|
Background
Adverse drug reactions (ADRs) are responsible for many debilitating side effects and are a major cause of death following drug therapy. It is now clear that a significant portion of these ADRs as well as therapeutic failures are caused by genetic polymorphism and genetically based inter-individual differences in drug absorption, disposition, excretion or metabolism.
Genotyping for Cytochrome P450 Polymorphism:
Recent advances in molecular biology have improved the understanding of genetic factors underlying many ADRs. Until recently, investigations in this field have generally centered on gene coding for drug-metabolizing enzymes. Inactivating mutations have been found in gene coding for enzymes belonging to the cytochrome P450 (CYP) system, which is important in the hepatic metabolism of many drugs. Individuals with a lack of functional activity in these enzymes should be treated with very low doses of drugs metabolized by the same enzyme to avoid excessive drug levels and thus toxic effects. In recent years, research has been focused on gene coding for drug targets. As a result, most studies have been performed on single genes known to be or assumed to be functionally related to a given ADR. An alternative method is testing for complex single nucleotide polymorphisms that may be associated with ADRs, although the functional relationship between them may be completely unknown. As a consequence of influence from non-genetic factors in the development of ADRs, the association between a specific genotype and an ADR will always be less than 100%. Thus, there is a need for well-designed clinical trials to ascertain the extent of environmental influences on the ADRs for which a genetic basis has been implicated (Guzey and Spigset, 2004). This is in agreement with the observation of Pirmohamed and Park (2003) who stated that “ADRs are common and many are suggested to have a genetic predisposition. There has been intense research in the role of CYP enzyme gene polymorphisms in the cause of ADRs. The major impact to date of polymorphic CYP expression has been on pre-clinical drug development. The direct clinical impact of CYP polymorphisms on prediction of ADRs, however, has been limited mainly because published reports have been small and retrospective and their findings conflicting. Moreover, the clinical- and cost-effectiveness of pre-prescription genotyping for CYP polymorphisms has not been established. More investigation is needed before prospective CYP genotyping can become routine clinical practice”.
Kirchheiner and colleagues (2003) noted that antidepressants are characterized by a high rate of drug failure. There is evidence that genetic factors are contributing to the inter-individual variability in response to these medications. Genetic polymorphisms in drug metabolizing enzymes are well established and have significant effects on oral clearances or elimination half-lives of antidepressants. These differences can be compensated by adapting the individual dose to genotype in addition to other factors such as age, gender, weight, drug interactions, diseases (cardiovascular, hepatic, renal and respiratory) as well as environmental influences on drug metabolism (e.g., diet and smoking). Genetic variability is found in molecular structures of antidepressant effects. Furthermore, Kirchheiner, et al. (2003) noted that studies on response of antidepressants have revealed influences of polymorphisms in neurotransmitter receptors and transporters altering sensitivity of patients to treatment with antidepressants; however, results were often contradictory.
Kirchheiner and Brockmoller (2005) stated that the genetic coding for the CYP enzyme 2C9 (CYP2C9) carries many inherited polymorphisms. Those coding for R144C (*2) and I359L (*3) amino acid substitutions have both significant functional effects and appreciable high population frequencies, and their in vivo consequences have been examined in humans in relation to drug metabolism -- pharmacokinetics, drug responses as well as outcomes of clinical trials in subjects with different CYP2C9 genotypes. Tentative estimates of how CYP2C9 genotyping might be applied to dose adjustments in clinical therapy were based on dose-related pharmacokinetic parameters such as clearance or trough drug concentrations. Mean clearances in homozygous carriers of the *3 allele were below 25% of that of the wild type for S-warfarin, tolbutamide, glipizide, celecoxib, and fluvastatin. In the more frequent heterozygous carriers (genotype *1/*3), the clearances were between 40 and 75%. In these cases in which individual dosages are derived from clinical drug effects, such as for oral anticoagulants, the pharmacogenetics-based dose adjustments showed a good correlation with the genotype-specific empirically derived doses. In addition to its role in pharmacokinetics, CYP2C9 contributes to the metabolism of fatty acids, prostanoids, and steroid hormones, and it may catalyze potentially toxic bioactivation reactions. However, the current understanding of the role of CYP2C9 in biotransformation of endogenous signaling molecules and in drug toxicity is relatively meager. These investigators concluded that the concept of therapy based on genotyping for CYP polymorphisms has not been assessed in prospective, randomized, controlled studies in which one group is dosed according to genotype while another group is dosed in a usual manner. It is unlikely that CYP2C9 genotyping will become routine practice unless its clinical value is supported by such rigorous assessments.
The following is a list* of some CYP2C9-metabolized drugs with narrow therapeutic ratios (drugs with a narrow difference between therapeutic and toxic concentrations) with serious toxic effects:
-
Angiotensin II Blockers (e.g., irbesartan, losartan)
-
NSAIDS (e.g., diclofenac, ibuprofen, naproxen, piroxicam)
-
Oral Hypoglycemic Agents (e.g., glipizide, tolbutamide)
-
Sulfonylureas (e.g., amitriptyline, celecoxib, fluoxetine, fluvastatin, glipizide, rosiglitazone, tamoxifen, tolbutamide, S-warfarin).
Diagnostic genotyping tests for some CYP enzymes are now available commercially. The AmpliChip (Roche Diagnostics, Basel, Switzerland), cleared by the United States Food and Drug Administration (FDA) through the 510(k) process, is a microarray consisting of many DNA sequences complementary to two CYP genes applied in microscopic quantities at ordered locations on a solid surface (chip). The AmpliChip tests the DNA from a patient's leukocytes collected in a standard anti-coagulated blood sample for 29 polymorphisms and mutations for the CYP2D6 gene and 2 polymorphisms for the CYP2C19 gene. CYP2D6 metabolizes about 25% of all clinically used medications (e.g., antiarrhythmics, antidepressants, beta-blockers, dextromethorphan and morphine derivatives), while CYP2C19 metabolizes several important types of drugs including anti-epileptics and proton-pump inhibitors. The AmpliChip was cleared for marketing for CYP2D6 testing in December 2004, and for CYP2C19 testing in January 2005.
The AmpliChip is marketed for use in screening of patients who are to be treated with drugs metabolized by the CYP system that have a narrow therapeutic ratio with serious toxic effects.
The following is a list* of CYP2D6-metabolized drugs with narrow therapeutic ratios (not an all inclusive list):
-
Antidepressants (e.g., amitryptiline, clomipramine, desipramine, imipramine, paroxetine)
-
Antipsychotics (e.g., chlorpromazine, haloperidol, perphenazine, risperidone, thiroidazine)
-
Beta Blockers (e.g., carvediol, propafenone, timolol).
