Predictive Biomarkers and Personalized Therapy: Use of Pharmacogenetic Testing in a Scandinavian Perspective

ABSTRACT

Precision medicine has significantly advanced through the development of predictive biomarkers based on pharmacogenetic (PGx) testing. These tests identify interactions between drugs and genetic variants that influence patient responses to treatments. Understanding genetic variations in drug‐metabolizing enzymes, receptors and transporters and their impact on pharmacokinetics and pharmacodynamics allows for the prediction of drug effects and side effects, enabling tailored treatments for different patient groups. This review focuses on drugs metabolized by cytochrome P450 (CYP450) enzymes, for example, citalopram and clopidogrel or transported by the solute carrier organic anion transporter family member 1B1 (SLCO1B1), for example, atorvastatin and simvastatin, with PGx dosing guidelines, in the context of consumption in Scandinavian countries. A major barrier to the widespread adoption of PGx tests in clinical practice has been healthcare professionals' uncertainty about their efficacy, complexity in result interpretation and questions regarding the evidence base. However, recent studies have demonstrated PGx testing has the potential to improve treatment outcomes, reduce adverse drug reactions and achieve cost savings. These findings underscore the potential of PGx testing as a valuable tool in clinical decision making, promoting its use in a pre‐emptive manner to enhance patient care.

Summary.

Pharmacogenetic (PGx) testing has advanced precision medicine by identifying drug–gene interactions that affect patient responses. Understanding genetic variations in drug‐metabolizing enzymes, receptors and transporters helps predict drug effects and side effects, enabling personalized treatments. This review highlights drugs metabolized by CYP450 enzymes (e.g., citalopram and clopidogrel) and transported by SLCO1B1 (e.g., atorvastatin and simvastatin) in Scandinavia. Adoption of PGx tests faces barriers like healthcare professionals' uncertainty, result interpretation complexity and questions regarding the evidence base. However, recent studies show that PGx testing can improve treatment outcomes, reduce adverse reactions and save costs, promoting its use in clinical decision making to enhance patient care.

1. Introduction

Precision medicine is based on the last 20–30 years of development in ‘omics’ technologies, which have been instrumental in the paradigm shift towards a more stratified and individual approach to medical treatment [1]. The use of pharmacogenetics, referred to as ‘PGx’, is a central tool for identifying predictive biomarkers based on genetic profiling. These predictive biomarkers can thus be helpful in predicting the effects and side effects of drugs [1, 2] based on pharmacokinetic and pharmacodynamic parameters. The PGx test can determine whether minor variations in the genes responsible for drug‐metabolizing enzymes, receptors and transporters affect their functionality. These variations can thereby serve as predictive biomarkers to identify individuals who are more likely to respond favourably to specific medical treatments, leading to improved clinical effectiveness and reduced side effects [1, 2]. The US (FDA) and European (EMA) health authorities and a number of international clinical organizations [1, 2, 3, 4] and pharmaceutical companies have increasingly focused on the use of predictive biomarkers and their implementation in clinical practice. This has resulted in recommendations and guidelines for clinical development and use of predictive biomarkers. Therefore, the FDA and EMA have issued PGx‐based annotations [1, 2, 4] for a large number of different drugs used in daily clinical practice and prescribed by general practitioners [5, 6]. Furthermore, several drugs, particularly those used in cancer treatment, for example, trastuzumab or pertuzumab for HER2‐positive breast cancer, have associated diagnostic tests based on biomarkers [1]. These tests provide essential information for the safe and effective use of the corresponding drug or biological product [1, 2]. These are referred to as companion diagnostics or CDx and are subject to regulatory control and approval [1, 2]. The utilization of predictive biomarkers in clinical practice based on PGx tests in Denmark has been discussed in several instances [7, 8]. However, it has not gained significant traction in clinical practice despite a growing body of evidence supporting the integration of PGx in clinical decision‐making processes [8, 9, 10]. Similar considerations are also taking place in both Sweden and Norway [11, 12]. Although there is recognition of the potential benefits of PGx in improving drug therapy and reducing adverse effects, several challenges need to be addressed to achieve widespread implementation of PGx [10, 12, 13, 14]. The European Ubiquitous Pharmacogenomics (U‐PGx) project was an initiative funded by the European Union's Horizon 2020 program [10]. Its primary goal was to integrate pharmacogenomics (PGx) into routine clinical practice across Europe. For example, in the Netherlands, initiatives have been taken to implement a so‐called PGx passport, containing a comprehensive report of a person's genetic information, including genetic variations that may affect their response to certain medications [15]. Healthcare providers can use this information to guide treatment decisions and improve patient outcomes by reducing the risk of adverse drug reactions and optimizing drug efficacy [10]. The PREPARE study, which was initiated by the U‐PGx project [10, 16], recently published in the Lancet [16], found that patients whose doses were adjusted according to their genotype on a panel of 12 pharmacogenes experienced 30% fewer serious side effects assessed in several medical specialties, including oncology, cardiology, psychiatry and general medicine. Furthermore, a pilot study started in 2023 in the United Kingdom to pre‐emptively explore the use of PGx testing for patients taking statins, antidepressants and proton pump inhibitors (PPIs) to adjust the dose and ensure that they receive the right dosing [17] and similar initiatives have been initiated elsewhere [18]. In addition, many ‘summary of product characteristics’, which is a detailed document approved by regulatory authorities, describes the properties and approved conditions of use for a drug including PGx‐based information. However, they do not provide any guidelines for clinical decision making and use of PGx. This article presents an overview of the status and implementation of predictive biomarkers in clinical practice, with a focus on Scandinavian countries, along with the barriers to their clinical use. It focuses specifically on drugs metabolized by cytochrome P450 (CYP450) drug‐metabolizing enzymes or transported by the solute carrier organic anion transporter family member 1B1 (SLCO1B1), which plays a crucial role in the uptake of statins in liver cells.

