predictivehealth-2011

predictivehealth-2011 - Human Genomics in the 21st Century:...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Human Genomics in the 21st Century: Using Genetic information to Predict Health and Prevent Disease Muin J. Khoury MD, PhD Director, Office of Public Health Genomics Centers for Disease Control and Prevention NEJM Outline Genomics 2011: the excitement of scientific Genomics discovery discovery Genomics and health 2011: characterizing Genomics genetic influences on health and disease through gene-environment interaction Genomics and predictive health 2011: How do Genomics we actually use genetic information to improve health and prevent diseases? The Human Genome at 10 NEJM May 2010 2007-2011: GWAS! Pennisi E, Science 2007; 318:1842-43. Functional Classifications of 465 Trait­Associated SNPs and the SNPs in Linkage Disequilibrium Manolio TA. N Engl J Med 2010;363:166­176. Examples of Previously Unsuspected Associations between Certain Conditions and Genes and the Related Metabolic Function or Pathway, According to GWAS Manolio TA. N Engl J Med 2010;363:166­176. GWAS for Age­Related Macular Degeneration Klein et al, Science 2005; 308:385­389. We live in The “Ome” Era: The Human Genome is Just the Beginning Genome Transcriptome Epigenome Proteome Metabolome Nutrigenome Microbiome “ The Incidentalome!” 2008: Invention of the Year Time, November 10, 2008 Proliferation of Personal Genomic Tests The Complete Human Genome Sequence Faster and Cheaper …… Who When How Long How Much Human Genome Project (NIH) 2003 13 Y $ 3 Billion Craig Venter 2003 (Celera) 13 Y $ 100 Million James Watson 2008 2M $ 1 Million Stephen Quake 2010 5D < $ 50 Thousand Commercial 2010 Days-Weeks $ 5-10 Thousand http://journals.plos.org/plosbiology/suppinfo/pbio.0050254/pbio.0050254.sd001.h The Personal Genome: An Almost Reality A 40 year old man with family history of heart disease and early sudden death Ashley E et al. Lancet May 1, 2010 Samani NJ et al. Lancet May 1 , 2010 Outline Genomics 2011: the excitement of scientific Genomics discovery discovery Genomics and health 2011: characterizing Genomics genetic influences on health and disease through gene-environment interaction Genomics and predictive health 2011: How do Genomics we actually use genetic information to improve health and prevent diseases? Nature vs. Nurture? “There are two causes of asthma, the There environment and genetic variants. Each accounts for about 50 per cent of the risk of disease” Cookson W and Moffat M. Making sense of asthma Cookson genes. Editorial NEJM 2004;351: 1794 genes. Nature vs. Nurture? “There are two causes of asthma, the There environment and genetic variants. Each accounts for about 50 per cent of the risk of disease” Cookson W and Moffat M. Making sense of asthma Cookson genes. Editorial NEJM 2004;351: 1794 genes. “100% of any disease is environmentally 100% caused and 100% of any disease is genetic. Any other notion is based on a naive view of causation” Rothman K. Modern Epidemiology 1986 Gene-Environment Interaction in Cardiovascular Disease “Some vegetarians with Some 'acceptable' cholesterol levels suffer myocardial infarction in the 30's. Other individuals...seem to live forever despite personal stress, smoking, obesity, and poor adherence to a Heart AssociationHeart approved diet" R.A. Hegele (1992) G E N O M E C V D TIME From M Bouchud G E N O M E C V D ENVIRONMENT TIME From M Bouchud G E N O M E C V D DIET PHYSICAL EXERCICE ENVIRONNMENT SMOKING TIME From M Bouchud INTERMEDIATE PHENOTYPES BLOOD PRESSURE HYPERTENSION G E N O M E LDL, HDL, TRIG DYSLIPIDEMIA GLUCOSE DIABETES BMI OBESITY DIET PHYSICAL EXERCICE ENVIRONNMENT SMOKING TIME From M Bouchud C V D Importance of Gene-Environment Interaction (From Khoury et al. Am J Hum Genet 1988;42:89-95) Disregarding interactions weakens gene­disease associations Em p iric vide nc e e nc e t n Effe c s S iz e s Empiric al eal Ev ido n e ffe co s ize s and t ample s ize re quire me nts fo d gGe neas s oAs s o c iatiomple x fo r Valid ate r e ne tic tic c iatio