Basic & Clinical Pharmacology & Toxicology, 2015, 116, 438–444
Doi: 10.1111/bcpt.12341
Pharmacogenomic Biomarker Information in FDA-approved Paediatric Drug Labels Therasa Kim1, Nayoung Han2, Minji Sohn2, Jung Mi Oh2, Eui-Kyung Lee3, Eunhee Ji4 and In-Wha Kim2 1 College of Pharmacy, Chonnam National University, Gwangju, South Korea, 2College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul, South Korea, 3School of Pharmacy, Sungkyunkwan University, Suwon, Gyeonggi-do, South Korea and 4College of Pharmacy, Gachon University, Incheon, South Korea
(Received 16 August 2014; Accepted 6 October 2014) Abstract: Gene maturation differs between paediatric and adult populations, and the extrapolation of adult pharmacogenomic information to paediatrics is not always appropriate. We sought to determine the extent of paediatric pharmacogenomic trial translation into US FDA-approved labels and to evaluate needs for biomarker studies. Using FDA’s Table of Genomic Biomarkers and Drugs@FDA website, 38 pharmacogenomic biomarkers in 56 drug labels were identified with possible application in paediatrics. Of these 56 drugs, biomarker comparison against ‘Very Important Pharmacogenes (VIPs)’ defined in PharmGKB’s database revealed a total of eight VIPs labelled among 41 drugs. One hundred and thirty-nine product reviews posted on the FDA website under the Best Pharmaceuticals for Children Act and Paediatric Research Equity Act between October 2007 and July 2014 were examined. Review screening identified 43 drugs with ‘pharmacogenomic’ content, of which only three were true genotyping study reviews for proton pump inhibitors, all evaluating CYP2C19 polymorphisms. Pantoprazole was the sole drug labelled with pharmacogenomic information obtained specifically from paediatric trials. Clinicaltrials.gov was searched to further evaluate the current availability of pharmacogenomic studies in the paediatric population. Of the 33,132 trials registered on Clinicaltrials.gov, 137 were labelled as paediatric pharmacogenetic and pharmacogenomic studies. Pharmacogenomic studies directly conducted in paediatric patients are lacking, and thus, pharmacogenomic biomarker information based on adult studies is commonly presented in FDA-approved labels for use in paediatric patients. Considering differences in gene expression and physiological maturation between paediatric and adult populations, studies investigating pharmacogenomic effects specifically in paediatric patients should be conducted whenever significant biomarkers are available.
In a new era of genome-based personalized medicine, pharmacogenetic or pharmacogenomic information has the potential to predict variability in the pharmacokinetics and pharmacodynamics of therapeutic drugs, contributing to the optimization of drug efficacy and avoidance of adverse drug effects [1]. Today, open-source databases such as the Drugs@FDA [2] and PharmGKB [3,4] websites provide pharmacogenomic information that can be utilized to improve efficacy and safety of drug use [5]. This information is primarily derived from adult populations and has limitations in its application to the paediatric population. Diseases differ between adults and children across a wide range of ages, suggesting that the relationships between genotype and phenotype will differ between adults and children, and also within the paediatric population [6]. Physiological or metabolic maturation is a complex and continuous process that occurs from the prenatal period to mid-childhood to adolescence [7], and could modify pharmacogenomic effects at any given stage of development. Consequently, the extrapolation of adult pharmacogenomic information to paediatric patients may most likely not be straightforward, especially for neonates and infants [8]. Age-dependent changes in pharmacokinetic Author for correspondence: In-Wha Kim, College of Pharmacy, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 151-742, South Korea (fax +82 2 880 9560, e-mail [email protected]).
and pharmacodynamic parameters play a crucial role in paediatric drug response, and a pharmacogenomic approach to the treatment of paediatric disease requires an understanding of gene variation in relation to age [9,10]. Considering such heterogeneity, the incorporation of pharmacogenomics into paediatric drug development studies is essential to ensure appropriate application of genomic parameters. Many of the currently approved drugs available on the market have not been adequately studied in paediatric populations, and an even fewer number of pharmacogenomic paediatric studies are available. However, both the United States and the European Union now require paediatric studies as part of the drug approval process, which facilitates the participation of children in clinical drug development studies [11]. Consequently, a relatively large number of paediatric studies have been conducted and reported since the passage of the Food and Drug Administration Amendments Act (FDAAA) of 2007 [12]. In 2012, the Food and Drug Administration Safety and Innovation Act (FDASIA) brought permanency to the Best Pharmaceuticals for Children Act (BPCA) and the Paediatric Research Equity Act (PREA) [13]. The Office of Paediatric Therapeutics at the United States FDA now provides medical, statistical and clinical pharmacology reviews of paediatric studies on their website [14]. With a growing body of information on genomic biomarkers, the incorporation of such pharmacogenetic information
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
into drug labels and its appropriate use is becoming increasingly important in paediatric patients as well. Thus, theoretically, a greater number of paediatric pharmacogenomic studies validating the clinical application of biomarkers in this population should also become available. In this study, we sought to (i) identify drugs approved in the paediatric population with pharmacogenomic biomarker information available in the FDA-approved drug label, (ii) compare labelled pharmacogenomic markers to ‘Very Important Pharmacogenes (VIPs)’ defined in PharmGKB, (iii) review the influence of BPCA and PREA on the conduction and incorporation of pharmacogenomic studies in paediatric drug labels and (iv) survey the current status of paediatric pharmacogenetic and pharmacogenomic studies available. Based on these analyses, we ultimately aimed to provide an evaluation of potential needs present in paediatric pharmacogenomic biomarker studies. Method Analysis of pharmacogenomic information in paediatric FDAapproved drug labels. The FDA website provides the ‘Table of Pharmacogenomic Biomarkers in Drug Labelling’, listing FDAapproved drugs with pharmacogenomics information in their prescription label [15]. Drug labels for medications in this table were searched using Drugs@FDA [2] to (i) determine the approval for use in the paediatric population and (ii) extract pharmacogenomics information for the identified paediatric drugs in a two-step process. In the first stage, the labels were screened for the occurrence of one or more of the following key words: gene, genetic, genotype, DNA, RNA, variant, haplotype, polymorphism, deficiency and SNP. In the second stage, labels identified were further screened based on a set of selection criteria that identified descriptions of human genetics or genomics such as gene expression changes, gene deficiencies, deletions or insertions, genetic mutations or genetic polymorphisms and variations. Topical and diagnostic medications were excluded from the analysis, and drugs with multiple formulations or multiple indications were included only once in the overall label count. To conduct a comparison of the labelled pharmacogenomics biomarkers to established VIPs, the PharmGKB website (http://www. pharmgkb.org/index.jsp) was used to search for pharmacogenomics genes that were associated with the drugs identified [4]. PharmGKB’s database provides genomic information based on various annotations, literature reviews, pathways and information automatically retrieved from DrugBank [16]. ‘VIPs’ are defined in PharmGKB as genes having extensive demonstrated relationships between variants and drug response or metabolism and are expected to have clinical significance [17]. There are 54 VIPs identified in PharmGKB’s database, including genes such as CFTR, CYP2C19, CYP2C9, G6PD, HLA-B, TPMT and VKORC1. Pharmacogenomics biomarkers identified in the drug labels were compared specifically against VIPs reported for these agents. Analysis of pharmacogenomics studies in the paediatric population. Publically available medical, statistical and clinical pharmacology reviews of paediatric studies, conducted under Section 505A and 505B of the Federal Food, Drug and Cosmetic Act as amended in 2007 [12] and Section 505A and 505B of the Federal Food, Drug and Cosmetic Act as amended in 2012 [13], were analysed to identify trials pertaining to pharmacogenomics. Studies were classified as performed in accordance with BPCA, PREA or both legislations. The aforementioned methodology employed to screen drug labels was again utilized to screen reviews, and
439
‘pharmacogenomic’ studies identified were further inspected to evaluate the incorporation of paediatric pharmacogenomic information into drug labels. In addition, pharmacogenomic trials conducted under BPCA and/or PREA were identified for paediatric approved drugs discovered under the FDA ‘Table of Pharmacogenomic Biomarkers in Drug Labelling’. Only paediatric drug studies submitted to the FDA between October 2007 and July 2014 were included in the analysis. To further determine the overall availability of paediatric pharmacogenomic studies, the Clinicaltrials.gov website (http:// clinicaltrials.gov/) was searched for paediatric trials using the key terms ‘pharmacogenetics’ and ‘pharmacogenomics’ [18,19].