The following is a list* of CYP2C19-metabolized drugs with narrow therapeutic ratios (not an all inclusive list):
-
Anti-epileptics (e.g., amitryptiline, clomipramine, cyclophosphamide, diazepam, imipramine, moclobemide, phenytoin, phenobarbitone, primidone, R-warfarin)
-
Proton Pump Inhibitors (e.g., lansoprazole, omeprazole, pantoprazole).
* These lists are based on information excerpted from the “Cytochrome P450 Drug Interaction Table”, Department of Medicine, University of Indiana, 2005.
Randomized controlled trials are needed to ascertain if the AmpliChip will lower the incidence of ADRs by detecting patients with CYP2D6 and CYP2C19 mutations. The effectiveness of the AmpliChip in reducing toxic effects and improving health outcomes would need to be compared with standard methods of therapeutic drug monitoring (e.g., by monitoring clinical response or by measuring serum drug concentrations). Moreover, it is still unclear that the AmpliChip would eliminate the need for simultaneous use of other methods of therapeutic drug monitoring (including serum measurements) because factors other than genetic polymorphisms such as those described earlier have a significant effect on drug pharmacokinetics and pharmacodynamics.
Juran and colleagues (2006) stated that "[e]ven though the AmpliChip CYP450 has been approved by the FDA, its practical clinical utility has not yet been determined, and there is a paucity of data related to gastrointestinal and liver diseases. An understanding of the principles and opportunities provided by this new category of diagnostic test is key before planning the necessary studies to evaluate the usefulness of AmpliChip CYP450 in gastroenterologic clinical practice".
A special report (BlueCross BlueShield Technology Evaluation Center, 2004) on genotyping for CYP polymorphisms to determine drug-metabolizer status stated that diagnostic tests to identify specific polymorphisms that may be linked to increased/reduced effect or serious adverse effects are now available. Whether such testing, and for which drugs, improves patient outcomes is not yet known. While genotyping for the CYP enzymes would only need to be performed once per patient and the results could be used to consider other drugs metabolized by the same enzymes, whether genotyping is clinically useful would need to be determined for each drug. Even drugs of the same class may variably rely on specific CYP enzymes. For example, the plasma level of the selective serotonin reuptake inhibitor (SSRI) fluoxetine is significantly affected by CYPP2D6 polymorphisms, whereas the SSRI sertraline appears to be little affected and may depend more upon CYP2C19 for metabolism. The report stated that while the potential of pharmacogenetic studies is intriguing for many clinical applications, it is still unclear which are most likely to provide clinical benefit in the near future.
Ingelman-Sundberg (2005) stated that the polymorphism of CYP2D6 significantly affects the pharmacokinetics of approximately half of the drugs in clinical use, which are CYP2D6 substrates. The consequences of the polymorphism at ordinary drug doses can be either ADRs or no drug response. Predictive CYP2D6 genotyping is estimated to be beneficial for treatment of about 30 to 40% of CYP2D6 drug substrates representing about 7 to 10% of all clinically used drugs, although prospective clinical studies are needed to determine the exact benefit of drug selection and dosage based on the CYP2D6 genotype. Furthermore, Sanderson, et al. (2005) assessed the strength and quality of existing evidence about CYP2C9 gene variants and clinical outcomes in warfarin-treated patients in a meta-analysis (11 studies with a total of 2775 patients). These investigators concluded that patients with CYP2C9*2 and CYP2C9*3 alleles have lower mean daily warfarin doses and a greater risk of bleeding. The authors also stated that while testing for gene variants could potentially alter clinical management in patients commencing warfarin, evidence for the clinical utility and cost-effectiveness of genotyping is necessary before routine testing can be recommended.
In a review article on “Drug Metabolism and Variability among Patients in Drug Response” published in the New England Journal of Medicine, Wilkinson (2005) stated that “there have not yet been prospective clinical trials showing that knowledge of a patient’s genotypic profile before prescribing drugs either increases drug efficacy, prevents or reduces adverse drug reactions, or lower the overall costs of therapy and associate sequelae …. For now, however, the individual patient is probably best served by an alert physician aware of the possibility that a genetic polymorphism in drug metabolism may be a potential factor in unexpected drug response”.
In a multi-center 6-week study, Fux, et al. (2005) examined the impact of CYP2D6 polymorphism on the tolerability of metoprolol in a primary care setting. The adverse effects studied comprised effects related to the central nervous system, cardiovascular effects, and sexual dysfunction. The dosage of metoprolol was determined on an individual basis and could be adjusted on clinical grounds. The indication for treatment was hypertension in about 90 % of cases. CYP2D6 genotyping covered alleles *3 to *10 and *41 and the duplications. Possible ADRs of metoprolol were systematically assessed over the study period using standardized rating scales and questionnaires. The final study population comprised 121 evaluable patients: 5 ultrarapid metabolizers (UMs) (4.1%), 91 extensive metabolizers (EMs) (75%), 21 intermediate metabolizers (IMs) (17%), and 4 poor metabolizers (PMs) (3.3%). Plasma metoprolol concentrations normalized for the daily dose and metoprolol/alpha-hydroxymetoprolol ratios at steady state were markedly influenced by CYP2D6 genotype and displayed a gene-dose effect. The median of the dose-normalized metoprolol concentration was 0.0088 ng/ml, 0.047 ng/ml, 0.34 ng/ml, and 1.34 ng/ml among UMs, EMs, IMs, and PMs, respectively (p < 0.0001). There was no significant association between CYP2D6 genotype-derived phenotype (EMs and UMs combined versus PMs and IMs combined) and ADRs during treatment with metoprolol. There was a tendency toward a more frequent occurrence of cold extremities in the PM plus IM group as compared with the EM plus UM group (16.0% versus 4.2%, p = 0.056; relative risk, 3.8 [95% confidence interval, 1.03 - 14.3]). These investigators concluded that CYP2D6 genotype-derived phenotype was not significantly associated with a propensity for ADRs to develop during treatment with metoprolol. However, the results concerning tolerability of metoprolol in PMs were inconclusive because of the small number of PMs enrolled.