2. Pharmacogenetics of CYP450 and SLCO1B1: Clinical Implications and Impact on Treatment Outcomes

CYP450 enzymes are primarily expressed in the liver and play a central role in the oxidative biotransformation of 70%–80% of all clinically used drugs [5]. The CYP450 family of enzymes and in particular CYP2C9, CYP2D6 and CYP2C19 are crucial for identifying and understanding differences in drug responses and side effects [19, 20]. Genetic polymorphisms and variations in the genes through point mutations can result in reduced or absent CYP450 enzyme activity, leading to drugs not being metabolized and consequently to unintentional overdose and introduction of side effects [19, 20, 21]. On the other hand, the gene may also be overexpressed by copy number variation, leading to increased CYP450 enzyme activity and thereby rapid inactivation of the drug with a lack of pharmacological effect and vice versa for prodrugs.

Thus, a PGx test can be a useful tool in identifying the pharmacological relationship between a drug and a genetic variant of a CYP450 enzyme, often referred to as ‘drug–gene interaction’ (DGI), which can affect a patient's response to drug treatment. Based on the DGI activity scores for metabolic activity [2, 3, 9], these are classified into five different phenotypes: ‘poor metabolizer’ (PM), ‘intermediate metabolizer’ (IM), ‘extensive metabolizer’ (EM; normal activity) and ‘rapid and ultra‐rapid metabolizer’ (RM and UM). The same considerations also apply to the SLCO1B1 transporter, and the PGx test can identify whether a person has ‘normal function’ (NF), ‘intermediate function’ (IF) or ‘low function’ (LF) of the transporter. Increased exposure to statins has been observed in persons with IF and LF, which has resulted in the occurrence of myopathy and rhabdomyolysis [3], especially with high doses of statins. Table 1 shows the percentage distribution of the different phenotypes in Caucasians for various CYP450 enzymes and SLCO1B1. However, it should be mentioned that genotype/phenotype distribution is dependent on ethno‐geographic origin [22].

TABLE 1.

Phenotype distribution (%) among Caucasians.