Results Pharmacogenomic biomarkers in FDA-approved drug labels approved for paediatrics. A total of 140 unique drugs were identified in the FDA Table of Genomic Biomarkers after excluding topical, diagnostic and duplicate active ingredients. Of these, 56 drugs (40.0%) were determined to be approved for use in the paediatric population and listed Pharmacogenomics genes of interest in the FDAapproved labels (Table 1). For these 56 agents, a total of 38 pharmacogenomic biomarkers were available with the cytochrome P450 (CYP) enzyme family being identified most frequently (CYP2D6 – 17.1%, CYP2C19 – 13.2%, CYP2C9 – 2.63%) and G6PD being the second most common biomarker. Comparison of FDA-approved pharmacogenomic biomarkers to PharmGKB. One hundred and seventeen of the above-mentioned 140 drugs were related to specific genes when searched for in PharmGKB. A total of 361 genes/genetics were documented with 31 genes classified as VIPs. The CYP enzyme family was identified most frequently (CYP3A4 – 3.58%, CYP2D6 – 3.21%, CYP2C19 – 2.96%), and ABCB1 was identified as the second most common pharmacogenomics-related gene. In the comparison of genomic biomarkers from FDA-approved labels with VIPs presented in PharmGKB, 41 of the 56 paediatric drugs had labelled biomarkers that were VIPs including HLAB, CYP2D6, CYP2C19, CYP2C9, G6PD, TPMT, F5 and VKORC1 genes (eight in total). Of the remaining 15 drugs, three did not have defined VIPs and 12 had labelled biomarkers that did not include VIP genes (table 2). Pharmacogenomic markers related to drug indication or hypersensitivity reactions were excluded in the summary. Influence of BPCA/PREA on paediatric pharmacogenomic drug labelling. After excluding topicals, diagnostics and duplicate active pharmaceutical ingredients, 139 drugs remained of the 195 products studied under BPCA and PREA as of July 2014. For these drugs, review screening resulted in identification of 43 drug reviews with ‘pharmacogenomics’ content, of which only three involved true genotyping studies for the proton pump inhibitors pantoprazole, lansoprazole and rabeprazole. CYP2C19 polymorphisms were evaluated for all three agents.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
THERASA KIM ET AL.
440
Table 1. Paediatric drugs with FDA-labelled pharmacogenomic biomarkers as of July 2014. A total of 38 biomarkers identified for 56 paediatric approved drugs. Therapeutic class Antidotes Cardiology Dermatology Endocrinology
Gastroenterology
Haematology
Infectious disease
Metabolic disorder Neurology
Oncology
Rheumatology Psychiatry
Pulmonary Reproductive Transplantation
Drug Sodium nitrite Carvedilol Metoprolol Tretinoin Dapsone Atorvastatin Pravastatin Rosuvastatin Esomeprazole Lansoprazole Omeprazole Pantoprazole1 Rabeprazole Succimer Warfarin Abacavir Chloroquine Maraviroc Nalidixic Acid Nitrofurantoin Quinine Sulphate Sulfamethoxazole and trimethoprim Voriconazole Carglumic acid Velaglucerase Alfa Carbamazepine Clobazam Phenytoin Arsenic Trioxide Busulfan Cisplatin Everolimus Fulvestrant Imatinib Mercaptopurine Rasburicase Tamoxifen Thioguanine Celecoxib Carisoprodol Aripiprazole Atomoxetine Clomipramine Diazepam Fluoxetine Fluvoxamine Paroxetine Perphenazine Pimozide Risperidone Thioridazine Valproic acid Venlafaxine Ivacaftor Drospirenone and ethinyl estradiol Mycophenolic acid
BPCA/PREA/ OTHER
Paediatric age in study
G6PD CYP2D6 CYP2D6 PML/RARA G6PD LDLR LDLR LDLR CYP2C19 CYP2C19 CYP2C19 CYP2C19 CYP2C19 G6PD VKORC1, CYP2C9, PROC HLA-B G6PD CCR5 G6PD G6PD G6PD G6PD
OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER PREA BPCA PREA BPCA, PREA PREA OTHER OTHER
All 2 months to 17 years 6–17 years 1–16 years 12–17 years 10–17 years 8–18 years 10–17 years 1 months to 17 years >1 year 1–16 years All 12–18 years >1 year All
PREA OTHER OTHER OTHER OTHER OTHER OTHER
3 months to 13 years All 16–18 years 3 months to 12 years >1 months 1.5–12 years >2 months
CYP2C19 NAGS GBA HLA-B, HLA-A CYP2C19 HLA-B PML/RARA Ph Chromosome TPMT ERBB2, ESR1 ESR1 KIT, BCR/ABL1, PDGFRB, FIP1L1/PDGFRA TPMT G6PD ESR1, PGR, F2, F5 TPMT CYP2C9 CYP2C19 CYP2D6 CYP2D6 CYP2D6 CYP2C19 CYP2D6 CYP2D6 CYP2D6 CPY2D6 CYP2D6 CYP2D6 CYP2D6 POLG, NAGS, CPS, ASS, OTC, ASL, ARG CYP2D6 CFTR (G551D) CYP2C19 HGPRT
OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER BPCA, PREA OTHER OTHER
>12 years All 4–17 years >6 months 2–17 years All 4–18 years All All 1–17 years 1–8 years 2–17 years
OTHER OTHER OTHER OTHER OTHER OTHER BPCA, PREA OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER PREA OTHER
All All 2–10 years All 2–17 years >16 years 1–17 years 6–15 years 10–17 years 2–17 years 7–17 years 8–17 years All >12 years 2–17 years 5–17 years All 10–18 years 12–18 years 6–17 years 14–18 years 5–16 years
Biomarker
BPCA, Best Pharmaceuticals for Children Act; PREA, Paediatric Research Equity Act. Paediatric pharmacogenomic BPCA/PREA trial resulting in label modification.
