Background/Methods/Introduction
Human Genome Project
Genetic Components of Disease
Genetic Maps
International Hap/Map Project
Other Approaches
Proteomics
Metabonomics
Pharmacogenetics/Pharmacogenomics
Research and Development
Treatment and Clinical Practice
Pharmacogenetics of Drug Metabolism
Genetic Variation
Pharmacogenetics: Drug Receptors and Membrane Transporters
Pharmacogenomic/Genetic Tests
Ethical, Legal, and Social Implications
Disparities in Treatment
Conclusion and Recommendations (Adopted AMA Directives)
References
Note: This report, written in response to Resolution 422 (A-05), was presented as CSAPH Report 4 at the 2006 AMA Annual Meeting; it represents the medical/scientific literature on this subject as of June 2006.
The clinical research enterprise, technologies, and data required to implement genetically based preventive medicine on a personal basis are not yet available; however, some pharmacogenetic elements are sufficiently developed to inform clinical decision-making. Social, ethical, legal, and regulatory issues and other barriers confront the development and implementation of genetically based, personalized health care. In the United States, pharmacogenomics and personalized medicine are being introduced into a fragmented health care system confronted by significant disparities in health status, health care delivery, and access. Additionally, a significant percentage of the population has low health literacy and a poor understanding (or fear) of science and technology.
This narrative review provides an overview of the current scientific status of genomics, pharmacogenomics, and other measures (eg, proteomics, metabonomics [or metabolomics]) as applied to clinical medicine. The report also briefly notes some ethical, social, and regulatory issues that have been raised. The report does not review or evaluate specific technologies that are available for sequencing DNA, identifying mutations or variants, analyzing gene expression, or the bioinformatics platforms used to capture and analyze genomic, proteomic, or metabonomic data. The report also does not discuss the potential medical liability implications of the genomics revolution.
Methods
Literature searches were conducted in the PUBMED database for English-language articles published between 2000 and February 2006 using the search term personalized medicine or pharmacogenomics, in combination with drug and metabolism, receptor, or transporter. In addition, the Web sites of the Human Genome Project, Pharmacogenomics Research Network, SNP Consortium, International HapMap Project, and the National Cancer Institute were consulted for relevant information.
Introduction
Individualization of therapy is not a new concept in medicine, but rather one that has been central to its practice since the writings of Hippocrates.1 The mapping of the human genome, coupled with the abilities to measure the function of genes and their expressed products (proteins), offers the possibility that a patient’s genome can be analyzed in order to choose appropriate diagnostic and therapeutic strategies, including preventive measures.2
The Human Genome Project and related activities (http://doegenomes.org/) have created an expanding library of DNA sequence information, genes, and their variation. Currently, high per-patient cost does not permit the wide-scale use of genomic approaches in large clinical trial populations. Nevertheless, the efforts to establish population-based genetic maps and to develop technologies enabling capture of an individual’s genomic features, coupled with the use of sophisticated imaging and bioinformatics platforms, are being touted as the way to develop prevention strategies and treatment plans tailored uniquely to the individual patient.3,4
Human Genome Project. The principal reason for publicly funding genome research is the prospect of preventing, diagnosing, and treating diseases. The completion of the Human Genome Project has provided the reference framework.5-7 The human genome contains more than 3 billion chemical nucleotide bases (adenosine [A], cytosine [C], thymine [T], and guanine [G]), virtually all of which (99.9%) are the same in all people. Approximately 30,000 genes exist, but functions are unknown for more than half. Repetitive sequences that do not code for proteins (sometimes called "junk DNA") make up at least 50% of the human genome and may assist in reshaping it. Stretches of up to 30,000 C and G base repetitions often occur adjacent to gene-rich areas, forming a barrier between the genes and the "junk DNA." These CG islands may help regulate gene activity.
Progress in genetic knowledge and gene sequencing has renewed attention to the use of racial and ethnic classifications in health care. Genetic evidence does not support the concept that homogenous groups (races) can be distinguished by major biological differences. Although it is easily recognized that sufficient time has passed for selection pressure to work on some visible traits (eg, skin color, hair, and facial structures), these phenotypic differences reflect mostly differential selection by climate in various parts of the world.8 The categories of race and ethnicity are predominantly social and cultural constructs, rather than meaningful indicators of actual genetic difference.9
Any two randomly selected individuals will differ on average by approximately one DNA base change every 1000 to 1200 base pairs. Such single nucleotide polymorphisms (SNPs) account for the majority of variation between individuals. Genetic variations at a specific site are termed polymorphic if they occur at frequencies ≥1%. Each human racial or ethnic “group” contains more genetic variation within its own group than exists between groups. Because SNPs are inherited, some allelic variations in disease susceptibility and in drug responsiveness correlate with race and ethnicity (as commonly understood). These differences in response to drug therapy, however, are not due to “race” but rather to the distribution of polymorphic traits among population groups. Traits that are strongly associated with specific “races” stem from reproductive isolation. Because of geographic and social isolation of populations over hundreds or thousands or years, some patterns of genetic variation occur in individuals with a shared lineage comprising a socially defined subpopulation (eg, Tay-Sachs disease; sickle cell anemia trait). Although frequencies of drug-related polymorphisms vary among ethnic-related groups, no group is homogenous. Back to top
Currently, patients tend to encounter genetic testing primarily in the context of serious disease. In rare monogenic diseases, mutations in a single gene are both necessary and sufficient to produce the clinical phenotype and to cause the disease. The impact of the gene on genetic risk for the disease is the same in all families with the mutation. More than 1400 genes have been identified for approximately 1200 Mendelian disorders through positional cloning (www.geneclinics.org). The majority of current molecular diagnostic assays test for mutations in these types of disorders.
