing CLIA requirements and also to provide specific Quality assurance guidance.
So the recommendation addresses of further testing process including the pre analytic and post analytic spaces of molecular genetic testing and the recommendations also address laboratories regarding authorized persons and particularly in situations where direct consumer genetic testing it evolves.
Also the recommendations addressed laboratory practices to ensure comfort and agility. Laboratory personnel qualifications and competency and the factors that should be considered and the quality management system involved in molecular genetic testing.
So on the biochemical genetic testing front, in 2009 a CLIAC work group was formed and worked very hard to formulate and put regarding test cases for biochemical genetic testing and a newborn screening. As of peppery 2010, the committee provided recommendations for biochemical at the genetic testing and the newborn screening that is performed for diagnosis of newborns so similar to the practice issue of what molecular genetic testing the CLIAC recommendation, including the total testing process, personnel qualifications and competency factors that should be considered before introducing new tests and implementing the quality management system approach and biochemical genetic testing and newborn screening.
A copy of the CLIAC recommendations is included in your binders and for the audience, you can download the CLIAC recommendations from our web site. So curly CDC are in the process are in developing a new guidelines to include these CLIAC recommendations.
I would like to highlight some of the recommended practices that have been or will be included in the CDC and MMWR guidelines. So the MMWR document for molecular genetic testing testing addresses heritable diseases as well as aspects of tests that encompasses those molecular genetic test procedures and other methods.
And the upcoming four MMWR four biochemical and newborn screening will address genetic testing for screening diagnosis and management of metabolism to include biochemical genetic testing and this includes the diagnostic biochemical testing performed for presumptive cases out of newborn screening and newborn screening for metabolism. And this will also address the biochemical aspects that encompass both biochemical genetic test procedures and other test methods.
So the recommended practices for the P analytic -- pre analytic phases encompass these practice issues. First of all, the information that laboratories should provide to users of their services and users of the laboratory services include the health care providers, patience, pagers and other individuals who may request a -- patients and payers and other individuals who may request a laboratory test results.
And the ability for informed consent and laboratory practices pertaining to test request. Specimen submission, handling and referral as well as pre analytics system assessment. Also a little bit more in that first issue.
So as a general principle of, lavatory's should provide information on the genetic tests that were performed -- laboratory should provide the genetic information and this is, first of all, the selection and request across the genetic tests and this is typically because genetic testing is -- For most as agents and conditions is using laboratory developed test. So without Laboratories making the specific test information available to their users, it will be difficult for providers to access this information. Let alone to correctly use these genetic tests.
They should provide information to facilitate a puppet collection handling Transport and submission of patient specimens -- appropriate collection of handling, transport, and submission of patient specimens and the require laboratories receiving patients specific information to start with and this may include the patient's race, ethnicity, a family history and other clinical information.
So it is recognized that laboratories should inform and communicate to their tester requestors. So that laboratories can facilitate the prompt initiation of patient testing. Alternate the laboratories should be responsible for providing information that is necessary for informed decision making by the patients and their health care providers.
So here is a brief summary of the information that laboratories should provide for each molecular or biochemical genetic test that is performed. So just quickly, this includes test methods to be used. The specific analytic performance specifications, clinical validity information and of the mutations of the test for whether the test is performed using FDA-approved systems or clearance systems to facilitate a proper Transport and submission. And the types of patient information by the laboratories to have implications for family members availability of the laboratories consultation as well as information whenever possible.
You do see that this list covers a lot of information proposed for the genetic test registry and I think it is helpful to find out that it is already recognized through the lab requestors that they provide this type of information.
In terms of informed consent for molecular and biochemical genetic testing, laboratory should provide users with information necessary to make informed decisions and this includes the pieces of the information. Whether informed consent is required or not, if you recognize that unless mandated, generally it is not considered the job of the laboratory. We're informed consent is required lavatories are responsible for assisting health-care providers in determining the proper levels of informed consent and those laboratories should provide appropriate means to facilitate the documentation of informed consent.
For newborn screening, the recommendations are that explicit parental consent is not necessary for mandated public health newborn screening that meet the expected criteria. The new tests that do not meet the accepted criteria, explicit parental consent should be required but also parental and provider education should be to go to the newborn screening programs regardless of consent requirements and research use of tested specimens and should have prepared a human subjects protections procedures.
For the analytic testing phase, it addresses the astonishment or verification of the specification as well as documentation of available information on clinical validity the MMWR provide specific guidance for quality control procedures in molecular genetic testing and the upcoming MMWR for biochemical genetic testing and newborn screening will address the specific quality assurance issues in this area as well as other specific analytic issues including the reagents.
They should follow the recommended practices for alternative Performance assessment when these programs are not billable.
So the post analytic phase overall should provide information for interpretation of the test results by health-care providers and other individuals who may use genetic test results. The content must comply with the general CLIA requirements and should include the recommended additional information event I would refer you to the wall like a genetic testing MWR and the CLIAC recommendations for additional information.
In terms of the report retention, molecular genetic test reports should be kept as long as possible, at least 25 years and biochemical genetic testing reports should be reclaimed for a least 21 years. The retention of the newborn screening test report is subject to CLIA and applicable state requirements.
The retention of test records must also comply with CLIA and other applicable requirements in terms of retention of tests for the specimens for molecular genetic testing, the specimens that are stable should be retained as long to meet the Laboratory at code needs to conduct Quality assurance and quality approvement and should be kept until the next testing to love the opportunity to identify possible errors in patient testing and take corrective action.
Pour biochemical genetic testing, specimens should be reclaimed as long as possible. At least until after the final results reporting. And it's possible these specimens, after completion of patient testing should also be cleaned until the next PT or alternative Performance assessment.
The retention of newborn screening specimens is subject to federal, state, and local requirements and the upcoming practices in the upcoming MMWR will also address post analytics systems assessment for the biochemical genetic testing and newborn screening.
In terms of personnel qualification, Laboratory directors must meet the CLIA qualification requirements for high complexity testing. Technical Supervisors for molecular or biochemical genetic testing should have an equivalent qualification to a CLIA requirement for clinical cytogenetics or have current certification and genetic testing approved by HHS.
For newborn screening they must meet the CLIA certification requirements and should have for years of experience in newborn screening and should also meet any applicable additional state requirements.
Clinical consultants, General supervisors and testing personnel must all meet applicable CLIA qualification requirements and should have training that is relevant for the testing their laboratories perform.
So the intended users of these practice guidelines certainly include laboratories that perform molecular or a biochemical genetic testing for newborn screening for inherited metabolic diseases. We also expect that these documents will provide a useful resource for these economic laboratory entities for example, users of laboratory services to include their utilization of genetic testing services for professionals such as a laboratory inspector and standard settings organizations and professional societies where they develop new standards and guidelines for federal and state agencies in consideration of genetic testing quality issues. For IVD manufacturers and also for the general public to help them better understand quality pertaining to genetic testing.
So currently we are working on, to promote or provide information dissemination for the molecular genetic testing, MMWR, by developing all of the specific educational informations for just the stakeholders such as laboratories, health care professionals and consumers appear also by providing answers to frequently asked questions on the web site and that is why we continue education activities.
And for the upcoming MMWR for biochemical genetic testing and newborn screening this is currently a collaborative effort that we get input from CMS and FDA and we expect to publish this in 2011.
So we hope that these guidelines will help to improve the quality of laboratory genetic testing to enhance the oversight for genetic testing and the current framework and improve the test care outcomes for patients to receive genetic testing. If you need informations regarding the review recommendations on biochemical DEC testing or newborn screening or on the molecular genetic testing MMWR you can go to our division at's website that is on the right-hand side of the screen, and this is available on the CDC's MMWRs sites.
The development of these guidelines has seen huge efforts. And we would like to acknowledge CLIAC 1997 as well as the CLIAC testing work group, the biochemical testing workgroup and last but not least we also need to acknowledge the CDC participants. I am delighted to see some of the participants in this effort here in this room. Dr. Andrea Gonzales and Dr. [ Indiscernible ] Pratt both precipitated and Dr. Mansfield from FDA from the CLIAC genetic testing work group.
So I have some questions and issues that we really appreciate input from this committee so before we go out to the committee for questions, can ask, are there questions so far?
We have limited time so let's just take a couple of questions here and then I suggest that you go through these but we have to provide some of those of line.
Let me start with Liz.
I just wanted to kind of correct one little thing that has been said for anybody who was brought pressure shot up when she said the FDA is regulating [ Indiscernible ] we actually are not doing that now.
[ laughter ]
Thank you.
I actually am going to skip to the first bullet and Q3 specific areas that you need clarification in the post analytic phase, you state that the laboratory report should be electronically compatible and in the informatics will that would be defined as not adequately explicit and what does that mean? A PDF paper file and electronic health record would technically be compatible? There are huge issues with this and I will not go into great detail about this relating to deficiencies and our current ability for electronic information systems to represent genomics data and other types of data.
I think you are really serious about this there has to be a lot of attention to what exactly you really mean here. Also in post analytics in clarification, 6A, genetics consultation you indicate that the CLIAC articulates genetic counseling services and again as we have reported out of this committee in the access, there are significant issues relating to genetic codes the reimbursement and I think if you really are encompassing this -- and this is a recommendation for CMS that this is really critical then I think it is incumbent to address some of the issues of how we actually coffin's the genetic counselors particularly using the genetic counseling code that was greeted but was elected not to be covered by CMS.
And then lastly the issues relating to personnel. This has been a recurring problem and this may reflect some ignorance on my part in actually reading the CLIA requirements relating to these different categories. But over and over again they're talking about being a MB, DO, or DPN and it is interesting to think about how we podiatrist would have any impact but they could run in laboratory it looks like as I read this but in none of these areas are individuals with certification as Ph.D.es in biochemical or molecular genetics been articulated as knowledgeable and a laboratory supervisor and, in fact, the only place that Ph.D. s are reference relates to clinical consolidation and while we do have a handful that are providing clinical consultations, the majority are working in laboratory conditions and many of them are directors of those programs. I think that is something that really needs to be clarified with in these recommendations. But that is something that has also been talked about at this committee and I believe is also reference did not access report from a few years ago -- referenced and our access report from a few years ago.
I do not know but under CLIA you can be a Ph.D. actually and be able to be a consultant or supervisor and that needs to be reflected in that document. Thank you for pointing that out.
[ Captioners Transitioning ] Please hang up the phone so the new captioner can dial in. Thanks.
I wanted to recognize the amount of work to come out with some extremely valuable recommendations that will be useful for everybody in the laboratory community. I want to thank you for that. The second comment that I have, have you talked to anybody in the [ Indiscernible ] program to see what are some of the most citations for these laboratories? Maybe develop a FAQ to the specific areas to provide more guidance.
Well, we are trying to -- well, reach out to [ Indiscernible ] organizations and accredited organizations to see how we can collaborate and promote the use of these -- the use of these MMWR recommendations. I should say some of the MMRW recommendations were developed to address the deficiencies identified through laboratory inspection. For example, the lack of documentation of [ Indiscernible ] performance assessment and other deficiencies. It is our hope that by providing recommended practiced for those issues we're not only helping accrediting organizations, but also helping the laboratory community to develop improvements in these areas.
The way that we have worked in the past of trying to educate laboratories of good practices is through workshops at the annual meetings. Like an early bird or some other companion organization. Maybe you can line up with some of the professional societies and provide a workshop. You just need to get a room. You will be amazing how many will attend these sessions, make it very interactive with the individuals. That will bring [ Speaker/Audio Faint or Unclear ] to able to learn more.
Yes. We have been doing that. We provided an early bird session at last year's session for molecular pathology meeting. We also provided updates at the meetings of American college of medical genetics and the [ [ Indiscernible ] screening and genetic symposium and other meetings. We plan to do that, not only to promote the document but also for the upcoming MMWR.
Thank you very much. We really appreciate it. This is clearly important work. It's great to see all of the incredible amount of effort it's taken to bring this to fruition. We look forward to see you how this gets implemented. I would encourage you if you have additional comments and thoughts to feel free to get them back to her directly. Thank you so much for joining us. We really appreciate it. [ Applause ]
At each of our meetings we set aside time for public comment. We both welcome and value the comments we receive as part of the public forum. So we move to that part of our program now. I will ask our presenters to stay to the five-minute limit. You should have in your folders written comments, which will be made part of the record. Let's begin.
Martin, welcome.
Hello.
Very good, thank you. My role has changed, I'm the chief of staff. I have been in that role for about two years. -- providing research tools and tools for applied science and evermore useful in the world of health, health research and healthcare. So if you are anything like me, you are getting pretty darn hungry. I will try to keep this quick for all of our benefit. My remarks relate to this afternoon's session on implications of the affordable human genome. Last week I attended the American society of clinical oncology conference in Chicago. As you know, cancer is a disease of the DNA and the field of onology incorporates biomarker analysis into many decisions. Presentations on analysis were extremely well attended at the conference. I can tell you that over a thousand people were in that room. There's a tremendous thirst. People are interested and looking for ways to interact with the technologies. Today's diistic tools are good, but there's room to improve on them. JenThis meeting about the implications of the affordable human genome couldn't happen a more crucial moment. I would like to cut straight to some considerations and suggestions that you might take in mind as a committee. And then what I will do is back up and talk about some of the actions that are underway at life technologies and a little bit of the rationale for these recommendations.
So these suggestions, first, about four here. First is to establish rapid biomarker review mechanisms to ensure responsiveness in incorporating it into decision rules. Second, a consideration of evaluation criteria for genomic technologies used in medicine. Emphasis on the raw accuracy of sequence detection events. We also encourage the continued facilitation of a development of information exchange networks incorporating genomic patient data. This should build on other efforts to establish electronic health records. It would be helpful to integrate support rules in a broader HIT initiative. Fourth, include training and education. So with that I will explain about what Life technologies has been doing. We've been doing evaluations aimed at bringing whole genome analysis into practice. We announced a small pilot earlier this year through a partnership to use whole genome sequencing to find novel options for 14 triple negative breast cancer patients. We're developing -- we hope to make a difference in those patients' lives as we do so. Last week we announced the genomic cancer care alliance. This alliance of eight entities will evaluate the use of whole genome sequencing as a method to find cancer options. The study will use whole genome sequencing. A dedicated tumor board will meet to discuss every patient, developing treatment plans that are Taylored. That study will begin enrolling later this year. We are the primary funding sponsor for this partnership.
Our second expectation is that genomic inquiry may prove more powerful [ Speaker/Audio Faint or Unclear ]. Clinicians are allowing the genome to speak and are open to treatment options based op what they find. The findings may suggest the use of drugs that are currently approved for other indications.For complex tumors a combination therapy approach may be recommended. And these treatments may be different from the decision rules that are currently in place. Just to work within my five minutes here I will pass some other stuff here. To ensure that biomarkers are measured and comparable standards may be needed nor technologies. And it's essential that users of the technology be able to discern true genetic variance from test art facts. To that end, as the FDA explores quality criteria for this sort of instrumentation it's imperative to consider raw data.
There are two measures of accuracy. One is the accuracy of the raw data coming from a sequencing detext event. The other is the accuracy of the data after analysis, processed data accuracy. We believe that raw accuracy is the driver of quality and cost for genetic information. With high raw accuracy coverage is more about seeing deeply into the biology, not about compensating for error rates. Another element is facilitating data exchange standards.As a thought on this, as diseases become evermore stratified patients with specific biology may be further and further apart from one another. You may need to compare patients across different medical systems. This introduces a different need. It's about being able to see across a broad population. The medication education part we've discussed before. I will leave that. We believe that whole genome sequencing is compatible. We believe this will not be a net costed a. And that it will be lead to better health outcomes and better cost. There's much work to be done. HHS has a major role to play to ensure that genomic medicine is available. Thanks for consideration. I'm here for the next two days. I'm happy to talk with you.
We have time for one or two comments. Good. Thank you so much.
Sure.
We appreciate that. We take into consideration as we go into the afternoon. Is [ Indiscernible ] here? Let me move on to Mark, who we welcome back from the association for molecular biology. Always appreciate your input.
