Objectives of Pharmacogenetics

Transcription

1 Instructors Introduction to Pharmacogenomics Russ B. Altman, MD, PhD Caroline Thorn, PhD Russ B. Altman, MD, PhD Professor of Genetics, Bioengineering & Medicine Stanford University Define Pharmacogenetics The role of genetics in drug responses. F. Vogel, 1959 CP

2 Objectives of Pharmacogenetics 1. Identify variation in response 2. Elucidate molecular mechanisms 3. Evaluate clinical significance 4. Develop screening tests 5. Individualize drug therapy pharmgkb.org CP Goals of course Learning Objectives I Understand informatics challenges in pharmacogenetics & pharmacogenomics Understand the basic sciences behind pharmacogenomics: pharmacology & genomics Understand how bioinformatics intersects with clinical informatics 1. Understand basic pharmacology concepts: drug discovery, pharmacokinetics, pharmacodynamics, adverse events 2. Understand role of genetics in response to drugs 3. Define pharmacogenomics vs. pharmacogenetics 2

3 Learning Objectives II 4. Understand categories of phenotypic information 5. Learn classical examples of pharmacogenetics 6. Learn current applications of PGx 7. Understand social, ethical, legal issues associated with PGx 8. Learn how to use PharmGKB ( Complexity of Relationships in Pharmacogenetics Genomic Information Coding relationship Molecules Pharmacologic activities Drug Response Systems Variations in genome Role in organism Isolated functional measures Protein products Molecular & Cellular Phenotype Observable phenotypes Alleles Molecular variations Drugs Integrated functional measures Genetic makeup Treatment protocols Observable phenotypes Clinical Phenotype Physiology Individuals Non-genetic factors Environment Omics X-omics = study the entire collection of X Genomics = study all genes Proteomics = study all proteins Pharmacogenomics = study all pharmacogenes What is a pharmacogene?! Pharmacogene Any gene involved in the response to a drug Thus, Pharmacogenomics = study of all genes involved in the response to a drug. 25,000 genes??? Unknown number of pharmacogenes 3

4 Human Genome Sequence completed in ,000,000,000 bases of DNA Divided into 23 chromosomes In females, all diploid (two copies) In males, X and Y are haploid (one copy each) 25,000 genes 99.8% of genome identical in humans What was promise of genome? New prognostic tools (predicting things that may happen) New diagnostic tools (identifying things that are happening) New therapeutic tools (creating things to affect what is happening) ONLY 78 POSITIONS (shown here) VARY IN 25,000! Pharmacogenomics relates to therapeutics, sometimes called personalized medicine. 4

5 Genetics Bimodal Distribution in drug response Simple Mendelian Genetics Categorical traits (e.g. presence/absence of a disease) associated with a single gene Two alleles A, a (allele = version of gene), one from mom, one from dad AA, aa = homozygous diploid Aa = heterozygous diploid IF Dominant Allele A: AA = Aa (same phenotype) IF Recessive Allele a: aa = (alternate phenotype) Frequency e.g. AA & Aa Single factor Large effect e.g. aa CP More genetics Unimodal Distribution in drug response Quantitative trait = not categorical, but a continuous variable (e.g. height) Complex trait = depends on more than one gene (quantitative traits usually complex) Example: 4 genes with two alleles each: Aa, Bb, Cc, Dd Many possibilities: ABCD, abcd, abcd, abcd, abcd, Abcd, ABcd, ABCd, etc Each possibility may be associated with a different quantitative trait value Frequency Many factors Small effects CP

