University of Groningen
Towards personalized management of drug interactions: from interaction to
drug-drug-gene-interaction
Bahar, Akbar
DOI:10.33612/diss.112160601
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Bahar, A. (2020). Towards personalized management of drug interactions: from drug-drug-interaction to drug-drug-gene-interaction. University of Groningen. https://doi.org/10.33612/diss.112160601
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Discussion and Future Perspectives
8
Main Findings
Avoidance of potential DDIs is a clinically relevant issue when trying to improve the quality of drug prescribing since DDI can inadvertently lead to therapeutic failures or toxicities. In the Netherlands, to prevent DDI induced ADR, the electronic prescription system is complemented with a DDI surveillance tool which works during drug prescription and before drug dispensing to monitor a potential DDI. The resulting computerized DDI alerts are evidence-based and regularly updated by the distributors of the Z-index (publisher of the G-standard and partner of KNMP-The Hague NL) and the Health Base Foundation (publisher of Pharmabase and partner of Pharma Partners, Oosterhout, NL)1. The application of DDI alerting systems has been reported to increase patient
safety, improve prescribing practice, and decrease healthcare spendings2. In contrast, some
studies reported that warnings provided by the DDI alerts are often ignored either by physicians or pharmacists2,3. There are many reasons for such overriding behavior including excessive alerts,
the fact that benefit outweighs risk, disapproval of patients to treatment modification, confidence of prescribing clinicians on their knowledge regarding a specific DDI, impractical recommendations provided by the systems, disagreement with other DDI compendia, complicated technologies, interruptions of working process, patients previously tolerating adverse reactions resulting from DDI, and unclear clinical relevance2,4-6. The absence of certainty around clinical relevance
may be because the assessment of various potential DDIs rely on theoretical considerations and extrapolations of one reported case of an interaction to other drugs within the same therapeutic class2. Moreover, some potential DDIs are only supported by low quality of causal evidence (in vitro
studies or a case report) and lack of well-designed studies with control groups to support decision making in handling the potential DDIs7. Hence, the need to identify and to estimate the burden of
such DDIs as well as to assess their potential clinical implication is high in order to further improve the quality of the DDI alerts and DDIs managements.
Burden and Management of DDI
In chapter 2, we estimated the frequency of potential DDIs ignored by the health practitioners using a cross-sectional study in the Dutch three-generation population Lifelines cohort as part of the PharmLines Initiative. Since the medication data in the Lifelines cohort were collected by using self-reported questionnaires, there was a need to validate the potential DDI information by using the University of Groningen prescription database IADB.nl as a reference tool before being used in a pharmacoepidemiological study. We found that one to two per hundred drug users in the Lifelines cohort were exposed to a potential CYP2D6/2C19/2C9-mediated DDI at least in a short period in time during entry of the cohort study. The concordance analysis between the Lifelines cohort and the prescription database IADB.nl on the combination of substrates and inhibitors of CYP2D6/2C19/2C9 indicated that the best concordance level was achieved using a time window of three months which agrees with previous validation studies8,9. Further analysis based on the type of
interacting medication showed that self-reported information on the combination of two regularly used drugs were more reliable (good agreement) than the combination of drugs used routinely and episodically (fair to moderate agreement). However, the latter was still more reliable than the combination of two drugs used episodically (poor to moderate agreement).
chapter EIGHT
228
8
Among the most commonly presented potential DDIs were co-prescriptions of metoprolol (a CYP2D6 substrate) with strong CYP2D6 inhibitors such as paroxetine and fluoxetine. The combination is likely observed in clinical practice because cardiovascular and psychiatric diseases are frequent comorbid diseases10. Therefore, in chapter 3, we further analyzed the burden of
these potential DDIs among the high-risk senior population. We found that despite the application of the DDI alerting systems, metoprolol-paroxetine/fluoxetine combinations were still commonly co-prescribed (2039 incidences)5. The contradictory assessment of two main knowledge-bases of
the Dutch DDI alerts about the clinical impact of the DDIs might be the culprit5. We also found that
switching either metoprolol or paroxetine/fluoxetine to other safer drugs was uncommon practice5.