This approach is currently not feasible to establish genetic risks for complex multifactorial human diseases (eg, hypertension, diabetes, coronary artery disease, cancer). For these and other conditions, variations in a number of genes encoding different proteins predispose individuals to developing a clinical phenotype that is influenced by environmental and lifestyle factors.10 Many individual drug responses also are multifactorial traits.
Genetic Maps. Multiple kinds of genetic maps assist in localizing genetic components of disease phenotypes to specific chromosomal regions. A microsatellite map based on identification of tandem repeats of short nucleotide stretches can map traits or diseases that segregate in large families. These markers assist in scanning the entire human genome to map genes for specific diseases, phenotypes, and traits.
Another approach has been to identify SNPs. Through the efforts of the SNP Consortium (http://snp.cshl.org) and Human Genome Project’s analysis of clone overlap, nearly 1.5 million SNPs were identified and used to begin generating a high-density genetic map. The SNP Consortium used DNA resources from a pool of samples obtained from 24 individuals representing several racial groups. This is a subset of the DNA reference panel for SNP identification collected by the National Human Genome Research Institute. The anonymous, voluntary DNA contributions were obtained, with informed consent, specifically for this use.
The number of identified human SNPs has grown to more than 10 million; nearly 5 million of these have been validated, but fewer than 700,000 have assigned population frequency estimates.11 The SNP map provides a public resource for defining haplotype variation across the genome and should help researchers identify multiple genes associated with complex diseases such as cancer, diabetes, cardiovascular disease, and mental illness. It may assist in identifying novel targets for diagnostic and therapeutic interventions.
International HapMap Project. When sperm and egg cells are being formed, the chromosome pairs undergo a process known as recombination. The members of each chromosome pair come together and exchange pieces. The result is a hybrid chromosome containing pieces from both members of a chromosome pair, and this hybrid chromosome is passed on to the next generation. Over the course of many generations, segments of the ancestral chromosomes in an interbreeding population are shuffled through repeated recombination events. Some of the segments of the ancestral chromosomes occur as regions of DNA sequences that are shared by multiple individuals. Genetic variants that are near each other tend to be inherited together (ie, linked). These regions of linked variants are termed haplotypes. The SNPs are linked to regions coding for disease susceptibility, drug metabolizing enzymes, etc, so polymorphisms can be used to “map” disease genes.
Because humans are a relatively young species, most of the variation in any current human population comes from the variation present in the ancestral human population. As modern humans spread throughout the world, the frequency of haplotypes began to vary from region to region through random chance, natural selection, and other genetic mechanisms. As a result, a given haplotype can occur at different frequencies in different populations, especially when those populations are widely separated and unlikely to exchange much DNA through mating. Additionally, mutations create new haplotypes, which have not had enough time to spread widely beyond the population and geographic region in which they originated.
The International HapMap Project (www.hapmap.org) was designed to identify common haplotypes in four populations with African, Asian, and European ancestry from different parts of the world. It also identified so-called "tag" SNPs that uniquely identify these haplotypes.12 By analyzing an individual's tag SNPs, researchers will be able to identify the collection of haplotypes in a person's DNA. The goal is to utilize the haplotype map of the human genome sequence for future genetic association studies of human diseases as well as drug efficacy and safety.13 Another effort seeks to provide a contiguous map of allelic variation of the major histocompatability complex (MHC Haplotype Project; http://www.sanger.ac.uk/HGP/Chr6?MHC). Linkage scans and association-mapping studies have identified the MHC as influencing most, it not all, autoimmune conditions.14 Back to top
In addition to studying genetic sequences, identifying SNPs and haplotypes, and investigating their associations with disease risk, other analytical techniques beyond genomic structure and gene expression have emerged, including proteomics and metabonomics.
Proteomics. Proteomics is the study of the structure and function of proteins, including the way they work and interact with each other inside cells.15 Proteomics measures the quantitative and qualitative changes in cellular or tissue protein expression and explores protein-protein and protein-ligand interactions. Protein analysis in biological samples is complicated by the large number of relevant proteins whose cellular expression is dynamic and whose concentrations may be minute. Differential protein analysis compares the expression profiles of proteins (proteomes) of cells, tissues, or body fluids in one condition (eg, disease, injury, drug therapy, intoxication) to a standard--or normal--proteome. Functional proteomics concerns the manner in which proteins interact and how these interactions determine function. Technologies include the coupling of protein separation and enrichment to mass spectrometry; surface-enhanced laser desorption and ionization (SELDI) ProteinChip surface technology; automated, miniaturized 2Dgel electrophoresis; protein microarrays; and activity based probes.15,16 Potential clinical applications include biomarker or drug target discovery, monitoring for target organ toxicity, tumor profiling, and antibiotic sensitivity profiling.17-19
Metabonomics: Metabonomics (or metabolomics) is concerned with the analysis and measurement of endogenous metabolites. Originally defined20 as the “quantitative measurement of time-related multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modification,” metabonomics encompasses the application of nuclear magnetic resonance spectroscopy, high-pressure liquid chromatography, gas chromatography, and/or mass spectrometry analyses coupled with pattern recognition tools and multivariate statistical methods to evaluate endogenous metabolites in biofluids and tissues. This technology represents a potentially powerful method for determining the systemic response to toxicity or disease.
