ࡱ> kmj Nbjbj 'fooF4!!!8/"$S",**"""""p#p#p#*******,.Z*p#p#p#p#t#*#""*###x#"""*#p#*###?!##n*$*0*#/#/###R*p#p#p#d !! E:   TESTIMONY OF DAVID EWING DUNCAN BEFORE THE SECRETARYS ADVISORY COMMITTEE ON GENETICS, HEALTH, AND SOCIETY SESSION OF PERSONAL GENOME SERVICES PERSONAL GENOMIC INFORMATON: A CONSUMERS PERSPECTIVE JULY 8, 2008 Good morning and thanks to the committee and organizers for inviting me to speak and to share some of my experiences and thoughts on the topic of personal genomic testing. Im here this morning as a consumer who has been tested by all of the major genomic testing services and quite a bit more. Im not your typical consumer, however. Im a journalist covering biotechnology, and an author thats writing a book called Experimental Man: What One Mans Body Reveals About His Future, Your Health, and Our Toxic World. For the book Im having not only my genes tested, but also environmental impacts on my body, along with tests on my brain, organs, blood, cells, microbes, proteomes, and other assorted omes. Im also the director of a new program at the University of California at Berkeley called the Center for Life Science Policy that launches this fall. In conjunction with the book, the Center is developing The Experimental Man Project, an educational effort that includes a website, forums, and other activities. The experiment Im running is for one man to take and analyze a wide array of new high-tech tests that aim to forecast the future health outcomes of a healthy individual. I started this project with a visit to my physician, who declared me to be a healthy white male, 50 years old. I also come from a mostly healthy family that lives a long time, sometimes over 90 years old. The committee has asked me to answer several questions about my experiences as a consumer of genomics. These questions can be broken down into three categories: expectations, tests and results, and reactions and thoughts. Under expectations I was asked: What were your reasons for pursuing personal genome services? I had three primary reasons. One was as a journalist and communicator seeking an effective way to report on genomics and genomic testing. A closely related reason is a curiosity common to early adopters of new technology, the sort of person who bought an I-Phone soon after it was launched. A third consideration was that I might discover an unexpected insight about my future health. The next question is: What sort of information did you anticipate receiving from these services? Given the newness of the science and its power to offer insights to individuals, I had rather low expectations. Mostly, I expected to receive confirmation that I am essentially healthy, and that my family is healthy with one exception Ill explain in a moment. I am in that category of patient that doesnt think about or worry much about getting sick. The next set of questions, about tests and results, is where I will spend the bulk of my presentation. I have had nearly two million genetic markers tested, mostly Single Nucleotide Polymorphisms (SNPs). I also have had some deletions, insertions, and copy number variations tested. Most of these markers were run on genomic arrays made by Illumina and Affymetrix, with a few tests run on other custom chips and mass spectrometers. Individual labs and companies have sequenced a number of complete genes, and Im hoping later this year to have my entire genome sequenced. A number of genomic testing services have run my tests and offered analysis, including companies, university researchers, and others. Most of these tests are not yet available to consumers. My genomic investigation so far has cost about $16,000, with the bulk of the tests performed for me pro bono by companies and labs, or paid for by publications that I write for. The cost of running a full genome sequence will boost the total to well over $100,000, although this cost is rapidly coming down, from over a million dollars just a year or two ago. I am not alone in participating in this experiment. My family has joined me for part of the testing: my parents, who are quite healthy at age 75 and 76; my brother, who is 48 years old and has a wife and two children; and my 19 year-old daughter, one of my three children. This morning, Im going to talk mostly about three direct-to-consumer services: Navigenics, deCodeme, and 23andme. Details about the three companies are discussed elsewhere in these hearings, although it is important to note that from a consumer perspective, the sites offer services both similar and different. All three companies use SNP-array chips that test hundreds of thousands of SNPS, but offer analysis of only a few dozen SNPs, which the companies promise to add to over time. The companies all use as sources the thousands of studies conducted and published by genetic researchers in peer-reviewed journals. So far, the companies have opted to offer mainly common diseases that affect large numbers of people, rather than rare diseases. All three companies offer risk factors for individual SNPs pertaining to diseases, and also lifetime risks that combine the results of individual SNPs. Navigenics focuses exclusively on risk factors for disease, offering results for 17 maladies. As part of their $2500 fee they provide access to live genetic counselors on the phone, a feature the other two services do not offer. DeCodeme offers results for 25 diseases, but also provides results for traits such as eye and hair color, and an analysis of a customers genetic ancestry. DeCodemes parent, deCode Genetics, is a publicly-traded company that develops drugs and is a world-leader in conducting the scientific studies used by the other sites. 