Advances in pharmacogenomics and individualized drug therapy: exciting challenges that lie ahead

Healthy Subjects With Extreme Patterns of Performance Differ in Functional Network Topology and Benefits From Nicotine

Do subjects with atypical patterns in attentional and executive behaviour show different brain network topology and react differently towards nicotine administration? The efficacy of pro-cognitive drugs like nicotine considerably varies between subjects and previous theoretical and empirical evidence suggest stronger behavioural nicotine effects in subjects with low performance. One problem is, however, how to best define low performance, especially if several cognitive functions are assessed for subject characterisation. We here present a method that used a multivariate, robust outlier detection algorithm to identify subjects with suspicious patterns of performance in attentional and executive functioning. In contrast to univariate approaches, this method is sensitive towards extreme positions within the multidimensional space that do not have to be extreme values in the individual behavioural distributions. The method was applied to a dataset of healthy, non-smoking subjects (n = 34) who were behaviorally characterised by an attention and executive function test on which N = 12 volunteers were classified as outliers. All subjects then underwent a resting-state functional magnetic resonance imaging (fMRI) scan to characterise brain network topology and an experimental behavioural paradigm under placebo and nicotine (7 mg patch) that gauged aspects of attention and executive function. Our results indicate that subjects with an atypical multivariate pattern in attention and executive functioning showed significant differences in nodal brain network integration in visual association and pre-motor brain regions during resting state. These differences in brain network topology significantly predicted larger individual nicotine effects on attentional processing. In summary, the current approach successfully identified a subgroup of healthy volunteers with low behavioural performance who differ in brain network topology and attentional benefit from nicotine.

A Transcriptional Regulatory Network Containing Nuclear Receptors and Long Noncoding RNAs Controls Basal and Drug-Induced Expression of Cytochrome P450s in HepaRG Cells

Cytochrome P450 (P450) enzymes are responsible for metabolizing drugs. Expression of P450s can directly affect drug metabolism, resulting in various outcomes in therapeutic efficacy and adverse effects. Several nuclear receptors are transcription factors that can regulate expression of P450s at both basal and drug-induced levels. Some long noncoding RNAs (lncRNAs) near a transcription factor are found to participate in the regulatory functions of the transcription factors. The aim of this study is to determine whether there is a transcriptional regulatory network containing nuclear receptors and lncRNAs controlling both basal and drug-induced expression of P450s in HepaRG cells. Small interfering RNAs or small hairpin RNAs were applied to knock down four nuclear receptors [hepatocyte nuclear factor 1α (HNF1α), hepatocyte nuclear factor 4α (HNF4α), pregnane X receptor (PXR), and constitutive androstane receptor (CAR)] as well as two lncRNAs [HNF1α antisense RNA 1 (HNF1α-AS1) and HNF4α antisense RNA 1 (HNF4α-AS1)] in HepaRG cells with or without treatment of phenobarbital or rifampicin. Expression of eight P450 enzymes was examined in both basal and drug-induced levels. CAR and PXR mainly regulated expression of specific P450s. HNF1α and HNF4α affected expression of a wide range of P450s as well as other transcription factors. HNF1α and HNF4α controlled the expression of their neighborhood lncRNAs, HNF1α-AS1 and HNF4α-AS1, respectively. HNF1α-AS1 and HNF4α-AS1 was also involved in the regulation of P450s and transcription factors in diverse manners. Altogether, our study concludes that a transcription regulatory network containing the nuclear receptors and lncRNAs controls both basal and drug-induced expression of P450s in HepaRG cells.

The pharmacokinetics of dexmedetomidine during long-term infusion in critically ill pediatric patients. A Bayesian approach with informative priors

The purpose of this study was to assess the pharmacokinetics of dexmedetomidine in the ICU settings during the prolonged infusion and to compare it with the existing literature data using the Bayesian population modeling with literature-based informative priors. Thirty-eight patients were included in the analysis with concentration measurements obtained at two occasions: first from 0 to 24 h after infusion initiation and second from 0 to 8 h after infusion end. Data analysis was conducted using WinBUGS software. The prior information on dexmedetomidine pharmacokinetics was elicited from the literature study pooling results from a relatively large group of 95 children. A two compartment PK model, with allometrically scaled parameters, maturation of clearance and t-student residual distribution on a log-scale was used to describe the data. The incorporation of time-dependent (different between two occasions) PK parameters improved the model. It was observed that volume of distribution is 1.5-fold higher during the second occasion. There was also an evidence of increased (1.3-fold) clearance for the second occasion with posterior probability equal to 62 %. This work demonstrated the usefulness of Bayesian modeling with informative priors in analyzing pharmacokinetic data and comparing it with existing literature knowledge.

The CYP2D6 gene determines oxycodone's phenotype-specific addictive potential: implications for addiction prevention and treatment

We propose a hypothesis for predicting addictive potential of oral drugs, in general, and oxycodone's addictive potential, in particular. We hypothesize that a patient's CYP2D6 phenotype determines oxycodone's addictive potential, in part, via genotype-specific regulation of its clearance; although, other possible modulators of oxycodone's addiction potential exist. For example, brain CYPs related to phenotype could be involved. To pilot test our hypothesis, we used a mathematical model which postulates that oxycodone's addictive potential is given by: LAP=E/(ka/ke), where LAP represents addictive potential, E represents euphoric potency, ka is the absorption rate constant of drug from the gastrointestinal tract, and ke is the systemic elimination rate constant of drug by all processes responsible for its removal from plasma. Using CYP2D6 phenotype-specific oxycodone pharmacokinetic parameter values derived from published data, our hypothesis predicted that the canonical order of oxycodone's addictive potential was UM>EM>IM>PM, with corresponding LAP values of 0.24, 0.21, 0.17, and 0.15 respectively. Our hypothesis about oxycodone's addictive potential may provide a unifying approach useful for both personalized medicine dosing and predicting addictive potential of oral drugs in humans, since it is based on both oxycodone's pharmacogenetics and pharmacokinetics.