ns o f c o ns d is e as e MJ J P A. Ioannidis, TA. Trikalinos, s Khoury. AJ E 2006 J PA. Ioannidis, TA. Trikalinos, MJ Khoury. AJ E (in press) Selected Genetic Variants and type II Diabete Gene Chrom UK UK (N=13,965) (N=13,965) DGI DGI (N=13,781) (N=13,781) FUSION FUSION (N=4,808) (N=4,808) All FTO CDKAL1 16 6 1.23 1.16 1.03 1.08 1.11 1.12 1.17 1.12 HHEX CDKN2B IGF2BP2 10 9 3 1.13 1.19 1.11 1.14 1.20 1.17 1.10 1.20 1.18 1.13 1.20 1.14 SLC30A8 TCF7L2 KCNJ11 PPARG 8 10 11 3 1.12 1.37 1.15 1.23 1.07 1.38 1.15 1.09 1.18 1.34 1.11 1.20 1.12 1.37 1.14 1.14 Genomic Locations of Proven Signals of Nonautoimmune Forms of Diabetes. McCarthy MI. N Engl J Med 2010;363:2339­2350. Odds Ratios of Associations from http://www.genome.gov/gwastudies/ GWAS Number of Associations 50 40 30 20 10 // 0 1.2 T Manolio 1.4 1.6 1.8 2.0 2.2 2.4 // 3.04.05.06.0 20 20. 3456 Odds Ratio (upper inclusive bound) Methodologic Challenges in Human Genome Epidemiology Publication bias Type 1 errors 5 4 3 2 DISEASE/GENE 1 False positives!! The more you look the more you find just by chance Cumulative odds ratio Nephropathy/ACE ,5 ,4 ,3 Alcoholism/DRD2 HTN/Angiotensinogen ,2 Parkinson/CYP2D6 Lung cancer/GSTM1 ,1 Schizophrenia/DRD3 ,05 ,04 ,03 ,02 Down dementia/APOE Lung cancer/CYP2D6 40 100 50 300 500 200 400 2000 1000 4000 3000 5000 Total genetic information (subjects or alleles) 10000 Methodologic Challenges in Human Genome Epidemiology Type 2 errors False negatives! You may not be able to “detect” effects even if they exist S ample S ize ne e de d fo r de te c ting ONE inte rac tio n fo r a dic ho to mo us trait and a 10% e xpo s ure G E N O M E C V D DIET PHYSICAL EXERCICE ENVIRONNMENT SMOKING TIME G E N O M E C V D DIET PHYSICAL EXERCICE ENVIRONNMENT SMOKING TIME G E N O M E C V D DIET PHYSICAL EXERCICE ENVIRONNMENT SMOKING TIME Candidate Genes Chosen in Major Pathways Apoptosis, Cell cycle, Apoptosis, Cellular growth and differentiation differentiation DNA Repair DNA Metabolism of free radicals/Oxidative stress radicals/Oxidative CAPN10, IL10, IL1B, IL4, IL4R, ITGB3, PPARG, TGFB1, TNF, VDR PPARG, OGG1, XRCC1 OGG1, CAT, NOS2A, NOS3, PON1 Blood pressure Blood regulation, Cardiac function function Hemostasis Hemostasis Nutrient Metabolism ACE, ADRB1, ADRB2, NOS2A, ACE, NOS3 NOS3 F2, F5, FGB, ITGA2, ITGB3, F2, NOS3, SERPINE1 NOS3, Cellular adhesion, Cell Cellular migration/motility migration/motility Immunity and Immunity Inflammation Inflammation Xenobiotic Metabolism CCL5, CCR2, CXCL12, F2, FGB, CCL5, ITGA2, ITGB3, SERPINE1 ITGA2, CCL5, CCR2, CXCL12, FCGR2A, CCL5, IL10, IL1B, IL4, IL4R, MBL2, NOS2A, PPARG, TGFB1, TLR4, TNF, VDR TNF, ABCB1, ADH1B, ADH1C, ALAD, ABCB1, CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2C19, CYP2C9, CYP2E1, CYP3A4, NAT2, NQO1, PON1 NAT2, ACE, ADH1B, ADH1C, ADRB1, ADRB2, ACE, ADRB3, ALAD, CAPN10, CAT, CBS, CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2C19, CYP2C9, CYP2E1, CYP3A4, MTHFR, MTRR, NOS2A, NOS3, NQO1, PPARG, SERPINE1, TNF, VDR TNF, Chang M, et al. Am J Epidemiol (in press) Genotype-Phenotype Studies “Genes, Behavior, and the Social Environment: Moving Beyond the Nature/Nurture Debate” “Recent knowledge is pushing Recent scientists to look beyond single agents of health and disease. By breaking out of their disciplinary “silos” and embracing a broader systems view, based on the understanding that health outcomes are the result of multiple determinants—social, behavioral, and genetic--that work in concert through complex interactions, the best health outcomes from research may be yet to come” yet IOM Report August 2006 Outline Genomics 2011: the excitement of scientific Genomics discovery discovery Genomics and health 2011: characterizing Genomics genetic influences on health and disease through gene-environment interaction Genomics and predictive health 2011: How do Genomics we actually use genetic information to improve health and prevent diseases? So What Do You Do With Genes When You Find Them? McCarthy et al. Nat Rev Genet 2008 Gene-Based Medicine in 2010? (now revised to 2020) Condition