Moreover, in this elderly group, the dose of metoprolol prescription is generally low, regardless its combination with paroxetine/fluoxetine5.
Using this information, we designed an observational retrospective cohort study to investigate the potential impact of metoprolol-paroxetine/fluoxetine combination among senior persons using the IADB.nl with two proxy parameters i.e. early termination or dose modification of metoprolol after the initiation of paroxetine/fluoxetine on metoprolol prescriptions (chapter 4). Compared to citalopram (which is a weak CYP2D6 inhibitor) co-prescription, the metoprolol-paroxetine/fluoxetine combination was not significantly associated with the need of early termination and a dose modification of metoprolol in elderly11. However, if the combination was
compared with metoprolol-mirtazapine (which has no inhibitory capacity on CYP2D6) combination, metoprolol-paroxetine/fluoxetine co-prescription was significantly associated with the need of early termination but not a dose modification of metoprolol, especially in the subgroup of female older persons11. Therefore, it seems that the prescribing doctors prefer stopping over dose
modification of metoprolol when there is a sign of metoprolol related toxicities.
By compiling and screening all studies related to the impact of metoprolol-paroxetine/ fluoxetine combinations, we systematically evaluated the clinical relevance of this DDI (chapter 5). We found three case reports, four experimental and two observational studies that reported on the clinical outcomes of the DDI12. Though not all studies agreed on the clinical impact, the majority
of the studies reported that the DDI may have potentially deleterious impacts12. Therefore,
avoiding the DDI should be considered because alternative safe combinations of beta-blockers and antidepressants with comparable effectiveness are available. However, if the combination is necessary, dose reduction of metoprolol as well as close monitoring of the metoprolol related side effects is crucial especially in the high risk population12. A relevant limitation of the included
studies was that they did not consider patients’ CYP2D6 phenotype status which can potentially modify the clinical severity of the DDI12. CYP2D6 has an essential role to metabolize metoprolol
and therefore, the phenotype status of CYP2D6 is essential13. The magnitude of
metoprolol-paroxetine/fluoxetine interaction on patients with deviating phenotypes might be different than in the population with the normal phenotype. A study from Goryachkina et al. indicated that among acute myocardial infarction patients using metoprolol and starting paroxetine 20 mg (n=17), two patients, who were CYP2D6 intermediate metabolizers (IM), required a dose modification because of the appearance of metoprolol related side effects such as postural hypotension and excessive bradycardia14. Phenoconversion was most likely responsible for these side effects since paroxetine
8
may change the phenotype status of CYP2D6 from IM to be poor metabolizer (PM). It was reported that persons with only one active allele are more easily phenoconverted to become PM than those with at least two functional genes (NM)15. This illustrates that the severity of the DDI depends on
the catalytic activity of the main metabolic enzymes.
Pharmacogenetics of drug interactions
We further reviewed systematically the impact of polymorphisms on three main drug metabolic enzymes (CYP2D6, CYP2C19, and CYP2C9) mediated DDIs and drug-drug-gene-interactions (DDGIs) (chapter 6). The clinical impact of a CYP-mediated pharmacokinetic DDI is greatly affected by the genetic status of this metabolic enzyme16. The greater the role of CYP catalytic activity to
the substrate disposition, the stronger the magnitude of DDI that might be exerted by an inhibitor drug. Therefore, the extent of the DDI magnitude can mostly be explained following the sequence of NM > IM > PM16.