Prediction of drug response and safety in clinical practice

Many clinicians hoped that the completion of the Human Genome Project would result in "individualized drug therapy," i.e., determining the right medication at the right dose 100% of the time based upon the individual's genetics. The pharmacogenomic prediction of drug efficacy and safety has not become a reality due to continuing realization of the complexity dictating the human-drug interaction. New methods of metabolomics, proteomics, and transcriptomics that account for this complexity hold promise for translational researchers hoping to increase drug efficacy and decrease drug toxicity.

Functional evaluation of genetic and environmental regulators of p450 mRNA levels

Variations in the activities of Cytochrome P450s are one of the major factors responsible for inter-individual differences in drug clearance rates, which may cause serious toxicity or inefficacy of therapeutic drugs. Various mRNA level is one of the key factors for different activity of the major P450 genes. Although both genetic and environmental regulators of P450 gene expression have been widely investigated, few studies have evaluated the functional importance of cis- and trans-regulatory factors and environmental factors in the modulation of inter-individual expression variations of the P450 genes. In this study, we measured the mRNA levels of seven major P450 genes (CYP1A1, CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP3A4 and CYP3A5) in 96 liver biopsy samples from Chinese population. Both trans-acting (mRNA levels and non-synonymous SNPs of putative regulator genes) and cis-acting (gene copy number and functional SNPs) factors were investigated to identify the determinants of the expression variations of these seven P450 genes. We found that expression variations of most P450 genes, regulator genes and housekeeping genes were positively correlated at the mRNA level. After partial correlation analysis using ACTB and GAPDH expression to eliminate the effect of global regulators, a UPGMA (Unweighted Pair Group Method with Arithmetic Mean) tree was constructed to reveal the effects of specific regulation networks potentially masked by global regulators. Combined with the functional analysis of regulators, our results suggested that expression variation at the mRNA level was mediated by several factors in a gene-specific manner. Cis-acting genetic variants might play key roles in the expression variation of CYP2D6 and CYP3A5, environmental inducers might play key roles in CYP1A1 and CYP1A2 variation and global regulators might play key roles in CYP2C9 variation. In addition, the functions of regulators that play less important roles in controlling expression variation for each P450 gene were determined.

Physician liability: the next big thing for personalized medicine?

Liability is likely to be a major driver for the future direction and implementation of personalized medicine, spurring the adoption of genetic tests and other pharmacogenomic technologies, in some cases appropriately, and in other cases prematurely or as inefficient defensive medicine. While all entities in the personalized medicine chain will face liability risks, physicians will be at the greatest risk owing to their lack of defenses, limited experience in dealing with genetics and the growing disparities within the profession in implementing new medical technologies. The history of liability for genetic testing, primarily in the prenatal testing context, suggests that liability will often be both unpredictable and influential in changing medical practice. It is critical to anticipate and attempt to prevent such liability risks in a proactive manner so to minimize the disruptive impact that liability can cause.

Pharmacogenetics of cytochrome P450 (CYP) in the elderly

The genetics of cytochrome P450 (CYP) is a very active area of multidisciplinary research, overlapping the interest of medicine, biology and pharmacology, being the CYP enzyme system responsible for the metabolism of more than 80% of the commercially available drugs. Variations in CYP encoding genes are responsible for inter-individual differences in CYP production or function, with severe clinical consequences as therapeutic failures (TFs) and adverse drug reactions (ADRs), being ADRs worldwide primary causes of morbidity and mortality in elderly people. In fact, the prevalence of both TFs and ADRs strongly increased in the presence of multiple pharmacological treatments, a common status in subjects aging 65 years and over. The present article explored some basic concepts of human genetics that have important implications in the genetics of CYP. An attempted to transfer these basic concepts to the genetic data reported by the Home Page of The Human Cytochrome P450 (CYP) Allele Nomenclature Committee was also made, focusing on the current knowledge of CYP genetics. The status of what we know and what we need to know is the base for the clinical applications of pharmacogenetics, in which personalized drug treatments constituted the main aim, in particular in patients attending a geriatric ward.

Basic properties and molecular mechanisms of exogenous chemical carcinogens

Exogenous chemical carcinogenesis is an extremely complex multifactorial process during which gene-environment interactions involving chronic exposure to exogenous chemical carcinogens (ECCs) and polymorphisms of cancer susceptibility genes add further complexity. We describe the properties and molecular mechanisms of ECCs that contribute to induce and generate cancer. A basic and specific property of many lipophilic organic ECCs including polycyclic aromatic hydrocarbons and polyhalogenated aromatic hydrocarbons is their ability to bioaccumulate in the adipose tissue from where they may be released in the blood circulation and target peripheral tissues for carcinogenesis. Many organic ECCs are procarcinogens and consequently need to be activated by the cytochrome P450 (CYP) system and/or other enzymes before they can adduct DNA and proteins. Because they contribute not only to the cocarcinogenic and promoting effects of many aromatic pollutants but also to their mutagenic effects, the aryl hydrocarbon receptor-activating and the inducible CYP systems are central to exogenous chemical carcinogenesis. Another basic property of ECCs is their ability to induce stable and bulky DNA adducts that cannot be simply repaired by the different repair systems. In addition, following ECC exposure, mutagenesis may also be caused indirectly by free-radical production and by epigenetic alterations. As a result of complex molecular interplays, direct and/or indirect mutagenesis may especially account for the carcinogenic effects of many exogenous metals and metalloids. Because of these molecular properties and action mechanisms, we conclude that ECCs could be major contributors to human cancer, with obviously great public health consequences.