Good morning.Sequencing technology is advancing at a rapid pace with the cost and turn around time of processing a simple sampling decreasing, buts idea of sequencing an individual's entire genome. In time whole sequencing will make targeted molecular tests less effective.AMP commends the committee for addressing the related challenges of policy and practice. As the technology advances AMP's concerns focus on the clinical applications. The add vent or adoption of the technology is not controversy, but how labs use it, and how physicians use it to inform decision making. On a broad level currently marketed molecular diagnostic tests, the current state of the art may consider whole genome sequencing as a screening test. As all testing technology advances AMP believes that reimburse policies should be modernized to represent the value of the information. The wealth of data obtained in whole sequencing creates new practice questions that pathologists will have to address. AMP believes that cornerstone will be the assessman of a clinical utility. How can 3 billion base pairs in the sequence be coupled to clinical utility? The answer is that there will be difficulty for molecular pathologies to associate meaning. An effective approach will depend on a multidiscipline research agenda. Also there should be a central repository to commit clinical and analytical data to further inform the interpretation and clinical utility of results. The vast amounts of data will require investments in bioinformatics technology. Measures to standardize the data for entry into EMRs and to simplify the reporting of sequencing results will be needed. When healthcare providers request whole sequencing they may describe a phenotype or symptoms. AMP is concerned that this presents a quandary. Members question whether this is responsible and appropriate to only report on the test indication, or whether there's a duty to report all findings. Members will need to consider whether they should mask the nonrelevant data and only report on the test requisition. They will consider whether to then report data as new evidence as new information becomes available. AMP encourages the committee to consider the appropriateness of masking data irrelevant to the test ordered. AMP has provided comments to the past on the issue of DNA patent's. DNA patents on specific sequences may restrict the reporting and interpreting of the full results of such testing. This links to our concerns about clinical utility and the duty to report all results. Confounding these issues are the anticipated communication gaps among the laboratories, physicians and patients. Molecular pathologists will have the added responsibility of educating about the complexity of genetic associations, risk information and laboratory decisions about which data to report and in what matter. They will have the challenge of communicating the significance of this information to patients. With other areas AMP has recommended that the committee continue to explore the provider, education, and training needs. AMP is aware that the committee has released a draft report on the education and training of healthcare professionals. AMP wishes to clarify that next generation sequencing techniques should not be viewed as whole genome methods even though the same technology is used. [ Speaker/Audio Faint or Unclear ] versus whole genome sequencing is another test using evolving technology, but used in a targeted manner. Lastly, AMP values the role of whole genome sequencing. AMP members realize to truly capture [ Indiscernible ] the results must be compared and contrasted. Similar to our previous certains molecular pathologists will have the responsibility of determining what should be shared with the patient. The committee will need to address the ethical challenges.We thank you for your attention. We look forward to partnering with the committee to explore these challenges. We have forked a working -- formed a working group.
Thank you very much. We'll be talking about a variety of those issues over the span of the rest of the day? Any comments or queries?
Thanks so much. Hopefully you'll be joining us for the discussion.
Yep.
Is Dr. Schafer here? Any others who had something they wanted to say as part of the public comment? Mark?
Yes. This is actually reflecting back on our task to come up with something for the imperative effectiveness [ Indiscernible ] tomorrow. There was an article by Dr. Angriss behind Tab 6 that called for phenotyping. I had added that to my list of things to talk about that could be added to our CER Christmas list. For your consideration I would like to put whether or not a comment in investing in feasibility of collecting phenotype data in a standardized way would be something that would be worth while to propose to the secretary from the committee. I apologize I had neglected to mention that before. But I think this is a very powerful article. I think there could be an opportunity here.
Any -- for those of you that have had a chance to read that, any thoughts about that? Mark, I suspect we need to hear more -- to understand the implications and what we actually would want to say. Do you want to formulate that tonight?
Absolutely.
Those of us who did not have a chance to read it will have a chance to look through it at the -- as you said, in Tab 6. All right.
Rochelle, one thing. We had some comments from Kathy this morning regarding the ACLU case. I didn't know if you had some additional thoughts to share --
If you don't mind.
Sure.
It's a good follow-up on the representative from AMP. AMP is the named plaintiff in this case that was just decided by the district court in New York. Now Kathy mentioned that the case concerned product patents and the process patents and the validity of both. One thing that I was uneasy about is she acted like the court was acting improper by not citing past cases. Courts don't talk about cases irrelevant to the question asked. In the past questions they've been on utilities, how inventive the invenges has to be, none of them have asked the questions in the AMP case. There really wasn't very much for the court to cite. The Supreme Court case on this is man made microorganisms. The case that he found was [ Indiscernible ], which was also decided by his own court, the 7th direct in New York in 1911. That was a case about adrenaline. That's since been looked at by other courts, which judge sweet cited. The issue about whether a natural occurring substance is patentable is if it's different in kind when it's isolated from when it occurred in nature. Judge sweet found that's not true in the case of DNA. So far as diagnostics were concerned he did not think that gene products were patentable. And then on the question of simple association, the association between the BRCA mutation and breast cancer he said that pended on Bilski.
Now I did agree with Kathy that the chances that the federal court will affirm the decision are miniscule. There's one other issue in the case, which the judge dismissed. He found that these were not patentable under patent law. Free expression, the ability to communicate with your doctor is inhibited if these genes are patented. The judge didn't decide that issue because he dismissed the case on the ground that these patents were invalid. Were the federal circuit to reverse this issue would come before the court. I don't know what the courts would say about that. I think most people think it's a pretty unlikely claim to win.
I also agree with Kathy that the district court was really writing for the Supreme Court's attention. The question is whether or not the Supreme Court would take the case. For the Supreme Court there's a fourth issue, that's standing. You can't just bring any case to court because you feel like it. You can only come to court if there's a direct injury to you personally. It's not clear that the plaintiffs that were brought before the district court have standing. If the Supreme Court were to look at it they might just decide not to take it. They seem to be having a lot of trouble with Bilski. That's where we are.
Thanks for that update. We will continue to follow that with interest.
A couple of things before we break. One, I wanted to acknowledge David on the phone. We appreciate you being here. We need a count of those interested in going to dinner? Anyone who is interested would you please raise your hand so we can get an accurate tally? Alison, are you counting? Is that it? Okay. Thanks, everyone, for an interesting session. We will take a break for an hour and be back at 1:00. Thank you.
[ SACGHS on lunch break until approximately 1:00 Eastern Time Zone ] All right folks pooh we are going to embark on what is going to be a fascinating discussion so welcome back.
The rest of our afternoon is really dedicated to the policy issues that surround the anticipated development of the affordable genome sequencing and our colleagues have been working to assemble a terrific set of speakers except for the last one.
[ laughter ]
Where they were desperate.
[ laughter ]
It does is an exciting development and raises a host of opportunities and challenges and some issues that we began to hear about in the public comments earlier so with a reminder that a lot of the materials are in in I will turn it over to this afternoon discussion. Thank you both.
So thank you to the Committee for the opportunity to organize on this work and then want to thank Steve and Kathy and many others for the opportunity so far. I will try to fool you that I am going to be witty and not saying very much right now.
I will say that many great geneticists have had a motivation underlying their commitments to the field. Some of the founder's 6 feet World War II and others have trouble --ing family histories in some of you do not know I am a triple and what of my other siblings is eight co-chair on this committee and she is known around my family as [ Indiscernible ]. The other of the triplets lives here in D.C. and on Pennsylvania in the big white house. If you could recognize my twin by his good choice in pies and I can only say that I all along with John Kennedy and Rush Limbaugh have been working for that birth certificate since last year is so I hope allows me to share more ties with my brother.
There have been waves of popularity that have afflicted the field of human genetics, eugenics and they appealed to the social part of them and I never would have got married if I did not produce my certificate.
And then, of course, of the early interest in public health issues and the broad interest in the community and those things and later areas of popularity include the integration of genetics into the medical mainstream and, of course, the human genome project which commanded great headlines and produced gains. We have Nell answered I guess the era of hold genome sequencing with billboards being the most popular and this one that was actually on the tabloid.
There has been a dearth of hard facts proven clinically and many false starts in human genetics. In fact, with respect to Lee Hood, the four Ps are practically unproven, probably wrong and oversold, potentially dangerous and wasteful, and personally useless. I would hope that as we evaluate all this information be could reinstate the high ground and try to show humility and respect.
For me, the core issue in this affordable genome charge has affected us for hundreds of years while it is true that the White House never saw a greater brain and then when Thomas Jefferson dined alone he was no human geneticist. Our results were over a century ago and now more show that all men in fact, all people are not created equal. A major impact of the whole genome sequencing will be to Iran quickly prove that and identified said -- will be to quickly prove that and identify that for those living in this country. The results and how they will be used for the benefit and not harm of the individual from conception to death are the core teams and challenges that drive me.
We may adopt policies and additions or write provisions in a report that are not a logical result of the data that we have considered. We have done this before in my estimation. Why? Because they serve the principles and are the right thing to do. For me, the state legislation that preceded it for two decades were examples of that making the advances that will surely result in full genome knowledge and personalization of medicine available to all is another great causes. Affordability of good and hopeful technology which really means the availability of new and insightful messages is as close to a non enumerated right in this country as any I know pretty imagine asking Jefferson or Lincoln who probably had a connective tissue disorder or FDR or all those present to gained ambition and energy what they're genomes tell us. There is no gene to be found by any means [ Indiscernible ] or potential for success in our world, and our genomes whereby hold genome sequencing. There may be clues to better treatment to happen lives for some answer me great promise and hope and we will hear today from a cell culture and our country now dominated by traditions and the moral obligation to first do no harm but.
Are charged must be to critically analyze the notion of genome sequencing and what is or could be and make it a benefit to us our families and all of our neighbors the.
Thank you and I am really pleased to do be part of this.
Good afternoon. I am very pleased to be here. Uh oh, this is a PC and I am a Mac person as you can see [ laughter ]
So we can to sequencing but I do not know how to work a PC.
So Paul has set the tone and I would like to give a little bit of philosophy even in 2006, eons ago, they actually saw all of this and said that Gene no medicine is the basic to help and we want to intervene but how do we know who and what symptoms and what to do? So this is a great way to make a diagnosises at who is at risk and this would be summit make the orders of magnitude more effective and in 2009 they released their top-10 health industry issues.
And UC that genetic testing and some discussions alluded to that and all of its pilots' bid technology is a powerful health extender. And then in 2010 they released everything there that is germane to this discussion on whole genome sequencing the technology backbone and labor shortages and there are more astronauts in this country than genetisicists.
They have much more fun.
[ laughter ]
In the old days in 1935 we knew that there was a code that said in a visitation by an Angel [ laughter ]
And Henry can attest to that. I did not make that up. Today, what are the gaps that have to be fulfilled from genomics content to clinical context and organizational, not even as many as estimated and, of course, the individual and societal needs.
So this is what I call my slide, on the rate is Nirvana -- right is Nirvana and you will understand why eight. Should we talk about everything that is actionable and we can fix its? And then I said no because then there is nothing to talk about because we are somewhere in between. So we have the speakers to address their expertise and some of the public's that we heard in the ugly duckling stage that will occur in the next year or two.
Net out a few logistics' -- now a few logistics. So be on time and if that does not work, Mike Quinn and I will jump up and if you ignore us, then our chair will come get you.
Brevity is the soul of wit, therefore you should be brief.
You are allowed one or two burning questions after each speaker and they have to be clarifying and without the clarification they make no sense a.
So now I will end with a poem. All of you are familiar with the first stanza but not the last one. Still thou are blest, compaired with me -- and with that I will turn it over to the next speaker.
Thank you.
So the first speaker is Deedrick who is the chief executive officer of the Institute for individualized medicine which is an industrial size transitional genomics House and Dietrich has worked on many different things and prior to this work and prior to the border and undocumented people legislation in Arizona he was a senior member of TGEN and later moved to be the deputy director of that institute and he will tell us today about whole genome sequencing.
Thank you and thanks for the invitation to be here today. I have been passed and you a brief introduction of next generation sequencing and we can all anticipate that will be feasible to get a reasonably inexpensive and reasonably accurate sequence and set the stage for the following speakers.
So why would you want to sequence a cold genome, first of all? There are a few applications and what is in the research stages and I will try to give you a couple of teasers up why you'd want to do that simply so I can frame the contents. I have also been asked to give you some definitions surrounding whole genome sequence so when you hear about the different nuances you can sort of have an understanding to tabulate. I also want to give you a sense of the trajectory of the technology and we all recognize that the field is moving very quickly but perhaps more quickly than we would anticipate or in some cases that we would like it. How do we extract maximum value from the technologies and I will say how you can specifically sequence those and what you want to do that.
And then just touched on in the vein of nomenclature we talk about a second generation or third generation sequencing what types of technologies are we talking about and what are the different applications and the shoes of around those.
Finally I will not talk about assembly analysis and interpretation of the genome but clearly this is the night that I think this oversight body needs to grapple with as we turn on these very powerful technologies and I would just leave it at that.
Why do we want to sequence a whole genome? There is incredible value given that all human diseases has a genetic component and articulate of that around both monogenic and chronic, complex disorders has value so what is that value? The value comes in terms of being able to predict what your risk is in achieving a differential diagnosis on the diagnostic side. And then on the therapeutic side really understand the core pathogenesis or the biological networks under all of these diseases is the core of developing knowledge base, a very tightly targeted therapeutic.
So what are we going to be doing with the technology coming on line and I will talk to you about how you reach a price point and we are talking about in the early part of next year. What can that power in terms of research save. We have a huge backlog of clinical samples that is very will -- will characterized . What we can do is take lots of individuals with those diseases and versus lots of individuals without those diseases and simply sequencer to allow the entire genome and extract that inherits inheritability.We can find not only the common it variant floating around in the population but those more rare variants as well as modifications of the genome and all of these technologies that we are talking about today to articulate epigenetic variation and also copy number variants. So they comprise the vast majority of heritability and what we can also do is understand the creation of individuals with is these diseases and we know that optimism spectrum disagree is an umbrella diagnosis -- autism spectrum of disorder is an umbrella diagnoses and at one end it is aviary high functioning and, on the other hand, we have kids who have classically been labeled with mental retardation and we have nuances within that umbrella diagnosis upset these technologies have the ability to tease out a specific subclasses of those diseases which will form the basis of a new molecular nomenclature and the foundation for a new wave of exposure epidemiology and drug development. So that is what we loosely call personalize medicine right now but these technologies will power that classification.
On the diagnostic side in terms of articulating heritable risk we have had this with companies were really you'd try and whip in an entire population of individuals which you see on the upper left-hand side of the slide began to understand which of those individuals has a higher risk for a specific mutation. This is still largely in the research phase. But how can you classify across population individuals who have the highest heritable risk for a common disease and treat those individuals differently across their lifetimes?
Is still largely people been in the research there needs to be research on how would you take this bye% of people with the highest terrible risk for this disease and tell them strategically -- of 5% of people with this heritable risk and diagnose them earlier and what drugs they may respond to prove this is one application I think of full genome sequencing.
We have seen that genome and sequencing technology is have an incredible ability and on the research side, the full genome sequencing can be a better insight into the biological circuits so that you can team to look at them or move more quickly into these common diseases.
This is a teaser also of the future to, but you can also start to a debate all of the types of molecular data and -- aggregate of the types of molecular data and this is a group around release using the human genome to perturb networks with the network been highly on drug development.
And you can envision a day in five or ten years this group will be asked to look at this would happen to be taken individual genome and banging up against the network in the sky that is a steady state never come how the human body behaves and get an output of what drugs the individual might respond to specifically and uniquely without any president and how do you regulate this type of activity?
Specifically in a cancer space and you will hear some information about this today. How you use the cancer gene known to stratify the standards of care today to develop experimental therapies so that individual patients can have an improved outcome? It is not as far off as you think.
There is a lot of exciting step in the research space and I think as a group we recognize that and are comfortable with the fact that these tools have a credible or value. I think where we start to get concerned is how do we in the face of declining costs for Human genome sequencing and an incredible body that can be applied to a human genome sequence, how do we in a very short timeframe become comfortable with to agree that type of information? In 2001 we had the first human genome sequence and today they announced last week in Boston that we can deliver a CLIA certified human genome sequence for $10,000.
We have been doing fancy genetic resources and human diseases for the last ten or 20 years and that information is aggregated and largely reached a sort of a wall in the research space and it is a matter of applying disinformation to this genome at minimal cost effectively accurately and safely.