Furthermore, a more complex interaction is obtained when there is an interplay of gene polymorphism and a perpetrator drug (DDGI) in affecting the metabolism of a substrate16. DDGI
may involve the disposition of a drug metabolized with multiple pathways by multiple impairment schemes due to a non-active allele and an inhibitor. The obstruction of one biotransformation pathway by an allelic polymorphism makes the role of the secondary elimination pathway more important than in the normal condition. Hence, the subsequent alteration of the minor metabolic pathway by an inhibitor may produce a substantial increase of substrate blood concentration and generate a more severe interaction than the expected DDI. It can be seen in a case report which recorded an opioid toxicity in a hydrocodone (a CYP2D6 and CYP3A4 substrate) treated child with a decreased function of CYP2D6 who was co-administered clarithromycin (a CYP3A4 inhibitor)17.
The decreased metabolic activity of CYP2D6 due to genetic polymorphism makes the metabolic contribution of CYP3A4 becomes more important. Consequently, the addition of clarithromycin further impaired the biodegradation of hydrocodone. As a result, a very high concentration of hydrocodone led to the hydrocodone toxicity in the child.
The magnitude of DDGI also depends on the number of functional CYP alleles. PM and IM produce a greater extent of DDGI than NM because the less the contribution of main metabolic pathway, the greater the metabolic contribution of the secondary pathway. It can be seen in the interaction between voriconazole (a CYP2C19/3A4 inhibitor and substrate) and tacrolimus (a CYP2C19 and CYP3A4/5 substrate) in different CYP2C19 genotypes18. Voriconazole increased the AUC values of
tacrolimus by 4.4, 6, 6.5-fold in patients with CYP2C19 NM, IM, and PM, respectively, compared to patients with CYP2C19 NM using tacrolimus alone. In NM, the inhibition of CYP3A4 by voriconazole is less because tacrolimus metabolism via CYP2C19 is intact as well as voriconazole is also metabolized effectively by CYP2C19. However, a decreased metabolic function in CYP2C19 IM and PM lead to a more pronounced metabolic role of CYP3A4 in the disposition of tacrolimus. Moreover, the magnitude of the voriconazole AUC is also higher in CYP2C19 IM and PM than NM which leads to a stronger CYP3A4 inhibition in CYP2C19 IM and PM. As a consequence, the tacrolimus AUC is also substantially higher in PM and IM than NM. Hence, polymorphism has a pivotal role in determining the impact of DDI and DDGI.
chapter EIGHT
230
8
We then explored the influence of genetic polymorphisms and drug interactions using real-world-data. In chapter 7, we presented an inception cohort study about the associations between DGI, DDI and DDGI to (es)citaloporam therapy. We found that patients with predicted reduced metabolic function of CYP2C19 had a higher risk of switching and/or dose reduction of (es)citalopram than those with normal metabolic function. Additionally, CYP3A4 with predicted reduced-function-allele seemed to potentiate the effect sizes. Yet, in the condition of CYP2C19 NM, CYP3A4 with predicted decreased metabolic function seemed not to have an important effect on (es)citalopram treatment. Fudio et al. reported a comparable trend of outcomes for CYP2D619. CYP2D6 with
reduced-function-allele has a limited impact on (es)citalopram metabolism when CYP2C19 has normal catalytic activity. However, it may increase the effects of decreased metabolic function of CYP2C19 on (es) citralopram disposition. It might imply that it is important to know pre-emptively the predicted phenotype status of not only CYP2C19 but also CYP3A4 and CYP2D6. Within our relatively limited sample size (316 patients), we found that more than half of them had a deviating genotype related drug interaction either in the form of DGI or DDGI. These high numbers were probably unnoticed by clinicians since they had no information about the genetic status of the patients at the time of (es)citalopram prescription. Consequently, it might potentially pose considerable health impact as well as economic burden. Therefore, it is important to have information of the genetics status of multiple enzymes involved in drug disposition at the time of prescription.
Future perspectives and need for research
The magnitude of drug interactions is determined by the contribution of intrinsic and extrinsic factors. Significant alteration of these factors might lead to changes in the clinical implication of drug interactions. Yet, current guidelines to manage drug interactions have not considered the concurrent influence of the intrinsic and extrinsic factors, which potentially produces complex drug interactions, on the severity and management of drug interactions.