23andme offers results on 78 characteristics, blending disease with a wide range of other traits. They offer a rating system that assesses tests they designate as preliminary research or established research. I want to mention two other online genetic services San Francisco-based DNA Direct, which offers only individual genetic tests in common use by physicians, and leans towards a focus on customers who have a family history of disease or some other indication, though anyone can take their tests. They provide results that include an analysis from a physician and extensive genetic counseling, as I learned when I took one of their tests. The Coriell Institute, a nonprofit based just outside of Philadelphia in Camden, New Jersey, will launch later this year as a more academic exercise. They will run a genome-wide array of up to 100,000 people for no cost they are supported by grants starting with physicians in the Philadelphia area. They have convened an independent oversight committee made up of scientists, academics, ethicists, patient advocates, and others to choose what genetic tests to offer. I have given them my spit and hope to get results soon. I will briefly describe my results from Navigenics, deCodeme, and 23andme for three sample diseases. More extensive results are available at www.experimentalman.com. First is age-related macular degeneration. This ailment does not run in my family, so I came out with average or below-average risk factors both for individual SNPs and for lifetime scores. The lifetime scores for the three companies varied by as much as five times, but the actual numbers showed a uniformly lower-than-average risk, which was reassuring. The companies each cite a different average lifetime risk for the general population, ranging from 1.2% to 8.0%. I did have one outlier result among my SNPs -- Navigenics gave me a score that is a 6.98 times risk factor, which is quite high, and seems to contradict the other results. This result has an asterisk attached to it that says: The 95 percent confidence interval for this odds ratio crosses 1, which means the effect is statistically insignificant, a statistical explanation that laypeople might not understand. My results for type II diabetes also made me feel good. I am mostly in the medium to low risk range. Only one SNP result out of 19 SNPs offered by the three companies is in the high-risk category. It is a bit puzzling, however, that out of these 19 SNPs collectively offered, only four appear on all three sites. Out of these four, two gave me different risk-scores for the same SNP. Four other SNPs appeared on two of the three sites; these results were all they consistent on the different sites. My lifetime risk scores were not too far off for the three sites, with Navigenics giving me a 21% lifetime risk of coming down with type II diabetes, deCodeme an 18.8% risk, and 23andme a 16.8%. This was a spread of about 4.2%, although all the scores made the same point: that my chances are lower than the average person of European ethnicity, which have a 25% lifetime risk, according to the sites. I have not yet completed a thorough quantitative analysis of my results comparing the three sites, but I can offer the following observations: The genotyping results (CLIA-covered lab tests) were very consistent among the three sites (that is, if one site said GG, the others did, too). Risk factor results provided for SNPs were generally consistent where the companies used the same SNP, with a few exceptions. The lifetime risk factors presented by disease were not always consistent. That is, if there was more than one SNP used for a disease, the results could vary from low to high risk. This is one reason the companies use lifetime or blended risk factors. The lifetime risk factors provided by the sites were usually consistent, with at least one exception for me heart attack. In my data from 2001, the only results that were out of the ordinary were some high-risks for heart disease, although in those early days, I could easily dismiss them because the science was so new and incomplete. Seven years later, however, as I began checking out my results, I admit to being slightly apprehensive about my heart particularly when a private test done by deCode Genetics before they launched their deCodeme site suggested that I had at least one relatively high-risk genetic marker for heart attack. Imagine the scene, where Im sitting at home logging on to each of the three websites in turn as they became available. Naturally, I clicked first on my heart attack results only to find that the three sites gave me different results. According to deCodeme, I have