The ability to generate large and complex data sets requires sophisticated methods to reduce and analyze data (eg, principal component analysis; pattern recognition and predictive model development). Metabonomic data can be used to classify compounds that cause hepatic and/or kidney injury, to identify potential biomarkers of toxicity, and in the evaluation of metabolic profiles and disease severity. Potential clinical applications broadly include enhanced physiological monitoring, drug safety assessment, and disease diagnosis.21-24 Back to top
Pharmacogenetics/Pharmacogenomics
While genomics offers the prospect of better identifying risk factors for chronic diseases, its other dimension involves the prospect of genetically tailored treatments. Pharmacogenetics is the effect of genetic variation on individual drug response. Pharmacogenomics is concerned with applying genomics to the study of human variability in drug response and the development of new drugs. The origin of pharmacogenetics dates to the 1950s and 1960s.25,26 Subsequently, numerous genetic polymorphisms in various drug metabolizing enzymes, drug transporters, receptors, ion channels and other types of cellular targets affecting both drug efficacy and drug safety have been documented.27,28 Pharmacogenomics has implications for research and development, treatment and clinical practice, and public policy.
Research and Development. Variable human response is a barrier to the successful clinical use of many drugs (see below). It also is a barrier to new product development. During drug development, candidate compounds are tested in subjects who have the disease or disorder in question. Approval for marketing is based on the average responses in that population. For drugs approved under this paradigm, some patients will respond adequately, some will not, and some will experience adverse reactions that limit use of the drug. Average response rates will vary markedly depending on the therapeutic target (eg, hypertension vs. Alzheimer’s disease vs. ovarian cancer). Overall, perhaps only about 50% of patients respond adequately to drugs in major therapeutic classes.29 In fact, many current medications are associated with a significant risk of drug toxicity or treatment failure. Meta-analysis of prospective studies in the United States suggests that serious and fatal adverse drug reactions (ADRs) occur in ~7% and 0.3% of hospitalized patients, respectively.30 Drug-associated morbidity and mortality in ambulatory settings and nursing homes adds to the burden. Some differences in drug response profiles correlate with race and ethnicity. Pharmacogenomics provides the conceptual framework to identify host-specific genetic factors underlying common (or rare) medication side effects and therapeutic failures that otherwise are assigned the label “idiosyncratic.”
One key to increasing the success of research and development efforts is to identify product failures earlier in the drug development process and reduce attrition, which may occur later in pivotal Phase 3 clinical trials. Thus, it has been suggested that pharmacogenomics can identify new drug targets and aid in hypothesis generation; or be used to establish biomarkers of drug activity and mode of action.31 Valid biomarkers can potentially be used to enrich study populations or aid in patient stratification. By stratifying patients by biomarker status in early efficacy (Phase 2) clinical trials, populations with a high probability of response can be identified, thereby simplifying the design of Phase 3 trials and increasing their probability of success.
Alternatively, rather than screen subjects before investigative trials, samples can be obtained to establish “pharmacogenetic profiles” afterwards. Those who tolerated and responded to a drug can be compared with those with ADRs or therapeutic failure. The genetic profile then becomes either a predictor of drug efficacy or of an ADR. There also is the potential to “rescue” a promising therapy that may have been abandoned because of serious toxicity in a small fraction of susceptible patients, if the genetic link can be identified. To develop “personalized medicines” there must be a commitment to prospective pharmacogenomic testing in Phase 3 trials designed specifically for biomarker validation and development of companion genetic tests.
Current requirements by the Food and Drug Administration (FDA) to enroll sufficient numbers of women and minorities in clinical trials, and to carry out subgroup analysis based on sex, racial, or ethnic categories, are intended to facilitate the identification of variables affecting drug response. The FDA’s recent guidance on voluntary pharmacogenomic data submission may facilitate scientific progress and the use of pharmacogenomic data in drug development.32 When the technology is more fully developed, sponsors and researchers who are interested in recruiting individuals with a particular genotype should involve individuals from all ethnic groups, rather than using a convenience sample comprising individuals exclusively from the ethnic group with the highest frequency of the genotype of interest. Back to top
Treatment and Clinical Practice
Therapeutic and toxic responses generally correlate with plasma drug concentrations. Disposition processes (eg, drug absorption, distribution, metabolism, and excretion) determine the pharmacokinetic behavior of the drug in individual patients and the time course of the drug/plasma concentration profile.