6 Brief History of PGx Genetics rises in late 1800 s early 1900 s with interest in agriculture/husbandry Initial analysis of medical applications in the 1950 s Mendelian simple trait analysis of families Definition of dominant/recessive/x-linked patterns of inheritance in human disease Recessive genes causing disease = rare disease. History of PGx 1970s: analysis of clinical phenotypes related to known genes E.g. hemoglobinopathies leading to sickle cell anemia, thallesemia, clotting disorders Leads to an interest in finding genetic causes of other diseases that show familial heredity patterns s/60 s classic PGx examples Glucose-6-phosphate-dehydrogenase Severe anemia in some African-Americans upon taking primaquine, later found in 400 million Africans. Isoniazid (INH) for tuberculosis Slow and rapid metabolizers, related to N-acetyltransferase variants (more than 100) Unusually long anesetheisa with succinylcholine Prolonged effects due to atypical cholinesterase 1960 s s: CYP2D6 Discovery that response to debrisoquine (blood pressure medicine) and opioids (pain medications) was related to the level of cytochrome p450 activity Cytochrome 450 found in all animal kingdom, involved in metabolism of foreign substances Later found to be CYP2D6, with many (> 80) polymorphisms in human population 6

7 1970 s/80 s: Rise in use of genetic linkage analysis If two genes (or genetic features) are (1) close to one another on a chromosome, and (2) each have at least two different alleles, then It is likely that a disease allele for one gene would be inherited together with a particular allele of the other gene. If the other gene has an easy to measure phenotype, then the presence of the disease allele could be inferred from the presence/absence of this phenotype [usually checked in parents/children]. P p P p Recombination Q q q Q Exchange strands of DNA to make two new hybrid DNAs. Recombination Linkage Example P p q Q P p Chance of recombining increases with distance between genes (or genetic markers) q Q Mutate A to get a AB a B A/a = easily measured phenotypes for each allele B = normal allele b = disease allele P p q Q Mutate B to get b a b b is always associated with a, while B is only sometimes. 7

8 But which genes/markers are easy to phenotype? Initially, easy to measure traits like: eye color, blood group 1970 s: Restriction fragment length polymorphisms (RFLP) provided 100 s of markers throughout genome. 2000s: Genome sequence provides millions of markers These provided the easy to measure phenotypes that allowed genes to be associated with linked genetic regions (loci). Enzyme cuts GGGAGG But does not cut GGGGGG RFLP ACATAGAGGGAGGAGACCAGA ACATAGAGGG + AGGAGACCAGA ACATAGAGGGGGGAGACCAGA ACATAGAGGGGGGAGACCAGA Thus, easy phenotype = 2 fragments vs. 1 fragment. Human Genome The sequencing of the full human genome provides millions of potential markers for these types of studies. (We will review these later today) Routine application of PGx? Pharmacogenetics is NOT routinely used clinically. Why? G6PD tests available and cost effective, but not used Other considerations to be addressed later in course. 8

9 SCIENCE Oct 24, 2003 PGx in Developing World G6PD Some anemia is self-limited, other versions profound and life threatening Enzyme based tests exist and evaluated Still not clearly cost-effective overall PGx in Developing World ABCB1 (transporter molecule with variations) Higher variation in West Africa May be due to different environmental exposures Affects levels of HIV drugs PGx in Developing World Malaria Need cheaper and more effective tests DNA tests are emerging that predict organism resistance to drugs Testing of metabolism enzymes may be important for evaluating new drugs in proper populations. 9

10 Conclusions Pharmacogenomics still a relatively new field. Human genome provides full information to catalog human variation, gene locations, marker locations Initial pharmacogenetic examples are simple, with one gene-one drug. Future lies in complex drug responses involving many genes and quantitative response. Pharmacology: Basic Principles Russ B. Altman, MD, PhD Professor of Genetics, Bioengineering & Medicine Stanford University Different types of drugs Lipinski s Rules Oral (Most common) Pills & Coated pills (drugs in solid base) Capsules (powder in digestible container) Intravenous Mostly for compounds that can not tolerate digestive system (e.g. Vancomycin) Intramuscular For slow release and uptake (e.g. progesterone) Subcutaneous For quick absorption (e.g. insulin) Christopher Lipinski created rules to predict which drugs would fail because of poor pharmacokinetics. Molecular mass > 500 Da High lipophilicity (hydrophilic) More than 5 hydrogen bond donors More than 10 hydrogen bond acceptors 10