1
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
For paediatric patients who received pantoprazole, genotyping for CYP2C19 was conducted in six studies. Of 226 patients genotyped, six were poor metabolizers (CYP2C19*2/ *2). In these poor metabolizers, the pantoprazole dose-normalized AUC was more than six times higher than that of extensive (CYP2C19*1/*1) or intermediate (CYP2C19*1/x) metabolizers. The review team ultimately revised the pantoprazole paediatric label to include the following statement in ‘Pharmacogenomics Section 12.4’: ‘For known paediatric poor metabolizers, a dose reduction should be considered’. In the case of lansoprazole, the limited number of poor metabolizers and heterozygous extensive metabolizers in the neonatal study group precluded any conclusion on the relationship between the CYP2C19 genotype and lansoprazole pharmacokinetics. Additionally, CYP2C19 genotyping was performed for rabeprazole in adolescents and children 1–11 years of age. The lack of significant impact of CYP2C19 metabolizer status on the clearance of rabeprazole was consistent with current labelling, and as such, no labelling update was recommended. Of the 56 paediatric drugs with labelled pharmacogenomics biomarkers, only nine (16.1%) had paediatric studies that were conducted under BPCA and/or PREA (table 1). All other studies had been conducted in children and adolescences outside of the BPCA and PREA process. Of the nine drugs studied under paediatric trials for BPCA and/or PREA, only pantoprazole was labelled with pharmacogenomics information obtained specifically from paediatric trials. Pharmacogenomic information from paediatric trials for the remaining drugs did not translate to modifications in the label for reasons including non-significant difference from adult population, lack of information and non-significant association or clinical effect. Survey of pharmacogenetic/pharmacogenomic paediatric trials. From a total of 33,132 trials registered on Clinicaltrials.gov, 470 and 339 clinical trials containing pharmacogenetic and
441
pharmacogenomics analyses, respectively, were identified. One hundred and thirty-seven trials were of paediatric population, but only 60 mentioned the investigation of specific genes. Some examples include exploration of the association between morphine and CYP2D6 and UGT2B7, tacrolimus and CYP3A5, efavirenz and CYP2B6, and dexmedetomidine and CYP2A6. When sorted by therapeutic class, the majority of the trials (69%) examined drugs in the areas of oncology, psychiatry, analgesic, immunology and infectious disease (fig. 1). The age distribution of the study participants is as follows: 44 from birth, 2 from 1 month, 23 from 1 year, 29 from 6 years and 39 from 12 years of age.
Discussion Human development from the prenatal period through adolescence is a dynamic process. Newborns and infants demonstrate large variability in drug response and metabolizing capabilities [20]. The genetic effect of a polymorphism on adults may not be observed in paediatric patients, dependent upon a number of factors including the ontology of drug-metabolizing enzymes, transporters and receptors [21]. While the effect of genetic variations on drug metabolism is recognized, much less attention has been given to developmental pharmacogenomics, even though some genes are expressed much more abundantly in the earlier years of life compared to the adult years and others vice versa [10]. One of the most widely acknowledged examples of gene switching is represented by the CYP3A gene family. The level of CYP3A7 expression is detectable as early as in 50 days of gestation in foetal livers, and its expression begins to decline after the first post-natal week, reaching undetectable levels in most individuals by 1 year of age [22]. Conversely, hepatic CYP3A4/3A5 expression begins to dramatically increase at about 1 week of age, reaching 30% of adult levels by 1 month [22]. Accordingly, total CYP3A protein expression over the entire develop-
Table 2. Comparison of pharmacogenomic biomarkers in FDA labels to ‘Very Important Pharmacogenes (VIPs’) in PharmGKB for paediatric approved drugs. Twelve of 56 paediatric drugs are labelled with non-VIP genes. Drug
Biomarker
Arsenic Trioxide Atorvastatin
PML/RARA LDLR
Busulfan Everolimus Fulvestrant Imatinib
Ph Chromosome ERBB2, ESR1 ESR1 KIT, BCR/ABL1, PDGFRB, FIP1L1/PDGFRA CFTR (G551D) CCR5 HGPRT LDLR LDLR POLG, NAGS, CPS, ASS, OTC, ASL, ARG
Ivacaftor Maraviroc Mycophenolic acid Pravastatin Rosuvastatin Valproic acid
Very Important Pharmacogene KCNH2 ABCB1, ACE, CYP2C19, CYP2C8, CYP2C9, CYP2D6, CYP3A4, CY P3A5, HMGCR, NR1I2, SLCO1B1, UGT1A1 CYP2C19, GSTP1, MTHFR CYP2C8, CYP2D6, CYP3A4, CYP3A5 CYP3A4 PDGFRB, ABCB1, C YP2D6, CYP3A4, CYP3A5, SLC22A1 CYP3A4, CYP3A5 CYP3A4, CYP3A5 CYP2C8, CYP3A4, CYP3A5 ACE, ABCB1, CYP3A5, HMGCR, MTHFR, SLCO1B1 CYP2C19, CYP2C9, CYP3A4, CYP3A5, HMGCR, SLCO1B1 ABCB1, CYP2A6, CYP2B6, CYP2C19, CYP2C9, HLA-B, NR1I2
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
442
THERASA KIM ET AL.
Fig. 1. Paediatric pharmacogenetic/pharmacogenomic studies identified under Clinicaltrials.gov as of July 2014.
mental period remains constant. However, because CYP3A7 and CYP3A4 exhibit differences in substrate specificity and catalytic efficacy, inequalities in metabolic capacity during development are observed [22]. In addition, many drug-metabolizing enzymes are less developed in children than they are in adults. An example is UGT [23], a liver enzyme essential for the body’s elimination of xenobiotics and endogenous substances such as toxins and bilirubin. Only 1% of the normal adult level of hepatic UGT activity is present at birth [24]. Post-natal maturation of UGT is birth related and increases rapidly to adult levels by 14 weeks, regardless of the gestational age at birth, perhaps in relation to the methylation or acetylation status of the gene locus. Therefore, clinical pharmacogenomic trials in paediatric patients should be conducted whenever possible, especially when ontogeny may play a role in the impact of the established pharmacogenomic parameter. In this study, among the 140 drugs with pharmacogenomic biomarker information in the FDA-approved drug label, 56 drugs with paediatric indications were identified, involving 38 biomarkers that could potentially have application in paediatric patients as well. In our PharmGKB analysis, we conducted a comparison of labelled pharmacogenomic biomarkers to VIPs listed in PharmGKB’s database. However, this investigation does not evaluate the appropriateness of FDA-labelled biomarker suitability in paediatrics, but rather provides a summary of potential genes open to further investigation depending on clinical impact. Furthermore, caution should be taken in the interpretation of discrepancies between FDA-labelled genes and VIPs considering possible redundancy control in gene sets, which may reflect a broader selection of clinically significant biomarkers for labelling. Vivar et al. [25] have reported on the risk of ignoring overlap or redundancy in pathway databases and designating several pathways as significant when in fact, the biological mechanisms are not truly independent. Only nine of the 56 drugs were studied under BPCA and PREA in paediatric patients, with pantoprazole being the sole drug incorporating pharmacogenomic information from
paediatric studies into the product label. This suggests that product labelling of pharmacogenomic information on drug dosing, response and adverse drug reactions in paediatric patients is highly dependent on pharmacogenomic studies in adults. In similar context, a comprehensive screening of trials submitted under BPCA and PREA between October 2007 and July 2014 did not result in the detection of any additional paediatric trials leading to specific pharmacogenomic label modifications, and only three drugs involved true biomarker genotyping studies. Despite reauthorization and permanency of BPCA and PREA legislation, there appears to be a limited number of paediatric pharmacogenomic studies investigating the application of genomic biomarkers in this population. However, it is questionable whether this observation is an accurate reflection of study insufficiency, or whether there are actually a considerable number of such trials but they have been conducted outside of the BPCA and PREA process. There is also the possibility that these ‘OTHER’ paediatric study findings were non-significant and may have eliminated the need for additional trials to be conducted under such legislation. Nonetheless, we do not consider this to be likely considering the small number of paediatric pharmacogenetic/pharmacogenomic studies listed under Clinicaltrials.gov, which indicates there is a lack of studies investigating the effect of genomics on drug response or adverse drug reactions aimed specifically at children and adolescents. A lack of pharmacogenetic studies in paediatric patients presents several concerns for drug use in this population. For one, pharmacogenomic information may not be available for diseases that are more prevalent in childhood, such as attention deficit/ hyperactivity disorder, acute lymphocytic leukaemia and growth hormone deficiency [26]. Another problem is that the ontogeny of the process affected by the gene polymorphism will not be clearly identified in this population. Therefore, in the age groups where development is continuing, drug dose may be inaccurate when parameters derived from adult populations are used for pharmacogenetic-based dose modifications [27]. The primary objective of a drug label is to guide the prescriber in safe and effective medicine use. Drug efficacy and
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
safety are likely to be complicated in result of environmental and endogenous factors with the influence of many genes [28]. Changes with age and disease progression are particularly dynamic and complicated in paediatric drug response [29]. Therefore, the incorporation and appropriate use of pharmacogenetic information will contribute to improvements in drug use and safety in paediatric patients. While some progress has been made, there still exists a large need to incorporate specific pharmacogenomic biomarker testing in paediatric patients in drug development studies. Although drug labels that contain pharmacogenetic information derived from adults imply that this information can be applied to paediatric patients, this should be done cautiously considering reasons stated throughout this discussion. To determine urgent areas of need where pharmacogenomic information in paediatric labels is improperly extrapolated from adult studies, a true assessment evaluating labelled biomarker appropriateness should be conducted for approved, paediatric drugs. As previously stated, our PharmGKB analysis is not intended to evaluate suitability but suggests potential biomarkers that may warrant further investigation if clinical significance is present in the paediatric population. Also, an additional detailed analysis of studies conducted outside of the BPCA and PREA process is likely required to accurately decipher areas of need in paediatric pharmacogenomic studies. These paediatric studies may have led to modifications is drug labelling or may have identified non-significant relationships that were unrecognized in our study. Conclusion By utilizing public databases, we attempted to determine the degree of paediatric pharmacogenomic trial translation into FDA-approved drug labels, in reference to studies conducted under BPCA and PREA legislation. We conclude that pharmacogenomic studies directly conducted in paediatric patients are lacking, and pharmacogenomic biomarker information based on adult studies is presented in FDAapproved labels for use in paediatric patients without cautions related to doing so. Therefore, studies investigating effects of pharmacogenomic markers specifically in paediatric patients should be conducted in future clinical trials, during the drug development stage and whenever significant biomarkers are known and available. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government Ministry of Science, ICT and Future Planning (No. 200900081414). The authors would like to acknowledge the assistance of Drs. Padmaja Mummaneni and Gilbert Burckart of the Office of Clinical Pharmacology, US Food and Drug Administration. Conflict of Interest The authors declare no competing interests.