In our research, we directed our focus on the contribution of the three main phase I drug metabolizing enzymes. Yet, drug interactions might also be facilitated by phase II drug metabolizing enzymes and drug transporters which are also subjects to genetic polymorphisms20,21. Drug
interactions might be influenced by multiple pharmacogenes involving both genes related to drug metabolizing enzymes and transporters simultaneously22. Furthermore, genetic polymorphisms
can also change the pharmacodynamic response of a drug. Polymorphisms on drug targets might influence the severity or even direction of drug responses23. However, drug interaction studies
rarely consider simultaneous contributions of pharmacogenes related to pharmacokinetics and pharmacodynamics of drug in assessing the impact of particular drug interactions. Consequently, the knowledge about combined contribution of these gene variants on the impact of drug interactions is still limited and requires more scientific explorations.
Another important intrinsic factor that may affect the pharmacokinetics of drugs is the disease itself, either as the main disease or as a comorbidity. Diseases such as kidney failure, liver diseases, metastatic cancer or chronic inflammation have been reported to modify the magnitude of drug interactions24. In clinical practice, the information about patients’ diseases are commonly available
8
polymorphism, these diseases may generate complex drug-drug-gene-disease interactions (DDGDI) which may produce a wide range of inter-individual variability on drug exposures24. The clinical
manifestation of DDGDI can be illustrated in an opioid intoxication report because of the use of clarithromycin and voriconazole (CYP3A4 inhibitors) by a 62-year-old CYP2D6 UM patient with acute renal failure who was treated with codeine (a CYP2D6 and CYP3A4 substrate)25. The inhibition
of CYP3A4 makes the CYP2D6 the only available metabolic pathway of codeine. However, because of the gene duplication, CYP2D6 has an increased capacity to convert codeine to morphine. The blood concentration of morphine was then increased about 20 to 80 fold compared to the expected range. Moreover, the patient also had a decreased renal function. Therefore, the excretion of glucuronide metabolites of morphine was impaired and led to their accumulation. Morphine-3-glucuronide and morphine-6-glucuronide are also active metabolites and have opioid activities. As a consequence, the patient developed coma and respiratory depression25. A comprehensive review about DDGDI
can be read elsewhere24.
In order to achieve personalized management of drug interactions, clinical predictors covering genetic and non-genetic factors potentially influencing the magnitude of drug interactions should be identified and their contributions should be quantified as well as validated. Once those factors can be elucidated, they can be incorporated in a modelling and biosimulation study such as physiologically based pharmacokinetic (PBPK) in order to predict their combined effects on drug pharmacokinetics quantitatively26. This method may accommodate a lot of information which
may influence drug pharmacokinetics such as drug related aspects and variation in physiological properties of the individual patient26,27. Such a model can give guidance to providing clinical dosing
recommendation for individuals exposed to complex drug interactions27. PBPK have been shown to
successfully estimate the impact of CYP2D6 mediated DDGI and therefore, can be exercised in other clinical schemes28.
Moreover, for routine implementation, multiple genes testing using panel-based pharmacogenomics should be done prospectively and the genetic status of the pharmacogenes should be incorporated in the electronic health records of patients. Nonetheless, despite increasing published evidence reporting the importance of pharmacogenetics guided pharmacotherapy, some reports indicated that this personalized concept of medication is still infrequently adopted by clinicians29,30. Costs of pharmacogenomic testing and lack of knowledge are among the most cited
barriers hindering the implementation by clinicians31,32.
The coverage of pharmacogenetic testing costs by health insurance companies is still limited to several genes related to adverse drug reactions. Since the current evidence is still composed of small studies with limited samples, large randomized clinical trials with the aim to investigate the clinical utility and cost-effectiveness of preemptive implementation of panel-based screening for selected multiple relevant pharmacogenes in a particular age group (elderly) or condition (kidney failure) are needed30,33,34. The restriction to a particular group of genes and a high risk
population might help to improve the value of a pre-emptive pharmacogene testing strategy, and therefore might boost its implementation, in case of scarcity in the available healthcare budget35.