11 Two key concepts Pharmacokinetics: how a drug gets in and out of the body. Kinetics = movement of the drug. What the body does to drug. Pharmacodynamics: how the drug acts on the body to produce good effects (as well as bad effects). What the drug does to the body. Drug in Tissues of Distribution Distribution Drug Administered Drug Concentration in Systemic Circulation Drug Concentration at Site of Action Pharmacologic Pharmacodynamics Effect Clinical Response Absorption Pharmacokinetics Toxicity Elimination Pharmacodynamics Efficacy Drug Metabolized or Excreted Giacomini, 1999 Pharmacokinetics Pharmacodynamics Pharmacokinetics: ADME Absorption--passive, active, or cotransport? Distribution--fat or water soluble? Metabolism Phase I oxidation/reduction Phase II conjugation (methylation, sulfation, acetylation, glucuronidation) Elimination--renal excretion to urine, liver excretion to bowel Pharmacodynamics: target Target = the place in the cell that the drug targets for its effect Most common target classes: G-protein coupled receptors Large family of receptors (1-3% of genome) that detect signal, and start intracellular cascade in response. 160 receptors with known ligand 200+ receptors with unknown ligands = OPPORTUNITY 11

12 Top Drug Targets (1997) GPCR Diversity G-protein coupled receptors Enzymes (non-protease) Ion channels Non-human ribosomes Biotherapeutics Proteases Symporters Pumps Types of interactions with target Competitive inhibition Competes with natural compound to bind target Noncompetitive inhibition Binds target independently from natural compounds Activation Competitive Noncompetitive Measuring pharmacodynamics Can measure the action of a drug in many ways: Molecular assays (e.g. drug binding, target function) Cellular assays (cellular changes, e.g. expression) Organ-based assays (e.g. liver cell response) Organism-based (e.g. clinical measures, e.g. desired effect of drug) 12

13 Pharmacogenes Now, we can define pharmacogenes more precisely: Genes that are involved in either the PK or PD of a drug = pharmacogenes Defining P-etics vs. P-omics Pharmacogenetics = study of individual gene-drug interactions, usually the gene that has the dominant effect on a drug response. (SIMPLE relationship) Pharmacogenomics = study of the full set of PK/PD genes, often using highthroughput data (sequencing, expression, proteomics) (COMPLEX interactions) Informatics Challenge in PGx Standards for data representation & exchange Genotype Phenotype High throughput data Analysis of literature Find pgx papers Extract gene-drug interactions 13

14 Informatics challenges II Integration of data from multiple sources Gene sequence, protein sequence SNP databases Drug info databases Representation and manipulation of pathways PK pathways PD pathways Informatics challenges III Standard terminologies for PGx Genes Drugs Diseases Symptoms (Adverse events) Machine learning for associating genotypes to phenotypes Informatics challenges IV Intro to PharmGKB Creating databases that have adequate security and privacy mechanisms Information systems for point of care delivery Russ B. Altman, MD, PhD Principal Investigator 14

15 NIH relevant mission Building a premier web-based knowledge base (PharmGKB) that rapidly disseminates accurate and detailed definitions of genotypes and phenotypes in pharmacogenetics, along with tools and resources. PharmGKB Mission 1. To develop, implement, and disseminate a public genotype-phenotype resource focused on pharmacogenetics and pharmacogenomics. In the short term, this resource will facilitate basic research. In the long term, it will impact how medicine is delivered. This resource will serve a broad community including geneticists, molecular biologists, pharmacologists, physicians, policy makers and the lay public. 2. To perform high quality informatics research in support of this goal, and to catalyze scientific discovery in both pharmacogenetics/pharmacogenomics and biomedical informatics. Make core contents clear 15