What Bell you would be created for a person if we were able to do that? Was a sequence for $10,000. what value would be created for a person ? How would you use that? You can imagine, for example, using a genome sequence to supplant metabolic screening when a baby is born the. You can imagine using that here to pull this information to give information to people -- using that heritable risky permission to give to people so they can manage their help throughout their lifetime. And you can also using that information to this data by the screening paradigms'. We have heard a lot of debate among women and mammography. And would about a couple of those percentage of people who happen to be in a high-risk category. How do we start by their behavior better to improve outcomes for the subgroup? And finally when a person present with their disease how do we use the genome to give them the right dose of the right drug to increase outcomes.
So this is my attempt in this the city that genome has value in a medical setting and we need to understand how to use it. There is another obligation, what is the germ line genome, the one that you have been board with and the second is the canter genome and it is the disease of a single cell in that cell changes and forces it to replicate unchecked.
You will hear something wonderful from a group called the cancer care alliance doing that from the point of diagnosis and pulling that sell out in sequencing and an understanding how it will respond to drugs going for Woodson there is a series of applications among sequencing into ratifying care between these incredibly important.
I would like to call its conversion, the point at which the genome becomes accurate and cheap and the content about using it in a clinical setting pay we have seen the beginnings recently with the direct to consumers wait and certainly in the monogenic phase.
I believe it will happen in a short time. And you'll see it had a sweet spot in two years and we really need to figure out what to do with it quickly. I talked about cancer sequencing and the way could -- if you get diagnosed with cancer and you see your oncologist and they figure out whether it is: cancer or breast cancer and they basically pick the first drug on the list and if you fail they take the second. This is a paradigm starting to change and largely this is the way it works.
What we can do, I think you'll see activities start here again. Be able to stratify -- not change the standards of care the stratify the standards of care and see this individual because they carry a mutation or a hypometabolizer into this and a better way for maximum outcomes.
So those are some teasers, so we sit applications and on the clinical side you'll see that bucket into diagnostics and therapeutics and how quickly is it moving?
Until recently -- until just the middle of the last decade sequencing was down on the gene by gene basis. It was hundreds of thousands of dollars for every single gene and we recently had a technology evolution that has allowed that increase exponentially produce a sort of the point at which second-generation sequencing technologies hit and I describe what those look like it or on the purge of having third -- on the verge of having third generation technologies hit and then fourth generations which will include accuracy are probably two years away.
But the technology is evolving faster than Moore's Law. What are those technologies? The first generation technologies, you have a strand of DNA. And these are highly accurate and they are the gold standard for diagnostic platforms but are largely an affordable if you think about the cold genome consequence -- the full sequence.
The second technologies are the ones that are currently in use in the research settings and are stable and commercially available. They break down into the Biosystems platform, complete genomics platform, the illumina platform, and the plonator. These can do those reactions in parallel and I will show you some of those. They are reasonably accurate for research purposes but they are not accurate for clinical purposes.
Third generation technology, these are single molecule and some Kolbe's order of 2.5 generation technology and these are technologies where you are looking at a single molecule of DNA and measuring what is added on to it using fluorescence.
Fourth generation technologies largely to write useful lessons -- Do not use fluorescence and use the dip in Corporation and these can go much faster and have the potential for much longer lead length and there will talk about that in the second but third and fourth generation technologies have the promise. This is not commercially available yet. They have a promise to be very affordable for a genome but the are largely unproven and we do not know what the accuracy is.
A couple of definitions. I am sure it -- we will run through them quickly, DNA extension. You may hear things about engineering DNA polymerase, this is a naturally occurring enzyme. Nucleotides of the building blocks of DNA and they can before -- be tagged and green is a and red means that you incorporated a C and these are visualized using laser imaging strategies.
You can either image a single base been incorporated or take a picture of a set of different bases being incorporated.
Bold coverage is important term, this is a number of times you have sequenced a region of the genome in a specific -- the more times that the sequence a region of the genome the better idea that you have of the accuracy and the accuracy improves and a more full coverage that you have the more gaps you cross in the genome. Read length is important because it sure read lengths are difficult to assemble so it may stick in three or five different places and also these short read lengths, people like Walter read lengths in general. Paired end reads is the way to give ground shorts the read length -- to get around short read lengths. A library is basically taking the human genome and shattering it and putting it into eighth genome reaction so it means a lot of small fragments of the human genome.
Partitioning or selection is an important concept rather than shattering the whole genome and putting it in a machine, pulled out certain regions and sequence the news, you can sequence either one gene or a set of genes. This allows you to a sequence what you want to sequence at reasonable costs. The cost per base, the cost per genome. Accuracy is a critical metric that will come up over and over again throughout this talk. Throughput, how much can you push through in a day or in a run and how many gigabases an unspecified period of time.
Assembly, how would you take the output of sequencing a library and assemble it into something that actually looks like something a strand of DNA or a chromosome. Storage. There is the time of the debt and you need lots of storage and confrontational if the structure -- there is a town Monday and you need lots of storage . And interpretation I will let them talk about that in the question and answer period.
First-generation technology, I will be brief. Basically this is papillary of electro precess -- capillary and basically everybody uses this isn't if you talk about DNA based diagnostics in a laboratory setting, bracket one, bracket two, this is the machine. Has incredibly long reeds and is the gold standard. Is very expensive but it has very high accuracy.
First-generation technology. Second generation technologies.
Basically these are the ones being used in the research space. These are platforms like the life Technologies platforms, Illumina, George Church's freeware, and these platforms can sequence again libraries of fragments that can generate millions to billions of sequence in days which I guess if you think about it could be fast or slow but the in general have the short read lengths that make assembly difficult and are played by moderate rates of one and 1,000 which is if you think about sequencing 3 billion genome leaves you with a 3 million errors that they have to wade through and interpret and are essentially false positives.
So the notion here is that you can get a genome but it is not perfect. And data analysis is difficult. There are basically two different types of assays pick there are assays on Long Emulsion PCR and Bridge PCR. Basically this is an Illumunia assay and you take your library and you do the reaction in the bubble and land them on the microscope slide and you have hundreds of these bubbles in real time and it is basically you take the library and anchoret on the slide and this variations on a theme.
There are strategies here for partitioning and selection NDP do not want to sequence the whole genome you can capture the region of interest and there are none too ways that you can do that. One is this technique were you take a short span that experience of DNA single stranded DNA and shattered the genome and mix them together and the stock goes down with mint and captured everyone's out of the human genome and another is a technology called rain dance and you use those little fat bubbles and output that right onto the sequensor.
There really are two front runners right here in the commercial marketplace, D Biosystems and what is the Illumina platform and we will see shortly then they can the sequence 300 gigabases of DNA and can sequence of roughly three genomes which is the coverage that you need on either one of these platforms. With the average read length of 100 days.
You look at accuracy which is the key metric that we will all need to keep in mind and you will see the number of colored takes in here represent errors and these have accuracy rates of about one in 10,000 relative to other platforms which might have accuracy rates of one and 100 which leads to tons of false positives and can lead to do tell someone they're going to die every horrible disease and they won't.
Finally third and fourth generation technologies are single molecule technologies we put a strand of DNA into a tiny channel to measure and corporations in real time using a detector that fits in the platform and a tethered a single strand on and watch those and incorporate and a fourth generation technologies which are probably between one and two years out in commercial use really promised to get a genome at a very low-cost and these are companies that you'll hear about, Oxford Nanopore and Ion Torrent which is the world's smallest pH meter and Genia which is another single molecule and Siegel channel and a company which is eight Solid State Technology and a promises very small read lengths and potentially low error rates. Lots of data, pedabytes of data and it can cost between $100,001,000,000 to store when you think about processing all of this data and how do you interpret all of this data?
We have put a solution but I am not going to talk about that today where you can get a sample, from a genome or part of the genome and deliver that to a person's doctor and interpret that in a scalable AP there are other folks working on this problem. But this is the real challenge, once you get to the price point and accuracy, how do we make sense a bit so we do not hurt people yet reduce suffering so I will end there.
[ applause ]
Are there any brief clarifying questions for Dietrich?
I guess you were clear [ laughter ]
Thank you.
The next speaker is Cliff Reid who is executive officer of complete genomics and he got a BS in physics with there is not much BS from MIT and a Ph.D. from the second great Stanford university systems engineering place and he had 25 years of working in growing companies and he will discuss with us the accuracy of whole genome sequencing and how that data by initially be best managed.
Thank you and good afternoon. Thank you very much for giving me an opportunity to present here with you today. What I am going to do very briefly is talk to you about two topics that will really drill us down a bit into the topics that Dietrich has already introduced be the first topic will be about the quality of whole genome sequencing and the second will be about managing the date tepe I will discuss relieve the same topics in each of these two categories but I will make some simple obligations and a few calculations for you and then discuss the implications for both discovery research and for critical applications.
First of all, let's dive into quality and the main thing I want to do here is contrast the quality of a whole genome with the quality of retargeted test. And these are fundamentally and profoundly different things and the profound difference comes from the fact that they are measured by it genomes are calculated -- but genomes are calculated and let me walk you through the process. All of the technologies applied the same three steps to sequencing a complete human genome.
The first step is to break up DNA into fragments but this is actually a requirement imposed upon us by mother nature because it did not know how to pick DNA out of the self and so it actually fragments itself. This is something that we will live with for the next decades to come. So once we have all of these fragments of DNA we make measurements of them using whatever topology happens to be available.
That we take all of these individual measurements that are a series of ACTGs and we have to jigsaw puzzle the whole thing back together again be on the measurement side, all measurement technologies have some air raids produced to have very high error rates -- some error rates and we do not have the physics to exactly measured the molecules. And today the kind of error rates to see army be one and a hundred or one and 8,000 all technologies have a error rate.
So the act of putting these technologies back together again has been informed by some very well understood methods of information. And it is how do you generate a reliable result of a on reliable measurement? The reason that our systems work that you can talk over the phone in the other person can hear you is because of this concept because when we talk on the phone what happens is our voices cut up into little packets and is sent over a network and is reassembled on the other side and invariably some of those packets disappeared yet when we talk on the phone we hear each other perfectly. We incorporate redundancy into the phone networks so that when your boys to send across the wire is actually sent with redundant information so that an unreliable delivery channel results in an reliable center. This is what our e-mail networks are built on and most systems are built on some form of this concept and we have built DNA sequencing on the basis of this same concept.
So today we use a lot of redundancy in your CD player is 8 to 14 soul the redundancy there is just about a factor of two and we use a lot more than a factor of two. So we use a factor of 20 to 42 wash out the errors.
Because we're using this calculation to 80 rhetoric construct we are able to take these and reliable measurements and produce reliable results predictable accuracy of the cap to the results really depends on three things, it depends on the quality of the underlying measurement, it depends on the number of the underlying measurements. The amount of redundancy. So that we can wash out errors in the process. This kind of a voting process, voting the bad guys off the islands. And it also depends on the quality of the software because unfortunately the voting strategies are not post form and there's not one right answer. There are a lot of difference voting strategies that can be applied.
Let me give you an idea of the complexity of our software. I just checked yesterday and we have about 1.75 million lines of code in the Assembly and 10 million to up the system. But this voting algorithm is about one with 79 million lines of code and how did you validate that? And within that code in the new simply suffer in the community you can decide which kind of errors you want to make. It false positive versus false negatives. So the software is very complex and will remain a complex for the foreseeable future.
So what does this mean cracks again, with some of the nice comments that Dietrich may appear the most accurate Human genome that was ever published was published at the beginning of this year and had an error rate of one per 1,000. What does that mean? We kind of flow that through a genome and there are 3 million base positions in the genome. So if you follow randomly what were assigned of one per 100,000 base is you would end up with 30,000 errors across that genome and it looks like a round have, maybe two-thirds does not do very much but in particular there is a 1% of the genome that is great important that codes the proteins. So let's talk about the part that coot's for proteins.
If we were to spread these 30,000 errors you would get about 30 dozen errors in the coding region pit. That is about 1% of the genes will have an error rate and for research purposes that is terrific the reason is in the research study take 100 samples and did you go to any single gene, if 1% has an error the probability that in a given gene and even five of the 100 copies for your 100 samples, that is one in 10 billion.
So again, these large numbers works out that in the Research world these kind of accuracies are excellent and enable us to draw correct research results from data that has a error in something like 1% of the genes. So that is where we are today, best genome published is completely unacceptable for clinical use and if this was your error we would see thank you but no thank you. So we need to independently validate and use these to go back and verify every change in the genome that we think is medically important and that prevented from having high critical utilities of the question is what would it take? What would it take to make full genome sequencing a clinical tool? That is a judgment call and this committee and other policy setters will have to come to some conclusions about that.
Was prudes one error per 10 billion basis, down by a factor of 100, what would happen then the? We would take this new error rate, this one in 10 billion base error rate and that corresponds to 300 errors and that is 3 errors in the entire coding and that is about 99.99% accurate. That is starting to be a clinical results. I would say that is probably a diagnostic results for many applications but not all.
Here I am much more pessimistic and I had built the systems. I think we're going to spend between two and four years before we see this, did the industrial community and I think that will happen in a laboratory environment with industrial said drink equipment and supercomputers in people in white coats walking around really enforcing regulations and before that happens, a point of care, I think we could double it and I think we are six to ten years away.
So there is a time delay built in here and one of the key challenges is to have over a million lines of software and how do you validate it? How are you sure that those things are right? We are on our way but I think we are in the reference lab model which is a vault that will clearly hit the market first before these are critical quality results.
Are you going to change gears on you and start talking about the size. There is a lot of conversation about the size of the data. I think there is something here, in fact I think this was not a problem two years ago because they could not produce enough data to cause this to be a data management problem and this will not be a problem in four years. If we just ignore it it will go away and we all love those problems and let me explain to you why.
Here is the flow of data and all of them fell basically the same way. What I have characterized for you is the size in gigabytes in to put a another metric on it the dollars to store that amount of data using the cloud and this is the Amazon web services system.
So when you sequence one complete human genome, these are my -- are microscopes and the images are about 3,000 gigabytes. So this is three terabytes and store that on Amazon for about a year causes about $10,000 paid a very important observation be will come back to but that is the raw data and the raw data in our system is so big wheat -- we just process this on the fly otherwise it would be economically prohibitive the.
The next phase of storage is the reads at that map to those individual segments of the genome and these are about 10,000 smaller. And then comes the genome cell, in another factor of ten small but here's the real way. You do not care of the 6 billion bases because there is a preference genome and there on 99 put 9% of the same and all you care about is the variants Indy want to know what this difference -- and all you want to know is the defense and that is about 300 megabases and it's very nicely on your thumb Drive.
So where is the Data Management problem here? The reason is this, people do not use the variants right now, they want the reads Kawai? Because there are errors in the variants and the reads explained that errors. So we have enabled a researcher to sit this one looks suspicious, how did you get it? He does not want the raw images because three years later he will have an answer to his question but he can go back to the reads and stack them up and look at the voting sole reads are being kept along and that is about to go away because as soon as they go to a 10 to the minus seven, the reads are useless and all of the important information is captured in the variants. We just do not trust them yet but we will.
In fact, our customers are moving to a speech by -- to a variant formula because they can make good research discoveries without ever having to mess around with his huge collection of data that would cost $1,000 to store for a couple years. My final comment is that producing as the processing comes down it is going to cost more to store the data then it will to resequence it and it turns out that DNA is incredibly cheap storage mechanism so we are on the way to having this data management problem going away.
Quickly I am going to wrap up in another version of the amendment is the interpretation did that -- another version of Data Management is the interpretation data and what is going on inside of the genome. I pulled a number out, this is the simplest of the variant. There are about 25 million entries and is not carried and we do not know which ones are right and a whole new category is coming on line and this is the stuff that needs to be carried properly.
Some believe the scientific literature is largely uncan read and we know a lot of it is wrong and was done with early technology. So this ends up being the more interesting of the data management problems as we project our thinking five years into the future.
Many other issues that I will not touch on today and some we will talk about some more in other talks and maybe in the Q&A session but validation. The difference whole genome sequencing methods and with normal genomes we can go back to using the expensive subsidy methods and spot checks and that works pretty well however with cancer genomes is much harder because they have their own set of challenges quality control is a big issue and the whole 23andMe thing and that is a simple procedure. The knowledge bases are the key issues in curating them and then electronic health records are getting the data in a standard format.
Huge issues around data quality and the command to bid but the actual body of genomes themselves I do not think we are agreed to have much of the data management problem. So my conclusion, first of all on the research side the revolution is under way. We are taking thousands of complete new.