Therefore, the pharmacogenes panel should be quite flexible to allow additional genetic variants if the cost of genetic testing is decreasing and new clinically relevant evidence is published35.
chapter EIGHT
232
8
Furthermore, awareness and knowledge of health care practitioners about internal and external factors as well as their interactions which might influence the severity of drug interactions need to improve. The teaching and training subject of drug interaction should be expanded to include not only the mechanisms, clinical relevance and management of bimodal drug interactions such as DDI or DGI, but also complex drug interactions such as DDGI and DDGDI.
Since patients often carry multiple actionable pharmacogenes and the trend of polypharmacotherapy is increasing especially in elderly population with multiple diseases, clinicians are very likely to encounter either DDI, DGI, DDGI or DDGDI in healthcare facilities. The multimodal drug interactions are preventable when the computerized surveillance system can be used to predict these incidences. Yet, current management of DDIs with DDI alerts is still limited to only handle a binary drug interaction and lacks the ability to detect or manage complex drug interactions. Therefore, the DDI alerts incorporated in computerized physician order entry should be improved to be able to screen and combine all information about potential factors influencing the impact of drug interactions such as multi-prescriptions from different health care facilities as well as non-prescribed drugs, genetics, comorbidities, laboratory data (kidney or hepatic function), life style (smoking, drinking alcohol), or patient specific information (allergy, pregnancy, age, sex, BMI)2. The drug interaction clinical decision support should have an advanced ability to produce
clinical information about the severity of drug interactions as well as to provide case specific recommendations according to potential drug interaction risks. Nevertheless, more research is still required to provide sufficient evidence which can be used to assess clinical relevance of complex drug interactions and to support the development of guidelines on practical management of the complex interactions. Big databases which cover information about the clinical data and genetic information of patients, their medication, as well as treatment outcomes offer possibility to study the clinical impact of the complex interaction. The outcomes of the research should be then incorporated into the knowledge-bases of the DDI alerts to support the implementation of personalized management of drug interactions.
To conclude, in our work we showed that potential DDIs might still happen in daily clinical practice despite the use of the DDI alerting systems. To further improve medication safety, assessments of the clinical relevance of these potential DDIs is essential notably for older and vulnerable patients. Scientific explorations should consider the genetic background of the patient since polymorphisms in drug-metabolizing enzymes could substantially modify the outcome severity of DDIs. Even more, if a genetic polymorphism and an effector drug co-occur this may result in a complex DDGI responsible for a greater inter-individual variability in the magnitude of interactions than bimodal DDI or DGI. Consequently, generalizations made in pharmacotherapeutical management of drug interactions in clinical practice may be inappropriate. In the near future, pharmacogenomics will therefore become a fundamental backbone to support the implementation of personalized management of drug interactions.
8
References
1. van Roon EN, Flikweert S, le Comte M, et al. Clinical relevance of drug-drug interactions. Drug Safety 2005; 28: 1131-1139.
2. Smithburger PL, Buckley MS, Bejian S, Burenheide K, Kane-Gill SL. A critical evaluation of clinical decision support for the detection of drug–drug interactions. Expert opinion on drug safety 2011; 10: 871-882.
3. van der Sijs H, Mulder A, van Gelder T, Aarts J, Berg M, Vulto A. Drug safety alert generation and overriding in a large Dutch university medical centre. Pharmacoepidemiol Drug Saf 2009; 18: 941-947.