Genes Genes Prostate Ca Alzheimer’s Alzheimer’s Heart disease Colon Cancer Lung Cancer HPC1, 2, 3 RR Lifetime 0.5 APOE,FAD3,XAD 0.3 APOE,FAD3,XAD APOB,CETP 2.5 FCC4,APC 4.0 NAT2 6.0 Collins FC, New Engl J Med 1999;341:28-37. Collins 7% 10% 70% 23% 40% Gene-Based Medicine in 2010? Prevention Strategies Based on Genetic information? Increased Risk for Prevention Strategies Heart disease Colon Cancer Lung Cancer Tertiary: Cholesterol drugs + Lifestyle changes changes Secondary: Increased Increased surveillance for early detection detection Primary: Behavior Behavior modification for smoking cessation smoking Genetic Tests as a Public Health Issue Case Study 1: Prostate Cancer Susceptibility Testing 48 year old white male in good health, father diagnosed with localized prostate cancer at age 68 Concerned, he got tested using deCODE Prostate Cancer Genetic Test: Relative risk = 1.88 High risk prompted early PSA test by primary care PSA – high normal at 2.0ng/ml High risk prompted urologist to perform TRUS­guided biopsy Positive ­Gleason score of 6 Radical prostatectomy with nerve sparing Case Study 2: Dr Oz “Dr. Oz found out he's Dr. 30 percent less likely than the average man is of developing prostate cancer. Which means, he can be a little less diligent about scheduling regular prostate examinations. "Think of the trade-off," he says. "Thanks to this test, I don't have to have rectal exams have Selected Loci Associated with Prostate Cancer Region p­value Risk Allele Freq. 8q24 (loc1) 6.7 10­16 0.1 1.49 (1.34­1.64) 1.83 (1.32­2.53) 10q11 8.7 10­14 0.38 1.20 (1.10­1.31) 1.61 (1.42­1.81) 8q24 (loc2) 4.7 10­13 0.50 1.13 (1.02­1.26) 1.46 (1.30­1.64) 17q21 1.5 10­10 0.52 1.25 (1.13­1.34) 1.47 (1.31­1.65) 11q13 4.1 10­10 0.50 1.18 (1.08­1.28) 1.48 (1.27­1.74) 10q26 1.7 10­7 0.25 1.14 (0.94­1.38) 1.40 (1.16­1.69) 7p15 3.2 10­7 0.76 1.18 (1.07­1.31) 1.54 (1.37­1.73) Odds ratios Heterozygotes Homozygotes NCI CGEMS data, courtesy N Chatterjee, November So What is Going on Here? What do these odds ratios mean? Are they What reliable?(clinical validity) reliable?(clinical Are these numbers actionable? What do you Are do with this information? (clinical utility) do What would you tell individuals contemplating What such testing? such And what would you tell those already tested? Imagine this scenario repeated over multiple Imagine diseases in clinical practice? What is the net balance of benefits and harms on a population basis? basis? The Debate About Prostate Cancer Screening “A Model for PSA Screening Outcomes for Low to High Risk Men” K Howard et al. Arch Int Med 2009;169:1603 “Participation in screening considerably increases the likelihood of having prostate cancer diagnosed; Yet, few of these men die of prostate cancer, and death rates are similar in screened and unscreened men.” “The present estimates provide a sobering illustration of the frequency of harms men are likely to experience if they participate in PSA screening. The risk of having a false alarm rises strongly with age and with increasing familial risk.” Multidisciplinary Evaluation of Genomic Multidisciplinary Information for Improving Health Information Each intended use ACCE Framework Four components • Analytic Validity • Clinical Validity • Clinical Utility • ELSI Analytic Validity Defines the ability of a test to accurately and reliably identify or measure the analyte or mutation of interest Multidisciplinary Evaluation of Genomic Multidisciplinary Information for Improving Health Information Each intended use ACCE Framework Four components • Analytic Validity • Clinical Validity • Clinical Utility • ELSI Clinical validity Defines the ability of a test to detect or predict the phenotype or particular clinical outcome Multidisciplinary Evaluation of Genomic Multidisciplinary Information for Improving Health Information Each intended use ACCE Framework Four components • Defining the risks and benefits associated with introduction into practice Likelihood of improved health outcome Clinical Validity • Clinical Utility Analytic Validity • Clinical