S now and the discoveries that are going to take place I think are going to be extraordinary and the data management problem is on its way to resolving itself and in the interim, thank goodness that Amazon exist so we do not have to go by those disks.
And then on the clinical side it breaks into two areas. On the markers side we're just going to see the golden age of protest. We have moved on of the cockpit area of common variant and that is where we will see the major discoveries made and once the discovery is made abundantly is easy to the point because those are markers that easy to measure, easy to look up and we will see kind of a golden age of genetic markers and in the whole genome sequencing arena there are two that we see clearly prefers the fall is cancer genomes pick it is mind-boggling what is going on inside the cancer genome. That will be in a renown Fort whole genome sequencing for a long time to come and markers are not going to tell us what is happening inside of cancer.
The other one that is full of controversy, full of issues that bodies like us are going to have to help the community to make is the universal genetic panel and we are in the Haiti right now out of complete the genome sequencing bid costs are down and quality is up and the research committee is going crazy.
So I will stop there and take time for questions.
Questions?
Thanks, Cliff. A very good presentation. I had a couple of good questions, but how do you actually determine [ Indiscernible ], what do you compare?
The way that community determines the accuracy of a complete human genome is bye spot checking its and if you go back to the first generation you can get a good highly accurate results so you go in and you spot check the genome. You look at the most suspicious places, the places are this exactly like the reference but the place where technology thinks it found something novel in the genome preview go back to the rich and a sample and snip that DNA and send it up for small but very expensive sequencing and see how many mistakes the new technology makes and that is how we calculate those numbers.
And you mentioned the different platforms get different kinds of errors with the errors are not random?
They are not random and they are not well characterized. We know that the Illumina platform has great difficulty with insertions and deletions and we know that the life platforms is on two day snips. But that is the bill that we know about the systematic platforms but there are systematic errors and I think everybody has to step up and it meant that there are systematic errors and we step up and mentioned that there are systematic errors but we do not know where they are yet, we agreed that out but there are errors in all these platforms and will continue to be could you get into quantum mechanics factors when you start measuring these things. A third of a nanometer across and we are decades away of having the physics to hold that the suckers down and measure them.
We need to live in this world of exploiting to generate reliable results. We can do that to help quality. I am confident of that.
Will bold coverage actually resolve those errors?
If they are systematic full coverage does not [ Indiscernible ] the errors and then you need to characterize those errors and then say whenever you have the disease determining variant that has that characteristic, when it is whole genome sequencing, you see in the ballot, grab that piece of DNA and send it over to the box and sequence it so we can construct a valid system.
Thank you.
Thank you.
You're welcome.
[ applause ]
The next speaker is Martine Reese who is the co-founder of an advanced software company and prior to founding Omicia he founded Niomorphic and provided the foundation of a successful commercialization and Martin is a championship ping-pong player, a warrior is a band and bears -- warriors fan and bears fans and is foolishing supporting Germany in the World Cup.
Thank you for inviting me to speak today. I will give you a little bit of an overview of what we can do. Oh Steve they asked me to talk about preparing and managing whole genome Dana and this is wonderful for me because it is about how to generate data, the data quality and I am trying to run you through some issues stomachs of the biggest challenge really is and I when Hunter% agree with click on this one is how to distill the genome data. In the future we will have the sense of personal genomes and we have 4 million variants for each individual genome and that is the number roughly we have to try to interpret and then it bast amount of literature today.
Said the big question is, how do we take personal genome sequencing all the way to diagnostics? So to put it in context what I will talk about, we've heard about the machine and how to make it sort of a beautiful introduction there and done that we call bioinformatics which is about assemblyman and this idea of getting the quality of. And but I will talk about is the analysis and interpretation of that data.
What do I mean? There is information about variants in the literature and I can use that and then what I need to do is take this and generate reports. This is what I call technical interpretations complex Genetics and genomics data.
And then ultimately, of course, is the conical interpretation. But that comes basically on top of a piece sort of have a genome report. So what I'm going to do is focus on three main things which is the genome quality and we have heard a lot about this and second the integrated approach that is really needed. And then lastly, the critical interpretation.
The way I'm going to do this is walking through an experiment that we did where we just use the publicly available first ten.
S that were available and I want to thank everybody for giving out that information's of here are the ten.
S and we see different ethnicities and pitifully we have six different platforms that deal with -- and beautifully we have six different platforms from 2006, 2008, 2009. We need to be very careful because technology is moving so fast that the error rates and everything is much better today. So the first thing that we do is compare a medium sequence with the referenced genome and then what we generate are in the variant types and all we need to do is compare variants with each other and if we do this we see in nicely structured tree and if you look at the ethnicity you see that there are Caucasian, Korean, Chinese, and African-American.
That is encouraging because of all the different technologies that we can see and we still get that very nice tree. You see at a very high level that is very good and there's actually one genome and the same individual that a sequenced twice on two different platforms so what we can do now is action look in detail, what are the differences between these existing technologies? Would you can see here is if you look carefully there are actually 500,000 different variants and use of the first to published genomes in the published literature.
Another thing that actually illustrates the complexity that we are facing increasing this to look at the total number of variants. 3 million, 3 million, and 4 million these are actually a bit easy variants that we have to interpret prigs of the complexity of freely interpreting 4 million -- of interpreting a 4 million. So we have these again for the systematic errors that we are not even touching today to try to interpret.
What that translates into and I am just using this sequence as if it is correct we have 21,000 coding variants.
So you can look at Wall Notes databases to try to see what can we learn about that does to give you some scaled this is the online database for a lot of the genetic [ Indiscernible ] and what we have is roughly 2000 genes from last year. And with in there we have 16,000 all -- allelic variants and 15,000 is orders in that database is. And we have a lot of software to now link the genomes to the genome. And as we ask the questions, what do we actually see? Here is what you get to produce get roughly 100 of these variants that have some clinical function present for a much in each of these ten published genomes. What we see is heterozygous and 30 in homozygous. And we said oh, they have a lot of the diseases and it should be very severe but a lot of these are called the genomes and help the individuals so we were a little bit surprised about Pat.
These are the summary and this shows the complexity ever interpret the in the genomes. The good news is in that all of these alleles we did not find any [ Indiscernible ] disorder variants so that is the results that we got.
And now if you think about this we have 21,000 coding variants and 10,000 other changing the protein sequence. These are the most important to look at and 100 of them overlap only so the question remains what to do with the next 99%?
There is a lot of variants that have no obvious classification. The next thing that we can do is actually go into the literature to build a knowledge base and basically pulled out all of the genes that have been linked to diseases and then we rolled it all out into the Harrison [ Indiscernible ] internal medicine textbook for diseases in general and we rolled up all of the genes from the literature into that of -- and we try to be as accurate as we can.
So this is the distribution of the genes in the literature rolled up into the categories. So if you do that, what you can then do is look at the Pops for each of the ten individual genomes and what are the carrying? You see with in the ten genomes there is one of variation and that might be due to we are comparing African genomes against the reference genome and the reverence Gino itself was mostly assembled using caucasian samples and we called this the genome load or variance.
If you look specifically what you can see here again here are the categories of diseases in this is the total number of genes per category and this is the proteins sequence changing with varying load that we can find the specific genome did so this demonstrates pretty clearly how complicated it can be to try to interpret this and again this is only in the diseased gene.
[ Captioners Transitioning ] Please hang up the phone so a new captioner can dial in. Thanks.
[ Relief captioner standing by, trying to obtain audio, need customer to hang up the phone on their end so I may dial in ]
Here's another example which is well known. I like this one, this is for one of the genomes just made up. What you see is it's not like you have the gene or you don't. Among these five markers you have to interpret. You have to interpret that. For us that means you have to build rules, so when we do this we implement the rule, it can go into the clinical setting. Let me try to give you some take-home messages from our work. Quality, assessment and control. From a clinical point of view, the first thing is [ Indiscernible ] testing. If I give you a genome I want to make sure this is really your genome. We need another technology, an array, to make sure that the genome that I'm getting is the right one. The other one is Concordance. A lot of times the sequencing technologies this the papers report the error rate compared to Concordance to DNA chips. The DNA chips were built for single [ Indiscernible ] variance, which were easy. They were picked because they were easy to assay. We looked at some of the error rates, it turns out they are really easy positions to query in the genome. A lot of these error rates are high estimates of what going on the rest of the genome. The genome is very different. Sometimes there's repetitive areas [ Speaker/Audio Faint or Unclear ] error rates in these regions go up. There's differences in the quality. We're just starting to look into that. Very important, the verification experiment. And then data security, of course. The second point is the integrated system. What we did during our analysis, it's critical to do the variance ranking. Today there are not really good tools out there to do anything. We need to really work on systems to do this, to do the ranking.
Clearly when you have the genome that can be reinterpreted later on. I agree with the vision of [ Indiscernible ] genetic reports. And then the information goes to the treating physician. For us it's just a different way of supporting the individual chain. And then interpretation. If I see the person I can look some interesting genes that I want to look at that for that person, and then integration of EHRs. Thank you. [ Applause ]
Questions for Martin? Thank you, Martin.
All right. Moving right along.I first met Greg when he was chief of [ Indiscernible ] at [ Indiscernible ]. He gave a very understandable talk on genetics 101. Knowing that, I invited Greg here to talk to us. He will tell us about what the doctor and various stakeholders do with it. Greg?
Thank you. And thank you all for having me back in front of you. As I have said before, it's often difficult to decide which hat I am wearing. For the first part I'm wearing the guy from the place that no one lives hat. The second half I'm wearing the research hat. So I also think I drew the short straw for the talk today. I get to talk about clinical utility, which everyone has a difficult time with. I think that a lot of my slides in the packet -- it's hard with that many speakers to not have too much redundancy. The technologies are allowing us to sequence DNA at an every increasing pace. This relates it in terms of cost and time.
I just make the point that there is a push, I think, generally to try to get the latest and greatest technology out there in front in of the healthcare provider and the patient. We have to be cautious about the push of getting this out into the environment. This is just in the last couple of weeks, vastly exceeding anyone's estimates here. So there are huge questions associated with the availability of whole genome sequencing. It's vastly different from a start-up company, from a researcher, from an onologist, from an average patient.
I went back and read a report by a task force that was put together and made the recommendations for the predecessor of this group. And they set forth three criteria for the evaluation of genetic tests.This issue of clinical utility, I have been several of the meetings where we talk about clinical utility versus personal utility. Clinical utility in this report brought in elements of personal utility into it. Those are enumerated in the first part of the list. But clearly in this report they envisioned it encompassing aspects of personal utility.
So this definition that has arisen is falling into an environment in primary care that is challenging to say the least. I think that primary care in the U.S. right now is under a tremendous amount of stress. I think the patients of primary care providers, particularly in rural areas, are a bit disenfranchised.
What are solutions from the president of primary care providers? They don't rest in genomics. They rest in in the bread and butter things. Doing these things in the system that we have right now. Simple things like measuring blood pressures and glucoses. So this, unfortunately, for many primary care providers is the perspective on genomics. For your rank and file primary care provider you are there. There are folks in the primary care community who are actively questioning the value of expending money on genomics due to this lack of evident utility. We have to be cognisant of this in this group.
These folks have a perspective that is rooted in the word of evidence-based medicine. It's a rather imperfect field, it has its problems. And one of the things that these evidence-based folks have in common is -- supports the use of a proposed new technology in healthcare. You probably can't read this slide. This is just one of the criteria put forth called the sort criteria. A grading mechanism for the evidence that exists. At the the top are studies that show improvements in patient oriented outcomes. Second, studies that show improvements in surrogate markers. The end, studies that are associative and don't show improvements in outcomes. What do I mean? Outcomes that matter to patients to help them live longer and better lives. When considering the issue of clinical utility we have to be cautious about aligning our fairly nuanced view and what the rank and file healthcare provider who is facing an evermounting number of patients and lower reimbursing and a growing number of things to do and how they view utility. I put those together. These are gross generalizations.
Morbidity, mortality, surrogate markers, effectiveness, efficacy. I think the primary care clinician is rooted in the world of effectiveness than in the world of efficacy. There's peril in being too firmly rooted in either of these two. We have to be cognisant when considering the definition, this distinction between the EM world and the world of genomics. How are we doing in terms of matching evidence with expectation for evidence? Recently probably many are you familiar with this conference that the NIH held last year.
As a screening tool in primary care office settings for reducing morbidity and mortality for a wide variety of common conditions. The bottom line is there was not that much evidence out that would lead to a high bar. The intersection at this point and time is actually quite small. I think this is a huge barrier. Why? Because we focus most of our attention on getting the early stages of science to where you can begin to grapple with issues. Expenditure is in this TO to T2 phase. This is where the vast amount of our focus has been. I would argue this is where the actual key to adoption will lie for these technologies ultimately.
How do we get to patient oriented outcomes? That's a lot of that T0-T1 work to go. We should have an eye to the latest stages of research. So at this point I think we've heard this. We've had this change in requirements. We're achieving the DNA sequence production is not the bottle neck for figuring these things out and gaining clinical application. It will be this world of analysis and the next step is this integration into actual clinical analysis and use.
I don't need to belabor this, there are a whole lot of variants out there that might be relevant to human disease. There's a project that drives this point home. They're looking at a series of candidate genes. Look to see if they can use that information to help predict risk. And this is older data from the study. You can see this huge number of variants popping up. And making sense of what is wheat and what is chaff is a large challenge. Yes, we're in the bubble phase. I think we're going to face challenges in data storage capacity all the away long from the you cannily duckling -- ugly duckling phase on.
I think this is going to be an issue. I just love this. Just mind boggling right now what storage capacity is needed. Here's another slide. They have their own substation for handling the power requirements for their data storage facility. Which is just remarkable. This has been touched on already. Even if you get beyond 10 to the minus 7th you still have a number of errors. As we get to individualized e Joels it will be increasingly hard to figure out what is useful information. So we have this informational bottle neck. We have a regulartively -- relatively simplistic view at this time of how a complex disease comes about. Really there's probably tens of hundreds of variants in some cases, probably tens of behaviors, unknown number of environmental influences. We're just beginning to sort these out.
First we need to figure out what we're talking about with conditions and exposures.You can look at an activity that was put together called Geneva, they're bringing together folks and beginning to grappling with issues of phenotyping. It is not a trivial exercise. This issue can be from two perspectives. You look across the definitions we have and harmonize them moving forward. The other -- that turns out to be quite daunting. This is just a table for cigarette smoking. It's really quite complicated. The other approach is top/down.This, of course, limits your ability to use already collected cohorts. This is an example coming out of the project for nicotine, doing a top/down look. We know now that [ Indiscernible ] polly morphisms are not the entire game, or not even close to the entire game. We need to work on sorting out what variants we're talking about. Increasingly we get to the point where we have rare variants, small numbers of people, and aggregating the number of people we need together with the variants to figure out what we should do with those folks will be a big challenge.
Environment, coming out of the Geneva project. Sought to come up with better ways of measuring the environmental influences. I think we're now just beginning to be able to touch this. There was just a paper out a week or two ago looking at common environmental variations, common environmental differences and the potential effect on snip markers relevant to breast cancer. So models moving forward. You could use existing cohorts for these studies. I think there's need to consider large perspective studies moving forward, this is from the common fund website, talking about that. I think it's difficult to envision how to sort this information out without a concerted effort to develop the infrastructure to harness our [ Indiscernible ] systems to help in realtime doing research on whole whole -- whole genome sequence variation and what it means.Then I think we have a whole raft of other questions to tackle. Will society act on this information in productive ways? This is a press release from the last couple of weeks. I would suggest you take a look at this website. They set up a snip panel, deployed at the Henry Ford healthcare system and looked at how patients reacted to this type of information. It's eye openingive I think opening. -- I think this issue is important to look at.
At this point in time we also have the issue of: Can the rubber meet the road? This is a great paper, it looked into this in some detail. Our record systems was not prepared for Whole Genome Sequence data. We would like to get to the point in the future where we can integrate clinical decision support with the rest of the information in EHRs. Yes, we're at ten years. Yes, there are growing pains. I think there's tremendous progress potential moving forward. We need to work through the bumps in the road. We need to keep our eye on utility. I would like to thank the folks that have contributed slides. Thank you all, again. [ Applause ]
Any clarifying questions for Greg? All right. Seeing none, Emily is up next. She is one of a handful of counselors in the country that quickly embraced genomic counseling and put together a curriculum. So I asked Emily because remember I said practicing Jen practicing geneticcologists are rare.So here is Emily. She will tell us about how we educate and make our patients understand all of this.