4. Van Der Sijs H, Aarts J, Vulto A, Berg M. Overriding of drug safety alerts in computerized physician order entry. Journal of the American Medical Informatics Association 2006; 13: 138-147. 5. Bahar MA, Hak E, Bos JH, Borgsteede SD,
Wilffert B. The burden and management of cytochrome P450 2D6 (CYP2D6)-mediated drug–drug interaction (DDI): co-medication of metoprolol and paroxetine or fluoxetine in the elderly. Pharmacoepidemiol Drug Saf 2017; 26: 752-765.
6. Taylor L and Tamblyn R. Reasons for physician non-adherence to electronic drug alerts. Medinfo 2004; 11: 1101-1105.
7. Wang Y, Bahar MA, Jansen AM, et al. Improving antibacterial prescribing safety in the management of COPD exacerbations: systematic review of observational and clinical studies on potential drug interactions associated with frequently prescribed antibacterials among COPD patients. J Antimicrob Chemother 2019.
8. Lau HS, de Boer A, Beuning KS, Porsius A. Validation of pharmacy records in drug exposure assessment. J Clin Epidemiol 1997; 50: 619-625. 9. Sediq R, van der Schans J, Dotinga A, et al.
Concordance assessment of self-reported medication use in the Netherlands three-generation Lifelines Cohort study with the pharmacy database iaDB. nl: The PharmLines initiative. Clinical epidemiology 2018; 10: 981.
10. Hare DL, Toukhsati SR, Johansson P, Jaarsma T. Depression and cardiovascular disease: a clinical review. Eur Heart J 2013; 35: 1365-1372. 11. Bahar MA, Wang Y, Bos JH, Wilffert B, Hak
E. Discontinuation and dose adjustment of metoprolol after metoprolol-paroxetine/ fluoxetine co-prescription in Dutch elderly. Pharmacoepidemiol Drug Saf 2018; 27: 621-629. 12. Bahar MA, Kamp J, Borgsteede SD, Hak E, Wilffert B. The impact of CYP2D6 mediated drug–drug interaction: a systematic review on a combination of metoprolol and paroxetine/fluoxetine. Br J Clin Pharmacol 2018; 84: 2704-2715.
13. Rau T, Heide R, Bergmann K, et al. Effect of the CYP2D6 genotype on metoprolol metabolism persists during long-term treatment. Pharmacogenetics and Genomics 2002; 12: 465-472. 14. Goryachkina K, Burbello A, Boldueva S, Babak S, Bergman U, Bertilsson L. Inhibition of metoprolol metabolism and potentiation of its effects by paroxetine in routinely treated patients with acute myocardial infarction (AMI). Eur J Clin Pharmacol 2008; 64: 275-282. 15. Storelli F, Matthey A, Lenglet S, Thomas A,
Desmeules J, Daali Y. Impact of CYP2D6 functional allelic variations on phenoconversion and drug– drug interactions. Clinical Pharmacology & Therapeutics 2018; 104: 148-157.
16. Bahar MA, Setiawan D, Hak E, Wilffert B. Pharmacogenetics of drug–drug interaction and drug–drug–gene interaction: a systematic review on CYP2C9, CYP2C19 and CYP2D6. Pharmacogenomics 2017; 18: 701-739.
17. Madadi P, Hildebrandt D, Gong IY, et al. Fatal hydrocodone overdose in a child: pharmacogenetics and drug interactions. Pediatrics 2010; 126: e986-9. DOI:10.1542/ peds.2009-1907 [doi].
18. Imamura CK, Furihata K, Okamoto S, Tanigawara Y. Impact of cytochrome P450 2C19 polymorphisms on the pharmacokinetics of tacrolimus when coadministered with voriconazole. J Clin Pharmacol 2016; 56: 408-413. DOI:10.1002/jcph.605 [doi].
chapter EIGHT
234
8
19. Fudio S, Borobia AM, Piñana E, et al. Evaluation of the influence of sex and CYP2C19 and CYP2D6 polymorphisms in the disposition of citalopram. Eur J Pharmacol 2010; 626: 200-204.