	CYP2C9	CYP2C19	CYP2D6		SLCO1B1
EM	64.0	39.1	84.2	NF	72.3
IM	20.8	26.9	6.2	IF	25.5
PM	15.2	2.6	5.4	LF	2.3
RM	—	26.9	—	—	—
UM	—	4.6	4.2	—	—

Note: CYP450‐genotypes: UM ‘ultra‐rapid metabolizer’, RM ‘rapid metabolizer’, EM ‘extensive metabolizer’ (normal activity), IM ‘intermediate metabolizer’ and PM ‘poor metabolizer’. SLCO1B1 genotypes: NF ‘normal function’, IF ‘intermediate function’ and LF ‘low function’. Grey shaded means not applicable.

Source: Data modified from [22].

Over the past two decades, there has been extensive development and validation of clinical dosing guidelines for DGI for a variety of drugs. These guidelines can be accessed through the Pharmacogenomics Knowledge Base (PharmGKB; https://www.pharmgkb.org/) [3] and were created in collaboration with The Clinical Pharmacogenetics Implementation Consortium (CPIC) [23] and the Dutch Pharmacogenetics Working Group (DPWG) [24, 25]. However, other clinical societies have also contributed to the guidelines that can be seen at the PharmGKB homepage. PharmGKB contains information about drugs, genes, diseases and drug annotations in addition to 181 PGx‐based clinical dosing guidelines. The clinical dosing guidelines offer widely accepted recommendations for dose adjustment, dose monitoring and the use of alternative drugs based on genotype/phenotype scores [3]. Initiatives to harmonize the clinical guidelines of the DPWG and the CPIC aim to align pharmacogenomic standards globally, optimizing patient outcomes and ensuring consistent application of personalized medicine [26].

Table 2 gives an overview of PGx annotations from PharmGKB and FDA relating to CYP450 and SLCO1B1 and the corresponding consumption of drugs expressed as total number users and prevalence of use in Denmark, Norway and Sweden in 2023.

TABLE 2.

Pharmacogenetic information and consumption of PGx drugs often used in clinical practice in Denmark, Sweden and Norway.