The mechanisms of variable response to tamoxifen have been the subject of much scrutiny in the published literature. Early studies attempting to link a clinical response to tamoxifen therapy with plasma tamoxifen concentrations reported no statistically significant differences in outcomes between women who received 20 mg of tamoxifen daily and those who received 40 mg of tamoxifen daily, even though women in the 40 mg tamoxifen group had higher plasma tamoxifen concentrations than those in the 20 mg tamoxifen group. These results have been reported as evidence that plasma tamoxifen concentration is not a predictor of clinical outcome. Because there is evidence that tamoxifen is converted to anti-estrogenic metabolites, one hypothesis is that altered patterns of metabolism of tamoxifen might contribute to inter-individual variability in effects (Jin, et al., 2005). Plasma concentrations of the active tamoxifen metabolite endoxifen are associated with the cytochrome P450 (CYP) 2D6 genotype.
Goetz, et al. (2005) stated that polymorphisms in tamoxifen metabolizing genes affect the plasma concentration of tamoxifen metabolites, but their effect on clinical outcome is unknown. These investigators determined cytochrome P450 (CYP)2D6 (*4 and *6) and CYP3A5 (*3) genotype from paraffin-embedded tumor samples and buccal cells (living patients) in tamoxifen-treated women enrolled onto a North Central Cancer Treatment Group adjuvant breast cancer trial. The relationship between genotype and disease outcome was determined using the log-rank test and Cox proportional hazards modeling. Paraffin blocks were obtained from 223 of 256 eligible patients, and buccal cells were obtained from 17 living women. CYP2D6 (*4 and *6) and CYP3A5 (*3) genotypes were determined from 190, 194, and 205 patient samples and in 17 living women. The concordance rate between buccal and tumor genotype was 100 %. Women with the CYP2D6 *4/*4 genotype had worse relapse-free time (RF-time; p = 0.023) and disease-free survival (DFS; p = 0.012), but not overall survival (p = 0.169) and did not experience moderate to severe hot flashes relative to women heterozygous or homozygous for the wild-type allele. In the multi-variate analysis, women with the CYP2D6 *4/*4 genotype still tended to have worse RFS (hazard ratio [HR], 1.85; p = 0.176) and DFS (HR, 1.86; p = 0.089). The CYP3A5*3 variant was not associated with any of these clinical outcomes. The authors concluded that in tamoxifen-treated patients, women with the CYP2D6 *4/*4 genotype tend to have a higher risk of disease relapse and a lower incidence of hot flashes, which is consistent with their previous observation that CYP2D6 is responsible for the metabolic activation of tamoxifen to endoxifen. They noted that "[t]hese findings have the potential to improve the ability of physicians to select the optimal hormonal therapy for the treatment of ER-positive breast cancer. Further studies are needed in women receiving tamoxifen to fully define the effect of CYP2D6 genetic polymorphisms and medications that inhibit CYP2D6 on tamoxifen response."
An assessment of CYP2D6 pharmacogenomics of tamoxifen treatment conducted by the BlueCross BlueShield Association Technology Evaluation Center (2008) evaluated the evidence for CYP2D6 genotyping and tamoxifen treatment efficacy. The hypothesis examined by the assessment is that CYP2D6 poor metabolizers, whether by genotype or by co-administration of CYP2D6 inhibitory medication, have reduced tamoxifen metabolism and lower endoxifen levels compared to better metabolizers, and as a result have poorer clinical outcomes. The reviewers stated that this hypothesis is based on the assumption, not yet supported by evidence, that some level of endoxifen is necessary for tamoxifen efficacy and that this level is not achieved in genotypic and functional CYP2D6 poor metabolizers and possibly not in some intermediate metabolizers. However, the reviewers found no scientific evidence for a significant association between endoxifen and clinical outcomes. In addition, they reported several limitations on the association of genotype with clinical outcomes. The reviewers stated, "Because tamoxifen metabolism is complex and CYP2D6 does not appear to account for all variability in endoxifen levels, it is conceivable that polymorphisms in other tamoxifen metabolic pathway enzymes may affect active metabolite levels, and direct measurement of the metabolite(s) itself may be the better predictor of benefit from tamoxifen treatment. However, since it takes eight weeks for tamoxifen metabolites to reach steady-state concentrations, measuring metabolite levels is not practical for clinical applications outside of a retrospective study." Furthermore, the reviewers noted that multiple enzyme genotypes may be needed to confidently predict tamoxifen versus aromatase inhibitors treatment benefit; however, there are little data at present to recommend any genotype combinations. The assessment concluded, "There is insufficient evidence to permit conclusions regarding the use of CYP2D6 genotyping for directing endocrine therapy regimen selection for women at high risk for or with breast cancer."
In a review on pharmacogenomics and individualized drug therapy, Eichelbaum, et al. (2006) stated that "[t]here is also a growing list of genetic polymorphisms in drug targets that have been shown to influence drug response. A major limitation that has heretofore moderated the use of pharmacogenetic testing in the clinical setting is the lack of prospective clinical trials demonstrating that such testing can improve the benefit/risk ratio of drug therapy". Furthermore, Humphries and Hingorani (2006) noted that the full potential of the field of pharmacogenetics will only be realized with much further work.
An assessment of CYP450 genetic testing by the Canadian Coordinating Office for Health Technology Assessment (Palylyk-Colwell, 2006) concluded: "No published studies show that patient outcomes can be predicted or altered by knowledge of DME status in the absence of other confounding variables. Prospective studies are needed to assess the benefits and potential risks of this technology in guiding drug selection and dose adjustment."
An assessment published by the Canadian Agency for Drugs and Technology in Health (Ndegwa, 2007) on "Pharmacogenomics and Warfarin Therapy" concluded that "[p]rospective studies are needed to determine whether pharmacogenomic testing improves patient outcomes, identify which subgroups of patients may benefit, and clarify the risks and costs associated with the use of these tests. Several randomized controlled trials are currently evaluating the impact of pharmacogenomics on dosing accuracy, time to achieve and maintain target international normalized ratio (INR), incidence of bleeding or thromboembolic events, and monitoring requirements".
In August 2007, the FDA updated the product label for warfarin to include information on the impact of genetic variations in CYP2C9 and VKORC1 on wafarin metabolism; however the FDA does not require the use of these genetic tests in dosing individual patients initiating warfarintherapy. On January 28, 2008, the FDA cleared the Infiniti 2C9-VKORC1 Multiplex Assay (AutoGenomic, Inc., Carlsbad, CA) for detection of Warfarin sensitivity. Guidelines for pharmacogenomics-based warfarin dosing are under development.