Thank you. That's very generous. Thank you for inviting me today. It's an honor to be included with all of these experts today. Thank you to the speakers who spoke before.
I was asked to speak on future events. My talk will be somewhat speculative. I want to bring in practical examples to examine. This summarizes the key points. How is it applied to clinical practice? How it will change the practice of medicine. Will it change the practice of medicine? If so, how? I will address informed consent, communication of results, approach to variants, clinical and patient education.
This is somewhat of a futuristic event. So here is Sarah. She is 32. She is coming into the OBGYN. She has some questions about breast cancer screening and family planning. Her nephew has [ Indiscernible ]. She has questions about screening. The OBs a genome. In a few weeks they return for follow-up. The doctor is able to log in to the portal. Based on the family history the OB can query a breast cancer and preconception panel. She has-- this testing identifies she's a carrier for three other things that could affect a future pregnancy. There's accessible support to help the OB to take this information and incorporate it with her personal history and come up with some recommendations. The OB can counsel Sarah. This would mean that she would have a mammogram at 35. They also discuss with the patient some of the considerations about these variants and what this could mean in the future. The OBs the husband's genome and remembers them to genetic counseling, to discuss CF and other conditions that might impact a future pregnancy. This was a fun example for me to play with. How do we get from where we are here today to where we need to be?
We need affordable testing, of course. We need more information about clinical validity and utility for disease associations and snip associations, and information about the ethical, legal and social implications. We need data interpretation systems and clinical decision support systems. We need someone to pay for this. We need professional societies to recognize change in practice and provide education. We need informed and educated consumers, of course. When we look at this list, there are many more I did not identify here, and think about the things that we needs a healthcare systems and as a society to put together to effectively integrate this into practice it's clear that the affordable testing is coming first. We have some gaps we need to address.
When thinking about -- before we can talk about informed consent and risk assessment we have to ask why someone would order this, or what they would hope to learn? What is the motivation? Will this change patient management? Will this result in an action or behavior change? There's a few different models. One model is when a clinician orders a genome and inquiries part of a panel screening for a particular condition. Another model could be more of a preventive health model. If an adult presents for testing then the clinician could pull up a list of results that impact someone's risk for treatable or modifiable diseases. There's the direct to consumer model that can address both -- can overlap with diagnostic or targeted treatment and preventive health models.
Informed consent and consumer education. I think these go hand in hand. Over time as the complexity of information on consent forms increases so does the length of that form. That makes sense. Studies have shown the longer the form the less likely it is that someone will read it and understand it. This is not a unique problem to testing. It's something we have to grapple with if we want to incorporate Whole Genome Sequence into the clinical setting.
How someone presents and the consent conversation will have a little bit to do with their motivations. I think the basic principals of risks and benefits that we cover remain the same, but there's an additional layer to that. We need education for clinicians and consumers. I think we need -- the current research for Whole Genome Sequence consent is nearly an hour. These are long consent forms. This is not sustainable in our current model. Particularly if we're thinking that primary care providers will have a role in this. We need more education for our public.
Before I talk about risk assessment, risk communication and results communication, I will raise the question: Is Whole Genome Sequence different than what we already do in genetic medicine? Does Whole Genome Sequencing ask something different of us as clinicians and educators than what we already do and expect? We can put this question up against different domains. One would be knowledge. Our skill set to manage uncertainty for ourselves and the clinicians in the room, for our patients. This is something that doctors do every day. Is it different with Whole Genome Sequence? And our skill set having to do with communication. Is there something different about Whole Genome Sequence in the nature of the uncertainty than other uncertainties that we already deal with in healthcare? Who is involved in communication? I thing that genetics professionals have a key role here, particularly initially in the next few years as Whole Genome Sequencing becomes more available. I think they have the responsibility and a role to be involved in leading and participating in clinical research to help identify infrastructure and identifying best practices for risk communication for clients.
I do think that as the data and decision support and aids increases, whether it's from a lab or another company, as it gets better and better the requirement for the clinician to have a high level of expertise would decrease over time. Perhaps over time we'll see a distribution of the responsibility. How do we communicate results and risks? I don't have an easy answer for this. What information needing to be con vaned to someone? What is the minimal information? And what is the best way to present it? It comes down to communication. Is it something different than we're already doing? In what content do we have to present results? Should it be in the context of other medical information for this could be meaningful for the clinician and the client? And how do we address the variants of insignificant things? If you just spend a few minutes per variant that's a few hours, not to mention prep and follow-up time.
I told you I would try to put this in a practical context, here we go. In 2009 I was working at the [ Indiscernible ] and at that time the Cleveland clinic was offering it free to clinicians. Individuals could elect direct to consumer testing. They could also elect to contact a counselor afterwards. We scheduled appointments on the phone, we obtained family history, talked about the results, how they would get me their results, and did contracting to figure out what their priorities were. We found it was most effective to put the conversation about testing and what the results could be for someone in the context of all of the medical information that we gather. We gathered medical history, lifestyle, family history, risk assessment. And then talked about the onowe Mick results, emphasizing the potential of the results. We did document this and share this with a managing provider.
This took a lot of time. This took significantly more time than a traditional case. It's a challenge to work with the direct to consumer testing company. They're not set up to Lee ayes with clinicians. There's this concept of needing to confirm resulting, particularly if you are using a company that says their information should not be used for medical management purposes. It's really challenging to talk to people about the different types of genetic tests and results that you can get back. And conveying concepts of different degrees of accuracy and validity. That's really hard. Trying to talk to someone about a high pen trant [ Indiscernible ] mutation that has clear medical management implications.
It's a challenge that right now we have limited information about clinical validity and utility for a lot of these resources. Limited tools for clinicians. Talking about healthcare professional education, it's been a challenge. We all know it's a challenge to get your average nongenetic provider excited about incorporating this into their practice.This has been a challenge for us. We have made recommendations to break down these barriers. Thinking about how challenging that has been for the past half century, think about incorporating education for low pen trant variants, variants where one might have good information about an association with disease, and another variant has poor data. The associated risk might change over time. This is something that I think about something a lot as one who is providing genetics education.
How do we close the gap? I think that genetic professionals will very involved. We need to continue to emphasize preclinical and continuing education. I think that laboratories and third parties, other companies and professional societies will have a role in the creation of genetics education materials, but also making them available to their members and to the people who need to receive them. You all can read my conclusion. I think affordable testing is coming soon. It will be affordable before we understand its importance. It I think it will be challenging to educate people. But it's an exciting time to work at [ Indiscernible ]. I think we also have the potential right now to really make significant contributions to society's understanding of genetics. People are concerned about it now. And people are interested now. I think we need to capitalize on the interest.
I would like to acknowledge these folks who took time with me to talk about applications happening right now. Thank you. [ Applause ]
Any questions for Emily?
Emily, the doctors with the [ Indiscernible ] and me test. After the education event how many declined?
Great question. I wish we had data on it. This was a sensitive issue. It was an employer offering testing to its employees. It was confidential. We didn't track that information. Although we certainly wish we could have. Thank you.
All right. So I think we've come to the end of the first part of this very illuminating discussion. Why don't we take a 15-minute break. And be back at 20 after 3:00.
[ SACGHS on 15-minute break until approximately 3:20 Eastern Time Zone ]
And then we want to have some good discussion, so Paul and [ Indiscernible ], take it away.
Let's see. Is Rich back yet?
Poke out there and give him a yell. So Rich is up. From the public comments every other word smacked of ethics, legal, social. It's my pleasure to introduce Rich. He's the directer of ethics research at the Cleveland clinic. This is actually his research expertise. He's expert on various [ Indiscernible ] on Whole Genome Sequencing, take it away, Rich.
Thank you.Thank you for the invitation. It's a real privilege to be here to speak. What a lot of what I talk has been anticipated by a lot of the other speakers. I guess I wanted to begin with a type of thought experiment. Imagine that we had the ability to access the world's greatest clinical genicsologist and access to primary care specialists that were able to help us to navigate the systems in which we live and help us to gain appropriate access based on the findings of the results. As we thing about what we would want to consider what would be on that list? If money were no object what sorts of genetic tests would we be seeking today? Personally, I have a hard time answering that question with even a single test. I find that remarkable that we're pushing so hard towards a movement to implement Whole Genome Sequence data into primary care when we can't even think of simple examples of situations in which the information would be profoundly helpful right now. I want you to hold that thought in mind as we go through and I make my comments.
Disclosures here. I have no financial relationships to disclose. Quite regrettable for me and my family. My lab is funded by NIH money, there's a cliff that is coming, I am in even bigger trouble. We know that genomic costs are coming down. Technology is expanding. This target of a $3000 genome and $1000 genome is soon on the horizon. People are doing a lot of innovation in regards to this and thinking about how to manage this information and annotate it in ways that might be accessible and useful. The work that George Church is doing is innovative in regard to that. We need to see projects like that involve taking risks and the administration. We need to see these types of projects done and funded more aggressively if we're ever going to get to the point of clinical utility associated with Whole Genome Sequence analysis.
I wanted to start with a quote from Ken.We're already seeing that with the direct to consumer companies. I think we're soon approaching that time. My questions that I want to explore are these here. As we approach this time in which Whole Genome Sequence data is cheap enough to generate and produce are physicians ready? What do we do with the results? Who is charged with doing this? How will we do this efficiently? How will remanage some of the -- will we manage some of the uncertainties that exist here? How do we make decisions about when to revisit that data? Reasees it. Let me start with the first. The question about whether physicians are ready. At the Cleveland clinic we had initiative that was meant to expand our efforts. A part of which involved something that Emily mentioned, the direct availability of testing to physicians that were interested in participating in that type of experiment as part of an employee benefit. We did a fair bit of assessment to determine how much physicians knew about genetic testing and clinical ideas. The important thing is this was a diverse sample of physicians at the Cleveland clinic. What we heard was a resounding statement that they felt they needed more education about genetics. These slides are a little bit difficult to read. Most physicians, nearly all, said to stay current they felt they needed to learn more about genetics, roughly about 95%. Increasing their familiarity would benefit their patients. They also thought that new genetic findings were changing their practice, having an impact on their practice of medicine. We also found when we asked them about their baseline knowledge that they reported they didn't feel well equipped to deal with questions they were receiving day-to-day. Here the data is that I'm familiar with recent research that effects my patients, 1 in 4 felt familiar, the majority felt not familiar with the leading edge research. They were not comfortable explaining genetic test results and felt that formal training in genetics didn't meet their needs as physicians. I suspect we will not get any different results if we were to conduct this study at other medical centers.
Physicians that are in specialty areas may be familiar with a few select tests. But by and large their familiarity overall is fairly limited. When we think about this the physicians aren't well prepared to counsel patients about current available genetic tests. Surely we shouldn't think they're well prepared to counsel patients about Whole Genome Sequence data that will be larger in terms of scope and clinical responses that would be indicated. That I think is probably the straightforward answer. Are physicians ready? No. There's a deeper point here. It's one that I want to make sure that I stress here. That is that the response to this situation is not simply to encourage more physician education, to make sure that physicians know more about genetics. That's obvious. The appeal of personalized medicine involves a paradigm shift of sorts. Move us away from a mode of delivering care, which is focused on addressing symptoms as we present themselves. A patient comes into the office because she's sick is seeking care for those symptoms that brought her to the clinic that day. What we're talking about in part when talking about personalized medicine is a shift towards more proactive preventive ways of addressing health problems in America. I think the bigger question is: Are physicians prepared for that foundational shift? I want to suggest that the answer to that is also no. That we don't have a healthcare system that supports physicians spending time counseling patients about disease prevention, that supports physicians sitting down and having a long discussion of health and the things that patients can do to address those health-related needs. And that lacking that infrastructure we will always struggle with questions about who will provide this information, who will counsel patients? I want to suggest that physicians are not only limited in their current education about genetics, but not well prepared to respond to this foundational paradigm shift that we're seeing, that is signaled by the appeal of Whole Genome Sequence data.
The second big item that I wanted to spend a little bit of time talking about is this question of how to pick and choose which results to return to patients if we do Whole Genome Sequence analysis. There are going to be many, many things that will come out of a Whole Genome Sequence that will be unanticipated, in the sense that they're not related to that particular patient's clinical presentation. Maybe it's a result that is related to an adult onset question. Maybe it's a recessive mutation that is important in terms of reproductive decision making. There's lots of things that might fall out of these types. I don't think it's realistic to believe we could sit down and counsel patients about the different types of diseases and genetic information that might be revealed in the course of a Whole Genome Sequence. ThatThat approaches the need for informed concert. If we're aspiring to put patients in a position to provide that consent I think we're working towards something that is unattainable. I think we need to adjust our standards here. I think we need to be comforting with something that is not as informed as similar sorts of decisions might be if we're evaluating single diseases or single genes. There's going to be some compromises with regard to patient care and informed consent.
So we're doing a study looking at some of these practical challenges associated with Whole Genome Sequence. This is a study in which we're seeking to describe patients and genetic professionals' attitudes and beliefs. And trying to identify what Colin initials believe -- Colin initials believe that patients should know prior to making a decision, as well as what patients would want to know before making a decision. Our aim here is to develop some practical guidance about how to talk to patients about testing, and how to make these decisions about what to prioritize. This is a study that is funded by the national human genome institute. Our approach is an a series of indepth qualitative interviews. These are patients that are often seen pass being on a genetic odyssey. They may have been referred from one physician to another physician with the expectation that they have some genetic disorder, but no one is able to provide a definitive diagnosis of the disorder. We're interviewing these patients that find themselves in the genetic odyssey. What we're doing is we're saying to those professionals: Imagine that your institute was planning to offer Whole Genome Sequence analysis as part of a clinical service. Many they were rolling out that type of testing at that institute, who would you want in the room to serve as expert advisors? After this interviewing we will do some surveys much patients and do some practical work. I will speak to you about the expert advisory groups. Again, a part of the aim is to identify what things professionals would prioritize in returning information from Whole Genome Sequence analysis.These advisors that we have attend a series of meetings, this is not a quick in and out type of process here. We go there on several occasions and begin to develop a rapport with them and ask them how they would deal with this. We're giving to them mock reports that are coming out of multiplexed arrayed. How would you make decisions about what is important to cover and what you might defer? And because of that I think the information that we're getting is quite unusual and unique in addressing the challenges.
At this point we've done 18 of these. I'm glad my wife is not in the room to hear that number. The expert advisory groups, some of the major findings so far, this is still in the early stages of data analysis. What we're learning so far is this: That genetic professionals, I should be more precise there, those considered leaders at academic medical centers do not believe that highly multiplex [ Indiscernible ] are ready for routine clinical use. As a companon to some remarks earlier, it doesn't get much better for the professionals either. They don't know what to do with this information either.
They worry that it will be difficult to put patients in a position to provide informed consent. They're worried about false positives.We also found that they struggled to articulate a clinical situation in which it would be appropriate to order Whole Genome Sequencing. They struggled to list higher and lower priorities. Often they would say that once the data is in hand and has some clinical implications you have to review everything, if it's reel vanlts to patient care in some broad way it's something you need to discuss with patients. They had a difficult time making these decisions. Part of this has to do with the culture of genetics. The approach of many is very conservative. You order a test based on a patient's clinical presentation. What bothered them the most about Whole Genome Sequence data is the scope. Why look for everything? Why not look for things that are indicated clinically?
So the point here is at the bottom. These are the medical professionals with the greatest experience. They are clearly urging caution with data. The last issue that I want to cover, I will skip in the interest of time. I want to say in regard to genetics we're all aware of the misconceptions that exist. We're all worried about the possibility that patients might overread genetic information, see it as more predictive than maybe it is. I continue to think there's a great need for research examining the social implications. We could also articulate a list of questions we consider to be important in this area. At the top of mine is this concern that we don't have good outcomes data related to the impact of genetic risk information on behavior. We don't know if personal awareness of information is predictive of improved health outcomes. We don't know if it's predictive of behavioral change. We certainly don't have any good quality data that is long-term perspective data looking at the impact of predictive genetic tests. We're working with a group at Baylor to generate some of this information. We have all of these really remarkable tools coming from clinical genetics. We don't have the infrastructure to add mother these tools -- administer these tools.