20. Board P, Blackburn A, Jermiin LS, Chelvanayagam G. Polymorphism of phase II enzymes: identification of new enzymes and polymorphic variants by database analysis. Toxicol Lett 1998; 102: 149-154.
21. Kerb R. Implications of genetic polymorphisms in drug transporters for pharmacotherapy. Cancer Lett 2006; 234: 4-33.
22. Neuvonen PJ. Drug interactions with HMG-CoA reductase inhibitors (statins): the importance of CYP enzymes, transporters and pharmacogenetics. Curr Opin Investig Drugs 2010; 11: 323-332.
23. Hoskins JM, Marcuello E, Altes A, et al. Irinotecan pharmacogenetics: influence of pharmacodynamic genes. Clin Cancer Res 2008; 14: 1788-1796. DOI:10.1158/1078-0432.CCR-07-1472 [doi]. 24. Storelli F, Samer C, Reny J, Desmeules J, Daali Y.
Complex Drug–Drug–Gene–Disease Interactions Involving Cytochromes P450: Systematic Review of Published Case Reports and Clinical Perspectives. Clin Pharmacokinet 2018: 1-27. 25. Gasche Y, Daali Y, Fathi M, et al. Codeine
intoxication associated with ultrarapid CYP2D6 metabolism. N Engl J Med 2004; 351: 2827-2831. 26. Jones H and Rowland-Yeo K. Basic concepts
in physiologically based pharmacokinetic modeling in drug discovery and development. CPT: pharmacometrics & systems pharmacology 2013; 2: 1-12.
27. Vieira MdL, Kim M, Apparaju S, et al. PBPK model describes the effects of comedication and genetic polymorphism on systemic exposure of drugs that undergo multiple clearance pathways. Clinical Pharmacology & Therapeutics 2014; 95: 550-557.
28. Storelli F, Desmeules J, Daali Y. Physiologically-Based Pharmacokinetic Modeling for
the Prediction of CYP2D6-Mediated Gene– Drug–Drug Interactions. CPT: pharmacometrics & systems pharmacology 2019.
29. Stanek E, Sanders C, Taber KJ, et al. Adoption of pharmacogenomic testing by US physicians: results of a nationwide survey. Clinical Pharmacology & Therapeutics 2012; 91: 450-458. 30. Bank PC, Swen JJ, Guchelaar H. A nationwide
survey of pharmacists’ perception of pharmacogenetics in the context of a clinical decision support system containing pharmacogenetics dosing recommendations. Pharmacogenomics 2017; 18: 215-225.
31. Horgan D, Jansen M, Leyens L, et al. An index of barriers for the implementation of personalised medicine and pharmacogenomics in Europe. Public Health Genomics 2014; 17: 287-298. DOI:10.1159/000368034 [doi].
32. Just KS, Steffens M, Swen JJ, Patrinos GP, Guchelaar H, Stingl JC. Medical education in pharmacogenomics—results from a survey on pharmacogenetic knowledge in healthcare professionals within the European pharmacogenomics clinical implementation project Ubiquitous Pharmacogenomics (U-PGx). Eur J Clin Pharmacol 2017; 73: 1247-1252.
33. Finkelstein J, Friedman C, Hripcsak G, Cabrera M. Pharmacogenetic polymorphism as an independent risk factor for frequent hospitalizations in older adults with polypharmacy: a pilot study. Pharmgenomics Pers Med 2016; 9: 107-116. DOI:10.2147/PGPM. S117014 [doi].
34. Elliott LS, Henderson JC, Neradilek MB, Moyer NA, Ashcraft KC, Thirumaran RK. Clinical impact of pharmacogenetic profiling with a clinical decision support tool in polypharmacy home health patients: A prospective pilot randomized controlled trial. PLoS One 2017; 12: e0170905. 35. Alshabeeb MA, Deneer VH, Khan A, Asselbergs
FW. Use of Pharmacogenetic Drugs by the Dutch Population. Frontiers in genetics 2019; 10.