A technology assessment by the California Technology Assessment Forum (CTAF, 2008) reviewed the scientific evidence for the use of genetic testing to guide the initial dosing of warfarin when initiating therapy for anticoagulation. The assessment stated that genotyping studies of patients at a stable, therapeutic dose of warfarin demonstrated that patients who have CYP2C9*2 and CYP2C9*3 require lower doses of warfarin on average than patients with the more common CYP2C9*1 allele; in addition, patients with the A haplotype of VKORC1 require lower doses of warfarin on average than patients with the B haploytpe. The assessment reported that these genetic variations have been shown to predict an increased risk of excessive anticoagulation and major bleeding among patients prescribed warfarin and that statistical models have been developed in an attempt to predict the dose of warfarin needed to achieve stable anticoagulation. The assessment examined the results of a small case series and three small, randomized clinical trials and reported that the model used by the case series did reasonably well at predicting the warfarin dose, but that patients with the CYP2C9*2 and CYP2C9*3 alleles were still at more than a four-fold risk of excessive anticoagulation; only one of the randomized trials used a model incorporating genotyping information from both CYP2C9 and VKORC1 and that study found almost no difference in outcomes between patients receiving warfarin using a pharmacogentic model and those treated according to a standard approach. The assessment stated, "Significant uncertainty remains in the field. There is no widely accepted, standard pharmacogenetic model to determine the starting dose of warfarin and new models are being developed in 2008. Current models only explain 50% - 60% of the variability in warfarin dosing and the remaining variability is unexplained. Only one of the randomized trials used a model incorporating information from both genes with variants known to influence warfarin dosing. Furthermore, all of the trials to date have been underpowered to meaningfully evaluate the effect of genotyping on major bleeding, the most important clinical outcome." The assessment concluded that the use of genetic testing to guide initial warfarin dosing does not meet Technology Assessment Criteria 3 through 5 for safety, effectiveness and improvement in health outcomes. Several large clinical trials are ongoing in both the United Stated and Europe to clarify the role of genetic testing in warfarin management.
Since their introduction in the late 1980s, SSRIs such as citalopram, fluoxetine, paroxetine, and sertraline have become the most commonly prescribed class of drugs for treating depression. However, the likelihood that a patient will experience relief from all symptoms of depression after 1 year of treatment is approximately 40 %, and adverse events cause 12 to 15 % of patients who start treatment to stop taking the drug. Following the recent FDA approval of a test to predict differences in the CYP450 gene, physicians and patients must decide if using such tests to choose a type or dose of an SSRI might improve the patient's response to treatment.
The Agency for Healthcare Research and Quality (AHRQ, 2006) released a new evidence report that found there is insufficient evidence to determine if current gene-based tests intended to personalize the dose of SSRIs improve patient outcomes or aid physicians or patients in making treatment decisions. The available studies indicated that the tests are largely accurate at evaluating differences in genes belonging to the CYP450 family that affect the rate at which a person metabolizes SSRIs. However, additional well-designed studies are needed to determine the usefulness of test results in the clinical setting. This report is the first step in the 2-step process of Centers for Disease Control and Prevention's Evaluation of Genomic Applications in Practice and Prevention (EGAPP) pilot project to evaluate and make recommendations regarding the use of gene-based tests.
The Evaluation of Genomic Applications in Practice and Prevention Working Group (EGAPP, 2007) found insufficient evidence to support a recommendation for or against use of cytochrome P450 (CYP450) testing in adults beginning SSRI treatment for non-psychotic depression. In the absence of supporting evidence, and with consideration of other contextual issues, EGAPP discourages use of CYP450 testing for patients beginning SSRI treatment until further clinical trials are completed. EGAPP found that use of genetic testing for CYP450 polymorphisms and impact on physician decision-making with regard to use of SSRIs is not known. EGAPP noted that, in the absence of evidence supporting clinical utility, widespread use of CYP450 genetic testing is potentially costly and may not lead to changes in treatment that improve patient outcomes.
The eradication rates of Helicobacter pylori by triple therapy consisting of a proton pump inhibitor and two antimicrobial agents are mainly influenced by bacterial susceptibility to antimicrobial agents and the magnitude of acid inhibition during treatment with a proton pump inhibitor (e.g., omeprazole, lansoprazole, rabeprazole). Tailored therapy using CYP2C19 pharmacogenomics with a proton pump inhibitor has been proposed as a method to help improve the efficacy of H. pylori eradication rates.
A review on the use of pharmacogenomics-based treatment of H. pylori infection conducted by the BlueCross BlueShield Association Technology Evaluation Center (2008) examined the scientific evidence of a pharmacogenomics-based treatment regimen for the eradication of H. pylori. The reviewers found one randomized, controlled study that met their inclusion criteria. The reviewers reported that this study found higher eradication rates after first-line treatment for the pharmacogenomics group compared with the standard treatment group, however, because of numerous variations in treatment protocol within the pharmacogenomics group, it was not possible to determine whether the improvement resulted from the tailored proton pump inhibitor dosages according to CYP2C19 genetic status, or if it was due to other variations in the treatment protocol unrelated to CYP2C19 status. Furthermore, the review noted that it was possible other clinical factors, such as clarithromycin resistance, or other treatment factors, such as length of antibiotic treatment, influenced eradication rates. In addition, the study was performed in a Japanese population and did not employ a diagnostic approach or a treatment regimen that is standard care in the United States. The review concluded, "The scientific evidence does not permit conclusions on whether the use of a pharmacogenomics-based treatment regimen for H. pylori improves eradication rates."
Genotyping for VKORC1:
Knowing how an individual will respond to warfarin would help in tailoring the dose needed to maintain appropriate anticoagulation. Toward that end, researchers studied variability in the recently discovered gene for the warfarin target, vitamin K epoxide reductase complex 1 (VKORC1). VKORC1 is the key enzyme of the vitamin K cycle and the molecular target of coumarins, which represent the most commonly prescribed drugs for therapy and prevention of thromboembolic conditions. Recent studies have identified variants of the VKORC1 gene as responsible for about one quarter of the inter-individual response to warfarin, and for significant inter-ethnic response variability. Whether or not this is sufficient to successfully direct initial dosing, achieve a shorter time to stable dose, and reduce bleeding events has yet to be shown in a prospective trial. There are several such prospective clinical studies that are currently ongoing both in the United States and Europe.
Wadelius and Pirmohamed (2007) explained that the most important genes affecting the pharmacokinetic and pharmacodynamic parameters of warfarin are CYP2C9 and VKORC1. These two genes, together with environmental factors, partly explain the inter-individual variation in warfarin dose requirements.
Although studies have shown that genetic polymorphisms in CYP2C9 and VKORC1 affect warfarin dosing, no randomized controlled trials have linked the use of pharmacogenomic testing to improvements in clinical outcomes. An assessment by the Canadian Agency for Drugs and Technologies in Health (Ndegwa, 2007) found that most of the studies performed to date have been of retrospective or cross-sectional design. Consequently, individuals who stop warfarin early because of adverse effects or those who have difficulty attaining a therapeutic maintenance dose may have been excluded. Furthermore, many studies were underpowered to investigate the risk of bleeding.