Systems biology and functional genomics approaches for the identification of cellular responses to drug toxicity

Extensive growth in the field of molecular biology in recent decades has led to the development of new and powerful experimental and computational tools that enable the analysis of complex biological responses to chemical exposure on both a functional and structural genetic level. The ability to profile global responses on a transcriptional level has become a valuable resource in the science of toxicology and attempts are now being made to further understand toxicity mechanisms by incorporating metabolomics and proteomics approaches. In addition, recent progress in understanding the extent of the genetic diversity within and between species allows us to take a fresh look at research on genetic polymorphisms that may influence an individual's susceptibility to toxicity. Whereas new technologies have the potential to make a sizeable impact on our understanding of the mechanisms of toxicity, considerable challenges remain to be addressed, especially with regard to the regulatory acceptance and successful integration of omics data. This review highlights recent advancements in the application of functional and structural genomics techniques to chemical hazard identification and characterization, and to the understanding of the interindividual differences in susceptibility to adverse drug reactions.

Global metabolic profiling and its role in systems biology to advance personalized medicine in the 21st century

Systems biology attempts to elucidate the complex interaction between genes, proteins and metabolites to provide a mechanistic understanding of cellular function and how this function is affected by disease processes, drug toxicity or drug efficacy effects. Global metabolic profiling is an important component of systems biology that can be applied in both preclinical and clinical settings for drug discovery and development, and to study disease mechanisms. The metabolic profile encodes the phenotype, which is composed of the genotype and environmental factors. The phenotypic profile can be used to make decisions about the best course of treatment for an individual patient. Understanding the combined effects of genetics and environment through a systems biology framework will enable the advancement of personalized medicine.

Metabolomics: a tool for personalizing medicine?

The concept of personalized medicine is based upon using personal genetic information to predict individual responses to drug therapy. However, environmental factors such as diet, energy status, gut microbiota, health status and age will influence the expression of one’s genetic potential. Metabolomics data from biofluid and tissue sample analysis hold information regarding a patient’s genotype and phenotype. Metabolomics data can be rapidly collected from biofluid samples over time, providing temporal metabolic analyses of patient samples. In addition to metabolic markers of a patient’s phenotype, metabolomics can provide markers of drug efficacy, toxicity and clearance.

From human genetics and genomics to pharmacogenetics and pharmacogenomics: past lessons, future directions

A brief history of human genetics and genomics is provided, comparing recent progress in those fields with that in pharmacogenetics and pharmacogenomics, which are subsets of genetics and genomics, respectively. Sequencing of the entire human genome, the mapping of common haplotypes of single-nucleotide polymorphisms (SNPs), and cost-effective genotyping technologies leading to genome-wide association (GWA) studies - have combined convincingly in the past several years to demonstrate the requirements needed to separate true associations from the plethora of false positives. While research in human genetics has moved from monogenic to oligogenic to complex diseases, its pharmacogenetics branch has followed, usually a few years behind. The continuous discoveries, even today, of new surprises about our genome cause us to question reviews declaring that "personalized medicine is almost here" or that "individualized drug therapy will soon be a reality." As summarized herein, numerous reasons exist to show that an "unequivocal genotype" or even an "unequivocal phenotype" is virtually impossible to achieve in current limited-size studies of human populations. This problem (of insufficiently stringent criteria) leads to a decrease in statistical power and, consequently, equivocal interpretation of most genotype-phenotype association studies. It remains unclear whether personalized medicine or individualized drug therapy will ever be achievable by means of DNA testing alone.

Monitoring gene changes during antiepileptic drug therapy to widen the safety window and reduce pharmacoresistance

A novel and potentially effective way to decrease pharmacoresistance is to widen an antiepileptic drug's (AED) safety window by altering its efficacy-toxicity relationship to reduce the frequency or eliminate the occurrence of drug-induced toxicities. The first step in this process is to identify specific AED toxicities that are common, objective, reliable to assess, and exhibit both interpatient and interdrug variability. Next, one or more of the fundamental sources of biological variability (e.g., DNA, RNA, proteins, or metabolites) is selected for study. By identifying a relationship between variations in these fundamental sources and specific AED toxicities, one aims to identify pathways underlying the genesis of the toxicity (mechanisms) or biomarkers related to elevated risk for AED toxicity (predictors). These types of studies have many methodological challenges. However, if properly conducted, the resulting data could be used to design interventions that directly block the toxicity mechanisms or preselect individuals who are at high risk of developing these toxicities. Both approaches would conceivably widen the safety window of efficacious AEDs and reduce the incidence of pharmacoresistance. This article describes a general framework for this approach and provides an example of the methodology using differential gene expression measured using RNA-based microarray technology in children experiencing VPA-associated weight gain.