Pharmacodynamics is the study of the biochemical and physiological effects of drugs and their mechanism of action, including the relationship between the dose (or concentration) of a drug and the effect it produces. While there are exceptions, the effects of most drugs result from their reversible association with a functional macromolecular component of the cell, termed a “receptor.” The receptor is that cellular component that interacts with a drug and initiates the chain of biochemical/physiological events leading to the effects of the drug.
A hallmark of drug disposition is large interindividual variability. Such variability is a major reason why patients differ in their responses to standard doses of a drug. A combination of genetic, environmental, and disease-state factors affect drug disposition, with the relative contribution of each, depending on the specific drug and disease. Genetic variations/mutations in genes that encode drug transporters, drug metabolism enzymes, and receptors are key factors in altering drug responses in some individuals. Back to top
Pharmacogenetics of Drug Metabolism
The pharmacogenetics of drug metabolism was first demonstrated in the 1950s with identification of polymorphisms in plasma pseudocholinesterase and N-acetyltransferase activity. Conceptually, human drug metabolism has two phases. Phase 1 reactions result in relatively minor chemical modifications of the parent compound. Generally, the metabolite is more polar or water soluble to facilitate renal excretion. These reactions may result in the formation of functional groups that serve as sites for phase 2 reactions, which are conjugation (synthetic) reactions in which the drug or metabolite is covalently linked with another molecule (eg, glucuronic acid, glutathione, sulfate, acetate, glycine, ribose, etc). Phase 2 metabolites also are more polar and less lipid soluble, and are preferred substrates for renal secretion transporters. With few exceptions (eg, morphine glucuronide; cancer antimetabolites), phase 2 metabolites are pharmacologically inactive.
An enzyme system (cytochrome P450) located in the smooth endoplasmic reticulum of the liver and several other organs, including the lung and gastrointestinal tract, mediates the oxidative metabolism of most drugs and environmental chemicals. Cytochrome P450 enzymes comprise a superfamily distributed across the living kingdom. About 270 gene families exist in various organisms. Nomenclature (http://www.imm.ki.se/CYPalleles) is based on the abbreviation CYP (for cytochrome P450), with a number designating the family (based on amino acid sequence homology), a letter designating the subfamily, and another number for the specific enzyme (eg, CYP3A4). The cytochrome P450 1, 2, and 3 families (CYP1, CYP2, CYP3) encode the enzymes involved in the majority of human drug metabolism reactions. Others are involved in the oxidation of endogenous compounds, or have roles in intermediary metabolism. Most CYPs involved in drug metabolism possess low substrate specificity; that is, they are capable of chemically converting an enormous range of lipid-soluble organic substrates.
The activity of a particular enzyme in vivo is determined by the inherent activity of the enzyme, its concentration, and the presence of agents that either inhibit enzyme activity or cause induction. Numerous disease factors, age, sex, nutritional status, adrenal and thyroid status, alcohol consumption, and smoking habits influence the average rate of hepatic drug metabolism.
Induction. The expression level of CYPs is not fixed. Many are responsive to environmental or drug- related influences, as are certain phase 2 enzymes (eg, GST, UGT).33,34 Induction occurs predominantly at the level of transcription mediated via nuclear receptors. In a few instances, elevation of enzyme level and enzyme activity occurs at post-transcriptional stages (ie, stabilization). Major inducers include persistent organic pollutants, cigarette smoke and polycyclic hydrocarbons, older antiepileptic drugs (barbiturates, phenytoin, carbamazepine), rifampin, and glucocorticoids.
Genetic Variation. A number of genetic polymorphisms exist for CYPs and phase 2 enzymes leading to altered drug metabolism phenotypes. Individuals who possess these polymorphisms are at risk of experiencing more adverse reactions or inefficacy at usual doses. Potential consequences of reduced or absent enzyme activity include an increase in the plasma concentration of the parent drug, reduction in metabolites, exaggerated and prolonged pharmacological effects, and an increased likelihood of drug toxicity. When gene duplication exists, consequences include an increased rate of metabolism, reduced bioavailability, and decreased plasma concentrations leading to therapeutic failure.
The gene encoding the CYP2D6 enzyme is on chromosome 22. Variants exist with reduced activity, reduced or absent levels, as well as gene duplication or amplification. Consequently, the phenotype can be “poor metabolizer” (PM), “extensive metabolizer” (EM), or fast or ultrafast metabolizer for the more than 100 drugs that have been shown to be preferred substrates of this enzyme.35 Some enzymes, notably CYP2C9 (which metabolizes warfarin) and CYP2C19, are encoded by genes that contain polymorphisms that alter structure and catalytic activity, rather than altering the level of enzyme expression.36 Comprehensive and current information on substrates, inhibitors, and inducers of CYP enzymes can be found at http://medicine.iupui.edu/flockhart/. See an abbreviated summary chart in the Table (PDF, 19 KB, requires Adobe® Reader®).