443
References 1 Weinshilboum RM, Wang L. Pharmacogenetics and pharmacogenomics: development, science, and translation. Annu Rev Genomics Hum Genet 2006;7:223–45. 2 U.S. Food and Drug Administration. FDA approved drug products. Drugs@FDA. July, 2014. http://www.accessdata.fda.gov/scripts/ cder/drugsatfda/index.cfm (last accessed on 6 July 2014). 3 Whirl-Carrillo M, McDonagh EM, Hebert JM, Gong L, Sangkuhl K, Thorn CF et al. Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther 2012;92:414–7. 4 The Pharmacogenomics Knowledgebase. July, 2014. https://www. pharmgkb.org/ (last accessed on 6 July 2014). 5 Tutton R. Pharmacogenomic biomarkers in drug labels: what do they tell us? Pharmacogenomics 2014;15:297–304. 6 Rieder MJ, Carleton B. Pharmacogenomics and adverse drug reactions in children. Front Genet 2014;5:78. 7 Leeder JS. Translating pharmacogenetics and pharmacogenomics into drug development for clinical pediatrics and beyond. Drug Discov Today 2004;9:567–73. 8 Strolin BM, Whomsley R, Baltes EL. Differences in absorption, distribution, metabolism and excretion of xenobiotics between the paediatric and adult populations. Expert Opin Drug Metab Toxicol 2005;1:447–71. 9 Bartelink IH, Rademaker CM, Schobben AF, van den Anker JN. Guidelines on paediatric dosing on the basis of developmental physiology and pharmacokinetic considerations. Clin Pharmacokinet 2006;45:1077–97. 10 Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Leeder JS, Kauffman RE. Developmental pharmacology–drug disposition, action, and therapy in infants and children. N Engl J Med 2003;349:1157–67. 11 Hoppu K, Anabwani G, Garcia-Bournissen F, Gazarian M, Kearns GL, Nakamura H et al. The status of paediatric medicines initiatives around the world–What has happened and what has not? Eur J Clin Pharmacol 2012;68:1–10. 12 U.S. Food and Drug Administration. Medical, statistical, and clinical pharmacology reviews of pediatric studies conducted under Section 505A and 505B of the Federal Food, Drug, and Cosmetic Act (the Act), as amended by the FDA AMENDMENTS Act of 2007 (FDAAA). June, 2014. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm049872.htm (last accessed on 6 July 2014). 13 U.S. Food and Drug Administration. Medical, statistical, and clinical pharmacology reviews of pediatric studies conducted under Section 505A and 505B of the Federal Food, Drug, and Cosmetic Act (the Act), as amended by the FDA Amendments Act of 2012 (FDASIA). June, 2014. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm316937.htm (last accessed on 6 July 2014). 14 U.S. Food and Drug Administration. Pediatrics. June, 2014. http://www.fda.gov/scienceresearch/specialtopics/pediatrictherapeuticsresearch/default.htm (last accessed on 6 July 2014). 15 U.S. Food and Drug Administration. Table of pharmacogenomic biomarkers in drug labels. June, 2014. http://www.fda.gov/Drugs/ ScienceResearch/ResearchAreas/Pharmacogenetics/ucm083378.htm (last accessed on 6 July 2014). 16 Law V, Knox C, Djoumbou Y, Jewison T, Guo AC, Liu Y et al. DrugBank 4.0: shedding new light on drug metabolism. Nucleic Acids Res 2014;42:D1091–7. 17 Eichelbaum M, Altman RB, Ratain M, Klein TE. New feature: pathways and important genes from PharmGKB. Pharmacogenet Genomics 2009;19:403. 18 Gillen JE, Tse T, Ide NC, McCray AT. Design, implementation and management of a web-based data entry system for ClinicalTrials.gov. Stud Health Technol Inform 2004;107: 1466–70.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
444
THERASA KIM ET AL.