Lots of folks that helped with our project. I want to make sure to acknowledge them. Sorry that I ran a little long on time. Thank you. [ Applause ]
Jim?
Fantastic. Great. I just wanted to -- ask you how you, um, might think about -- I just urge you to maybe think about, or maybe you could mention how it compares with what we do in every other aspect of medicine. When we order an MRI of the brain we find out things that are unanticipated. It's really no different from any other medical specialty. Do you think that's a reasonable analogy?
I don't, actually.
You know, I really didn't like your talk, actually. [ Laughter ] So why not? Why is it no different?
Okay, okay. So I think there's several dissimilarities. If you read the MRI today and six months from now it will have more or less the same interpretation. If an expert were to look at the same genetics data a year from now they will get a different result. The information has clear relevance for other individuals, their family.
We have encountered that before with infectious disease tests.
True. True. One of the reasons why we have a more [ Indiscernible ] informed consent is we want to slow down a bit and be clear that this is not quite the same as other tests that maybe they've consented to. It's not driven by a clinical presentation in all likelihood. One of the things that we want to cover is that we're ordering this test not because you have a particular symptom, because we want to anticipate --
But that is half of the testing we do in medicine already. We don't order cholesterols because we thing that somebody has a disease --
Yeah, yeah. There is a case on each side here. I do come down on the side of saying this is a place where if genetic exceptionalism makes sense, it makes sense here.
Thank you. Our last speaker is the chief scientific officer of the Los Angeles healthcare, he's known to us as our fearless leader, Steve, and honey.
Honey? !
Thanks. I will build on what the last few speakers have talked about. From an economist's point of view this is very simple. If you don't have a good measure of effectiveness it's hard to talk about the cost-effectiveness. In general we talked about in clinical medicine about economic evaluation. There's a lot of ways to do that. The most conventional way is a cost utility analysis, which is cost per quality. Which combines morbidity and mortality in a single measure, it's a health outcome. I thinks important to note that while we're talk being what might be cost-effective, cost-effective means it's a service that provides reasonable value. It does not mean that it's cost saving. Most of what we talk about is what kinds of tests provide that reasonable value.
Most of the tests when we say that they're cost-effective -- it really mean that is they're costed ative. This is a simple formula that I want you to work with me on. Remember that what we will look at are costs increases? Or decreasing? Do the health outcomes improve? Or not improve?
It's important to think broadly. We've heard a lot about the cost of the actual test. But that's just one small part what the costs are that-related to Whole Genome Sequences. We will talk about direct costs, which are the medical costs associated. I will talk more about them. There are also nonmedical direct costs, these are patient costs for transportation, other things. Indirect costs, which are productivity costs from loss of work or other activities. And intangible, grief and pain and suffering.
As we go through I will talk about direct medical costs primarily. It includes the cost of the tests, visits, hospitalizations, follow-up tests and other treatment. We need to look at the aggregate of all those. I think we've heard today about two very different basic uses of Whole Genome Sequencing. One is a focused clinical use. With we have information -- where we have information that is gathered for the purpose of managing a specific clinical condition. Somebody is presenting with some situation or condition or disease for which one wants additional information. I will call that focused clinical use. Much of what we are talking about is screening. This is testing, reporting, and using the information in a person who is otherwise asymptomatic.
The economics here are largely the same as for any other laboratory test. What is the information that you get from doing the test contribute to better decisions? Auto few variants make Whole Genome Sequence somewhat different. An economist would call this a sunk cost, once you do it it's done, you have the results, you can use it over and over. It's different in that way than many other tests. That means there are additional focused uses that may increase the health benefits, little inmental testing costs. There's issues with data storage and handling and clinical data support for the actual use of the tests, as we heard.
When these tests are used appropriately the issue is one of financial consider, not really economic ones. There's a variety of issues here. Who pays for the initial testing and access costs? How are the costs shared among payers? How are supports maintained? It's important to think about these things, we leave this to the finance people.
I want to talk about this problem: For almost all diseases that which is clinical apparent and meaningful is just the tip of the iceberg. For screening we want to look under the surface. For screening I mean just those that are asymptomatic, that do not present with a problem. -- without a problem. The problem with screening, I will talk from the perspective of someone who has been on the task force, the problem is to figure out whether early detection does or does not lead to change in behavior. There's the possibility for error, there's the hazards associated with follow-up. There's the question of how much benefit there is, and how much cost is incurred. The challenge is to identify those tests that provide clinical benefit and have minimal harm, so you have a net health benefit from doing the test. Remember these are asymptomatic populations.
Let's look at the six possible outcomes of screening. Screening test is negative, but the patient has the risk for the disease. That is a false negative. Someone might be inappropriately reassured. I will use examples from nongenomics, I think they're clearer. A[ Speaker/Audio Faint or Unclear ]. Second, the screening test is negative, the patient does not have the disease. That's a true negative. No health benefit. The patient is reassured. Think about knowing that you may be at lower risk for diabetes might lead to suboptimal behaviors. Your risk of diabetes is only half of what other people have, it doesn't mean that you don't have a significant risk at the same time. Outcome number 3, the screening test is positive, but the patient doesn't have the disease. It's subject to the risks and costs for the testing and anxiety. Screening outcome number 4-RBG the test is positive and the patient has the disease. They may end up getting -- they may be treated unnecessarily. Think of prostate cancer screening. 25% of men have prostate cancer, but the lifetime risk of death is only 3%. How many are treated for disease that would have never made it to the surface?
Number 5, testing positive and the patient is [ Indiscernible ] related to the disease. The outcomes of treatment are no different from treatment after symptoms materialize. We lengthen the treatment time. Think about the screening people for COPD, no net benefit, but increased costs.
Outcome 6. [ Speaker/Audio Faint or Unclear ] that would develop if the treatment was not started until after the symptoms were present. Think of examples of screening for colon cancers and treating it in an asymptomatic stage. It has health benefits, it may save costs. For five of the six there's no health benefits to the patient. These are not just costly. The patients incur no harms from screening and treatment.
So how does this get back to Whole Genome Sequencing? Let's think about how that exists. The challenge is the overwhelming number of tests being done simultaneously. You have to consider the benefits and harms in the context of having many, many, many, different observations. You will find many positive findings and many of those concerns that come from those will lead to follow-up. For most of those the harms are likely to exceed the benefits. The costs are likely to be substantial. In you look at what we -- if you look at what we recommend it's a fairly narrow set of things that passed the scrutiny of harm being less than the benefits on a population basis.
[ Captioner Transition ] [ Please hang up the phone line so that the relief captioner may retrieve an audio connection to the event, thank you ]
The first is for providers to do things are enormously greater than the forces for and them not to do it promotes of us are trained in clinical medicine with a noble ambition to do good and the failure to recognize and a good ability to ignore any harm. We have a cultural expectation that medical care can only do good, not harm and that more care is always better than less.
Both the public and medical professionals generally have high faith in technology. There are efficacy organizations that have potential sway over things like screening. Is the fear of litigation if one does not screen and there is the really the problem of the very near to detect problems -- the failure to detect problems.
Even if we look at quality measures most of them are about things that we should do and the quality measures include 13 specific measures that include the word screening and every one of them requires a screening and none of them that talks about unnecessary screenings orifices and there are other forces, if things to pay for, every dollar spent on health care is a dollar of income for someone and in the days of health care reform passed and present many people thought it was immoral to pay physicians to withhold care. It still is a major problem in this community.
Sold if the whole genome sequencing translates to unbridled use of screening, then the process of promoting prevention will be doing much harm and health care costs are likely to increase. So what do we do? We need to make sure that what we end up delivering our services that demonstrate health benefits and have clear clinical utility speed we need to develop financing coverage and reimbursement systems to recover those costs and a system to assure their appropriate use.
A lot of this comes from my good colleague Michael from the University of misery and I appreciate your attention and I will take a question or two if you have them.
[ applause ]
Direct questions? Go ahead.
Do you think that it is possible at this point to say, wait a minute, you know, implementation of whole genome sequencing into medical care is not proven, except are Due you think we should put the brakes on it and do you think we should?
It depends on what you mean by brakes. There are some areas where there are some clear indications for testing and there probably are some good examples year but we need to work, we need to have good examples where there is clinical utility and we need to make sure that we have a structure in place and these kinds of to the many issues that we talked about earlier and the equity issues. I do think that we need to slow down and look very hard because once these genies are out they are very hard to get back in the bottle. It is very hard to do with the ones that are already out and in practice. I think it is a cautionary tale and that is why groups that are looking at clinical utility doing the evidence based genomics the work I think they are with a critical and questions that we need to answer, whether the clinical standards, who should be doing it are critical questions so that we are not led by a technological and purity of that we can do it but to make sure that we actually get real value.
From the economics perspective there is real risk that these things are actually going to drive the extraordinary costs to the health-care system and it remains to be seen whether there really are the kind of cost savings for people hoping to do this.
Can I invite all the other speakers including Steve to visit the front of the room? And why don't you as they are walking up pose another question or make another statement.
This has been a really stimulating day and I guess I have a number of comments. But I think one underlying question is, I think we all agree -- I mean, I have heard from all of you about a sort of the present value of whole genome sequencing not being there but there is a project before gross in this area were the kinds of data that we get especially on clinical Liddy and utility and declining costs with thresholds. I want to know where that threshold is and kind of fast forward ten years from now, 20 years from now where we have the genome as Greg mentioned and keeping that presentation in mind tenures and how will we know we not? Or when will we know enough to cross that threshold for the masses, not for the management of individuals, people coming in with diagnostic conditions but for the whole scale testing of people for the genome so that we can use it sued the critical of -- clinical and validity and I have not seen that answer by any did you produce a we need more physician education, more evidence but when do we crossed that threshold for action between the movement between research and practice. A specific question to Greg because you put those two circles as opposed to genomics medicine. Is there isn't reason why did gnomic medicine should not be subjected? Is there some kind of underlying reason why genetics' would be different or exceptional from the rest? That is what the panel can add.
Now that that is out of the way, the presentation was really excellent from the screening perspective and on the issue of cost effectiveness we need to consider more than just the screening uses for the genomic information. It would be proven to consider a model that may go against what the gentleman was talking about this morning where you obtained sequence data, do not access the fully monty, if you will, and accessing information as need derives from the course of clinical care prepare for you extract a threshold said of variances that are specifically with a high degree of risks to the individual right at the outset. They are sort of no-brainers and then the things that are in the weeds you leave in the weeds and do not actually interpreted that person then in the appropriate way begins to extract information from their genome.
Or perhaps by clinical encountered the become diagnosed with a chemical complex condition and there is information about pharmcogenetics variance and some of the things that Steve was talking about in the other advantage of that is let's say that we're talking about a 21 year-old who decides they want to know about whole genome sequencing and they may not elect to know about their Alzheimer's risk until they are 35 or 40 because that would be vastly different than what it is when they're 21. It depends on what we find out about prevention for Alzheimer's but you would not want that information at 21 necessarily in the current clinical environment but you might want a 15 years from now -- want it 15 years from now.
Others? Did you have something else?
Are we in questions?
Questions, comments, statements.
This has been with like, one of the best sessions that we have had in my infinite number of years on this committee. I actually think that there is a note of optimism that I can uncharacteristically interject here and is based on our ignorance and greed okay? I think our ignorance may be our salvation here -- ignorance, okay? I think our ignorance may be our salvation here because the utility of the vast majority of this information is nill and that is a good reason why primary-care physicians are not that interested in genetics. They say they need to know more but much of that maybe they bought or selling.
When you try to come up with a list of finding whole genome sequencing and have actionable utility in the clinic is a pretty short list. I think there are a few but it is a pretty short list. That is why I am so interested in Richard's study because I think the way forward is that we simply have to figure out at the 30,000-foot level which of these findings may the -- make a difference. You do not need them to sit for two hours going through the whole whole genome sequencing but what you have this is a certain list of things that have been determined by some knowledgeable people. There would obviously be many iterations x cetera and the details of how you come up with that list are important.
That is an area where the Secretary's committee can have a real influence in helping formulate a mechanism by which that list is made and what I would appeal to us all to do is keep evidence at the forefront. Because I am an unrepentant advocate of evidence based medicine and we need to demand evidence that this stuff actually does any good and is beneficial and for those few trunks, nuggets in the genome we can find ways to implement. Does that make sense to people?
It does but let me add in historical context that because I would agree with your statement that today we know almost nothing about the human genome but I loved to draw the analogy between what you're doing right now in genomics and what happens in medicine centuries ago. The analogy that I like to drop is what we were building right now is a gene microscope with a wonderful analogy to the light microscope and it was bilks and was one of the great scientific tools of its time. The history of science is the history of scientific measuring tools and it is a test that allows it to measure what is right and what is wrong.
The microscope is no exception, it dramatically revolutionized many scientists and bought need to see how leaves work -- botony and there is one area that they had no effect on and that is medicine and even though Robert Hook identified the cell there was a thing that came up from the light microscope and 300 years and in the 1870's there was a second revolution in microscopy and the Germans invented dyes and by about 1879 was the first time that the microscope was good enough to look inside a cell and the discovered the bacterial cause of tuberculosis in each thank you.
There was more medical progress -- 1882 and there was more medical progress. We have had bad gene microscopes now for about 50 years and what has happened? O Steve there are about five actually -- okay, there are about five actually and nothing has happened in medicine and genealogy is revolutionized British the whole function of the human genome is revolutionized and we are about to change that. Collectively some of the folks on this panel and others are shipping the good gene microscope and turning it loose on the medical committees who knows how to use the measurement tools and given about five for ten or 15 years or three centuries.
I am very serious here, I think what you are confusing is our tremendous expertise in technology with an assumption that that will necessarily transform the way that we practice medicine and the reality is that the practice of medicine is inherently a messy and will be the rate limiting steps to determining critical utilities will not be transformed by having better the sequence and you have these issues of measuring the environment. And I think your analogy is possibly perfect and it may be three centuries --
We did the 50 years, we have the measurements.
My point is that the rules of clinical medicine are the very different than the rules of science and I think a great quote in Harold's region, about the 10th anniversary of the genome got to that and talked about the difference between science and medicine. Medicine is messy. And what we have learned over and over is when we do not insist upon data and evidence of benefits which is not going to be bought through more technology. When we do not insist on really strong evidence of benefits we can not hurt people, the stakes are much higher in medicine then they are in science.
So far I would argue that the discovery methodology has been to correlate phenotype with noise and for the first time we will start correlating phenotype with a signal.
I think it is a much smaller step for did what you think.
I will take my own question and then a question -- mark, do you have a question as well? Let me to mind first then.
I think I would post this to either Cliff War Martin and last week Jay Flattley said there were of these 1,000 individuals who had pulled genome analyses done -- 100,000. That is based on -- that is not sequences, people who have had whole genome genotypes and it is a good evidence that it is price sensitive and if you offer these things, maybe the critical example is -- clearly if you drop the price people will purchase its.
Since we're talking about affordability of the genome, what is the realistic expectation of that pricing open the next let's say five years?
So the prices today of a complete human genomes are about $10,000. I think what we will see is the technology being able to cut the price in half every year. It is not clear that that is what the markup price will go to because the market price can go higher and I think Jay has made this point as well. That is, if you look at the high end genetic tests they tend to be single-digit thousands of dollars and you may not need a $500, $500,000 genome in order to make that a valuable thing and in particular there is great consensus in the community that the first place quickly this is going to have an effect is in the cancer arena and there is a whole history of testing their supporting higher prices so we are beginning to move out of the cost base pricing into a value based pricing model.
So I see if you draw the street lined with a chart today with the continued improvement in 2014 it will cost a nickel. That is not going to happen. I think we will get down to between one and $2,000 I think it is likely that we could see a pause there and continued drops in behind the curtain technology might affect the pricing. That is kind of my speculative guess about the future of market pricing.
If I could briefly comment, I obviously am looking for to that and one alliance that we are involved in, clearly we as the interpreter are challenged to really deliver and really show that you can do something with the findings and again this is an exploratory trial that we are doing so I think it's for us to really get into it and try it out and what we can do and it's a challenge for us on the informatics side. I agree with what the critical part is in that is back to the analysis. So how good are our systems to interpret these very complex data. So I think some of the costs will be more and more in the analysis. That is my idea on that.