16 Stress PGx Flow & Mission DEMO Homepage Gene = ABCB1 Drug = Irinotecan Pathway = Irinotecan pathway VIP Gene = HMGCR Whole Genotype Data Who visits site? From where? Now at 40,000 visitors (unique IP address)/month Comparison 400,000/month at Protein Data Base 400,000+/day at Genbank 60+% gene oriented (from NCBI) 30% typing in drug names on site Miscellaneous 16

17 User profiles Where users come from? 17% 22%.COM OTHER 6%.ORG.GOV 3%.EDU 1%.com.net.edu.gov.org Other NCBI DrugBank resource from U. Alberta NIGMS Pharmacogenomics site Genecards resource Oakridge National Lab Human Genome info site 51% US > UK > Japan > Canada > Netherlands > Australia What do users type in Google? What are they doing? pharmgkb % pharmacogenetics % pharmacogenomics % bronchial spasm % glucocorticoids % methotrexate % bronchial spasms % 6-mercaptopurine % nicotine metabolism % irinotecan % cerebrovascular accident % mercaptopurine % isoproterenol Stanford 198 South Africa 0.1 Biomedical % Informatics Program xenobiotics % (5 clicks/visit) #1 Search #2 Browse genes #3 Browse drugs Specifically Looking at SNPs in pharmacogenes Looking for known PK/PD genes based on our curation for individual drugs Looking at pathways for candidate genes 17

18 Where data comes from? Genotypes - 90+% from PGRN Phenotypes - 80+% from PGRN Pathways - 90% from PGRN Literature annotations - 80% PharmGKB Curators PharmGKB Advisory Committee March 6, 2006 Complementary Efforts to PharmGKB Genotype Data dbsnp, jsnp (includes SNP consortium) NIEHS Environmental Genome SNPs Hapmap Human Genome Mutation Database HGVBase Locus specific databases (LSDBs) GeneCards PHARMGKB FOCUSES ON PHARMACOGENES AND HAS GENERALLY DENSER HIGHER QUALITY Pathway Data MetaCYC KEGG Reactome Biocarta And many others (including pathway display programs cytoscape, genmapp) PHARMGKB FOCUSES ON PK PATHWAYS (also PD PATHWAYS) Popular pathways: VEGF > ACE > Nicotine 18

19 KEGG Drugs DrugBank PubChem MedlinePlus RxNorm Drug data PHARMGKB AGGREGATES THESE Phenotype data All major NIH-sponsored trials with public data dissemination mission New NCBI dbpheno (e.g. GAIN, GWAS) PhenoFocus Model organism phenotype efforts (mouse, zebrafish, fly, yeast, etc ) PHARMGKB FOCUSES ON DRUG-RELATED PHENOTYPES, AGGREGATES/ANNOTATES/INTEGRATES. VIP Variant Pages No other resources annotating individual polymorphisms for pharmacogenomic significance. OMIM Thanks. russ.altman@stanford.edu PHARMGKB & PGRN LEADING IN ANNOTATING VARIANTS OF PHENOTYPIC IMPORTANCE 19

20 The typical clinical flow Clinical setting of pharmacogenomics Russ B. Altman, MD, PhD Professor of Genetics, Bioengineering & Medicine Stanford University In order to understand the opportunity for pharmacogenomics, one must understand the clinical setting and the flow of activities for physician-prescribers. Doctors learn these classes, and typically memorize information and use one or a few members of each class in practice. Drug classes Making the prescribing decision Most physicians think about drugs as grouped together by the disease that they treat, e.g. Antibiotics--treat bacterial infections Antivirals--treat viral infections Cholesterol medicines--lower bad cholesterol to avoid heart attacks Blood pressure medicines--lower blood pressure to avoid strokes & heart attacks Diabetes medicines--lower blood sugar And many others Patient comes to clinic with a problem. Physician takes history, examines patient, orders tests, and attempts to make a diagnosis. Physician may then evaluate the options for treatment with medications 20