Drug	ATC	DK	SE	NO	Biomarker	PharmGKB PGx annotations	FDA PGX annotations
		Number of users (prevalence: users/1000)
Amitriptylin	N06AA09	38 050 (6.41)	163 911 (15.58)	82 466 (15.02)	CYP2D6	UM/IM/PM UM/IM/PM	UM/IM/PM Sect. 3
Amitriptyline					CYP2C19	UM/RM/PM —	—
Aripiprazole	N05AX12	14 705 (2.48)	26 748 (2.54)	8227 (1.50)	CYP2D6	— PM	PM, Sect. 1
Atomoxetine	N06BA09	20 825 (3.51)	19 203 (1.83)	6948 (1.27)	CYP2D6	UM/IM/PM UM/PM	PM, Sect. 1
Atorvastatin	C10AA05	561 900 (94.71)	774 122 (73.57)	487 649 (88.84)	SLCO1B1	LF/IF LF/IF	LF Sect. 3
Brexpiprazole	N05AX16	370 (0.06)	4 (0.00)	192 (0.03)	CYP2D6	— PM	PM, Sect. 1
Celecoxib	M01AH01	1970 (0.33)	42 423 (4.03)	22 998 (4.19)	CYP2C9	IM/PM —	PM, Sect. 1
Citalopram	N06AB04	65 375 (11.02)	110 024 (10.46)	12 825 (2.34)	CYP2C19	UM/PM IM/PM	PM, Sect. 1
Clomipramine	N06AA04	2370 (0.40)	11 480 (1.09)	1519 (0.28)	CYP2D6	UM/IM/PM IM/PM	PM Sect. 3
Clomipramine					CYP2C19	UM/RM/PM UM/PM	—
Clopidogrel	B01AC04	146 155 (24.64)	105 777 (10.05)	56 108 (10.22)	CYP2C19	IM/PM IM/PM	IM/PM Sect. 1
Codeine	R05DA04	73 115 (12.32)	2084 (0.20)	15 576 (2.84)	CYP2D6	UM/IM/PM UM/IM/PM	UM, PM Sects 1 and 2
Escitalopram	N06AB10	26 640 (4.49)	173 827 (16.52)	112 393 (20.48)	CYP2C19	UM/PM UM/IM/PM	Sect. 3
Ibuprofen	M01AE01	601 275 (101.35)	154 276 (14.66)	235 028 (42.82)	CYP2C9	IM/PM —	Sect. 3
Lansoprazole	A02BC03	115 605 (19.49)	14 593 (1.39)	26 440 (4.82)	CYP2C19	UM/RM/IM/PM UM	—
Metoprolol	C07AB02	299 395 (50.47)	581 743 (55.29)	294 262 (53.61)	CYP2D6	PM UM/IM/PM	PM Sect. 3
Nortriptyline	N06AA10	12 220 (2.06)	1714 (0.16)	2681 (0.49)	CYP2D6	UM/IM/PM UM/IM/PM	UM/IM/PM Sect. 3
Omeprazole	A02BC01	92 725 (15.63)	706 168 (67.12)	36 956 (6.73)	CYP2C19	UM/RM/IM/PM UM	IM/PM Sect. 3
Ondansetron	A04AA01	25 400 (4.28)	46 894 (4.46)	22 395 (4.08)	CYP2D6	UM —	—
Pantoprazole	A02BC02	424 340 (71.53)	94 523 (8.98)	388 514 (70.78)	CYP2C19	UM/RM/IM/PM UM	IM/PM Sect. 1
Paroxetine	N06AB05	10 340 (1.74)	23 591 (2.24)	10 269 (1.87)	CYP2D6	UM/PM UM	Sect. 3
Pimozide	N05AG02	385 (0.07)	43 (0.0)	79 (0.01)	CYP2D6	— IM/PM	Sect. 1 test*
Sertraline	N06AB06	171 835 (28.96)	350 374 (33.30)	53 113 (9.68)	CYP2C19	PM PM	—
Simvastatin	C10AA01	177 570 (29.93)	221 700 (21.07)	130 234 (23.73)	SLCO1B1	LF/IF LF/IF	LF/IF Sect. 2
Tamoxifen	L02BA01	700 (0.12)	15 658 (1.49)	4951 (0.90)	CYP2D6	IM/PM IM/PM	IM/PM Sect. 3
Tramadol	N02AX02	103 495 (17.45)	39 942 (3.80)	221 326 (40.32)	CYP2D6	UM/IM/PM UM/IM/PM	UM, Sect. 1
Venlafaxine	N06AX16	45 295 (7.64)	89 480 (8.50)	36 958 (6.73)	CYP2D6	PM RM/IM/PM	PM Sect. 1
Warfarin	B01AA03	30 445 (5.13)	50 412 (4.79)	19 629 (3.58)	CYP2C9	— IM/PM	IM/PM Sect. 1

Note: Consumption in 2023 of drugs with PGx based dosing guidelines according to PharmGKB [3] in Denmark (DK), Sweden (SE) and Norway (NO) for the predictive biomarkers CYP2D6, CYP2C19, CYP2C9 and SLCO1B1. PharmGKB PGx annotations: CYP450‐phenotypes: UM ‘ultra‐rapid metabolizer’, RM ‘rapid metabolizer’, EM ‘extensive metabolizer’ (normal activity), IM ‘intermediate metabolizer’ and PM ‘poor metabolizer’. SLCO1B1 phenotypes: NF ‘normal function’, IF ‘intermediate function’ and LF ‘low function’. Genotypes in plain letter are from CPIC and in bold from DPWG. FDA annotations [4]: Sect. 1 ‘pharmacogenetic associations for which the data support therapeutic management recommendations’; Sect. 2 ‘pharmacogenetic associations for which the data indicate a potential therapeutic impact on safety or response’ and Sect. 3 ‘pharmacogenetic associations for which the data demonstrate a potential impact on pharmacokinetic properties only’. Data collected in October 2024. Consumption data expressed as total number of user and in brackets prevalence (number of users/1000 persons in the general population) were retrieved from Medstat (https://medstat.dk/) based on sale reported to the Register of Medicinal Product Statistics in Denmark, Legemiddelregisteret, Folkehelseinstituttet (https://statistikk.fhi.no/lmr); Norway and Statistikdatabas för läkemedel. Stockholm; Sweden: Socialstyrelsen (https://www.socialstyrelsen.se/statistik‐och‐data/statistik/statistikdatabasen/). Dash (—) indicates no recommendation.