Again, since only Greg answer my question and I was addressing the whole panel. So we believe in principle of evidence based medicine. What does that mean in factors? There are two measures that I am assuming with the costs and although it is complicated, the costs will go down and analytic performance will be better and better and we will have more data on people and will be able to Duke gena -- do genotype/phenotype correlation and the use of the genome and medicine will require additional studies and are we assuming that it is good because we have correlation studies that are being done. Are we assuming on average they use of a genome in the general population will, you know, increase and lead to better detection, early intervention save the lives without doing harm? And the question about that threshold between research and practice predict when it comes to me, the genome is a great research tool and will remain a research will for a long, long time but this seems that the price wars so to speak for the pressure on getting the genomes a reasonable price seems to imply with a utility. I cannot the imagine -- assuming that the genome you can buy it for $30 it is a great athletic performance and you have all of those annotated databases, is that the nab? When do we across a certain threshold?
This is a great technology that provide a lot of billable insights but on the clinical side is the intervention that actually -- that are critical for utility as opposed to just agnostics so you consider the diagnostic sketch so far and then we need the technology is on top of those in some of the technologies may be pretty simple maybe it will be, geez, this information will make a huge difference in people's willingness to exercise or eat healthy. We need to show that. We need to show that the benefits exceeds the harms. And if you are at lower risk, what does that do to people? You have to show that there is a real income and the benefit and we talked about pharmacogenomics and even with the reasonable basis for optimism such as anti coagulation the jury is still out and we will need a lot more information about the technologies that come after that in terms of clinical intervention and the things that will actually make the difference.
Don't you think that -- that we will reach clinical utility incrementally it as different genes are demonstrated to show benefit? So for example in the one example there were something like 3400 pharmacogenomically variant [ Indiscernible ] and they are not ready for clinical utility but, on the other hand, if we are talking about [ Indiscernible ] and [ Indiscernible ] perhaps those main reach utility paid one reasonable maunder and model Ford is to determine which parts of the genome makes sense and implement them as their demonstrated with good evidence.
Thank you.
I was not asking you that.
[ laughter ]
Just like at home.
Okay.
So we're mixing research and practice so I think that's does the genome qualify as a tool for coverage with development as to go along and hides the vast majority of the genome information because they're worthless at this point in time and as they learn more and covering that type of information for every bit of actual data today there are thousands if not hundreds of thousands and millions of not actionable data. And this will continue for years to come. It is not going to -- just because the genome is a $30 or $50 should we be investing at the same time in studies of outcomes. In addition to this sort of mad dash to sequence and getting the prices further and further down so that we can, you know, have the genome in our record and saying that now we deploy it for this purpose and that purpose. And what are the white models were doing this -- right models for during the.
Sam and Jim and Mark, let Sam go.
It seems that over time we will have greater and greater clinical utility because none of us will stop the important research that is going on but I want to take us into the current real world situation and I wonder if several of you could speak to this. Today in companies like mine are making decisions for coverage for one of the three Americans in terms of whether these tests have utility and we mentioned the several handfuls did so I have two questions so let's take breast cancer and we look at BRACA and what percent of women are getting stated the art treatment? We are taking the scions of CLIA -- the science of CLIA and what percentage do you think are getting sick of the art care and the second point I will make, what do you think that number is when it has been studied?
It is probably like everything else.
I.e. understand in general and the slower -- eye and a stand in general it is lower and probably depends on the environment that you in are in so I think the answer is lower than we would like it to be.
Surly it will be low but what percentage of women with breast cancer care to buy an ecology and I think that number is pretty high so here we have a sophisticated group of physicians and great science in the number you say is lower than you would like and then if we start looking at other drugs were the evidence is also there. Being used in millions, the second most common drug use a so we have to figure out today translate what we know to make sure it reaches clinical use more broadly.
The second comment is about the costs. You know, $30 genome, $100 genome I agree with you many of these tests are in the several thousand dollars. I do not think it is a matter of cost if you think that the average American the cost is nine dozen dollars, eight dozen dollars a year. We could all have our whole genome sequence even at $10,000 that will not change and we can have that as the database and maybe even one could advocate doing that and doing clinical outcomes research on that population is a very different type of research, not the controlled trials that we know so I do not think that costs and getting it down to 50 or 100 is the answer in the answer to me is clinical utility and what if you could speak to those issues and what we should do, the real-world setting of today in this next few years. Incredible things are happening and that is by 2014 Medicaid will be massively expanded and that is going to impact state budgets.
Many decisions need to be made field time and based on the knowledge that we have today -- realtime and based on the knowledge that we have today.
The clinical utility is repair beseem with that question I think there is no straight Ford answer to the question I think that is why we're sitting here struggling. I think in an ideal world is here with like to see some variants of foreign jam proposed and that is a situation where you take close to genes -- that Jim proposed and that is where it you take those closed jeans and package them together in some multiplex way that makes sense and maybe it is a cardiomyopathy panel or whatever and might be and you administer that prospectively maybe you randomly assign people to receive these results and other to undergo a discernible and not get a genetic testing results directly and maybe you follow them out and select the same sorts of things to show as you would from any other prospective cohort study. I think if we make headway that is the kind of data they are going to be expecting. I do not think there is a quick answer or quick approach but that is the data that in my opinion is most directly needed and we need that for high retargeted -- high retargeted limitations as well as the populations that would be using that testing. We need that type of targeted study in my opinion long before we have more general assessment of whole genome secret stayed any broad, undefined population.
I completely agree and I want to address this idea of the under recognize, of barge under recognize percentage of the population that to have a genetic risk right now that they should have been treated or tested. And I do not at all want to imply that I think we should give up on our traditional mechanisms for Genetics education. I think we need to continue with what we are doing. I think whenever possible we need to build support for our clinicians into the EHR, decision support and the promise that come up with the ICD9 codes to prompt that conversation and I think that we need to do something a little bit differently to get some of these messages across and there is only so much that the average person can remember in their head about genetics and we need to provide a little bit of more support for people to pick up on these things.
So we couple of different things. One is coming back to phenotyping and one not talk about this because I will have a chance to present something tomorrow on that but again the recognition that evidence is going to be held back by not understanding phenotypes well enough to be able to do the correlations that are needed and in some ways the human genome project has given us the hammer problem. There is a huge investment in sequencing capital investments and machines which means that we have this gigantic hammer and other areas. And just to carry that one step further Sam's question was a good introduction to this phenotype issue because in our institution of our cardiologists do not do genotyping and they compared the two and they say why should we do genotyping because what we're interested in is that aggregation and we should be doing that test. So genotyping may not always be the solution and we should be powerful about that. The second thing is that we have talked about evidence and we have talked about support, clinical decision support and you have heard me talk about the needs for that.
I think we also have to be cognizant of the fact that those in and of themselves have been shown not to be sufficient to be able to change care. It has to do with the culture of deliberate and this is what Jim was talking about. We're just beginning to get our hands around the science of the implementation of new knowledge and we do not do a very good job of that and Sam mentioned the breast care and did you come into the hospital for an MI how many leave with aspirin?
There are cultural aspects and our micro and macro systems. And if we try to do this -- once we have the evidence that will be sufficient or once and have the tools in place, we are not going to change practice because it has to do with implementation. This is a situation where genetic testing clearly has no relevance whatsoever repeated is applicable to anything that you're trying to do and removing the needle of improving care. The data is more complex and there will be more challenges in terms of managing the information but the basic cultural changes remain the same. And I think we have to pay attention to that. We cannot just say that developing the evidence is going to be enough to make people better.
So I have a couple of things not related but I actually would like to build a little bit on what Jim said because having a BRCA expression is an example of the complexity of this because it was not necessarily lead to the state of care. It may lead to a state of screening and you may go into a high risk screening program because there are so many different options and that is where the risk of screening comes in and some of the benefits that comes in with screen which is very different and has also involved over the years and tested again is a good example of how things have changed in terms of breast cancer treatment from when it was first introduced and the lower expression of those two and it was only and that she realized that she needed to have a higher expression and the tests were changed so these are really good examples.
I think the other thing and this is, you know as an advocate working as a cancer survivor one of the issues has been researched versus practice and how people understand it and you read the newspaper and what does that mean for me and everybody wants to know what it means for them and we do not always have the answers and in so many meetings I am and we are always talking about how to quantify what -- how do we tell them what we have done with all the research dollars or what the promise is and I think the analogy is really interesting that it can go many different ways so those are just some comments.
That was not a question at all.
I guess the point is how do you deal with the complexities of the evolution of what we happen to me in genetics and clinical Genetics and Gino Midases Asian studies and how do you tell that story and continue to get the support and take care of the patients.
Emily?
I agreed those are fantastic pleats and it is a good question I think that -- I brought this up in my talks the best I could and I think the issue on how we present this information -- I think it is how we think about this information in terms of the evolution of a single gene disorders, whole genome sequencing, and how we approach did and how it is similar or different to other areas of medicine that attitude and that belief I think colors the wave that we think about didactic content and how we want to provide decision support and colors how we set up research studies and it certainly affects how we communicate with people and how we think about the best way to share this information with clinicians and consumers.
Great point and I think it is something I am very interested in studying and Best practices with risk communication and thinking about the uncertainty. I think it was Rich who dropped this idea that the uncertainty that we think about with whole genome sequencing to meet it does seem to stand in some ways because -- to me does seem to be distinct in some ways and these numbers, it is a very different thing for people to understand.
Risk is the determinant and that is one of the things that people have the most tough time grasping did it indicates that there could potentially be another assault on their gene and is not just a high risk, not one assault, but two assaults paid so that is a critical piece that I do not believe people understand.
I think that we hesitate, people like Bob Greene has done great research and we do not know what the other factors are so we can say yes, there are environmental factors and other genes play a role and there are still so many unknowns and that can be a barrier sometimes to try and find more effective education.
Thank you. Paul?
To assess a follow-up to one of Mark's comments would is a better predictor of [ Indiscernible ] than anything else.
Do you want to answer?
Yeah, please.
I do not know the data but are cully -- are cardiology Group has done a study on that so we do practice what we preach.
That is good. I wanted to ask actually each member of the panel if they would pick it seems to me that one of the things that this could about the session is that there's a lot of issues in the research agenda as well as if there ever is a transitional moment in the clinical delivery agenda as well. And maybe there is some research subquestions as well.
Could you each maybe suggest the top and hopefully you will not all states critical utility and the evidence based medicine. But the cops searched question that you think remains to be addressed -- top search question that you think remains to be addressed in terms of the affordable genome.
I would actually argue that the clinical Liddy aspect of whole genome sequencing and what variants of phenotype you are talking about, I think that is one of the utility cities that need to be done.
I actually would agree with that and just to add to its I think that the question is a critically important one, too and that is to identify with it the perfect outcome measures can be used in evaluating these types of technologies. I would not be for something that would be in little bit more expensive in returns to the outcome measures speed suit the SEC changes in health-care related behaviors -- so things like changes in health-care related behaviors and whether they feel empowered to take control over their own help and whether they regret having had that type of experience, the psychological outcomes. I think it is that consolation of outcomes that we need to add into the mix and combined with traditional types of research, health research outcomes and that would be the one thing I guess I would add. It would be nice if the director was here to hear those types of comments.
[ laughter ]
How fortunate.
[ laughter ]
I completely agree did I absolutely think that clinical validity is our first step before we can really effective the look at outcomes and the impact of these associations and I really appreciate your point about thinking about the outcome is more broadly and I will discuss the size that while, of course, I kind of advocate for an agenda surrounding Research to risk mitigation education I think is the important and I think that looking at outcomes and long-term effects the only from a medical standpoint, how this change is conditioned behavior but also the perspective of personal utility and what people do with this information over their life course is really a big priority.
I think one of the most exciting things that affordable whole genome sequencing can do is to start the process of moving away from Association study. So Association studies our statistical studies not informed by biology. YouTube biology after the study to try to interpret -- reviewed to biology after the study to try to interpret results and the richness of the data set. Not that that was you honestly done, it was all that was unable to be done. And we did the best record over the past decade and a half or so with skip Association studies and there were 8Q good things but not as many as three all were hopes.
What's affordable whole genome sequencing does is it enables us to reintroduce and discover genetics and I think the first step in that direction is.
Being led with the disease studies that enable us to take the families who have reasonably rare genetic disease. Of unknown causes. And then simply too high quality whole genome sequencing of me before individuals, two parents, two children and be able to sequence 25 for $30,000 genes and chase down these heterozygous mutations that will enable us to see now we understand the genetic cause of this and then we can do things like testing and potentially understand the pathways.
So we're moving away from the sterile statistical analysis of genotype to phenotype the towpath for analysis and that is where we as a research community want to go and need to go and it will take a long time to get there with the exception of a handful of these low hanging fruit. They will be dramatic in their impact of these next five years.
Yes, from my point to a few I think the critical thing is that these surges, and markers and these are markers -- and what we get now is full of gene sequences. And we know of these genes from evidence and medical evidence. What we know now is cutting a much more complete picture for the genes. In my opinion the first applications are going to be in the diseases. We will actually sequence genes that have been rare genetic diseases 410 to 20 years and fined markers in their -- And find markers in there so I think that will be a very interesting area. On the pharmacogenomics one we have 100 genes now that we're using, curating from the literature and to show that instead of doing one thing at the time we're trying to do 100 things will all of these genes. Yes, we will find mutations that are unknown and what we are planning or hoping to do is then followed the studies on these individuals and see whether they work pretty I think one of the big difference is we really need to understand is that we have a complete picture we were looking for this for a very very long time because these SNPs were not. Functional. And I tell you in the database half of them were [ Indiscernible ]. So there is a very big difference here of the functional interpretation. So if you ask me, but I think we also need to have is a lot more functional validations studies on markers. So can we learn more? If we have a gene and we have heterozygous or homozygous can we validate that? I know and DRC1 and DRC2 there are a lot of cell based assays for validation studies in for example I will give you another clear example. If you have a normal legal at the beginning of the protein and you find [ Indiscernible ] so it is pretty much in the same position in that protein very likely it has the same not always the case but very likely. There is evidence that is coming and how do we treat that? Do we wait until we have 1 million find these rare nutations? We probably do not have the time so we need to start dealing with these are known mutations and make inferences a little bit earlier than that so I think that is what did not come out, the SNP panels today are markers and now we're going to really understand the full ID gene sequence, multiple mutations and we see a lot of them. The way that we do the ranking is that they have 100 loss of function gene so we're looking at does it make sense? And we have additional tests going for example, a panel tested for somebody who has a mutation in those genes. So this is what we're looking for in functional studies that have been but I do think we get the full gene sequences which is different to SNP panels been there is a huge difference between the two because we know from biology we know more that these imitations have a function so that is basically my comment.
I will take a little different tactic because clearly there is a lot of hard science that needs to be done that the kind of just heard about people we have to be prepared to use these technologies and I think the first question is would as the evidentiary standard? The evidence standards vary a lot for they were talking about people who do not have alternative therapy the bar is three different than the things we talking about in to read with screening and aging population so we need to have a lot more discussions about what those of the contrary standards should be and we need to have a lot more discussion about how did they get them out there and used appropriately. We need some macro level discussions and research about what the appeal of -- real economic implications are. This has huge complications to the macro how consistent and is not likely to be cost saving. People do need to still looking at how this is really played out when you're talking about not only very effective technologies but also very expensive technologies in many cases for drugs in the cancer arena. We need to think about all of that in terms of how did these things get used and how do we have the right to public discourse that will be talked about, not this use technologies but all of the technologies on an affordable health care system and those are hard discussions to have but I think we need to start bidding information together so we can at least be informed the.
Can we ask whether you have questions or adjusting the panel directly? Because if not then if no one has questioned with the panel we should receive permission to move on to the committee discussion.
Go ahead.
All I wasn't going to but I came up with a reason why I should.
First wanted to thank Paul for is the dean the sort of individual response. For illiciting the response. Sometimes I am accused of being a naysayer although I entered the field of genomics with a full positive force and a a public health agency that is skeptical about the use of genetics and I have been doing this now for 13 years and we work with all kinds of partners including state health departments and we are always scratching our heads the sort of when is the promise of genomics because we want to use it to save lives and improve health and there is always that balance between using what you know versus always going after the next thing and WGS, whole genome sequencing is going that exact thing so there is that tension and when we look at this a couple years ago from a transitional pathway perspective I think you showed the T1, T2, T3 and we tried to map of the research dollars and tried to look at the number of publications in various phases. In most of the genetic funding and research publications are either discovery or early translation and there is very little going on in the T2 and beyond. I call it the world less traveled because no one really publishes on how you actually implement it, how do you reach the health care systems? How do you reach the underserved cracks and here we are sort of an evolution in technology driving us more and more towards the new with of the implementation or with a variable implementation resources going into what we already know.