21 What does doctor consider? Efficacy of drug for this condition: is a drug likely to help the patient? Has it worked in the past? Side effects of the drug: is there a high rate of adverse reactions to the drug? Has this patient taken it before with problems or success? What does doctor consider? Compliance with drug dosing: Is the patient likely to take the drug? 10% to 90% noncompliance to all drug prescriptions. Noncompliance can be due to: Cost (expensive?) Perceived benefit by patient (symptom relief vs. not, e.g. hypertension meds) Dosing complexity (1/day vs. 4/day) Side effects (tolerable or not) What does doctor consider? Patient lifestyle: special elements of patient environment (based on their job, e.g.), use of over-the-counter medications, use of herbal medications, unusual diet, travel. Interactions with other drugs or diseases relevant to this patient Doctor makes a decision Gives patient a prescription for a drug Patient fills the prescription? Patient takes the medication on correct schedule, with/without food? Patient takes the medication for the time prescribed? Patient has special genetic background that changes expected response? 21

22 Key pharmacogenomics assumptions: Assuming that the (1) patient fills the prescription, (2) follows the dosing recommendations, (3) does not have an unusual diet, then Variation in response to the drug may be due to genetic factors (PD and/or PK). What is opportunity? High individual risk prescriptions, e.g. Serious infections: HIV, malaria, TB all potentially life threatening Cancer: because of resistance developing, would like to get prescription right on the first try. Depression: failure to adequately treat can be deadly (suicide) Major pain: major suffering of patient must be avoided What s the opportunity? Low individual risk, common disease, e.g. Diabetes, increases long term risks for cardiovascular disease, kidney disease, blindness High blood pressure, increases risks for stroke and cardiovascular disease Minor pain, can be disabling over long term What could we test for PGx? Genotype (ONLY DO ONCE) The whole genome? (Too expensive currently) Focused genetic tests Phenotype (MAY NEED MULTIPLE) Molecular Cellular Clinical NOTE: downstream phenotype measurements may be more accurate, since more functional 22

23 EXAMPLE: TPMT testing TPMT is an enzyme that metabolizes a drug called 6-mercaptopurine (6-mp). Variation in TPMT can lead to increased 6-mp side effects. Can measure the genotype of TPMT Can measure the activity of TPMT enzyme in red blood cells. More TPMT, less TGN Genotyping issues When to genotype? At birth (or establishment of care with doctor)? At time of prescribing? Where should genotype be stored? Measure and throw away Store in medical record Store in centralized database Weinshilboum (Mayo Clinic)

24 Where should we store genotypes? How do we deliver genotype information? Give to patient Store in local healthcare facility Store in central healthcare database With the government With a trusted third party Other? These choices have implications for privacy/security and patient comfort with information. Doctor needs genotype information to inform prescription decision in a timely manner. Too late = useless. Fax Phone Mail How do we explain the results to an MD who does not understand recommendation? Who pays for genetic tests? Individual Health care system Government Does this affect who has the right to see the information? Drug discovery and validation Russ B. Altman, MD, PhD Professor of Genetics, Bioengineering & Medicine Stanford University 24

25 How have we discovered drugs? Average time from project inception to drug launch: years Average total investment per LAUNCHED drug = $1 billion Average chance of project success: 1-3% at inception 7-8% if drug reaches preclinical testing 1. Basic science Generate hypotheses about potential drug targets based on basic research. E.g. A studied gene is mutated in some HIV-infected patients who never progress to AIDS. Can I develop a drug that mimics this mutation in other people, so that they also will not progress? 1. Basic science OR E.g. I have elucidated a series of genes involved in the development of cancer. Can I interrupt the development by blocking one (or more) of the genes? 2. Identify a lead compound Given the target, attempt to find a compound that binds it (binding assay) or interferes with its function (functional assay) Usually identified through screening: Using the target, develop an (ideally inexpensive) assay for binding/function Create (or purchase) a large library of compounds Test them in the assay, pull out the positives for further study. 25