19 ClinicalTrials.gov. July, 2014. https://clinicaltrials.gov/ (last accessed on 6 July 2014). 20 Leeder JS. Developmental and pediatric pharmacogenomics. Pharmacogenomics 2003;4:331–41. 21 Freund CL, Clayton EW. Pharmacogenomics and children: meeting the ethical challenges. Am J Pharmacogenomics 2003;3:399–404. 22 Hines RN, McCarver DG. The ontogeny of human drug-metabolizing enzymes: phase I oxidative enzymes. J Pharmacol Exp Ther 2002;300:355–60. 23 Cheok MH, Lugthart S, Evans WE. Pharmacogenomics of acute leukemia. Annu Rev Pharmacol Toxicol 2006;46:317–53. 24 Odell GB. “Physiologic” hyperbilirubinemia in the neonatal period. N Engl J Med 1967;277:193–5. 25 Vivar JC, Pemu P, McPherson R, Ghosh S. Redundancy control in pathway databases (ReCiPa): an application for improving gene-set
26
27
28 29
enrichment analysis in Omics studies and “Big data” biology. OMICS 2013;17:414–22. Russo R, Capasso M, Paolucci P, Iolascon A. Pediatric pharmacogenetic and pharmacogenomic studies: the current state and future perspectives. Eur J Clin Pharmacol 2011;67(Suppl 1): 17–27. Leeder JS, Kearns GL, Spielberg SP, van den Anker J. Understanding the relative roles of pharmacogenetics and ontogeny in pediatric drug development and regulatory science. J Clin Pharmacol 2010;50:1377–87. Ma Q, Lu AY. Pharmacogenetics, pharmacogenomics, and individualized medicine. Pharmacol Rev 2011;63:437–59. Stevens A, De Leonibus C, Hanson D, Whatmore A, Murray P, Donn R et al. Pediatric perspective on pharmacogenomics. Pharmacogenomics 2013;14:1889–905.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
Copyright of Basic & Clinical Pharmacology & Toxicology is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Doi: 10.1111/bcpt.12341
Pharmacogenomic Biomarker Information in FDA-approved Paediatric Drug Labels Therasa Kim1, Nayoung Han2, Minji Sohn2, Jung Mi Oh2, Eui-Kyung Lee3, Eunhee Ji4 and In-Wha Kim2 1 College of Pharmacy, Chonnam National University, Gwangju, South Korea, 2College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul, South Korea, 3School of Pharmacy, Sungkyunkwan University, Suwon, Gyeonggi-do, South Korea and 4College of Pharmacy, Gachon University, Incheon, South Korea
(Received 16 August 2014; Accepted 6 October 2014) Abstract: Gene maturation differs between paediatric and adult populations, and the extrapolation of adult pharmacogenomic information to paediatrics is not always appropriate. We sought to determine the extent of paediatric pharmacogenomic trial translation into US FDA-approved labels and to evaluate needs for biomarker studies. Using FDA’s Table of Genomic Biomarkers and Drugs@FDA website, 38 pharmacogenomic biomarkers in 56 drug labels were identified with possible application in paediatrics. Of these 56 drugs, biomarker comparison against ‘Very Important Pharmacogenes (VIPs)’ defined in PharmGKB’s database revealed a total of eight VIPs labelled among 41 drugs. One hundred and thirty-nine product reviews posted on the FDA website under the Best Pharmaceuticals for Children Act and Paediatric Research Equity Act between October 2007 and July 2014 were examined. Review screening identified 43 drugs with ‘pharmacogenomic’ content, of which only three were true genotyping study reviews for proton pump inhibitors, all evaluating CYP2C19 polymorphisms. Pantoprazole was the sole drug labelled with pharmacogenomic information obtained specifically from paediatric trials. Clinicaltrials.gov was searched to further evaluate the current availability of pharmacogenomic studies in the paediatric population. Of the 33,132 trials registered on Clinicaltrials.gov, 137 were labelled as paediatric pharmacogenetic and pharmacogenomic studies. Pharmacogenomic studies directly conducted in paediatric patients are lacking, and thus, pharmacogenomic biomarker information based on adult studies is commonly presented in FDA-approved labels for use in paediatric patients. Considering differences in gene expression and physiological maturation between paediatric and adult populations, studies investigating pharmacogenomic effects specifically in paediatric patients should be conducted whenever significant biomarkers are available.
In a new era of genome-based personalized medicine, pharmacogenetic or pharmacogenomic information has the potential to predict variability in the pharmacokinetics and pharmacodynamics of therapeutic drugs, contributing to the optimization of drug efficacy and avoidance of adverse drug effects [1]. Today, open-source databases such as the Drugs@FDA [2] and PharmGKB [3,4] websites provide pharmacogenomic information that can be utilized to improve efficacy and safety of drug use [5]. This information is primarily derived from adult populations and has limitations in its application to the paediatric population. Diseases differ between adults and children across a wide range of ages, suggesting that the relationships between genotype and phenotype will differ between adults and children, and also within the paediatric population [6]. Physiological or metabolic maturation is a complex and continuous process that occurs from the prenatal period to mid-childhood to adolescence [7], and could modify pharmacogenomic effects at any given stage of development. Consequently, the extrapolation of adult pharmacogenomic information to paediatric patients may most likely not be straightforward, especially for neonates and infants [8]. Age-dependent changes in pharmacokinetic Author for correspondence: In-Wha Kim, College of Pharmacy, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 151-742, South Korea (fax +82 2 880 9560, e-mail [email protected]).
and pharmacodynamic parameters play a crucial role in paediatric drug response, and a pharmacogenomic approach to the treatment of paediatric disease requires an understanding of gene variation in relation to age [9,10]. Considering such heterogeneity, the incorporation of pharmacogenomics into paediatric drug development studies is essential to ensure appropriate application of genomic parameters. Many of the currently approved drugs available on the market have not been adequately studied in paediatric populations, and an even fewer number of pharmacogenomic paediatric studies are available. However, both the United States and the European Union now require paediatric studies as part of the drug approval process, which facilitates the participation of children in clinical drug development studies [11]. Consequently, a relatively large number of paediatric studies have been conducted and reported since the passage of the Food and Drug Administration Amendments Act (FDAAA) of 2007 [12]. In 2012, the Food and Drug Administration Safety and Innovation Act (FDASIA) brought permanency to the Best Pharmaceuticals for Children Act (BPCA) and the Paediatric Research Equity Act (PREA) [13]. The Office of Paediatric Therapeutics at the United States FDA now provides medical, statistical and clinical pharmacology reviews of paediatric studies on their website [14]. With a growing body of information on genomic biomarkers, the incorporation of such pharmacogenetic information
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
into drug labels and its appropriate use is becoming increasingly important in paediatric patients as well. Thus, theoretically, a greater number of paediatric pharmacogenomic studies validating the clinical application of biomarkers in this population should also become available. In this study, we sought to (i) identify drugs approved in the paediatric population with pharmacogenomic biomarker information available in the FDA-approved drug label, (ii) compare labelled pharmacogenomic markers to ‘Very Important Pharmacogenes (VIPs)’ defined in PharmGKB, (iii) review the influence of BPCA and PREA on the conduction and incorporation of pharmacogenomic studies in paediatric drug labels and (iv) survey the current status of paediatric pharmacogenetic and pharmacogenomic studies available. Based on these analyses, we ultimately aimed to provide an evaluation of potential needs present in paediatric pharmacogenomic biomarker studies. Method Analysis of pharmacogenomic information in paediatric FDAapproved drug labels. The FDA website provides the ‘Table of Pharmacogenomic Biomarkers in Drug Labelling’, listing FDAapproved drugs with pharmacogenomics information in their prescription label [15]. Drug labels for medications in this table were searched using Drugs@FDA [2] to (i) determine the approval for use in the paediatric population and (ii) extract pharmacogenomics information for the identified paediatric drugs in a two-step process. In the first stage, the labels were screened for the occurrence of one or more of the following key words: gene, genetic, genotype, DNA, RNA, variant, haplotype, polymorphism, deficiency and SNP. In the second stage, labels identified were further screened based on a set of selection criteria that identified descriptions of human genetics or genomics such as gene expression changes, gene deficiencies, deletions or insertions, genetic mutations or genetic polymorphisms and variations. Topical and diagnostic medications were excluded from the analysis, and drugs with multiple formulations or multiple indications were included only once in the overall label count. To conduct a comparison of the labelled pharmacogenomics biomarkers to established VIPs, the PharmGKB website (http://www. pharmgkb.org/index.jsp) was used to search for pharmacogenomics genes that were associated with the drugs identified [4]. PharmGKB’s database provides genomic information based on various annotations, literature reviews, pathways and information automatically retrieved from DrugBank [16]. ‘VIPs’ are defined in PharmGKB as genes having extensive demonstrated relationships between variants and drug response or metabolism and are expected to have clinical significance [17]. There are 54 VIPs identified in PharmGKB’s database, including genes such as CFTR, CYP2C19, CYP2C9, G6PD, HLA-B, TPMT and VKORC1. Pharmacogenomics biomarkers identified in the drug labels were compared specifically against VIPs reported for these agents. Analysis of pharmacogenomics studies in the paediatric population. Publically available medical, statistical and clinical pharmacology reviews of paediatric studies, conducted under Section 505A and 505B of the Federal Food, Drug and Cosmetic Act as amended in 2007 [12] and Section 505A and 505B of the Federal Food, Drug and Cosmetic Act as amended in 2012 [13], were analysed to identify trials pertaining to pharmacogenomics. Studies were classified as performed in accordance with BPCA, PREA or both legislations. The aforementioned methodology employed to screen drug labels was again utilized to screen reviews, and
439
‘pharmacogenomic’ studies identified were further inspected to evaluate the incorporation of paediatric pharmacogenomic information into drug labels. In addition, pharmacogenomic trials conducted under BPCA and/or PREA were identified for paediatric approved drugs discovered under the FDA ‘Table of Pharmacogenomic Biomarkers in Drug Labelling’. Only paediatric drug studies submitted to the FDA between October 2007 and July 2014 were included in the analysis. To further determine the overall availability of paediatric pharmacogenomic studies, the Clinicaltrials.gov website (http:// clinicaltrials.gov/) was searched for paediatric trials using the key terms ‘pharmacogenetics’ and ‘pharmacogenomics’ [18,19].