Rieder, et al. (2005) analyzed genetic data from 186 American patients of European descent who were recruited from anticoagulation clinics and were receiving long-term warfarin therapy. They identified 10 single-nucleotide polymorphisms of VKORC1 that showed significant associations with warfarin maintenance doses and that had an overall frequency of more than 5% in this cohort. From these 10 single-nucleotide polymorphisms, the researchers inferred five common VKORC1 haplotypes (a haplotype is a set of closely linked genetic markers present on one chromosome that tend to be inherited together). Four of the haplotypes had independent associations with warfarin maintenance doses, two with a low-dose requirement and two with a high-dose requirement. Ultimately, subjects were linked with one of three haplotype groupings; the mean daily warfarin maintenance dose differed significantly across these three subgroups (2.7 mg, 4.9 mg, and 6.2 mg, respectively). The authors report that VKORC1 haplotype explained 25% of the variance in warfarin dose, and they replicated these findings in a larger European American population. The researchers also examined VKORC1 haplotype frequencies in Asian American and African American populations and found significant variability by race.
The U.S. Food and Drug Administration (FDA) has cleared for marketing the Verigene warfarin metabolism test, manufactured by Nanosphere, which detects variants of two genes, CYP2C9 and VKORC1, that can contribute to changes in warfarin metabolism (FDA, 2007).
Large ongoing studies of genes involved in the actions of warfarin, together with prospective assessment of environmental factors, will increase the capacity to accurately predict warfarin dose. Kamali (2006) stated that prospective studies that incorporate both CYP2C9 and VKORC1 genes and environmental factors in warfarin dose calculation will be needed to demonstrate the safety, cost-effectiveness, and feasibility of individualized dosing regimens.
ECRI Institute's Health Technology Trends (2007) reported that "[a]lthough genetic testing can currently identify who has these variants [CYP2C9 and VKORC1], more studies are needed to explore the precise starting does for these patients".
Anderson, et al. (2007) stated that pharmacogenetic-guided dosing of warfarin is a promising application of "personalized medicine" but has not been adequately examined in randomized studies. In a randomized trial, these investigators assessed genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation (n = 206). Buccal swab DNA was genotyped for CYP2C9 *2 and CYP2C9 *3 and VKORC1C1173T with a rapid assay. Standard dosing followed an empirical protocol, whereas pharmacogenetic-guided dosing followed a regression equation including the 3 genetic variants as well as age, sex, and weight. Prothrombin time INR was measured routinely on days 0, 3, 5, 8, 21, 60, and 90. A research pharmacist unblinded to treatment strategy managed dose adjustments. Patients were followed-up for up to 3 months. Pharmacogenetic-guided predicted doses more accurately approximated stable doses (p < 0.001), resulting in smaller (p = 0.002) and fewer (p = 0.03) dosing changes and INRs (p = 0.06). However, percent out-of-range INRs (pharmacogenetic = 30.7 %, standard = 33.1 %), the primary end point, did not differ significantly between arms. Despite this, when restricted to wild-type patients (who required larger doses; p = 0.001) and multiple variant carriers (who required smaller doses; p < 0.001) in exploratory analyses, results (pharmacogenetic = 29 %, standard = 39%) achieved nominal significance (p = 0.03). Multiple variant allele carriers were at increased risk of an INR of greater than or equal to 4 (p = 0.03). The authors concluded that an algorithm guided by pharmacogenetic and clinical factors improved the accuracy and efficiency of warfarin dose initiation. Despite this, the primary end point of a reduction in out-of-range INRs was not achieved. In subset analyses, pharmacogenetic guidance showed promise for wild-type and multiple variant genotypes.
The Invader UGT1A1 Molecular Assay:
Colorectal cancer (CRC) is one of the most common malignancies in western countries showing an increasing incidence, and has been associated with genetic as well as lifestyle factors. About 150,000 new cases of CRC are diagnosed each year in the United States, 40 to 50 % of which are metastatic. Irinotecan (Camptosar) is a chemotherapeutic agent approved as a combination therapy with 5-fluorouracil/leucovorin for the treatment of advanced CRC. The response to irinotecan is variable, possibly because of individuals' variation in the expression of the enzymes that metabolize irinotecan. Although multiple genes may play a role in irinotecan activity, the uridine diphosphate glycuronosyltransferase 1 family, polypeptide A1 (UGT1A1) enzyme has been strongly associated with irinotecan-related toxicity. The UGT1A1 gene is responsible for glucuronidation of the active metabolite of irinotecan. A common di-nucleotide repeat polymorphism in the UGT1A1 promoter region (UGT1A1*28) has been correlated with toxicity in cancer patients receiving irinotecan-containing therapy.
Lentz, et al. (2005) stated that hepatic metastases occur in about 50 % of patients with CRC. Since hepatic metastases are often inaccessible for surgery, chemotherapy of metastases is important. The most commonly used chemotherapeutic agents for hepatic metastases are fluorouracil, irinotecan, and oxaliplatin. Several enzymes are known to be involved in the metabolism of these drugs, and the activity of these enzymes varies greatly between individuals. The causes of this variation include genetic polymorphisms, different regulation between normal and cancer tissue, and the influence of chemotherapy on enzyme expression. The varying enzyme activity may have an important effect on the outcome of chemotherapy. Lentz, et al. (2005) reported that several studies confirm the influence of the activity of thymidylate synthase, thymidine phosphorylase and dihydropyrimidine dehydrogenase on the outcome of fluorouracil therapy for CRC, with higher enzyme activities predicting lower treatment efficacy. Although fewer studies are available regarding therapy of hepatic metastases, the same relationship between thymidylate synthase activity and outcome of fluorouracil therapy observed for primary CRC was found. For the other two enzymes, only a few studies are available, but the results indicate similarly that higher enzyme activity seems to be disadvantageous. Enzymes that are responsible for the activation, metabolism, and mechanism of action of irinotecan (e.g., CYP 3A4 and UGT1A1) also exhibit variable inter-individual activity. The authors concluded therefore, that there may be an association between enzyme activity and response to therapy. For example, in patients with CRC, higher enzyme activity of topoisomerase-I seems to be predictive of a better response to irinotecan. CYP3A4 and UGT1A1 activity levels might be predictive of irinotecan toxicity rather than efficacy. These authors stated that available data indicate the importance of the different enzyme activities on the outcome of chemotherapy of hepatic metastases in CRC. The authors noted that more information is needed, especially for the newer drugs irinotecan and oxaliplatin. However, the existing data are very promising in respect to the potential to guide dose and drug selection for more efficient and less toxic chemotherapy of hepatic metastases from CRC.