Pharmacogenomics and nutrigenomics: synergies and differences

The success of the Human Genome Project and the spectacular development of broad genomics tools have catalyzed a new era in both medicine and nutrition. The terms pharmacogenomics and nutrigenomics are relatively new. Both have grown out of their genetic forbears as large-scale genomics technologies have been developed in the last decade. The aim of both disciplines is to individualize or personalize medicine and food and nutrition, and ultimately health, by tailoring the drug or the food to the individual genotype. This review article provides an overview of synergies and differences between these two potentially powerful science areas. Individual genetic variation is the common factor on which both pharmacogenomics and nutrigenomics are based. Each human is genetically (including epigenetics) unique and phenotypically distinct. One of the expectations of both technologies is that a wide range of gene variants and related single-nucleotide polymorphism will be identified as to their importance in health status, validated and incorporated into genotype based strategies for the optimization of health and the prevention of disease. Pharmacogenomics requires rigorous genomic testing that will be regulated and analyzed by professionals and acted on by medical practitioners. As further information is obtained on the importance of the interaction of food and the human genotype in disease prevention and health, pharmacogenomics can provide an opportunity driver for nutrigenomics. As we move from disease treatment to disease prevention, the two disciplines will become more closely aligned.

Genetic risk profiles for cancer susceptibility and therapy response

Cells in the body are permanently attacked by DNA-reactive species, both from intracellular and environmental sources. Inherited and acquired deficiencies in host defense mechanisms against DNA damage (metabolic and DNA repair enzymes) can modify cancer susceptibility as well as therapy response. Genetic profiles should help to identify high-risk individuals who subsequently can be enrolled in preventive measures or treated by tailored therapy regimens. Some of our attempts to define such risk profiles are presented. Cancer susceptibility: Single nucleotide polymorphisms (SNPs) in metabolic and repair genes were investigated in a hospital-based lung cancer case-control study. When evaluating the risk associated with different genotypes for N-acetyltransferases (Wikman et al. 2001) and glutathione-S-transferases (Risch et al. 2001), it is mandatory to distinguish between the three major histological subtypes of lung tumors. A promoter polymorphism of the myeloperoxidase gene MPO was shown to decrease lung cancer susceptibility mainly in small cell lung cancer (SCLC) (Dally et al. 2002). The CYP3A4*1B allele was also linked to an increased SCLC risk and in smoking women increased the risk of lung cancer eightfold (Dally et al. 2003b). Polymorphisms in DNA repair genes were shown to modulate lung cancer risk in smokers, and reduced DNA repair capacity elevated the disease risk (Rajaee-Behbahani et al. 2001). Investigations of several DNA repair gene variants revealed that lung cancer risk was only moderately affected by a single variant but was enhanced up to approximately threefold by specific risk allele combinations (Popanda et al. 2004). Therapy response: Inter-individual differences in therapy response are consistently observed with cancer chemotherapeutic agents. Initial results from ongoing studies showed that certain polymorphisms in drug transporter genes (ABCB1) differentially affect response outcome in histological subgroups of lung cancer. Stronger beneficial effects were seen in non-small cell lung cancer (NSCLC) patients following gemcitabine and in SCLC patients following etoposide-based treatment. Several DNA repair parameters (polymorphisms, RNA expression, and DNA repair capacity) were measured in vitro in lymphocytes of patients before radiotherapy and correlated with the occurrence of acute side effects (radio-hypersensitivity). Our initial analysis of several repair gene variants in breast cancer patients (n = 446) who received radiotherapy revealed no association of single polymorphisms and the development of side effects (moist desquamation of the irradiated normal skin). The risk for this side effect was, however, strongly reduced in normal weight women carrying a combination of XRCC1 399Gln and APE1 148Glu alleles, indicating that these variants afford some protection against radio-hypersensitivity (Chang-Claude et al. 2005). Based on these data we conclude that specific metabolic and DNA repair gene variants can affect cancer risk and therapy outcome. Predisposition to hereditary cancer syndromes is dominated by the strong effects of some high-penetrance tumor susceptibility genes, while predisposition to sporadic cancer is influenced by the combination of multiple low-penetrance genes, of which as a major challenge, many disease-relevant combinations remain to be identified. Before translating these findings into clinical use and application for public health measures, large population-based studies and validation of the results will be required.

The role of cytochrome P450 enzymes in endogenous signalling pathways and environmental carcinogenesis

Some cytochrome P450 (CYP) heme-thiolate enzymes participate in the detoxication and, paradoxically, the formation of reactive intermediates of thousands of chemicals that can damage DNA, as well as lipids and proteins. CYP expression can also affect the production of molecules derived from arachidonic acid, and alters various downstream signal-transduction pathways. Such changes can be precursors to malignancy. Recent studies in mice have changed our perceptions about the function of CYP1 enzymes. We suggest a two-tiered system to predict an overall inter-individual risk of tumorigenesis based on DNA variants in certain 'early defence' CYP genes, combined with polymorphisms in various downstream target genes.

Can pharmacogenetics help rescue drugs withdrawn from the market?