Thiopurine methyltransferase (TPMT) is a cytosolic enzyme whose normal physiological role is unclear. It catalyzes the S-methylation of the thiopurines, azathioprine, 6-mercaptopurine, and 6-thioguanine, using S-adenosylmethionine (SAM) as a methyl donor. TPMT is present in most tissues, including blood cells. The TPMT gene is located on chromosome 6 and contains 10 exons. Various alleles associated with different (average) ethnic expression have been identified.37 Numerous studies have shown that TPMT-deficient patients are at very high risk of developing severe hematopoietic toxicity if treated with conventional doses of cytotoxic thiopurines.38
Many other phase 1 enzymes (butylcholinesterase, epoxide hydrolase, alcohol and aldehyde dehydrogenase, monoamine oxidase, catalase, superoxide dismutase, and dihydropyrimidine dehydrogenases) and phase 2 transferases (glucuronyl, glutathione, N-acetyl, sulfo−) exhibit pharmacogenetic variation One study of 27 drugs frequently cited in ADR studies found that 59% are metabolized by one or more enzyme with a variant allele associated with deficient metabolism.39 DNA microarray-based tests are now available to evaluate patients for the presence of several different CYP polymorphisms. Back to top
Pharmacogenetics of Drug Receptors
Most receptors are either nuclear or cell surface varieties. Clinically relevant polymorphisms of nuclear receptors affect glucocorticoid, aldosterone, androgen, estrogen, vitamin D, and retinoic acid receptors.40 Steroid receptor polymorphisms are associated with either resistance or sensitivity to endogenous/exogenous hormones, hypersecretion, or increased suppression of hormone release.
Binding to cell surface receptors triggers the response to neurotransmitters, some protein and polypeptide hormones, autocoids, and some environmental chemicals. These receptors include ion channels; membrane transporters; receptors linked to “second messenger” cascades, and receptors whose action depends on an integral enzyme (eg, insulin). Mutations in several different ion channel receptors (K+, Na+, and Ca++) are linked with an increased risk of long QT syndromes. Other examples include mutations in the sulfonylurea receptor (familial hyperinsulinemia), insulin receptor (insulin resistance), beta-2 adrenergic receptor (bronchial hyperresponsiveness), CCR5 receptor (resistance to HIV infection), and the angiotensin type II receptor (hypertension).40
Pharmacogenetics of Membrane Transporters
A wide variety of transporters enhances the cellular uptake of drugs, or functions to export drugs from the intracellular to the extracellular compartment.41-43 Transporters, whether they mediate uptake or efflux, are localized in key organs involved in drug disposition (ie, intestine, liver, kidney, placenta), as well as the brain, and are therefore critical modulators of drug absorption, tissue distribution and elimination. Genetic heterogeneity in these transporters may contribute to interindividual and population variability in drug disposition.43
These transporters include the organic anion transporting polypeptide family (OATPs), which facilitates hepatocellular accumulation of drugs prior to metabolism, and efflux transporters, which mediate drug excretion into the bile.44 Members of the organic anion transporter family (OATs) have important functions in the kidney, where they directly transport small hydrophilic organic anions (eg, beta-lactams, NSAIDs, diuretics, many HIV antivirals, some anticancer drugs) into the tubular fluid. The organic cation transporter family (OCT) and a subfamily (novel organic cation transporter family; OCTN) are capable of transporting organic cations (eg, cimetidine, quinidine; cardiac glycosides, verapamil) in a number of epithelial tissues, including intestine, liver, and kidney.
The multidrug resistance P-glycoproteins (P-gps; MDRs) are a family of ATP-dependent transporters involved in the cellular efflux of numerous endogenous and exogenous compounds. P-gp, the gene product of MDR-1, has been investigated most thoroughly. It is found on the canalicular surface of hepatocytes, the apical surface of proximal tubular cells, the brush border of enterocytes, and the epithelium of the brain choroid plexus, as well as the luminal surface of blood capillaries in the brain, placenta, ovaries, testes, and lymphocytes. It acts in a protective manner by excreting toxins and limiting their accumulation in critical organs. P-gp also mediates the transport of several immunosuppressive drugs and a variety of agents used in the treatment of cancer, hypertension, allergies, infections, neurologic disorders, and inflammation.43 Genetic polymorphisms have been associated with certain disease risks as well as altered disposition of digoxin, fexofenadine, cyclosporine, tacrolimus, and HIV protease inhibitors.
Members of the multidrug resistance-related protein (MRP) family act as cellular efflux pumps for hydrophobic compounds conjugated with glutathione, sulfate, or glucuronic acid (major phase 2 metabolites), and nonconjugated compounds such vinblastine, HIV protease inhibitors, and methotrexate.
A comprehensive Web resource of transporter polymorphisms is maintained at http://www.vanderbilt.edu/. Back to top
Some question whether pharmacogenetic tests that predict altered drug handling or response should be regulated in the same way as genetic tests that predict risks of disease in that they do not specifically test the patient for the presence or absence of a disease-specific gene mutation, nor will they provide any other significant disease-specific predictive information about the patient or family members.45 At the very least, pharmacogenomic tests should demonstrate analytical and clinical validity, as well as clinical utility. When risk factors are sought, the genotypes to be detected by a genetic test must be shown by scientifically valid methods to be associated with the occurrence of disease. The observations must be independently replicated and subject to peer review (replication of findings).