The Invader UGT1A1 molecular assay (Genzyme Corporation, Cambridge, MA) was cleared by the FDA on August 22, 2005. The test can be performed before starting irinotecan therapy and is designed to identify patients who may be at risk for ADRs to the chemotherapeutic agent by detecting a genetic variation in the UGT1A1 gene.
A clinical study (Innocenti, et al., 2004) indicated that patients with one of these variations in the UGT1A1 gene have a significantly greater risk of experiencing drug-related toxicity from irinotecan than those without it. The product labeling for irinotecan (Camptosar) was updated to recommend that a reduced initial dose should be considered for patients homozygous for UGT1A1*28 allele, although the precise dose reduction in this group of patients is unknown (Waknine, 2005). The product labeling, however, does not include a recommendation for assessment of UGT1A1 status prior to initiation of irinotecan therapy.
In a prospective study, Innocenti and colleagues (2004) assessed the association between the prevalence of severe toxicity and UGT1A1 genetic variation. A total of 66 cancer patients with advanced disease refractory to other treatments received irinotecan 350 mg/m2 every 3 weeks. Toxicity and pharmacokinetic data were measured during cycle 1. UGT1A1 variants (-3279G>T, -3156G>A, promoter TA indel, 211G>A, 686C>A) were genotyped. The prevalence of grade 4 neutropenia was 9.5 %. Grade 4 neutropenia was much more common in patients with the TA indel 7/7 genotype (UGT1A1(*)28 homozygous) (3 of 6 patients; 50 %) compared with 6/7 (3 of 24 patients; 12.5 %) and 6/6 (0 of 29 patients; 0 %) (p = 0.001). The TA indel genotype was significantly associated with the absolute neutrophil count nadir (7/7 less than 6/7 less than 6/6, p = 0.02). The relative risk of grade 4 neutropenia was 9.3 (95 %) for the 7/7 patients versus the rest of the patients. Pre-treatment total bilirubin levels were significantly higher in patients with grade 4 neutropenia (0.83 +/- 0.08 mg/dL) compared to those without grade 4 neutropenia (0.47 +/- 0.03 mg/dL; p < 0.001). The -3156G>A variant seemed to distinguish different phenotypes of total bilirubin within the TA indel genotypes. The -3156 genotype and the SN-38 area under the concentration versus time curve were significant predictors of absolute neutrophil count nadir (r2 = 0.51). These investigators concluded that UGT1A1 genotype and total bilirubin levels are strongly associated with severe neutropenia, and could be used to identify cancer patients predisposed to the severe toxicity of irinotecan. Furthermore, the hypothesis that the -3156G>A variant is a better predictor of UGT1A1 status than the previously reported TA indel requires further testing.
In an accompanying editorial, McLeod and Watters (2004) raised questions regarding the findings of Innocenti et al (2004): (i) is the relative risk of grade 4 neutropenia in a patient with the UGT1A1(*)28 homozygous genotype the same after 300 - 350 mg/m2 every 3 weeks (9.3-fold risk) as after the 100 - 125 mg/m2 weekly regimen? and (ii) does this marker retain its predictive power when irinotecan is used as part of a combined treatment? These researchers also noted that it would be difficult to perform randomized controlled trials to ascertain if there is a predictable, safe, and effective dose of irinotecan that can be given to patients with the UGT1A1(*)28 homozygous genotype, or if physicians should choose a non-irinotecan-containing regimen, since only 10 % of all patients have this genotype. Moreover, studies are planned to determine the impact of dosage on the safety of irinotecan in patients with either the UGT1A1 6/6/ or 6/7 genotype, which are the 2 most common genotypes in the general patient population.
While several investigators (And, et al., 2000, Iyer, et al., 2002, and Innocenti, et al., 2004) reported that testing of patients carrying the UGT1A1*28 polymorphism may detect their susceptibility to irinotecan-related toxicity, others (Marcuello, et al., 2004, Carlini, et al., 2005, Dervieux, et al., 2005, and Eichelbaum, et al., 2005) have questioned its clinical value. In a clinical trial, Marcuello, et al. (2004) examined the influence of the UGT1A1 gene promoter polymorphism in the toxicity profile, in the response rate and in the overall survival (OS) in 95 patients with metastatic CRC treated with an irinotecan-containing chemotherapy. Genotypes were determined by analyzing the sequence of TATA box of UGT1A1 of genomic DNA from the patients. Clinical parameters and genotypes were compared by uni-variate and multi-variate statistical methods. The more frequent ADRs were asthenia (n = 34), diarrhea (n = 29) and neutropenia (n = 20). Severe diarrhea was observed in 7/10 (70 %) homozygous and 15/45 (33 %) heterozygous in comparison to 7/40 (17 %) wild-type patients (p = 0.005). These results maintained the statistical significance in logistic regression analysis (p = 0.01) after adjustment for other clinical relevant variables. The presence of severe hematological toxicity increased from wild-type patients to UGT1A1(*)28 homozygotes, but without achieving statistical significance. No relationship was found between the UGT1A1(*)28 genotypes and infection, nausea or mucositis. In uni-variate studies, patients with the UGT1A1(*)28 polymorphism showed a trend to a poorer OS (p = 0.09). In the multi-variate analysis, the genotype was not related to clinical response or to OS. These investigators stated that the role of the UGT1A1 genotype as a predictor of toxicity in patients with CRC receiving irinotecan demands the performance of randomized controlled studies to determine if genotype-adjusted dosages of the drug can help to establish safe and effective doses not only for patients with the UGT1A1(*)28 homozygous genotype, but also for those with the most common UGT1A1 6/6 or 6/7 genotype.