3. Pharmacogenetics and Polypharmacy

Polypharmacy, often defined as the use of ≥ 5 medications, is becoming a growing concern as the incidence of multiple diseases and polypharmacy increases with age [21, 27]. This is linked to an increased risk of drug‐related problems such as side effects, drug interactions, hospitalization and mortality [9, 21, 28]. Recent studies in Denmark have shown that a large part of the Danish population, especially elderly nursing home residents, persons with diabetes and persons with chronic kidney disease, is treated with drugs and/or drug combinations for which PGx‐based dosing recommendations exist [6, 9, 29, 30, 31, 32]. The prevalence of drug use for drugs with PGx‐based dosing guidelines is also significantly higher in these groups than in the general population [6, 29, 30, 31, 32]. An important issue regarding polypharmacy is the concept of phenoconversion, which can further complicate the situation and potentially lead to a genotype–phenotype mismatch [19, 33]. This means that a person can be phenoconverted when taking additional drugs, and this is referred to as ‘drug–drug–gene interaction’ (DDGI). Phenoconversion has been shown to change a person's phenotype in polypharmacy patients [19, 21, 33], for example, from RM to EM or from EM to PM (see Figure 1). This emphasizes the importance of considering and accounting for both DGI and DDGI [19, 33] in a given clinical situation, particularly in cases of polypharmacy. However, although clinical decision support systems that provide recommendations on drug–drug interactions (DDIs) and some guidelines on DGIs [34] have been implemented during electronic prescribing, DDGIs have rarely been considered due to their complexity [35].

FIGURE 1.

Illustration of phenotype and phenoconversion. Note: Genotype scores for CYP450 metabolic activity are classified into five different phenotypes: ‘poor metabolizer’ (PM), ‘intermediate metabolizer’ (IM), ‘extensive metabolizer’ (EM; normal activity) and ‘rapid or ultra‐rapid metabolizer’ (RM or UM) shown in this fig. as RM. The same considerations also apply to the SLCO1B1 transporter: ‘normal function’ (NF), ‘intermediate function’ (IF) or ‘low function’ (LF) of the transporter. Phenoconversion has been shown to change a person's phenotype, that is, a person only taking Drug A can be phenoconverted when additional taking Drug B.

4. Examples of Drug Gene Interactions

4.1. Citalopram

CPIC (see https://www.pharmgkb.org/guidelineAnnotations) recommends the use of an alternative medication for individuals who are CYP2C19 UM and for those who are CYP2C19 PM. Otherwise, it should be considered to reduce the dose by 50% of the recommended starting dose and titrate to response for PM. The FDA recommends a maximum daily dose of 20‐mg citalopram for PMs, which is half the maximum recommended dose for adults with normal enzyme activity [4]. The product label for citalopram recommends that PMs start with an initial dose of 10 mg daily for the first 2 weeks of treatment; depending on the response, the dose can be increased to a maximum of 20 mg daily. In 2023 (see Table 2), there were 65 375 citalopram users in Denmark (DK), 110 024 in Sweden (SE) and 12 825 in Norway (NO). Based on the phenotype distribution among Caucasians shown in Tables 1 and 2, 6% and 4.6% will either be PM or UM CYP2C19. This means that approximately 1700 individuals in DK, 2860 in SE and 335 in NO taking citalopram may be CYP2C19 PMs, thus in risk of adverse effects reducing quality of life. Additionally, approximately 3000 individuals in DK, 5060 in SE and 5900 in NO taking citalopram may be CYP2C19 UMs and will therefore not achieve sufficient therapeutic dose. It should be noted that DPWG opposite CPIC provides recommendations for IM and PM (see Table 2).