a lower-than-average lifetime risk of about 42%, while Navigenics gives me a 62% lifetime risk. The average risk for a male Caucasian living in North America is about 49% -- which means that one site has told me that I have a lower risk, and the other a higher risk. 23andme uses a different method for calculating lifetime risk and averages, which I dont entirely understand. They use a single SNP that seems to confer a medium risk for me compared to those who have the lower-risk variant, and their life score for me appears to be a comforting 29.9%, compared to 42% and 62%. But they list the average risk for a male between ages 45 and 84 as being only 17% -- which is different than the 49% average risk given by the other two sites, and suggests that I have a 75% higher relative risk of getting a heart attack than their average which is alarming, I think. In comparison, Navigenics says that I have a 26% greater chance of a heart attack than the average Caucasian, while deCodeme says I have a 14% less chance. As a reporter, I have spent time with each of the companies going over these results, and I have some understanding of why I had variable results for heart attack. Briefly, here are some of the reasons: The sites used different SNPs and studies to determine overall scores. As far as I can tell, the sites use different methods for determining risk. deCodeme uses Relative Risk; 23andme uses what they call weighted odds ratios, while Navigenics apparently uses their own blend of relative risks and odds ratios. The sites use different methods for determining combined SNPs risk to produce a lifetime risk; each company uses its own algorithms. The companies rely on correlative SNPs that for most people give similar results, but not for everyone. The end result for me is a fair bit of head scratching as I try to ascertain my true risk for heart attack based on my genes. In discussions with my physician, he is willing to include genetic risk factors for my heart attack risk with other results such as my blood pressure and cholesterol levels in making a diagnosis, but he needs to have a consistent risk factor that he can trust. In articles Im writing about my project and in the Experimental Man book, Ill be providing much greater detail about my results, and those of my family. I will say here that perhaps the biggest surprise in the family results so far have been that my father and brother have a medium-high risk variant for Alzheimers Disease. My father shrugged at the news, since he is completely lucid and healthy at age 76, and his risk of Alzheimers is rising anyway as he gets older. The news didnt much bother my brother, either, since we have no Alzheimers that we know of in the family. I would like to mention that my brother suffers from a rare genetic disorder, Osteogenesis Imperfecta (OI), also known as brittle bone disease. His bones have the density of a very old man, and he has become disabled due to repeated breaks. My brother is the first member of the family to suffer from this disease in our family. We have had the two genes responsible fully sequenced in my brother and in me, and predictably he came out positive, and I tested negative. Im bringing this up because for most of his life, my brothers disease was a mystery that might have been solved early in his life on if a genetic test for OI had been available. The question is whether or not this test should be offered to someone like my brother before the disease manifests through a direct-to-consumer service, or through a physician. So far, none of the DTC sites are offering tests for rare genetic disorders who should offer these tests is an issue that this committee and other forums should continue to discuss. It is worth noting that as more data comes in and becomes validated, there will be a crush of information available that may defy attempts to organize and make sense of it for individuals. My excel chart for the experimental man project now has close to 1000 genetic markers and results listed, and, if printed out, would run over 24 feet long. To answer the final questions from the committee, Ill start with: Did you alter your behavior in light of test results? If so, how? I want to emphasize again that I am one person, and I am not the average consumer as a journalist covering this topic. My over-all response is that I did not alter my behavior because of the tests on the three websites, although subsequent tests that used advanced algorithms and computer models to incorporate genetics, traditional tests such as cholesterol, and ultrasound and CT scans did convince me to alter my diet. This modeling is not yet available to consumers, but will be within a year or two. In general, Id like to present some pluses from this consumers point of view for the genomic testing services I have used. This technology offers: Insight into ones personal and societal health that will improve as tests become better validated and consistent; Personal empowerment; A nudge to researchers and the health care industry and to regulators to commit time and resources towards making genetic testing more relevant for individuals; New avenues for genomic and