In a phase II clinical study, Carlini and co-workers (2005) examined whether germ-line polymorphisms within genes related to drug target (thymidylate synthase) or metabolizing enzymes (UGT isozymes) would alter response and toxicity to the combination of capecitabine plus irinotecan. A total of 67 patients with measurable CRC were treated with intravenous irinotecan (100 or 125 mg/m2) on days 1 and 8 and capecitabine orally (900 or 1,000 mg/m2, twice daily) on days 2 through 15 of each 3-week cycle. Genomic DNA was obtained from peripheral blood and genotyped using Pyrosequencing, GeneScan, and direct sequencing technologies. The overall objective response rate was 45 % with 21 patients (31 %) exhibiting grade 3 or 4 diarrhea and 3 patients (4.5 %) demonstrating grade 3 or 4 neutropenia in the first two cycles. Low enzyme activity UGT1A7 genotypes, UGT1A7*2/*2 (6 patients) and UGT1A7*3/*3 (7 patients), were significantly associated with anti-tumor response (p = 0.013) and lack of severe gastrointestinal toxicity (p = 0.003). In addition, the UGT1A9 -118 (dT)(9/9) genotype was significantly associated with reduced toxicity (p = 0.002) and increased response (p = 0.047). There were no statistically significant associations between UGT1A1, UGT1A6, or thymidylate synthase genotypes and toxicity or tumor response. The authors concluded that these data strongly suggest that UGT1A7 and/or UGT1A9 genotypes may be predictors of response and toxicity in CRC patients treated with capecitabine plus irinotecan. Specifically, patients with genotypes conferring low UGT1A7 activity and/or the UGT1A9 (dT)(9/9) genotype may be particularly likely to exhibit greater anti-tumor response with little toxicity. However, it is interesting to note that the allele frequencies of UGT1A7 gene in Taiwan Chinese are different from those in Caucasians and Japanese (Huang, et al., 2005).
In a recent review, Dervieux, et al. (2005) stated that several proofs of principle have established that pharmacogenetic testing for mutations altering expression and functions of genes associated with drug disposition and response can reduce the "trial-and-error" dosing and decrease the risk of ADRs. These proofs of principle include UGT1A1 and irinotecan therapy, as well as CYP450 2C9 and S-warfarin therapy. These evidences advocate for the prospective identification of mutations associated with drug response, serious ADRs and treatment failure. The authors stated that with the convergence of rising drug costs and evidence supporting the clinical benefits of pharmacogenetic testing, it will be important to demonstrate the improved net health outcomes attributed to the additional costs for this testing.
This in agreement with the observation of Eichelbaum, et al. (2005) who noted that there is also a growing list of genetic polymorphisms in drug targets that have been demonstrated to influence drug response. A major limitation that has moderated the use of pharmacogenetic testing in the clinical setting is the lack of prospective clinical trials showing that such testing can improve the benefit/risk ratio of drug therapy. Moreover, Gardiner and Begg (2005) stated that currently pharmacogenetic tests for drug metabolizing enzymes are rarely carried out in clinical practice, despite repeated claims that they may benefit patient care. They noted that "the only tests performed with any regularity in Australasia are for thiopurine methyltransferase and pseudocholinesterase, and CYP2D6 phenotyping in one center for patients on perhexilene. The low clinical utilization reflects a poor evidence base, unestablished clinical relevance and, in the few cases with the strongest rationale, a slow translation to the clinical setting".
Han, et al. (2006) determined if uridine diphosphate-glucuronosyltransferase 1A1, UGT1A7, and UGT1A9 polymorphisms affect the pharmacokinetics (PK) of irinotecan and treatment outcome of Korean patients with advanced non-small-cell lung cancer (NSCLC). A total of 81 patients with advanced NSCLC were treated with irinotecan (80 mg/m2) on day 1 and 8 and cisplatin (60 mg/m2) on day 1 intravenously of each 3-week cycle. Genomic DNA was extracted from peripheral blood and genotyped using direct sequencing. These researchers analyzed the association of UGT1A genotypes with irinotecan PK and clinical outcomes. All statistical tests were two-sided. In genotype-PK association analysis, UGT1A1*6/*6 (n = 6), UGT1A7*3/*3 (n = 6), and UGT1A9-118(dT)9/9 (n = 11) were associated with significantly lower area under the time-concentration curve (AUC) SN-38G to SN-38 (AUC(SN-38G)/AUC(SN-38)) ratio (p = 0.002, p =0 .009, and p = 0.001, respectively). In linkage disequilibrium analysis, the UGT1A7 variants were highly linked with the UGT1A1*6 (D' = 0.85, r2 = 0.63) and UGT1A9*22 (D'= 0.95, r2 = 0.88), which was substantiated in haplotype analysis. Patients with UGT1A1*6/*6 had lower tumor response and higher incidence of severe neutropenia. UGT1A9-118(dT)9/9 also showed a trend for high incidence of severe diarrhea, but not tumor response. In survival analysis, patients with UGT1A1*6/*6 had significantly shorter progression-free survival (p = 0.001) and overall survival (p = 0.017). The authors concluded that UGT1A1*6 and UGT1A9*22 genotypes may be important for SN-38 glucuronidation and associate with irinotecan-related severe toxicity. Specifically, UGT1A1*6 might be useful for predicting tumor response and survival outcome of Korean patients with NSCLC treated with irinotecan-based chemotherapy. These investigators also stated that "[a]lthough it is still hypothetical, we suggest that UGT1A1*6 and/or UGT1A9*22 genotypes might be important for predicting severe toxicity and treatment outcome after irinotecan-based chemotherapy. To confirm the data observed in this study, further larger studies are needed in an independent data set, preferably in a group of patients of similar ethnicity".
In an editorial that accompanied the study by Han, et al. (2006), Innocenti and associates (2006) stated that "[t]he study by Han et al provides evidence that the UGT1A1*6 polymorphism can be considered a biomarker of severe toxicity of irinotecan in Asians. The impact of this variant on efficacy in irinotecan-containing regimens should be prospectively investigated in patients of Asian descent….".
In a review on genetic polymorphisms of drug-metabolizing enzymes and drug transporters in the chemotherapeutic treatment of cancer, Bosch, et al. (2006) focused on the clinical significance of polymorphisms in drug-metabolizing enzymes (CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, CYP3A5, dihydropyrimidine dehydrogenase, UGT1A1, glutathione S-transferase, sulfotransferase [SULT] 1A1, N-acetyltransferase [NAT], thiopurine methyltransferase [TPMT]) and drug transporters (P-glycoprotein [multi-drug resistance 1], multi-drug resistance protein 2 [MRP2], breast cancer resistance protein [BCRP]) in influencing toxicity and effectiveness of chemotherapy. The authors stated that the clinical application of pharmacogenetics in cancer treatment will require more detailed information of the different polymorphisms in drug-metabolizing enzymes and drug transporters; and that larger studies, in different ethnic populations, and extended with haplotype and linkage disequilibrium analysis, are needed for each anti-cancer drug separately.