The use of a pharmacogenomic test is more likely to be cost-effective when severe clinical/economic consequences can be avoided, current methods of monitoring are inadequate, there is an unequivocal association between genotype and clinical phenotype, the test is rapid and relatively inexpensive, and the test is applied to a relatively common genetic variant.
The primary aim is to provide the clinician with pharmacogenomic tests that can be applied at relatively low cost in order to predict efficacy and adverse effects. As of 2004, fewer than 80 drug package inserts contained pharmacogenomic data. Where this information was provided, the gene was usually a drug-metabolizing enzyme or the information was related to the variability in viral genomes as predictors of response to antiviral therapy or drug resistance. Information to guide treatment decisions was found in only 25 package inserts representing 22 drugs.46 Back to top
Ethical, Legal, and Social Implications
An Ethical, Legal and Social Implications (ELSI) working group was established by the program advisory committee on the human genome in 1989. The working group provided overall guidance to the National Human Genome Research Institute ELSI Program and formed two task forces aimed at analyzing and developing recommendations about genetic information and health insurance, and genetic testing. The report of the task force on genetic information and insurance was published in 1993; the report from
the task force on genetic testing was released in 1997.47,48 Subsequently, the ELSI Working Group responsibilities were divided among different committees and at various levels within the government.
Many of the ethical concerns raised by pharmacogenomics are shared by genetic research into disease predisposition and screening, and the acquisition and storage of DNA samples. These have recently been reviewed and evaluated.49 General areas of concern include informed consent; privacy and confidentiality concerns associated with genetic (DNA) information; allocation of resources; equity; and control over research materials and data. The information issue for individual research subjects gives rise to a slightly different issue for minority groups participating in research. In addition to the return of individually meaningful information, groups participating in genetic research may want the return of information learned about their groups. ELSI issues relevant to pharmacogenomics or the use of genetic information to foster development of “personalized medicine” are not further evaluated in this report.
Disparities in Treatment. Numerous studies demonstrate that for those able to gain access to the health care system, disparities in treatment based on race and ethnicity are profound, even when controlled for insurance coverage, education, and income. Several factors (lack of access, lack of trust, assumptions and stereotypes, lack of culturally and linguistically appropriate care, environmental factors) affect racial and ethnic disparities in health care. Although these are not the focus of this report, they are relevant to the future implementation of personalized health care.
Lack of access to drugs often leads to declines in health status. The poor, uninsured, and elderly have less access to drugs and rely on safety net providers for pharmaceuticals. Because a pharmacogenomically derived drug is essentially based on market segmentation, the effects on the economy and pharmaceutical pricing are difficult to predict. Some concerns exist that revenue from the sale of pharmacogenomics-based drugs will be less because the market size for any new therapy will be smaller. It also has been postulated that profit-conscious companies will use pharmacogenomics to direct their drug development efforts toward genetic subgroups of people who can best afford them, further marginalizing already underserved populations and creating “therapeutic orphans.”
As market size decreases, the total cost of drug development will be spread over a smaller number of users, which will in turn increase the average prices of pharmacogenomics-based drugs. Diminished market size effect may motivate larger pharmaceutical companies to focus their resources on the development of drugs to treat more prevalent genotypes at the expense of less common genotypes. This pessimistic view suggests that the promise of individualized treatment may never be fully realized by pharmacogenomics because it is more cost-effective to research and develop therapies that will be used (and paid for) by the largest groups of people possible.50 On the other hand, this effect may be countered by an increase in the number of new drug targets acquired via genomics research. If biotechnology and pharmaceutical companies are unwilling to research and develop niche market drugs, governmental subsidies, akin to those under the Orphan Drug Act, may be needed. Back to top
Translating pharmacogenomic findings and genetically determined risk factors for chronic disease from the bench to bedside is a multidisciplinary problem, involving the private and public sectors. Philosophical, societal, cultural, behavioral, educational, drug development, communication, and clinical practice issues are all relevant. The technological task challenging the scientific community is twofold. The first challenge is identifying and conveying the relevant information to the clinician. This effort will require the establishment of large, well-annotated databases that link information on disease pathogenesis, functional genomics, disease susceptibility, proteomics, and genetic variation.
The need to develop high-throughput, sensitive, and cost-effective testing for genetic variation remains a challenge for molecular diagnostics. High-density genetic maps are critical for identifying the chromosomal regions and genes involved in the susceptibility to complex, chronic diseases that are influenced by environmental and lifestyle factors, and that are major causes of morbidity and mortality (eg, cardiovascular disease, cancer, diabetes). The new, multiplexed genotyping systems that are under development promise to assist in the genome-wide identification of genetic variants that predispose to multfactorial traits. In the future, DNA microarrays will be able to test for many disease risk factor and/or drug response genotypes simultaneously, thus avoiding the need to rely on self-reports of race and ethnicity as proxies. Exploitation of genetic variations opens up the prospects for identifying those at increased (and possibly early) risk for chronic diseases and for developing intervention strategies to disrupt the natural history of disease.