4.2. Clopidogrel

PharmGKB (see https://www.pharmgkb.org/guidelineAnnotations) recommends the use of an alternative platelet function inhibitor (P2Y12 inhibitor) to clopidogrel for the treatment of CYP2C19 PMs and IMs for thrombosis prophylaxis. FDA notifies that PMs and IMs have lower plasma concentrations of the active metabolite (clopidogrel is a prodrug), lower platelet inhibitory function (also indicated in the ‘summary of product characteristics’), and this can result in a higher risk of cardiovascular events [4]. Therefore, the use of an alternative platelet P2Y12 inhibitor should be considered. Recently, the American Heart Association issued a scientific statement advocating for genetic testing of CYP2C19 before prescribing clopidogrel [36]. This recommendation is based on extensive data from pharmacokinetic, pharmacodynamic, observational, meta‐analysis and clinical trials. Based on the number of total clopidogrel users shown in Table 2 and the frequency of being PM or IM (see Table 1), it can be calculated that approximately 3800 users are CYP2C19 PM and 39 300 are IM in DK. In SE, the numbers are approximately 2750 CYP2C19 PM users and 28 450 IM users. In NO, there are approximately 1460 CYP2C19 PM users and 15 100 IM users, respectively. These examples demonstrate that a relatively high number of patients can potentially be affected by their inherited genotypes. Similar considerations apply to the other drugs shown in Table 2.

5. Example of DDGIs

Furthermore, clopidogrel use is associated with an increased risk of gastrointestinal bleeding and is often administered with PPIs [31]. Recently, it has been shown that approximately 35% of clopidogrel users in the general population also take PPIs [31]. Among individuals with diabetes, this proportion is approximately 42% [31] while among nursing home residents, it is approximately 45% [29]. This is not without problems, as PPIs inhibit CYP2C19 activity and potentially can reduce the effects of clopidogrel and give rise to DDGI by phenoconversion (Figure 1). This finding prompted the FDA to issue a warning stating that PPIs can reduce the antithrombotic effect of clopidogrel by nearly 50% when used concurrently [37]. Consequently, it is recommended to use lansoprazole or pantoprazole in combination with clopidogrel, rather than the more potent inhibitors, that is, omeprazole and esomeprazole, if concurrently treatment with PPIs is deemed necessary. However, depending on the drug interaction tracker used in Denmark, Norway and Sweden, different recommendations for the combination of clopidogrel and PPIs are reported [31]. For instance, the combination of clopidogrel with esomeprazole, lansoprazole or omeprazole is classified as ‘serious, use alternate’ by Norwegian ‘Felleskatalogen’ (https://www.felleskatalogen.no/medisin/), whereas the Danish ‘Interaktionsdatabasen’ (https://www.interaktionsdatabasen.dk/) and the Swedish ‘Janusmed’ (https://janusmed.se/interaktioner) categorizes these combinations as ‘monitor closely’. However, it can be questioned whether clopidogrel users who are CYP2C19 IM or PM should take PPIs at all. PharmGKB, FDA, EMA and product summaries do not take DDGI into account for drugs with PGx labels, including clopidogrel [6, 31]. The same considerations apply to the use of citalopram and PPIs. Here, too, differences exist among the various drug interaction trackers: ‘Felleskatalogen’ rates the combination of citalopram with either esomeprazole or omeprazole as ‘serious, use alternative’, whereas the other two trackers do not assign this level of severity. Therefore, DDIs as well as DDGIs are recognized as a potential serious issue [35, 38]. Looking ahead, artificial intelligence presents a promising avenue for creating sophisticated algorithms capable of managing the complexities associated with both DGIs and DDGIs.