While initial reports described above suggested that UGT1A1*28 homozygotes were at high risk for worse irinotecan-related hematologic and gastrointestinal toxicity, more recent reports suggest that the magnitude of the problem (particularly the association with worse diarrhea) is not as great as was initially suspected. In a prospective study of 250 patients with metastatic colorectal cancer starting irinotecan, fluorouracil and leukovorin, the relative risk for grade 3 or 4 hematologic toxicity was significantly higher among UGT1A1*28 homozygotes (odds ratio 8.63, 95% confidence interval 1.31 to 56.55) (Toffoli, et al., 2006). However, the absolute magnitude of risk was relatively low (13.6 percent versus 1.7 percent for those with the wild-type alleles), and relevant for the first cycle only. Furthermore, there was no significant association between the presence of a UGT1A1*28 polymorphism and severity of diarrhea, or the need for irinotecan dose reduction.
One study found higher rates of neutropenia in persons homozygous for the UGT1A1*28 allele, regardless of whether the combination chemotherapy regimen included irinotecan (McLeod, et al., 2006). In a preliminary analysis of data from 520 patients with colorectal cancer enrolled in the United States Intergroup (INT) 9741 trial, which compared a variety of first-line oxaliplatin and irinotecan-containing chemotherapy regimens, the risk of grade 3 or 4 neutropenia was significantly higher for homozygotes (but not heterozygotes) regardless of whether they received irinotecan or oxaliplatin-based chemotherapy (36.2 percent versus 18.2 percent and 14.8 percent for homozygotes, heterozygotes, and wild-type alleles, respectively). Similar to the study by Toffoli, et al. (2006) described above, the investigators found no association between inheritance of UGT1A1*28 alleles and treatment-related diarrhea. UGT1A1*28 was also not a predictor of tumor response, time to progression, or overall survival.
In summary, although it has been advocated that pharmacogenetic testing of patients with CRC before chemotherapy with irinotecan may reduce the frequency of severe toxicities by allowing alternate therapy selections for patients carrying the UGT1A1*28 polymorphism, the clinical value of this testing (i.e., whether testing will lead to better health outcomes) has yet to be established by prospective, randomized, controlled trials. Only about 1 in 10 patients will be identified as being homozygous, and the excess risk of severe neutropenia that is attributable to the inheritance of this polymorphism appears to be small. There is a lack of consensus on whether initial dose reduction is needed for UGT1A1*28 homozygotes, and the precise dose reduction that is warranted in this patient population has not been determined.
Genotyping for HLA-B* 1502
Carbamazepine (brand names Carbatrol, Equetro and Tegretol) is used for the treatment of patients with epilepsy, bipolar disorder, and neuropathic pain. The use of carbamazepine is associated with a rare but severe and sometimes life-threatening skin reactions, which include toxic epidermal necrolysis and Stevens-Johnson syndrome, characterized by multiple skin lesions, blisters, fever, itching and other symptoms. The risk of these reactions is estimated to be about 1 to 6 per 10,000 new users of the drug in countries with mainly white populations. However, the risk is estimated to be about 10 times higher in some Asian countries. Studies have demonstrated a strong association between certain serious skin reactions and an inherited variant of an immune system gene, HLA-B* 1502, found almost exclusively among individuals with Asian ancestry (Hung, et al., 2006; Chung, et al., 2007).
In December 2007, the FDA announced that manufacturers of drugs containing carbamazepine have agreed to add to the drugs' labeling a recommendation that, before starting therapy with the drugs patients with Asian ancestry should receive genotyping of HLA-B* 1502.
HLA-B* 5701 Screening
Abacavir (Ziagen) is a nucleoside analogue reverse transcriptase inhibitor indicated for use in combination with other antiretroviral durgs for the treatment of HIV-1 infection. Review of reports of hypersensitivity in patients receiving abacavir (Glaxo Wellcome Inc., Research Triangle Park, NC) indicated that respiratory symptoms (including cough, dyspnea, and pharyngitis) have occurred in approximately 20 % of patients who have had hypersensitivity reactions. The frequency of the HLA-B*5701 allele varies in different populations, occuring in whites 5 - 8 %, Hispanics 4 - 7 %, Asians less than 1 %, Spaniards 1 - 4 %, and rarely in Sub-Saharan Africans. A delay in diagnosis of hypersensitivity can result in abacavir being continued or re-introduced, leading to more severe hypersensitivity reactions, including life-treatening hypotension and death.
In a double-blind, prospective, randomized study, Mallal, et al. (2008) examined whether HLA-B*5701 screening could prevent hypersensitivity reaction to abacavir. Patients who were infected with HIV-1 infection (n = 1956) who had not previously received abacavir were randomly assinged to undergo prospective HLA-B*5701 screening, with exclusion of HLA-B*5701 positive patients from abacavir treatment (prospective screening group), or to undergo a standard of care approach of abacavir use without prospective HLA-B*5701 screening (control group). All patients who started abacavir were observed for 6 weeks. Epicutaneous patch testing with abacavir was performed to immunologically confrim and enhance the specificity of the clinical diagnosis of hypersensitivity reaction to abacavir. The prevalence of HLA-B*5701 in this predominantly white study population was 5.6 %. Hypersensitivity reaction was diagnosed in 93 patients with a significantly lower incidence in the prospective-screening group (3.4 %) than in the control group (7.8 %). Of the patients receiving abacavir, 72 % were men, 84 % were white, and 18 % had not previously received antiretroviral therapy. Screening eliminated immunologically confirmed hypersensitivity reaction (0 % in the prospective-screening group versus. 2.7 % in the control group), with a negative predictive value of 100 % and a positive predictive value of 47.9 %. The authors concluded that HLA-B*5701 screening reduced the risk of hypersensitivity reaction to abacavir. Although the population in Mallal's study was predominantly white, other investigators have reported comparable sensitivity results of HLA-B*5701 for abacavir hypersensitivity in different ethnic groups, including blacks (Saag M, et al., 2007) and Spaniards (Rodríguez-Nóvoa, et al., 2007). In addition, Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents (2008) developed by the Department of Health and Human Services recommends:
-
Screening for HLA-B*5701 before starting patients on an abacavir-containing regimen to reduce the risk of hypersensitivity reaction (strength of recommendation: strong evidence with at least one randomized trial with clinical results);
-
HLA-B*5701-positive patients should not be prescribed abacavir (strength of recommendation: strong evidence with at least one randomized trial with clinical results);
-
The positive status should be recorded as an abacavir allergy in the patient’s medical record (strength of recommendation: strong evidence with clinical trials with laboratory results);
-
When HLA-B*5701 screening is not readily available, it remains reasonable to initiate abacavir with appropriate clinical counseling and monitoring for any signs of hypersensitivity reaction (strength of recommendation: optional with expert opinion).
|
Close Window