Observations over the later half of the last century have suggested that genetic factors may be the prime determinant of drug response, at least for some drugs. Retrospectively gathered data have provided further support to the notion that genotype-based prescribing will improve the overall efficacy rates and minimize adverse drug reactions (ADRs), making personalized medicine a reality. During the last 16 years, 38 drugs have been withdrawn from major markets due to safety concerns. Inevitably, a question arises as to whether it might be possible to 'rescue' some of these drugs by promoting genotype-based prescribing. However, ironically pharmacogenetics has not perceptibly improved the risk/benefit of a large number of genetically susceptible drugs that are already in wide clinical use and are associated with serious ADRs. Drug-induced hepatotoxicity and QT interval prolongation (with or without torsade de pointes) account for 24 (63%) of these 38 drug withdrawals. In terms of the number of drugs implicated, both these toxicities are on the increase. Many others have had to be withdrawn due to their inappropriate use. This paper discusses the criteria that a drug would need to fulfill, and summarizes the likely regulatory requirements, before its pharmacogenetic rescue can be considered to be realistic. One drug that fulfils these criteria is perhexiline (withdrawn worldwide in 1988) and is discussed in some detail. For the majority of these 38 drugs there are, at present, no candidates for genetic traits to which the toxicity that led to their withdrawal may be linked. For a few other drugs where a potential candidate for a genetic trait might explain the toxicity of concern, the majority of patients who experienced the index toxicity had easily managed nongenetic risk factors. It may be possible to rescue these drugs simply by careful attention to their dose, interaction potential and prescribing patterns, but without the need for any pharmacogenetic test. In addition, the pharmacogenetic rescue of drugs might not be as effective as anticipated as hardly any pharmacogenetic test is known to have the required test efficiency to promote individualized therapy. Multiple pathways of drug elimination, contribution to toxicity by metabolites as well as the parent drug, gene-gene interactions, multiple mechanisms of toxicity and inadequate characterization of phenotype account for this lack of highly predictive tests. The clinical use of tests that lack the required efficiency carries the risks of over- or under-dosing some patients, denying the drug to others and decreasing physician vigilance of patients. Above all, at present, prescribing physicians lack an adequate understanding of pharmacogenetics and its limitations. It is also questionable whether their prescribing will comply with the requirements for pretreatment pharmacogenetic tests to make pharmacogenetic rescue a realistic goal.

Can personalized drug therapy be achieved? A closer look at pharmaco-metabonomics

Between 1930 and 1990, several dozen high-penetrance, predominantly monogenic disorders were identified and characterized, which led some investigators to speculate that individualized drug treatment was just around the corner. Informative DNA tests were sought to determine genetic predisposition to toxicity and cancer, thereby identifying individuals in which a drug was likely to be effective and those at increased risk of drug toxicity. These assays represent the leading edge of phenotype-genotype association studies, which are a major goal of clinical pharmacology and pharmacogenomics. Because of the complexity of the genome, however, the task is more challenging than anticipated originally. In the past decade we have come to appreciate how difficult it is to determine unequivocally either an exact phenotype or genotype. In the near future it seems unlikely that, by themselves, either transcriptomics or proteomics will be particularly helpful in achieving individualized drug therapy. However, recent advances in metabonomics are exciting and show promise. In the future, and perhaps in combination with proteomics, metabonomics might complement genomics in achieving personalized drug therapy.

TPMT, UGT1A1 and DPYD: genotyping to ensure safer cancer therapy?

The Food and Drug Administration (FDA) has approved label changes for two anticancer drugs, 6-mercaptopurine (6-MP) and irinotecan, to include pharmacogenetic testing as a potential means to reduce the rate of severe toxic events. Comprehensive evaluation of the clinical benefit and cost effectiveness of screening strategies with these tests has not been completed. However, the FDA decided that evidence indicates sufficient benefit to warrant informing prescribers, pharmacists and patients of the availability of pharmacogenetic tests and their possible role in the selection and dosing of these anticancer agents. Reviewing the gene-drug-phenotype relationships of 6-MP, irinotecan and 5-fluorouracil reveals properties of these relationships that lead to a clinically useful pharmacogenetic test. Research in the near future should clarify the role of pharmacogenetic testing in reducing the risk of severe toxicity and determine how these same tests might identify a subset of patients who should safely receive higher doses of treatment to derive the same benefit as the rest of the patient population.

Questionnaire evaluation and risk factor identification for nonallergic vasomotor rhinitis

BACKGROUND

Chronic rhinitis, a prevalent primary care disorder associated with significant comorbidities, is often incorrectly diagnosed. As a result, inappropriate treatment leads to higher health care costs and poor clinical outcomes.

OBJECTIVE

To verify whether responses to specific questions may help in the diagnosis of nonallergic vasomotor rhinitis (NAVMR).

METHODS

A questionnaire listing allergic and nonallergic triggers was blindly distributed to 100 random new patients with chronic rhinitis in an allergist's office. Questionnaire and physician diagnoses were compared. Allergic triggers included cat, dog, feathers, other furry animals, and symptoms during the spring, summer, and fall seasons. Nonallergic triggers included temperature changes, diesel and car exhaust, tobacco smoke, perfumes and fragrances, incense, cleaning products, newsprint, hairspray, and alcoholic beverages, spicy foods, or eating.

RESULTS

The construct validity and internal consistency indicated that the questionnaire was a good tool for diagnosing NAVMR. Multivariate analysis revealed that the absence of outdoor symptoms in the spring (odds ratio [OR], 7.76; 95% confidence interval [CI], 1.18-51.22; P = .03), no parental history of allergy (OR, 5.16; 95% CI, 1.05-25.20; P = .04), no symptoms around cats (OR, 3.82; 95% CI, 0.59-24.83; P = .16), the presence of symptoms around perfumes and fragrances (OR, 4.875; 95% CI, 1.05-22.60; P = .04), and age at symptom onset (OR, 1.08; 95% CI, 1.02-1.14; P = .008) were predictive of NAVMR.

CONCLUSIONS

Specific questionnaire responses may help primary care physicians accurately diagnose NAVMR. Questions asked of patients by physicians to diagnose allergic rhinitis do not correlate well with physician diagnosis.