6. Future Perspectives of Pre‐Emptive Use of PGx

Conventional prevention strategies have been employed to mitigate the risks of potentially inappropriate medication and polypharmacy, including medication review, de‐prescribing, patient education and regular monitoring such as therapeutic drug monitoring or patient‐reported outcomes [38, 39, 40]. Integrating PGx pre‐emptively into clinical practice could potentially serve as an additional tool to enhance preventive efforts [18]. In clinical practice, two primary approaches to PGx testing have been adopted: reactive testing and pre‐emptive testing [18, 41]. Reactive testing involves performing PGx tests after a patient has been prescribed a drug and is experiencing side effects or is not responding as expected to the treatment. Pre‐emptive testing, on the other hand, involves conducting PGx tests before prescribing medications, often as part of routine clinical practice or during initial patient assessments. Incorporating these approaches into clinical practice can significantly enhance precision medicine, optimizing treatment outcomes while minimizing adverse drug reactions. For example, as demonstrated in the PREPARE study [16], in the management of polypharmacy patients [28], and the use of CYP2C19 genotyping to guide clopidogrel antiplatelet therapy in patients undergoing percutaneous coronary intervention [36], these strategies can lead to better health outcomes and substantial cost savings [28, 36]. In addition, PM statuses of CYP2C19 and CYP2D6 have been associated with adverse outcomes in children, adolescents and young adults with depression who are newly prescribed (es)citalopram, sertraline or fluoxetine. This suggests the utility of PGx testing, particularly in younger individuals, for guiding antidepressant treatment [42].

The typical cost of PGx single gene test ranges typically from $100 to $500 while the price of a panel test of several genes may reach double of a single gene test. Currently, PGx testing has shifted towards a pre‐emptive and panel‐based approach when drugs with PGx guidelines are prescribed, due to cost‐effectiveness [18].

Despite numerous initiatives and advancements in PGx implementation, such as those seen in the Netherlands and UK [15, 17], significant barriers still hinder the proactive use of PGx tests in daily clinical practice. The use of predictive biomarkers based on PGx tests in general practice in Denmark has been described previously [7, 8]. The primary barriers to widespread use of PGx tests in clinical practice and their inclusion in clinical decision making are primarily a matter of education and awareness about both the possibilities and the limitations of these tests [7, 8, 21, 43]. Additionally, the complexity associated with interpreting PGx test results and uncertainties about the evidence base constitute the most significant concerns in this realm, particularly questioning whether there exists adequate evidence to support the use of PGx testing [7, 21, 43]. Scientific evidence, throughout the years, has always been the gold standard of randomized clinical trials. However, there has been a lack of randomized clinical trials in the field of the benefit use of PGx testing. This has resulted in a slow uptake of PGx testing into clinical practice owing to scarce evidence for clinical utility of the PGx tests [21, 43]. The results of a pilot study on pre‐emptive PGx testing in the United Kingdom to optimize prescribing in primary care [17], along with the outcome of the PREPARE study, demonstrated that preventive PGx tests can reduce side effects by up to 30% [16], and other initiatives [10, 15] further support the notion of considering PGx as an adjunct to other clinical considerations. These considerations include the individual's medical history, current health status, concomitant medications, estimated glomerular filtration rate (eGFR) and other relevant factors, to make informed decisions about drug management [20]. By tailoring drug treatments to individual genetic profiles, PGx testing has the potential to improve treatment outcomes, reduce adverse drug reactions and achieve cost savings. Despite existing barriers such as the need for greater education and evidence, the growing interest and successful pilot studies underscore the potential of PGx testing to optimize patient care. It is now believed that there is increasing interest in the use of PGx tests in Scandinavia [12, 13, 14, 29, 31].

Author Contributions

All authors contributed to this ‘mini review’ according to journal author guidelines and the ICMJE definition of authorship.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available in Medstat online (https://medstat.dk/). These data were derived from the following resources available in the public domain: Legemiddelregisteret, Folkehelseinstituttet Norway, (https://statistikk.fhi.no/lmr) ‐ Statistikdatabas för läkemedel. Stockholm, Sweden (https://www.socialstyrelsen.se/statistik‐och‐data/statistik/statistikdatabasen/).