26 2. Identify a lead compound Usually, the lead compound(s) will not be ideal drug candidates Do not fit Lipinski rules High chance of toxicity (or demonstrated in animal studies) Does not have desired effect Myriad other problems. Lipinski s Rules Christopher Lipinski created rules to predict which drugs would fail because of poor pharmacokinetics. Molecular mass > 500 Da High lipophilicity More than 5 hydrogen bond donors More than 10 hydrogen bond acceptors 3. Optimize the lead Organic chemists create variations of lead (using Lipinski rules, e.g.) to eliminate problems. Can use combinatorial chemistry in which many variations of a backbone molecule are generated by systematically adding/removing different chemical groups Develop more focused assays Test the desired characteristics more accurately Can be more expensive, since not used for screening 4. Test optimized leads in animals [NOTE: Rats are not just small humans ] Nevertheless, must establish safety in animals (mice, rats, pigs, dogs, etc ) Check for metabolism of drug Check for toxicity, adverse reactions Perhaps, check for signs of efficacy. Get indication of dosage ranges (mg/kg) 26

27 5. Phase I clinical trial < 100 healthy people (usually paid) Start low dose, increase Check safety Liver, kidney blood tests Other, as indicated Evaluate pharmacokinetics (blood levels as a function of dose) Establish maximum tolerated dose (from below!) In parallel, work on formulation (purity, reproducibility) 6. Phase II clinical trial < 1000 patients with disease Continue to evaluate safety Establish optimal dosing Preliminary test of efficacy 7. Phase III clinical trial < 10,000 patients with disease Use formal statistical hypothesis test to evaluate the efficacy and safety of new drug compared to current best Needs to at least match current best Fully documented trial data submitted to the government agency that authorizes marketing of drug. 8. Phase IV clinical study Post-marketing surveillance After the drug is released, company must continue to monitor for safety. Especially important for rare (< 1/10,000) side effects 27

28 Phase IV withdrawals VERY expensive to pull drug this late: Chloramphenicol--antibiotic with rare bone marrow failure Grepafloxacin--antibiotic causes increased cardiac arrhythmia Vioxx -- arthritis medicine with increased rate of heart attacks Troglitazone--diabetes medicine with rare liver failure Viagra--erectile dysfunction medicine with rare heart attacks (NOT WITHDRAWN, TOO POPULAR?) Notes on drug development Cost of canceling a drug project increases exponentially as it progresses through steps. Thus, better to cancel a project early with any indication of problems, than to hope it all works out. These decisions currently made based on incomplete information. Valuable drugs may be cancelled that could be saved. Notes on drug development Drug companies are generally looking for reasons to cancel a drug, and the pipeline of targets is generally thought to be adequate. Adverse events that are 1/10,000 are not seen until post-market, and are therefore very expensive. More commmon adverse events (1/100-1/1000) will lead to cancellation in phase I or II. What about pharmacogenomics to save these? Can PGx save drugs? In principle, YES, but issues: Need pharmacogenomic information early in development, so studies can be focused: Choose subset of patients who will tolerate drug in phase I studies. Avoid lengthy additional studies (patents = last only 17 years) May need to co-develop a genetic test (e.g. Herceptin) 28

29 Can PGx save drugs? Off patent drugs Companies prefer one size fits all drugs Unclear economic model for fractured markets with one size fits some Orphan drug regulations exist currently to make it attractive for companies to develop drugs for small populations E.g. life-saving drug for very rare disease Will orphan drug laws apply to fractured markets? After 17 years (in US) a drug goes off the patent, and other companies can begin producing it. Who is responsible for post-marketing surveillance then? Who should followup on pharmacogenomic opportunities? Cost/Benefit Concerns for PGx If cost of the test > cost of adverse reaction, then why do it? E.g. Codeine & CYP2D6, 7% of whites do not metabolize into active metabolite Cost of information systems to support PGx data storage and decision support Cost of industrial processes to create multiple drugs vs. one size fits all Ethical Issues Will pharmaceutical companies focus on particular genetic polymorphisms for drug development and ignore others? What if these polymorphisms are associated with groups that are more/less economically advantaged? More on this later 29