Utility • ELSI Association vs. Classification: Relation Between Genetic Associations and Clinical Validity of Testing for Genetic Risk Factors AUC Analysis Pepe et al. Am J Epidemiol 2004;159:882 Genetic Associations: Beyond Kraft P et al. Nat Rev Genetics 2009 Odds Ratios Performance of Common Genetic Variants in Breast-Cancer Risk Models (Wacholder S et al. NEJM March 2010) Genomic Profiling to Assess Risk for Cardiovascular Health Genes included on genomic tests for CVD / “heart health” (June 2008) Genes included on the CVD panel ACE AGT AGTR1 APOB ACPOC3 APOE CB S CETP CY B A CYP11B2 NOS3 Factor II Factor V . , , SELE SOD3 TNF-α 9p21 Testing Laboratory Total of 29 genes (some with multiple variants) Outcomes of HD & stroke) Epidemiologic Evaluation (summary) Heart Disease Data for 28 genes ORs for 38 gene/variants Credibility Strong = 1 (9p21) Moderate = 13 Weak = 24 (4 are C,C,-) The largest, most The credible effect for each gene Stroke ↑ OR = 1.51 (SELE, weak) OR ↑ OR = 1.26 (9p21, strong) OR ↓ OR = 0.79 (9p21, strong) OR Data for 19 genes ORs for 20 gene/variants Credibility Strong = 0 Moderate = 10 Weak = 10 (6 are C,C,-) The largest, most credible The effect for each gene ↑ OR = 1.33 (F5, moderate) OR ↑ OR = 1.33 (LPL, moderate) OR ↓ OR = 0.92 (PON1, OR moderate) moderate) Theoretical Distribution of Genes in Persons With and Without Disease Value-Added of Genome Score for Type 2 Diabetes? A B C D Meigs et al N Engl J Med 2008;359:2208-19 Clinical Utility of Genomic Information? D. Altshuler, IOM workshop: Evidence-based Medicine and the changing nature of healthcare 2008; 89-90 Data from Diabetes Prevention Program (DPP) RCT results stratified by genotype Clinical Utility of Genomic Information? D. Altshuler, IOM workshop: Evidence-based Medicine and the changing nature of healthcare 2008; 89-90 “In my view, this is a much more uncertain enterprise, and if we are not careful, we may never know its real value, because it will become part of routine healthcare before we actually know whether it actually helps to improve people’s lives”. “an innovative and powerful genetic test that assesses your inherent risk of developing lung cancer, whether you are a current or ”Biomedical Risk Assessment as an Aid for Smoking Cessation?” A strategy for increasing strategy smoking cessation rates could be to provide smokers with feedback on the biomedical or potential future effects of smoking, Risk assessment includes Risk measurement of exhaled carbon monoxide (CO), lung function, and genetic susceptibility to lung cancer. Review of 8 clinical trials “Due to the scarcity of Due evidence of sufficient quality, we can make no definitive statements about the effectiveness of biomedical risk assessment as an aid for smoking cessation” smoking Bize et al. Cochrane Review 2008 Potential Benefits and Harms of Personal Genomics: Point-Counterpoint (Euro J Clin Invest, Jan 2010) Bottom Line: We need data!!!! “The absence of evidence is not evidence of absence” Is Evidence-based Medicine the Enemy of Personalized Medicine? “My clinical experience, supported by pharmacological My literature, indicates that CYP testing may benefit some patients using some psychiatric drugs (not SSRIs). SSRIs). Some subjects (less than one in a thousand) lack two CYPs that metabolize most antidepressants. After CYPs that identification, they can be correctly treated by relying on current pharmacological knowledge. Evidence-based current based medicine focuses on average patients, whereas personalized medicine focuses on unusual subjects” personalized J De Leon, Science, August 8, 2008 De Take Home Messages Genomics 2011: the excitement of scientific Genomics discovery continues discovery Genomics and health 2011: We need to Genomics characterize genetic influences on health and disease through gene-environment interaction Genomics and predictive health 2011: We need Genomics to rigorously study how to use genetic information to improve health and prevent diseases ...
View Full Document

Ask a homework question - tutors are online