Results Pharmacogenomic biomarkers in FDA-approved drug labels approved for paediatrics. A total of 140 unique drugs were identified in the FDA Table of Genomic Biomarkers after excluding topical, diagnostic and duplicate active ingredients. Of these, 56 drugs (40.0%) were determined to be approved for use in the paediatric population and listed Pharmacogenomics genes of interest in the FDAapproved labels (Table 1). For these 56 agents, a total of 38 pharmacogenomic biomarkers were available with the cytochrome P450 (CYP) enzyme family being identified most frequently (CYP2D6 – 17.1%, CYP2C19 – 13.2%, CYP2C9 – 2.63%) and G6PD being the second most common biomarker. Comparison of FDA-approved pharmacogenomic biomarkers to PharmGKB. One hundred and seventeen of the above-mentioned 140 drugs were related to specific genes when searched for in PharmGKB. A total of 361 genes/genetics were documented with 31 genes classified as VIPs. The CYP enzyme family was identified most frequently (CYP3A4 – 3.58%, CYP2D6 – 3.21%, CYP2C19 – 2.96%), and ABCB1 was identified as the second most common pharmacogenomics-related gene. In the comparison of genomic biomarkers from FDA-approved labels with VIPs presented in PharmGKB, 41 of the 56 paediatric drugs had labelled biomarkers that were VIPs including HLAB, CYP2D6, CYP2C19, CYP2C9, G6PD, TPMT, F5 and VKORC1 genes (eight in total). Of the remaining 15 drugs, three did not have defined VIPs and 12 had labelled biomarkers that did not include VIP genes (table 2). Pharmacogenomic markers related to drug indication or hypersensitivity reactions were excluded in the summary. Influence of BPCA/PREA on paediatric pharmacogenomic drug labelling. After excluding topicals, diagnostics and duplicate active pharmaceutical ingredients, 139 drugs remained of the 195 products studied under BPCA and PREA as of July 2014. For these drugs, review screening resulted in identification of 43 drug reviews with ‘pharmacogenomics’ content, of which only three involved true genotyping studies for the proton pump inhibitors pantoprazole, lansoprazole and rabeprazole. CYP2C19 polymorphisms were evaluated for all three agents.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
THERASA KIM ET AL.
440
Table 1. Paediatric drugs with FDA-labelled pharmacogenomic biomarkers as of July 2014. A total of 38 biomarkers identified for 56 paediatric approved drugs. Therapeutic class Antidotes Cardiology Dermatology Endocrinology
Gastroenterology
Haematology
Infectious disease
Metabolic disorder Neurology
Oncology
Rheumatology Psychiatry
Pulmonary Reproductive Transplantation
Drug Sodium nitrite Carvedilol Metoprolol Tretinoin Dapsone Atorvastatin Pravastatin Rosuvastatin Esomeprazole Lansoprazole Omeprazole Pantoprazole1 Rabeprazole Succimer Warfarin Abacavir Chloroquine Maraviroc Nalidixic Acid Nitrofurantoin Quinine Sulphate Sulfamethoxazole and trimethoprim Voriconazole Carglumic acid Velaglucerase Alfa Carbamazepine Clobazam Phenytoin Arsenic Trioxide Busulfan Cisplatin Everolimus Fulvestrant Imatinib Mercaptopurine Rasburicase Tamoxifen Thioguanine Celecoxib Carisoprodol Aripiprazole Atomoxetine Clomipramine Diazepam Fluoxetine Fluvoxamine Paroxetine Perphenazine Pimozide Risperidone Thioridazine Valproic acid Venlafaxine Ivacaftor Drospirenone and ethinyl estradiol Mycophenolic acid
BPCA/PREA/ OTHER
Paediatric age in study
G6PD CYP2D6 CYP2D6 PML/RARA G6PD LDLR LDLR LDLR CYP2C19 CYP2C19 CYP2C19 CYP2C19 CYP2C19 G6PD VKORC1, CYP2C9, PROC HLA-B G6PD CCR5 G6PD G6PD G6PD G6PD
OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER PREA BPCA PREA BPCA, PREA PREA OTHER OTHER
All 2 months to 17 years 6–17 years 1–16 years 12–17 years 10–17 years 8–18 years 10–17 years 1 months to 17 years >1 year 1–16 years All 12–18 years >1 year All
PREA OTHER OTHER OTHER OTHER OTHER OTHER
3 months to 13 years All 16–18 years 3 months to 12 years >1 months 1.5–12 years >2 months
CYP2C19 NAGS GBA HLA-B, HLA-A CYP2C19 HLA-B PML/RARA Ph Chromosome TPMT ERBB2, ESR1 ESR1 KIT, BCR/ABL1, PDGFRB, FIP1L1/PDGFRA TPMT G6PD ESR1, PGR, F2, F5 TPMT CYP2C9 CYP2C19 CYP2D6 CYP2D6 CYP2D6 CYP2C19 CYP2D6 CYP2D6 CYP2D6 CPY2D6 CYP2D6 CYP2D6 CYP2D6 POLG, NAGS, CPS, ASS, OTC, ASL, ARG CYP2D6 CFTR (G551D) CYP2C19 HGPRT
OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER BPCA, PREA OTHER OTHER
>12 years All 4–17 years >6 months 2–17 years All 4–18 years All All 1–17 years 1–8 years 2–17 years
OTHER OTHER OTHER OTHER OTHER OTHER BPCA, PREA OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER OTHER PREA OTHER
All All 2–10 years All 2–17 years >16 years 1–17 years 6–15 years 10–17 years 2–17 years 7–17 years 8–17 years All >12 years 2–17 years 5–17 years All 10–18 years 12–18 years 6–17 years 14–18 years 5–16 years
Biomarker
BPCA, Best Pharmaceuticals for Children Act; PREA, Paediatric Research Equity Act. Paediatric pharmacogenomic BPCA/PREA trial resulting in label modification.
1
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
For paediatric patients who received pantoprazole, genotyping for CYP2C19 was conducted in six studies. Of 226 patients genotyped, six were poor metabolizers (CYP2C19*2/ *2). In these poor metabolizers, the pantoprazole dose-normalized AUC was more than six times higher than that of extensive (CYP2C19*1/*1) or intermediate (CYP2C19*1/x) metabolizers. The review team ultimately revised the pantoprazole paediatric label to include the following statement in ‘Pharmacogenomics Section 12.4’: ‘For known paediatric poor metabolizers, a dose reduction should be considered’. In the case of lansoprazole, the limited number of poor metabolizers and heterozygous extensive metabolizers in the neonatal study group precluded any conclusion on the relationship between the CYP2C19 genotype and lansoprazole pharmacokinetics. Additionally, CYP2C19 genotyping was performed for rabeprazole in adolescents and children 1–11 years of age. The lack of significant impact of CYP2C19 metabolizer status on the clearance of rabeprazole was consistent with current labelling, and as such, no labelling update was recommended. Of the 56 paediatric drugs with labelled pharmacogenomics biomarkers, only nine (16.1%) had paediatric studies that were conducted under BPCA and/or PREA (table 1). All other studies had been conducted in children and adolescences outside of the BPCA and PREA process. Of the nine drugs studied under paediatric trials for BPCA and/or PREA, only pantoprazole was labelled with pharmacogenomics information obtained specifically from paediatric trials. Pharmacogenomic information from paediatric trials for the remaining drugs did not translate to modifications in the label for reasons including non-significant difference from adult population, lack of information and non-significant association or clinical effect. Survey of pharmacogenetic/pharmacogenomic paediatric trials. From a total of 33,132 trials registered on Clinicaltrials.gov, 470 and 339 clinical trials containing pharmacogenetic and
441
pharmacogenomics analyses, respectively, were identified. One hundred and thirty-seven trials were of paediatric population, but only 60 mentioned the investigation of specific genes. Some examples include exploration of the association between morphine and CYP2D6 and UGT2B7, tacrolimus and CYP3A5, efavirenz and CYP2B6, and dexmedetomidine and CYP2A6. When sorted by therapeutic class, the majority of the trials (69%) examined drugs in the areas of oncology, psychiatry, analgesic, immunology and infectious disease (fig. 1). The age distribution of the study participants is as follows: 44 from birth, 2 from 1 month, 23 from 1 year, 29 from 6 years and 39 from 12 years of age.