other omics research that impact individuals and subgroups; New pathways for developing therapeutics. Minuses of this new technology include: Its early days for this technology, and much of this information is a work in progress. Genomic association studies are not always applicable to individuals. Disease and non-disease results are sometimes listed together. There are no generalized standards for the validity of tests or for what exactly is meant by validity or for how risk factors are determined. Physicians are not yet trained in genetics There is a potential that this information will frighten some people. The costs are high and insurance seldom covers these tests, though undoubtedly costs will come down. A few thoughts and suggestions from one consumer: Consumers should be free to access their information and buy legitimate services. It is vital that the sort of discussions we are having today continue, and that all views from all stakeholders be heard. Perhaps there is a way to incorporate early adopters into what is basically an experiment into how to best use this new technology. The Coriell approach is one example of this from a nonprofit. It is important that standards and guidelines for tests and information be established, either voluntarily by companies and other testing sites, or not. A baseline needs to be established so that validation criteria, results and risk factors are consistent. As these tests become more clinically useful, who will pay? More resources are needed for translational research devoted to making genomics more relevant to individuals and subgroups. Disease markers should be handled differently and counseling offered for traits that might impact a persons health. Physicians either inside or outside companies should review disease markers, and alert consumers of serious findings. Companies should provide online locators for local physicians and counselors trained in genetics In conclusion, genetics is just the beginning. I suspect that this committee and others will be holding hearings for many years to come on not only genes, but also on new technologies that will dramatically impact how humans interact with the environment, and also tests that assess brain and body. More information is available the environmental man website. I also welcome The book. Thanks again to the committee and organizers for inviting me to speak. TESTIMONY OF DAVID EWING DUNCAN PAGE  PAGE 2 k u |  5 p * T 5?QS*t4JRs%T&s&t&&(ڸڸڸڭڭڞhOJQJ^JaJh!srhOJQJ^JaJh}hOJQJh;xhOJQJhIh6OJQJhIhOJQJhOJQJh8hOJQJh8hDOJQJh(@JOJQJ5k R0vB@R`gdgd $a$gd "C#%&:**[++,X-a/y13p4<5~5555}67p78;; ^`gd ^gdgd`gd(() )>)F)J)O)d)i)******** +'+A+C+[+\+j+p+++++++,,,--F-W-X-N0O0<5=5`5h5555555555555p6{6|6~66666777-7o7p7t77@hEhOJQJh5hOJQJhFThOJQJhRyUhOJQJhOJQJh`hOJQJJ;;=.?@AA CCD,DD.EZEEE%FfFF(GpGG H]HHIJJ]KK`gd@AAACCCCCCDDD*D+D-DDDDDDEE,E.E/EEEEEEEEEE#F$F&F/FGFfFgFFFFFFG GGG)GZGbGoGqG H H HHHSH[H\H^HHHIIIIIIIJJJJ^KnKKKKKKKKKKL Lh:zh>*OJQJhOJQJh:zhOJQJY L#L'LJLKLgL{LLvNxNyNzNNNNNNNNNNNɸh0JmHnHu h0Jjh0JUhh8hOJQJh5hOJQJhOJQJh:zhOJQJKMLL0NwNxNyNzN{N|N}N~NNNNNNNNNNNNNNNgdgdgd`gdNNNNN$a$gd% 0:p/ =!"#$%666666666vvvvvvvvv666666>6666666666666666666666666666666666666666666666666hH6666666666666666666666666666666666666666666666666666666666666666666666666662 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~ OJPJQJ_HmH nH sH tH D`D  (NormalCJ_HaJmH sH tH DA`D Default Paragraph FontRiR Table Normal4 l4a (k ( No List 4@4 e0Header  !66 e0 Header CharCJaJ4 @4 e0Footer  !6!6 e0 Footer CharCJaJ.)@1. e0 Page NumberPK!K[Content_Types].xmlj0Eжr(΢]yl#!MB;BQޏaLSWyҟ^@ Lz]__CdR{`L=r85v&mQ뉑8ICX=H"Z=&JCjwA`.Â?U~YkG/̷x3%o3t\&@w!H'"v0PK!֧6 _rels/.relsj0 }Q%v/C/}(h"O = C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xml M @}w7c(EbˮCAǠҟ7՛K Y, e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+& 8PK!\theme/theme/theme1.xmlYOoE#F{o'NDuر i-q;N3' G$$DAč*iEP~wq4;{o?g^;N:$BR64Mvsi-@R4Œ mUb V*XX! cyg$w.Q "@oWL8*Bycjđ0蠦r,[LC9VbX*x_yuoBL͐u_. DKfN1엓:+ۥ~`jn[Zp֖zg,tV@bW/Oټl6Ws[R?S֒7 _כ[֪7 _w]ŌShN'^Bxk_[dC]zOլ\K=.:@MgdCf/o\ycB95B24S CEL|gO'sקo>W=n#p̰ZN|ӪV:8z1f؃k;ڇcp7#z8]Y / \{t\}}spķ=ʠoRVL3N(B<|ݥuK>P.EMLhɦM .co;əmr"*0#̡=6Kր0i1;$P0!YݩjbiXJB5IgAФ޲a6{P g֢)҉-Ìq8RmcWyXg/u]6Q_Ê5H Z2PU]Ǽ"GGFbCSOD%,p 6ޚwq̲R_gJSbj9)ed(w:/ak;6jAq11_xzG~F<:ɮ>O&kNa4dht\?J&l O٠NRpwhpse)tp)af] 27n}mk]\S,+a2g^Az )˙>E G鿰L7)'PK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-!K[Content_Types].xmlPK-!֧6 1_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!\theme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] F f kR0 v B  @RC:""[##$X%a'y)+p,<-~----}./p/0;35.7899 ;;<,<<.=Z===%>f>>(?p?? @]@@ABB]CCMDD0FwFxFyFzF{F|F}F~FFFFFFFFFFFFFFFFFF0@0@0@0@@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@@0@@0@@0@@0@ScB113336(@ LN*-/0;;KNN+,.12%,.6!8@0(  B S  ?FFFFFFFFVy2FnqTUY!#@gzJ#q);Po;#[q#.(q#BgzJ#GZ ac M@#[ q#.X y2F#F[ Af2YPkn#r;Po;#'M@#I!y2F +gzJ#|8!#uM" ; #7gzJ#GOVq#uA ; #0SM@#;Po;#N;Po;# ; Q !#!Vz& ; #?';Po;##z';Po;#]n(;Po;#( ; #)gzJ#J+y2F#v1-;Po;#lS.y2FZ / a#p/y2Fw27xk#Af2.37xk#>Nv8M@# I:7xk#;Po;B[?@ABCDEFGHIJKMNOPQRSTUVWXYZ[\]^_`acdefghilRoot Entry F(Ln1Table4/WordDocument'fSummaryInformation(LP*DocumentSummaryInformation8bCompObj` F Microsoft Word 97-2004 DocumentNB6WWord.Document.8