A broad array of health care professionals, not just physicians, require education about genetics, and geneticists need to become more familiar with broader aspects of health care delivery and public health. Accurate content about genetics, genomics, and proteomics is essential in curricula for undergraduate medical students, training programs for residents, and continuing professional development programs for practicing physicians. Reliable and easily accessible information from independent sources will be important, for both doctors and their patients.
As population-based genomic variation becomes better understood, and specific genetic polymorphisms are associated with disease risks or protection, ethical, social and legal issues will continue to evolve as technology evolves. Molecular medicine will serve as an ever-expanding foundation of current and future medical practice. Organized medicine must be prepared to accept this challenge.
RECOMMENDATIONS (Adopted AMA Directives)
The folowing statements, recommended by the Council on Science and Public Health, were adopted as AMA directives at the 2006 AMA Annual Meeting:
- The AMA will maintain a visible presence in genetics and molecular medicine, including Web-based resources and the development of educational materials, to assist in educating physicians about relevant clinical practice issues related to genomics as they develop. (Directive)
- The AMA will promote the appropriate use of pharmacogenomics in drug development and clinical trials. (Directive)
Back to top
CSAPH home page
Reports by topic
Also see AMA's Genetics and Molecular Medicine Web site
1. Sykiotis GP, Kalliolias GD, Papavassiliou AG. Pharmacogenetic principles in the hippocratic writings. J Clin Pharmacol. 2005;45:1218-1220.
2. Califf RM. Defining the balance of risk and benefit in the era of genomics and proteomics. Health Affairs. 2004;23:77-87.
3. Williams RS, Willard HF, Suyderman R. Personalized health planning. Science. 2003;300:549.
4. Collins FS, et al. A vision for the future of genomics research: a blueprint for the genomic era. Nature. 2003;422:835-847.
5. Lander ES, Linton LM, Berrin B, et al. Initial sequencing and analysis of the human genome.
Nature. 2001;409:860-921.
6. Venter JC, Adam MD, Meyers EW, et al. The sequence of the human genome. Science. 2001;291:1304-51.
7. Marth G, Schuler G, Yeh R. Sequence variations in the public human genome data reflect a bottlenecked population history. Proc Natl Acad Sci USA. 2003;100:376-381.
8. Owens K, King MC. Genomic views of human history. Science. 1999;286:451-453.
9. Byrd MW, Clayton L. Understanding and eliminating racial and ethnic disparities in health care: a background and history. In: Unequal Treatment: Confronting Ethnic and Racial Disparities in Care.
Washington, DC: Institute of Medicine, National Academy Press; 2002.
10. Page GP, George V, Go RC, Page PZ, Allison DB. "Are we there yet?": Deciding when one has demonstrated specific genetic causation in complex diseases and quantitative traits. Am J Hum Genet. 2003;73:711-719.
11. Single nucleotide polymorphism database. Available at: http://www.ncbi.nlm.nih.gov/SNP/snp_summary.cgi.
12. The International HapMap Consortium. The International HapMap Project. Nature. 2003;426:789-796.
13. Hoehe MR. Haplotypes and the systematic analysis of genetic variation in genes and genomes. Pharmacogenomics. 2003;4:547-570.
14. Traherme JA, Horton R, Roberts AN, et al. Genetic analysis of completely sequenced disease-associated MHC haplotypes identifies shuffling of segments in recent human history. PLoS Genet. 2006;2(1)e9:81-91.
15. Liotta LA, Kohn EC, Petricoin EF. Clinical proteomics. personalized molecular medicine. JAMA. 2001;286:2211-2214.
16. Witzman FA, Wang M. Proteomics. In: Kalow W, Meyer UA, Tyndale RF, eds. Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group;. 2005: pp.413-440.
17. Hoehn GT, Suffredinin AF. Proteonomics. Crit Care Med. 2005;33(12 Suppl):S444-448.
18. Clinical proteomics research map. Available at: www.cprmap.com.
19. Colantonio DA, Chan DW. The clinical application of proteomics. Clin Chim Acta. 2005;357:151-158.
20. Nicholson JK, Connolly J, Lindon JC, Holmes E. Metabonomics: a platform for studying drug toxicity and gene function. Nat Rev Drug Discovery. 2002;1:153-161.
21. Lindon JC, Holmes E, Bollard ME, Stanley EG, Nicholson JK. Metabonomics technologies and their applications in physiological monitoring, drug safety assessment and disease diagnosis. Biomarkers. 2004;9(1):1-31.
22. Holmes E, Nichols AW, Lindon JC, et al. Chemometric models for toxicity classification based on NMR spectra of biofluids. Chem Res Toxicol. 2000;13:471-478.
23. Mortishire-Smith RJ, Skiles GL, Lawrence JW, et al. Use of metabonomics to identify impaired fatty acid metabolism as the mechanism of drug-induced toxicity. Chem Res Toxicol. 2004;17:165-173.
24. Lehman-McKeeman LD. Metabonomics. In: Kalow W, Meyer UA, Tyndale RF eds. Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group; 2005: pp.327-340.
25. Vesell ES, Page JG. Genetic control of drug levels in man: antipyrine. Science. 1968;161:72-73.
26. Mahgoub A, Idle JR, Dring LG, Lancaster R. Smith RL. Polymorphic hydroxylation of debrisoquine in man. Lancet. 1977;2:584-586.