Interpreting Disparate Responses to Cancer Therapy: The Role of Human Population Genetics

Increasingly, investigators are recognizing differences in tumor biology, drug metabolism, toxicity, and therapeutic response among different patient populations receiving anticancer agents. These observations provide exciting opportunities to identify the factors most important for predicting individual variability in pharmacologically relevant phenotypes and consequently for personalizing the delivery of cancer therapy. Although pharmacogenomic differences may explain some of these disparities, rigorous investigation of both genetic and nongenetic differences is important to identify the variables most important for optimal selection and dosing of treatment for an individual patient. For example, pharmacogenetic tests currently used in cancer therapy, such as genotyping UGT1A1 to reduce the incidence of severe toxicity of irinotecan and sequencing epidermal growth factor receptor from tumors to identify somatic mutations conferring sensitivity to tyrosine kinase inhibitors, were developed without initial identification of interpopulation differences. Although interpopulation variability in toxicity and efficacy of these agents has been observed, the basis for these population differences remains only partially explained. Here, we review concepts of human population genetics to inform interpretations of disparate drug effects of cancer therapy across patient populations. Understanding these principles will help investigators better design clinical trials to identify the variables most relevant to subsequent individualization of a cancer therapy.

Multi-target strategies for the improved treatment of depressive states: Conceptual foundations and neuronal substrates, drug discovery and therapeutic application

Major depression is a debilitating and recurrent disorder with a substantial lifetime risk and a high social cost. Depressed patients generally display co-morbid symptoms, and depression frequently accompanies other serious disorders. Currently available drugs display limited efficacy and a pronounced delay to onset of action, and all provoke distressing side effects. Cloning of the human genome has fuelled expectations that symptomatic treatment may soon become more rapid and effective, and that depressive states may ultimately be "prevented" or "cured". In pursuing these objectives, in particular for genome-derived, non-monoaminergic targets, "specificity" of drug actions is often emphasized. That is, priority is afforded to agents that interact exclusively with a single site hypothesized as critically involved in the pathogenesis and/or control of depression. Certain highly selective drugs may prove effective, and they remain indispensable in the experimental (and clinical) evaluation of the significance of novel mechanisms. However, by analogy to other multifactorial disorders, "multi-target" agents may be better adapted to the improved treatment of depressive states. Support for this contention is garnered from a broad palette of observations, ranging from mechanisms of action of adjunctive drug combinations and electroconvulsive therapy to "network theory" analysis of the etiology and management of depressive states. The review also outlines opportunities to be exploited, and challenges to be addressed, in the discovery and characterization of drugs recognizing multiple targets. Finally, a diversity of multi-target strategies is proposed for the more efficacious and rapid control of core and co-morbid symptoms of depression, together with improved tolerance relative to currently available agents.

Search for an association between the human CYP1A2 genotype and CYP1A2 metabolic phenotype

The genotype responsible for more than 60-fold interindividual differences in human hepatic CYP1A2 constitutive expression is not understood. Resequencing the human CYP1A1_CYP1A2 locus (39.6 kb) in five major geographically isolated subgroups recently led to the identification of 85 single nucleotide polymorphisms (SNPs), 57 of which were double-hit SNPs. Here, we attempted to correlate the CYP1A2 genotype with a metabolic phenotype. We chose 16 SNPs (all having a minor allele frequency > or =0.05 in Caucasians) to genotype 32 DNA samples (26 Caucasians, six Ethiopians) in which CYP1A2 metabolism had previously been determined. From 280 subjects (five locations worldwide) that had been CYP1A2-phenotyped, we genotyped the 10 highest, 14 lowest and eight intermediate DNA samples. Although no SNP was significant (P<0.05), possibly due to the small sample size, we found a trend for several of the six SNPs across the CYP1A2 linkage disequilibrium block associated with the trait. Five CYP1A2 haplotypes were inferred, two of which had not previously been reported; haplotype 1A2H10 showed the greatest association with CYP1A2 activity. Regulatory sequences responsible for the large interindividual differences in hepatic CYP1A2 gene basal expression might reside, in part, with some of these CYP1A2 SNPS but, in large part, might be located either cis (in nearby sequences not yet haplotyped) or trans in that they are not linked to the gene. We conclude that no SNP or haplotype in the CYP1A2 gene has yet been identified that can unequivocally be used to predict the metabolic phenotype in any individual patient.

Present status of outcome prediction of invasive coronary treatment by using genetic markers

A growing number of studies suggest that the outcome after invasive coronary treatment may be in part genetically determined. Here, we review the present status of outcome prediction of invasive coronary treatments by using genetic markers. Although some studies found an association between one or another genetic marker with one or another clinical endpoint, many other studies found no such relations; to date, none of the genetic markers that have been investigated in association studies are used in routine clinical practice to prospectively assess the prognosis following invasive coronary treatment or to decide upon therapeutic strategies. Many associations between genetic markers and certain clinical endpoints were initially reported in small studies but could not be confirmed in larger ones. Some of these discrepancies may be explained by publication bias. Some genetic variants may have true effects on clinical endpoints, which, albeit biologically interesting, do not bear much clinical relevance. On the other hand, many-if not most-studies that have been published to date are more or less grossly underpowered and very rarely report on the results of an a priori power analysis. Thus, there is still a need for further high-quality studies designed to investigate the specific contribution of genetic factors to the outcome after invasive coronary interventions.

Prediction of anticancer drug potency from expression of genes involved in growth factor signaling

PURPOSE

This study develops and evaluates a systematic approach to finding biomarker genes for predicting potency of anticancer drugs against tumor cells, focusing on gene families related to growth factor signaling.

METHODS

Cytotoxic potencies of 119 drugs against 60 neoplastic cell lines (NCI-60) were correlated with expression of 343 genes, including 90 growth factors and receptors, 63 metalloproteinases, and 92 ras-like GTPases as downstream signaling factors. Progressively more stringent criteria and predictive models aim at identifying the smallest subset of genes predictive of cytotoxic potency.