Discussion Human development from the prenatal period through adolescence is a dynamic process. Newborns and infants demonstrate large variability in drug response and metabolizing capabilities [20]. The genetic effect of a polymorphism on adults may not be observed in paediatric patients, dependent upon a number of factors including the ontology of drug-metabolizing enzymes, transporters and receptors [21]. While the effect of genetic variations on drug metabolism is recognized, much less attention has been given to developmental pharmacogenomics, even though some genes are expressed much more abundantly in the earlier years of life compared to the adult years and others vice versa [10]. One of the most widely acknowledged examples of gene switching is represented by the CYP3A gene family. The level of CYP3A7 expression is detectable as early as in 50 days of gestation in foetal livers, and its expression begins to decline after the first post-natal week, reaching undetectable levels in most individuals by 1 year of age [22]. Conversely, hepatic CYP3A4/3A5 expression begins to dramatically increase at about 1 week of age, reaching 30% of adult levels by 1 month [22]. Accordingly, total CYP3A protein expression over the entire develop-
Table 2. Comparison of pharmacogenomic biomarkers in FDA labels to ‘Very Important Pharmacogenes (VIPs’) in PharmGKB for paediatric approved drugs. Twelve of 56 paediatric drugs are labelled with non-VIP genes. Drug
Biomarker
Arsenic Trioxide Atorvastatin
PML/RARA LDLR
Busulfan Everolimus Fulvestrant Imatinib
Ph Chromosome ERBB2, ESR1 ESR1 KIT, BCR/ABL1, PDGFRB, FIP1L1/PDGFRA CFTR (G551D) CCR5 HGPRT LDLR LDLR POLG, NAGS, CPS, ASS, OTC, ASL, ARG
Ivacaftor Maraviroc Mycophenolic acid Pravastatin Rosuvastatin Valproic acid
Very Important Pharmacogene KCNH2 ABCB1, ACE, CYP2C19, CYP2C8, CYP2C9, CYP2D6, CYP3A4, CY P3A5, HMGCR, NR1I2, SLCO1B1, UGT1A1 CYP2C19, GSTP1, MTHFR CYP2C8, CYP2D6, CYP3A4, CYP3A5 CYP3A4 PDGFRB, ABCB1, C YP2D6, CYP3A4, CYP3A5, SLC22A1 CYP3A4, CYP3A5 CYP3A4, CYP3A5 CYP2C8, CYP3A4, CYP3A5 ACE, ABCB1, CYP3A5, HMGCR, MTHFR, SLCO1B1 CYP2C19, CYP2C9, CYP3A4, CYP3A5, HMGCR, SLCO1B1 ABCB1, CYP2A6, CYP2B6, CYP2C19, CYP2C9, HLA-B, NR1I2
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
442
THERASA KIM ET AL.
Fig. 1. Paediatric pharmacogenetic/pharmacogenomic studies identified under Clinicaltrials.gov as of July 2014.
mental period remains constant. However, because CYP3A7 and CYP3A4 exhibit differences in substrate specificity and catalytic efficacy, inequalities in metabolic capacity during development are observed [22]. In addition, many drug-metabolizing enzymes are less developed in children than they are in adults. An example is UGT [23], a liver enzyme essential for the body’s elimination of xenobiotics and endogenous substances such as toxins and bilirubin. Only 1% of the normal adult level of hepatic UGT activity is present at birth [24]. Post-natal maturation of UGT is birth related and increases rapidly to adult levels by 14 weeks, regardless of the gestational age at birth, perhaps in relation to the methylation or acetylation status of the gene locus. Therefore, clinical pharmacogenomic trials in paediatric patients should be conducted whenever possible, especially when ontogeny may play a role in the impact of the established pharmacogenomic parameter. In this study, among the 140 drugs with pharmacogenomic biomarker information in the FDA-approved drug label, 56 drugs with paediatric indications were identified, involving 38 biomarkers that could potentially have application in paediatric patients as well. In our PharmGKB analysis, we conducted a comparison of labelled pharmacogenomic biomarkers to VIPs listed in PharmGKB’s database. However, this investigation does not evaluate the appropriateness of FDA-labelled biomarker suitability in paediatrics, but rather provides a summary of potential genes open to further investigation depending on clinical impact. Furthermore, caution should be taken in the interpretation of discrepancies between FDA-labelled genes and VIPs considering possible redundancy control in gene sets, which may reflect a broader selection of clinically significant biomarkers for labelling. Vivar et al. [25] have reported on the risk of ignoring overlap or redundancy in pathway databases and designating several pathways as significant when in fact, the biological mechanisms are not truly independent. Only nine of the 56 drugs were studied under BPCA and PREA in paediatric patients, with pantoprazole being the sole drug incorporating pharmacogenomic information from
paediatric studies into the product label. This suggests that product labelling of pharmacogenomic information on drug dosing, response and adverse drug reactions in paediatric patients is highly dependent on pharmacogenomic studies in adults. In similar context, a comprehensive screening of trials submitted under BPCA and PREA between October 2007 and July 2014 did not result in the detection of any additional paediatric trials leading to specific pharmacogenomic label modifications, and only three drugs involved true biomarker genotyping studies. Despite reauthorization and permanency of BPCA and PREA legislation, there appears to be a limited number of paediatric pharmacogenomic studies investigating the application of genomic biomarkers in this population. However, it is questionable whether this observation is an accurate reflection of study insufficiency, or whether there are actually a considerable number of such trials but they have been conducted outside of the BPCA and PREA process. There is also the possibility that these ‘OTHER’ paediatric study findings were non-significant and may have eliminated the need for additional trials to be conducted under such legislation. Nonetheless, we do not consider this to be likely considering the small number of paediatric pharmacogenetic/pharmacogenomic studies listed under Clinicaltrials.gov, which indicates there is a lack of studies investigating the effect of genomics on drug response or adverse drug reactions aimed specifically at children and adolescents. A lack of pharmacogenetic studies in paediatric patients presents several concerns for drug use in this population. For one, pharmacogenomic information may not be available for diseases that are more prevalent in childhood, such as attention deficit/ hyperactivity disorder, acute lymphocytic leukaemia and growth hormone deficiency [26]. Another problem is that the ontogeny of the process affected by the gene polymorphism will not be clearly identified in this population. Therefore, in the age groups where development is continuing, drug dose may be inaccurate when parameters derived from adult populations are used for pharmacogenetic-based dose modifications [27]. The primary objective of a drug label is to guide the prescriber in safe and effective medicine use. Drug efficacy and
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
PHARMACOGENOMIC INFORMATION IN PAEDIATRIC DRUGS
safety are likely to be complicated in result of environmental and endogenous factors with the influence of many genes [28]. Changes with age and disease progression are particularly dynamic and complicated in paediatric drug response [29]. Therefore, the incorporation and appropriate use of pharmacogenetic information will contribute to improvements in drug use and safety in paediatric patients. While some progress has been made, there still exists a large need to incorporate specific pharmacogenomic biomarker testing in paediatric patients in drug development studies. Although drug labels that contain pharmacogenetic information derived from adults imply that this information can be applied to paediatric patients, this should be done cautiously considering reasons stated throughout this discussion. To determine urgent areas of need where pharmacogenomic information in paediatric labels is improperly extrapolated from adult studies, a true assessment evaluating labelled biomarker appropriateness should be conducted for approved, paediatric drugs. As previously stated, our PharmGKB analysis is not intended to evaluate suitability but suggests potential biomarkers that may warrant further investigation if clinical significance is present in the paediatric population. Also, an additional detailed analysis of studies conducted outside of the BPCA and PREA process is likely required to accurately decipher areas of need in paediatric pharmacogenomic studies. These paediatric studies may have led to modifications is drug labelling or may have identified non-significant relationships that were unrecognized in our study. Conclusion By utilizing public databases, we attempted to determine the degree of paediatric pharmacogenomic trial translation into FDA-approved drug labels, in reference to studies conducted under BPCA and PREA legislation. We conclude that pharmacogenomic studies directly conducted in paediatric patients are lacking, and pharmacogenomic biomarker information based on adult studies is presented in FDAapproved labels for use in paediatric patients without cautions related to doing so. Therefore, studies investigating effects of pharmacogenomic markers specifically in paediatric patients should be conducted in future clinical trials, during the drug development stage and whenever significant biomarkers are known and available. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government Ministry of Science, ICT and Future Planning (No. 200900081414). The authors would like to acknowledge the assistance of Drs. Padmaja Mummaneni and Gilbert Burckart of the Office of Clinical Pharmacology, US Food and Drug Administration. Conflict of Interest The authors declare no competing interests.