27. Wilkinson GR. Drug metabolism and variability among patients in drug response. N Engl J Med. 2005;352:2211-2221.
28. Weinshilboum R. Inheritance and drug response. N Engl J Med. 2003;348:529-537.
29. Spear BB, Heath-Chiozzi M, Huff J. Clinical application of pharmacogenetics. Trends Mol Sci. 2001;7:201-204.
30. Lazarou J, Pomeranz BH, Corey PN. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA. 1998;279:1200-1205.
31. Roses A. Pharmacogenetics and drug development: prospective pipeline strategies for safer and more effective drugs. Nat Rev Genet. 2004;5:645-656.
32. U.S Department of Health and Human Services. Food and Drug Administration. Guidance for Industry. Pharmacogenomic data submissions. March 2005.
33. Pelkonen O, Maenpaa J, Taavitsainen P, Rautio A, Raunio H. Inhibition and induction of human cytochrome P450 (CYP) enzymes. Xenobiotica. 1998;28:1203-1253.
34. Pascussi JM. The expression of CYP2B6, CYP2C9 and CYP3A4 genes: a tangle of networks of nuclear and steroid receptors. Biochim Biophys Acta. 2003;1619:243-253.
35. Panagiotidis G, Bertilsson L. Pharmacogenetics of drug metabolism: two clinically important polymorphic enzymes, CYP2D6 and TPMT. In: Kalow W, Meyer UA, Tyndale RF , eds. Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group; 2005: pp.51-69.
36. Okey AB. Variablity in induction of human drug metabolizing enzymes. In: Kalow W, Meyer UA, Tyndale RF, eds. Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group; 2005: pp.157-205.
37. Krynetski EY, Evans WE. Genetic polymorphism of thiopurine- S-methyltransferase: molecular mechanisms and clinical importance. Pharmacology. 2000;61:136-146.
38. McLeod HL, Yu J. Cancer pharmacogenomics: SNPs, chips, and the individual patient. Cancer Invest. 2003;21:630-640.
39. Phillips KA, Veenstra DL, Oren E, Lee JK, Sadee W. Potential role of pharmacogenomics in reducing adverse drug reactions: a systematic review. JAMA. 2001;286:2270-2279.
40. Weber W. Receptors. In: Kalow W, Meyer UA, Tyndale RF, eds. Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group; 2005: pp. 71-107.
41. Marzolini C, Tirona RG, Kin RB. Pharmacogenetics of drug transporters. In: Kalow W, Meyer UA, Tyndale RF, eds: Pharmacogenomics, 2nd edition. Boca Raton, FL: Taylor and Francis Group;. 2005; pp.109-155.
42. Marzolini C, Paus E, Buclin T, Kim RB. Polymorphisms in human MDR1 (P-glycoprotein): recent advances and clinical relevance. Clin Pharmacol Ther. 2004;75:13-33.
43. Lee W, Kim RB. Transporters and renal drug elimination. Annu Rev Pharmacol Toxicol. 2004;44:137-166.
44. Kullak-Ublick GA, Becker MB. Regulation of drug and bile salt transporters in liver and intestine. Drug Metab Rev. 2003;35:305-317.
45. Roses AD. Pharmacogenetics and the practice of medicine. Nature. 2000;405:857-865.
46. Zineh I, Gerhard T, Aquilante CL et al. Availability of pharmacogenomics-based prescribing information in drug package inserts for currently approved drugs. Pharmaxcogenomics J. 2004;4:354-358.
47. NIH-DOE Working Group on Ethical, Legal and Social Implications of Human Genome Research. Report of the Task Force on Genetic Information and Insurance. Genetic Information and Health Insurance. 1993. Available at: http://www.genome.gov/10001750. Accessed February 8, 2006.
48. NIH-DOE Working Group on Ethical, Legal and Social Implications of Human Genome Research. Final Report of the Task Force on Genetic Testing. Promoting Safe and Effective Testing in the United States; 1997. Available at: http://www.genome.gov/10001733. Accessed February 8, 2006.
49. Nuffield Council on Bioethics. Pharmacogenetics: ethical issues; 2003. Available at: www.nuffieldbioethics.org. Accessed February 7, 2006.
50. Rothstein MA, Epps PG. Ethical and legal implications of pharmacogenomic. Nat Rev Genet. 2001;2:228-231.
Resolution 422 (A-05), introduced by the Medical Student Section at the 2005 Annual Meeting and referred to the Board of Trustees, asks:
That the American Medical Association (AMA) continue to recognize the need for possible adaptation of the US health care system to prospectively prevent the development of disease by ethically using genomics, proteomics, metabolomics, imaging and other advanced diagnostics, along with standardized informatics tools to develop individual risk assessments and personal health plans;
That the AMA support studies aimed at determining the viability of prospective care models and measures that will assist in creating a stronger focus on prospective care in the US health care system; and
That the AMA support research and discussion regarding the multidimensional ethical issues related to prospective care models, such as genetic testing.
Content provided by: CSAPH
Copyright 1995-2008 American Medical Association. All rights reserved.