RESULTS

Comparing gene expression with drug potency across the NCI-60 yielded genes with negative and positive correlations (p < 0.001), indicative of a role in chemoresistance and chemosensitivity, respectively. Of 17 genes with multiple negative correlations, 8 are known chemoresistance factors, validating the approach. Negatively correlated genes clustered into two main groups with distinct expression profiles and drug correlations, represented by EGFR and ERBB2 (Her-2/Neu). Accordingly, no synergism was observed between EGFR and ERBB2 inhibitors. However, combinations with classical anticacer drugs were not correlated with EGFR and ERBB2 expression in four cell lines tested, suggesting complex interactions in combination treatments. Finally, a subset of only 13 genes was found to be sufficient for near optimal prediction of drug potency against the NCI-60.

CONCLUSIONS

Our approach using a small subset of genes reveals known and potential biomarkers in cancer chemotherapy, providing a strategy for genome-wide analysis.

Mitochondrial aldehyde dehydrogenase-2 (ALDH2) Glu504Lys polymorphism contributes to the variation in efficacy of sublingual nitroglycerin

Glyceryl trinitrate (GTN), also known as nitroglycerin, has been used to treat angina and heart failure for more than 130 years. Recently, it was shown that mitochondrial aldehyde dehydrogenase-2 (ALDH2) is responsible for formation of NO, the metabolite needed for GTN efficacy. In the present study, we show that the common G-to-A polymorphism in exon 12 of ALDH2--resulting in a Glu504Lys replacement that virtually eliminates ALDH2 activity in both heterozygotes and homozygotes--is associated with a lack of efficacy of sublingual GTN in Chinese subjects. We also show that the catalytic efficiency (Vmax/Km) of GTN metabolism of the Glu504 protein is approximately 10-fold higher than that of the Lys504 enzyme. We conclude that the presence of the Lys504 allele contributes in large part to the lack of an efficacious clinical response to nitroglycerin; we recommend that this genetic factor be considered when administering nitroglycerin to patients, especially Asians, 30-50% of whom possess the inactive ALDH2*2 mutant allele.

Challenges and opportunities in the application of pharmacogenetics to antiepileptic drug therapy

The recent surge of interest in pharmacogenetics has provoked considerable thought regarding its relevance to antiepileptic drug (AED) therapy. Initial studies have focused on genes whose products play a putatively important role in AED pharmacology, particularly drug transporter proteins, drug metabolizing enzymes and ion channel subunits. However, there is a lack of good correspondence between results from different laboratories, and more recent findings are awaiting attempts at confirmation. Thus, there are currently no AED treatment guidelines that are informed by pharmacogenetic data. In order to begin to have clinical impact, standards specific to the conduct of future AED studies must be established. Of particular importance are the need for accurate epilepsy classification, appropriate AED selection and clear and objective assessment outcome measures. In addition, general standards for analysis and interpretation of genetic association data must be better codified and applied consistently across studies. Finally, extensive clinical research networks must be formulated and large numbers of well characterized patients must be recruited. Further development of these critical factors will optimize chances for overcoming current challenges posed by AED pharmacogenetic research and ultimately allow the realization of improved, more rational therapeutic strategies.

Comparative genomics analysis of human sequence variation in the UGT1A gene cluster

Common polymorphisms within the human UGT1A gene locus are associated with irinotecan and tranilast toxicity. To uncover additional functional variation across this gene cluster, cross-species sequence comparisons were performed. Evolutionarily conserved segments (a total of 47.1 kb) were re-sequenced in 24 African-American, 24 European-American, and 24 Asian individuals, and 381 segregating sites (including 123 singletons) were identified. Highly conserved coding sites were less likely to be polymorphic than diverged sites (P<0.0001) but this pattern was not observed at non-coding sites (P=0.1025). Among coding variants, the distribution of those computationally predicted to affect function was skewed toward low frequencies. Some alleles occurred at similar frequencies in each population; others had wide disparities. Although strong linkage disequilibrium was detected among the hepatically expressed genes, the degree of linkage disequilibrium varied among populations. These results suggest that rare functional gene variants and inter-population variability must be considered in the interpretation of association studies between UGT1A and drug metabolism/toxicity phenotypes.

A call for the creation of personalized medicine databases

The success of the Human Genome Project raised expectations that the knowledge gained would lead to improved insight into human health and disease, identification of new drug targets and, eventually, a breakthrough in healthcare management. However, the realization of these expectations has been hampered by the lack of essential data on genotype--drug-response phenotype associations. We therefore propose a follow-up to the Human Genome Project: forming global consortia devoted to archiving and analysing group and individual patient data on associations between genotypes and drug-response phenotypes. Here, we discuss the rationale for such personalized medicine databases, and the key practical and ethical issues that need to be addressed in their establishment.

Pharmacodiagnostic testing in breast cancer: focus on HER2 and trastuzumab therapy

Pharmacogenomics is defined as research into inherited genetic variations that determine an individual's response to therapeutic agents. In oncology, pharmacogenomics based on somatic molecular alterations inherited by subsequent cancer cell generations forms the basis of molecular targeting of novel therapeutic agents. What has emerged from clinical experience with such agents is the need for appropriate pharmacodiagnostic approaches to ensure the drugs are correctly targeted. Given the broad range of pharmacogenomic agents currently under evaluation for cancer therapy, it appears that a rapid extension of pharmacodiagnostic profiling will be required in the next 5-10 years, if not sooner. If this is to be successfully achieved, lessons learned in the past, particularly during the development of HER2 (ERBB2) testing for directing trastuzumab therapy in breast cancer, may provide a valuable framework for the development of future pharmacodiagnostic assays system. This article reviews the biological and clinical rationale for targeting breast cancer with trastuzumab and the steps taken to validate and improve pharmacodiagnostic procedures for testing tumor HER2 protein expression and HER2 gene amplification. Attention is given to quality assurance and reproducibility of testing approaches and the optimal selection of patients for response to trastuzumab. This approach serves as a paradigm for the future development of pharmacodiagnostic tests in oncology.