443
References 1 Weinshilboum RM, Wang L. Pharmacogenetics and pharmacogenomics: development, science, and translation. Annu Rev Genomics Hum Genet 2006;7:223–45. 2 U.S. Food and Drug Administration. FDA approved drug products. Drugs@FDA. July, 2014. http://www.accessdata.fda.gov/scripts/ cder/drugsatfda/index.cfm (last accessed on 6 July 2014). 3 Whirl-Carrillo M, McDonagh EM, Hebert JM, Gong L, Sangkuhl K, Thorn CF et al. Pharmacogenomics knowledge for personalized medicine. Clin Pharmacol Ther 2012;92:414–7. 4 The Pharmacogenomics Knowledgebase. July, 2014. https://www. pharmgkb.org/ (last accessed on 6 July 2014). 5 Tutton R. Pharmacogenomic biomarkers in drug labels: what do they tell us? Pharmacogenomics 2014;15:297–304. 6 Rieder MJ, Carleton B. Pharmacogenomics and adverse drug reactions in children. Front Genet 2014;5:78. 7 Leeder JS. Translating pharmacogenetics and pharmacogenomics into drug development for clinical pediatrics and beyond. Drug Discov Today 2004;9:567–73. 8 Strolin BM, Whomsley R, Baltes EL. Differences in absorption, distribution, metabolism and excretion of xenobiotics between the paediatric and adult populations. Expert Opin Drug Metab Toxicol 2005;1:447–71. 9 Bartelink IH, Rademaker CM, Schobben AF, van den Anker JN. Guidelines on paediatric dosing on the basis of developmental physiology and pharmacokinetic considerations. Clin Pharmacokinet 2006;45:1077–97. 10 Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Leeder JS, Kauffman RE. Developmental pharmacology–drug disposition, action, and therapy in infants and children. N Engl J Med 2003;349:1157–67. 11 Hoppu K, Anabwani G, Garcia-Bournissen F, Gazarian M, Kearns GL, Nakamura H et al. The status of paediatric medicines initiatives around the world–What has happened and what has not? Eur J Clin Pharmacol 2012;68:1–10. 12 U.S. Food and Drug Administration. Medical, statistical, and clinical pharmacology reviews of pediatric studies conducted under Section 505A and 505B of the Federal Food, Drug, and Cosmetic Act (the Act), as amended by the FDA AMENDMENTS Act of 2007 (FDAAA). June, 2014. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm049872.htm (last accessed on 6 July 2014). 13 U.S. Food and Drug Administration. Medical, statistical, and clinical pharmacology reviews of pediatric studies conducted under Section 505A and 505B of the Federal Food, Drug, and Cosmetic Act (the Act), as amended by the FDA Amendments Act of 2012 (FDASIA). June, 2014. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm316937.htm (last accessed on 6 July 2014). 14 U.S. Food and Drug Administration. Pediatrics. June, 2014. http://www.fda.gov/scienceresearch/specialtopics/pediatrictherapeuticsresearch/default.htm (last accessed on 6 July 2014). 15 U.S. Food and Drug Administration. Table of pharmacogenomic biomarkers in drug labels. June, 2014. http://www.fda.gov/Drugs/ ScienceResearch/ResearchAreas/Pharmacogenetics/ucm083378.htm (last accessed on 6 July 2014). 16 Law V, Knox C, Djoumbou Y, Jewison T, Guo AC, Liu Y et al. DrugBank 4.0: shedding new light on drug metabolism. Nucleic Acids Res 2014;42:D1091–7. 17 Eichelbaum M, Altman RB, Ratain M, Klein TE. New feature: pathways and important genes from PharmGKB. Pharmacogenet Genomics 2009;19:403. 18 Gillen JE, Tse T, Ide NC, McCray AT. Design, implementation and management of a web-based data entry system for ClinicalTrials.gov. Stud Health Technol Inform 2004;107: 1466–70.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
444
THERASA KIM ET AL.
19 ClinicalTrials.gov. July, 2014. https://clinicaltrials.gov/ (last accessed on 6 July 2014). 20 Leeder JS. Developmental and pediatric pharmacogenomics. Pharmacogenomics 2003;4:331–41. 21 Freund CL, Clayton EW. Pharmacogenomics and children: meeting the ethical challenges. Am J Pharmacogenomics 2003;3:399–404. 22 Hines RN, McCarver DG. The ontogeny of human drug-metabolizing enzymes: phase I oxidative enzymes. J Pharmacol Exp Ther 2002;300:355–60. 23 Cheok MH, Lugthart S, Evans WE. Pharmacogenomics of acute leukemia. Annu Rev Pharmacol Toxicol 2006;46:317–53. 24 Odell GB. “Physiologic” hyperbilirubinemia in the neonatal period. N Engl J Med 1967;277:193–5. 25 Vivar JC, Pemu P, McPherson R, Ghosh S. Redundancy control in pathway databases (ReCiPa): an application for improving gene-set
26
27
28 29
enrichment analysis in Omics studies and “Big data” biology. OMICS 2013;17:414–22. Russo R, Capasso M, Paolucci P, Iolascon A. Pediatric pharmacogenetic and pharmacogenomic studies: the current state and future perspectives. Eur J Clin Pharmacol 2011;67(Suppl 1): 17–27. Leeder JS, Kearns GL, Spielberg SP, van den Anker J. Understanding the relative roles of pharmacogenetics and ontogeny in pediatric drug development and regulatory science. J Clin Pharmacol 2010;50:1377–87. Ma Q, Lu AY. Pharmacogenetics, pharmacogenomics, and individualized medicine. Pharmacol Rev 2011;63:437–59. Stevens A, De Leonibus C, Hanson D, Whatmore A, Murray P, Donn R et al. Pediatric perspective on pharmacogenomics. Pharmacogenomics 2013;14:1889–905.
© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)
Copyright of Basic & Clinical Pharmacology & Toxicology is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