I know the panel members may have different opinions but I'd just pose this out as an arrow out there to figure out what we should do now while all of this research and technology being develop.
I think something that we also need to recognize is that we talked a lot about the cost benefit and DRC1 was brought up over and over again and today it costs $3,000 we're talking about a hundred dollar genome and actually there are prints of the choice today using the sequences for these disorders with a large number of genes, cardiomyopathy being one of those so just wondering how do we have these, the capability of sequencing all of this information is a lot cheaper than the target that we keep talking about and then balancing the use of this information to actually not cause harm.
Because these technologies that we have seen and we have acknowledged that whole genome we are going to need an opportunity.
Thank you.
My question relates to how this whole discussion is likely to be altered by the explosion in computer based social networking. They did not have faced book and my concern is we are acting as if we have control and that these decisions will actually come from our constituencies rather than from a different direction.
A $100 genome, people spend more money voting for American Idol than that. What are the implications that this may become modification of something that we take very seriously but the rest of the world may not and to get a kind of participatory genomics, Mass genetics going on we see this happening in other areas that we used to call medicine that will miss the boat and was recognized that this conversation and the challenge to critical utility -- the challenge is not going to come from this discussion, is clean to come from a completely different direction and how do we both educate ourselves and also prepare the policy discussions to engage this in an informed way?
I completely agree with you. I think the price point is clear and if you look at the Direct to consumer companies out there one thing that we have learned is if the price point drops at some point many people will go for it. They do not know what they're getting they just do it for the fun of it and the experience in the industry that we're seeing is that the price point of $3,000, $1,000 is still a little bit too high and $300,000 people paper and $100,000 everybody pays for it in the important thing is I do think that this will happen and I do think that more and more people will actually participate in some of these studies and that is maybe we will do research in the future and the question is when it is happy and we talk about $30.100 bid that is not next year, the full cost of, again, and that is optimistically three years out but maybe even more. But I do agree that there is a big driver.
I teach at Stanford and there are students that are being offered this test and I do have to recognize my colleague here --
[ laughter ]
In response to the snide remarks that he has made.
[ laughter ]
What is interesting in its call quickly decided to do this and how shocked they were that I hesitated and this is generational and what we are facing is a kind of transformation on the way that people think about identity and shared identity that we have not confronted adequately in this conversation.
I will be quick. The analogy of the microscope and how technology proceeds our knowledge on how to use it is a great example but I would counter a little bit that, indeed, it opens up dislike the telescope opened up the skies and the microscope opens up the human body and all sorts of things and we think about the body differently because of it, it has led to an understanding of bacteria and infectious diseases that has had an impact on global help and a lot of people consider that deal lying fruit that we were able to develop antibiotics and able to save lives and in just a quick history of that did a lot in the reduction and the way that we deal with biomedicine are around the world.
We are seeing more chronic diseases and that model is not working quite so well and Richard you are raising the issue that the sequencing might offer a paradigm shift that maybe we will be looking at Madison differently and I do not know if it is or not or whether we are just able to look at reduction is bigger so a bit complement about this fabulous panel and I agree you have a whole range of expertise here and I am wondering what would it be like if we had a similar panel looking at the environment when we talk about gene environment and is there enough overlap to have that discussion together or is it really two separate circles?
Might you about the environmental elements that are going to make their way are did so early stage. We know so little about it. Our experience as bad as they had been in genomics are worse in environmental factors. I think it will be a very fruitful panel in 2016.
[ laughter ]
All right.
To answer your question from a NIH point of view it is an absolutely appropriate question but I would echo what Cliff said, NHGRI has several partnerships with the Institute of Environmental Health Sciences and various joint studies and really the technology for accurately capturing environmental data while they are advancing the our Lady in behind the economic data acquisition so I think this is the kind of panel I do not know -- it is not that these groups are not talking is that there needs to be a little bit of catching up and we are looking at ways to deploy both sets of technologies and we also do not want to do that prematurely the technology is not quite ready but absolutely something this group should be monitoring.
So no other questions from the panel? Let's give them a round of applause.
One more, I am sorry.
[ Speaker unclear due to low audio ]
For the panel directly?
Okay.
So let's thank the panel.
[ applause ]
And we will pick up immediately with Charmaine.
Okay, I really -- When Barbara started to talk I thought she took my question and I wanted to go back to the comment earlier about the need for us to understand genomics and that to me is part of the key and even understanding how to translate his information to the clinic and the underlying mechanism I was going to say is more than the genome. It depends on how we define environment and I know at NIEHS the focus has been on the physical environment and we need to expand it to think about the social environment, cultural environment and psychological environment and I agree that the Mendelian disorders are great models to help us try to understand that so my comment was going to be the need to include the social and behavioral sciences and understanding of the underlying mechanisms of diseases when you start thinking about health and society diseases.
I am going to turn back over to you two because I think the next step for us is to think about where we want to go from here. We have heard a lot of issues and clearly there is a lot of interest and we need to begin to fight the charge for ourselves as to what we want to do.
We are going to pass around -- this draft charge not knowing entirely what we were going to hear today but go ahead.
I would advocate that we need to keep in mind that we need to tell the secretary something about this.
[ laughter ]
We can think about and cover all of these fascinating issues but at the end of the day what we want is to tell the secretary what she ought to do and I would just -- I have ideas about that but we need to, I think the figure out how to operational lies the integration of what makes sense and how we do that and how she can help facilitate that. I just want to make sure we do not spend all our time thinking about this at the 80,000-foot level.
You stole my line today [ laughter ]
I feel like I'm always the one assigned to say, remember we need to give the secretary recommendations that she can implement and --
So we are converging, yeah.
And in space, too the next thing you know we will be close to each other.
Too frightening for you.
[ laughter ]
So let's talk about how we get to that point were we have something wise to see about what to do. Mark?
Just quickly looking over the list here I think captures what we have heard pretty well today but taking Jim and Sheila's comments to heart it seems that the secretary has some control over the portfolio of resources in one of the messages that I heard loud and clear is more than just sequencing. Maybe I heard it because I said it.
[ laughter ]
But it is more than just sequencing. I would advocate that if there is something that we can send from the committee to the Secretary about how we would perhaps balance a portfolio to more adequately address issues that do not relate necessarily to the technologies or just the sheer generation of information but address other issues that we have heard about. The social implications, the implementation issues, the issues of annotation, education. A lot of which relates to other reports that we have done in the past. We need to have some sort of a tangible investments to say we want you to think about this. In some ways it is akin to what was done in the genome project and maybe it should have been the other way around and maybe we should have given 5% to the sequencers because it probably would have been in a battle bit more even be that as it may I think we are the balance from my perspective. And as a consequence we probably are looking at a chaotic rollout that we are going to be trying to pick up the pieces from.
Do you have anything else that you want to talk about?
So in addition to the kinds of things that you're talking about I think I hold a variety of ethical informed consent issues and maybe that is what you meant by social implications.
I think I was trying to pick up on what's Charmaine had said and it is more than just the issues of informed consent or risk and that sort of thing and also reflecting the fact that you know, medical delivery is messy. And will this potentially worse and disparities cracks and what would be the implications of that so I think it is a bit broader.
I would say one of the big concerns about personalized Madison is to reduce the disparities or increase the disparities so I think that should be addressed in an explicit manner rather than being and put and it is not an explicit comic -- explicit and it is a clear concern.
There are a lot of dimensions in what is the access and the other is the information base that is going to be available for some of those committees when our studies have been primarily based inculcation committees and there is a whole series of issues surrounding how it's used by these different population.
So perhaps that has to be identified potential out disparities and solutions or actions that can help insure those disparities are address.
Let me get Sheila first.
Sorry I jumped right ahead of you.
I was looking at the next bullet down, too and I think some reference to the concept of value as well because as we said, this reference is the cost goes up and is that a good thing or bad thing and that depends on what is the value over and about affordability.
Of the health-care system. Yeah, Mark?
I had the same reaction when Paul, you asked all of them to articulate because the interesting thing was to layperson none of the next steps really reflected sort of the big picture items, they are really reflecting to say this is an extremely important technology but we think the implication of this in rear of Mendelian diseases is the risk of this and again Paul's statement is right on we may not have a choice. It was instructed to me to listen to a group across the range of enthusiasm of the whole genome sequencing to still focus on the things that we can still get our hands around and that probably needs to be reflected in a charge as well. I am not sure how I would articulate that but it was an interesting observation.
Paul and then Emily.
I wanted to ask my fellow committee members. One area that was interesting to me and I have seen this occur, is the comparison of this technology and its growth in tech investment and its potential, the social net working part may be unique the comparing it to let's say CT scans or MRIs and I wonder if we need to hear from people who have studied that or no Stop about that. Obviously there are incentives there and wanted information that he did not ask for generated, etc., etc. So there are some analogies both in the research side and the delivery side that I think are interesting. I wonder if that is something we should follow up on.
Just some reflection on the bulleted list here and asking myself the question, so what is different about WGS than what we have done so far and it is obviously the elements or sorted dealing with large-scale, more false positives, technology, informatics and all of these things and you look at the list and you take away WGS and put Genetics here. Yeah, I am concerned about quality, technologies, clinical validity and utility so we're kind of three visiting old grounds. There are plenty of these reports and works that this committee has done and I think, is it a question of just a scale? The complexity or all of the above? I think the task force is a good idea and you probably do not need more data to write a report. So I am sort of looking here for help on whether or not it is something different on this list and we have done before.
I have at least a couple things that we have not really talked about here. One is we have talked about there is a lot of data but you're also talking about the cost of the affordable genome coming down. It changes the efficiency with which this information could potentially be used and get out there. If you have these up-front costs and then you begin to use it and has the potential for really changing the dynamic of this information and how it can be used and how widely it can be used. And the implications that take us out of this one of the time kind of work into massive ability of information in which case we will have to figure out how to use that information smartly raises a whole set of ethical and social issues that we have not really dealt with it fully in a clinical context.
I wanted to specifically respond to this in a couple of people mentioned the ideas in economies that you do it all and just use it as an as needed basis but you mentioned a couple of issues that that begins to raise and there's a whole bunch more, who holds the information and who has access to its and that in and of itself would be sufficient to get a task force going for quite a long period of time. But there is that compelling economic argument and if that is really what is going to drive the discretion than in some ways maybe that is what the focus should be. We will just assume that this will be done for everybody and -- no big deal. How do we actually use this information so we do not think revved -- not bankrupt the system.
I do not think we can use this for everybody because $100 will be a cost per year for people so I might go back one step and make sure that that is an assumption that we can actually make.
I completely agree with that. I mean, for my patients $100 is a lot of money and what I think that makes us focus on is what is our role of the Secretary committee? If this is going to be something that sweeps through and they will get their whole genome sequence, so be it big we cannot make any difference.
There you go.
But I think our role is that we're supposed to make recommendations about what policies should be and hopefully those will have some implications for peers and we will have some tangible repercussions as well. So I am not saying that we should ignore that this will be a social movement and sweep through things but I do think that we have a role in trying to set the agenda for how medicine should go forward and should be practiced or if we don't then what are we doing here?
Barbara?
A tiny point.
I think I am going to vote for going forward with it and one of the distinction is not everybody went for a cold body scan and this does spill different because of a genealogy has been so popular that love that technology and social networking like Paul has talked about there will be a different uptake and agree a hundred dollars or thousand dollars is not affordable for everybody but this is a little bit different.
So I am hearing -- yes -- she luck, go ahead.
I'd just have one more comment that kind of ties it back together and I like your analogy for the imaging, whole body scan in that it does not just go from where it is sweeping through to the clinical practice and you do have those issues obligatory oversight and how will that be evaluated by the larger payer, CMS and how will that be addressed in terms of that government oversight as well and that is something that this committee can help in putting together and a lot of our other context that we discussed in general practice.
I wasn't -- I was not saying that we should not go forward with that I think really what I was saying is that we should be explicit about what -- how it fits in and with our charge being genetic health and society and with the societal implications are not that it's not going to happen and pull body scans are a totally different thing but we do have to be explicit and mindful of what is affordable to one person is not affordable to you.
We're hearing a whole variety of issues and I am not sure we're talking about a discreet project or are we talking about a very large report that would be like one of our real studies I would be very interested in getting your feedback before we actually set up a group and charge them with that, with beginning to work on this and obviously we need to refine it in little bit more but I am interested to hear your thoughts, are we talking about a substantial, major undertaking? I heard somebody saying, haven't we done all this before? Maybe we can just write it up, maybe we need a stapler.
A stapler or shredder?
[ laughter ]
We can extract a lot of this and assemble it and how much new is there here that is going to take some more background kind of work.
I think that comes back to the fundamental question that Jim and she space and what are we really wanted to tell the secretary and I do not have a clear suggestion in my head I suggested some things that we could tell the secretary but do we really have anything to say to the secretary and that in some ways depending on what is the wont to say will determine whether it is going to be a relatively distinct letter versus a major report.
Let me ask Eric -- you are raising your hand.
I am an ad hoc members so I do not have official business necessarily but one observation that I would make is that the field is moving faster than a committee like this can deal with.
You are right.
In four months or six months which is currently happening is a huge amount of time. Without necessarily knowing exactly what you might produce I might suggest that you released put into motion enough of a framework for trying to catch the issues that you can even identify now recognizing that they will be far more danced by the time that you catch those issues were as if you do nothing you will struggle to catch up and there are several suggestions on what you might want to do. And six or nine months from now it will even be more striking so without trying to define to precisely the product just if you had a framework for semi regularly updating yourself and trying to hold on to this fast-moving train I think the community would be in a better position.
So if you were advising us as the key issues that you think that we can grapple with -- And what I am hearing is the short-term, what would those be?
We heard several of them. I think clearly to start with the sheer dealing with the massive amounts of data. We're struggling with a basic science level and we heard lots of discussion here even once you get your hands around that, clinical utility and all of these issues as it enters the complexity of the medical care system and I heard the suggestion, look for analogies that this has come in such as whole body skinning and so forth -- whole body scanning and so forth and there is a whole field of the there who are medical biometicians and with the electronic medical records are intersecting with all of this to no data and thinking about how it is going to be put into an information system. You heard a little bit about that but I do not think you heard as much as with this out there so maybe he means a little bit from them. But if several people thought about it this would what your appetite but there are a number of additional things -- whet your appetite.
And may be tabled the idea of what the product is going to look like and four to six months every getting a little bit updated would be well within your interest just that it is happening fast and furious.
I am just thinking out loud here but I hate to see you over to find out what your products should be.
I don't think we're going to do that today but maybe --
And on green -- ongoing working group.
We can have a series of things.
And I think also one of the issues that are very important is not only with clinical utility but from a practical point of view, the data and the instruments and electronic health records, those are key questions that we can ask the Secretary and try to develop an effective approach to the question on the multidisciplinary research agenda and they can be housed and query and gauge the clinical utility of the data and as we continue maybe we can have several products, out maybe later to the Secretary and the issues that we have addressed and what are the new issues that are coming out.
I think a lot of ideas are on the table and the hour is getting late pick what I suggest is that we go ahead and form a task force to take these ideas and began to work with them and this is a topic that we may want to visit in October and focus on some short-term things that they might want to do -- I am hearing some longer-term social ethical types of issues paid maybe we can ask some -- and maybe we can ask for some volunteers. The see if we can get some additional members to work with them.
Jim, Andrea? Janice and Charmaine? Others.
Those of you who did not raise your hand are not spared.
I think Eric's point is well taken and I would suggest that you look for an outside member of that group that has bioinformatics experience because are curdy luck chronic health record environment is not capable of handling simplex genetic tests. So someone like Mark Kaufman who has had a lot of -- I'm sorry?
Eric, do you have some specific suggestions?
There are some people at Vanderbelt.
They are right in the thick of the storm.
We may get back to you with some others.
And certainly get other ex officios and at Hawk members. So we will start with that group and thank you for a stimulating afternoon and we are going to start tomorrow. A reminder, at 8:00 a.m. and those of you who are heading to dinner with this illustrious cloud, we will meet in the lobby.
Thank you everyone.
I want to thank Harris and Paul for a terrific job.
At what time, Steve?
6:15 p.m.
[ event concluded ]