Pharmacogenomics steps toward personalized medicine

The goal of personalized medicine is to maximize the likelihood of therapeutic efficacy and to minimize the risk of drug toxicity for an individual patient. One of the major contributors to this concept is pharmacogenomics. Marked interindividual genetic variation contributes significantly to both susceptibility to diseases, and response to drugs. Even though pharmacogenomics is not a new science, the translation of pharmacogenomics into clinical practice (i.e., personalized medicine) has not taken place at the same pace as science is delivering new results. It is felt that a large number of recent pharmacogenomic findings allow bold steps to be taken toward personalized medicine. This review collates a variety of examples that have great potential for immediate and effective introduction into clinical practice. In addition, other exploratory examples with a particular focus on drug safety and targeted cancer therapy are summarized.

Theophylline pharmacokinetics: comparison of Cyp1a1(-/-) and Cyp1a2(-/-) knockout mice, humanized hCYP1A1_1A2 knock-in mice lacking either the mouse Cyp1a1 or Cyp1a2 gene, and Cyp1(+/+) wild-type mice

OBJECTIVES

Pharmacokinetics of theophylline was investigated in Cyp1(+/+) wild-type mice, Cyp1a1(-/-) and Cyp1a2(-/-) knockout mice, and humanized hCYP1A1_1A2 mice lacking either the mouse Cyp1a1 or Cyp1a2 gene.

METHODS AND RESULTS

Animals received a single dose of theophylline (8 mg/kg i.p.), either alone or pretreated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; 10 microg/kg i.p.) 24 h prior to theophylline. We found that mouse or human CYP1A2 is the predominant enzyme for theophylline metabolism, the contribution of mouse or human CYP1A1 to theophylline metabolism is negligible, and another TCDD-inducible enzyme plays a minor role in 1-methyluric acid and 1,3-dimethyluric acid formation as well as enhanced theophylline clearance from the body. The half-life of elimination from plasma was more than four times longer in Cyp1a2(-/-) than Cyp1(+/+) mice and more than 10 times different after TCDD pretreatment. In humanized hCYP1A1_1A2 mice lacking the mouse Cyp1a2 gene, the half-life of elimination from plasma was two to three times longer than that in Cyp1(+/+) mice and four to five times different after TCDD pretreatment.

CONCLUSION

Replacement of mouse Cyp1a2 with a functional human CYP1A2 gene restored the ability to metabolize theophylline, and the metabolism changed to a humanized profile (i.e. 3-methylxanthine formation, not seen in the wild-type mouse). TCDD-pretreated hCYP1A1_1A2 Cyp1a2(-/-) mice exhibited enhanced theophylline metabolism and clearance, due to induction of the human CYP1A2 enzyme. Comparing the hCYP1A1_1A2 Cyp1a2(-/-) and wild-type mice with published clinical studies, we found theophylline clearance to be about 5 times and 12 times, respectively, greater than that reported in humans.

Priorities and standards in pharmacogenetic research

The current enthusiasm for pharmacogenetics draws much of its inspiration from the relatively few examples of polymorphisms that have marked and seemingly clinically relevant effects on drug response. In this regard, pharmacogenetic research has paralleled the study of human disease, which has enjoyed success in identifying mutations underlying mendelian conditions. Progress in deciphering the genetics of complex diseases, involving the interaction of multiple genes with each other and with the environment has been considerably less successful. In most instances, drug responses will probably also prove to be complex, influenced by both the environment and multiple genetic factors. For pharmacogenetics to deliver on its potential, this complexity will need to be recognized and accommodated, both in basic research and in clinical application of pharmacogenetics. As the attention of researchers begins to shift toward more systematic pharmacogenetic investigations, we suggest some priorities and standards for pharmacogenetic research.

Genetic diversity and new therapeutic concepts

The differences in medicinal drug responses among individuals had been known for quite some time. Some patients exhibit a life-threatening adverse reaction while others fail to show an expected therapeutic effect. Intermediate responses between the above two extreme cases are also known. In fact, it has been recently reported that approximately 100,000 deaths and more than 2 million hospitalizations annually in the United States are due to properly prescribed medications. This interindividual variability could be due in part to genetically determined characteristics of target genes or drug metabolizing enzymes. This has now been substantiated by a variety of studies. We know that "one size fits all" is not correct. Therefore, the application of pharmacogenetic concepts to clinical practice is an excellent goal in the postgenomic era. The successful completion of the human genome project provided necessary molecular tools, such as high-throughput SNP genotyping, HapMap, and microarray, that can be applied to develop proper therapeutic options for individuals. Recently, there have been considerable scientific, corporate, and policy interest in pharmacotherapy. However, identification of causal variations in a target gene is only a starting point, and the progress in this rapidly developing field is slower than expected. One major drawback could be due to the multigene determinant of drug response that requires a genome-wide screening. Additionally, application of pharmacogenetic knowledge into clinical practice requires a high level of accuracy, precision (risk/benefit ratio), and strict regulations. This is because the pharmacogenetic approach raises several ethical, moral, and legal questions. It is also necessary that both health professionals and the general public must be urgently educated. Despite these limitations, translation of pharmacogenomic data into clinical practice would certainly provide better opportunities to increase the safety and efficacy of medicine in the future.