The frequency of rs2231142 in ABCG2 among Asian subgroups: implications for personalized rosuvastatin dosing

Abstract

Statins are widely used medications for the primary and secondary prevention of cardiovascular diseases. Statin-induced musculoskeletal symptoms are the primary adverse drug events contributing to poor adherence to lipid-lowering therapy. Rosuvastatin is characterized by interindividual differences in systemic exposure among different patient population subgroups. The missense variant Q141K within ABCG2, highly prevalent in some Asian subgroups, results in decreased transporter efflux function and increased exposure to rosuvastatin. We aim to highlight the implications of ABCG2 genotype in prescribing rosuvastatin and the ramifications of interpopulation differences in Q141K frequencies in the starting dose of rosuvastatin in major Asian subgroups, using the most recent genetic-based guidelines. The high frequency of Q141K in Filipinos could warrant a lower starting rosuvastatin dose versus non-Filipinos. The Q141K genotype frequencies in Asian subgroups suggest significant interpopulation differences, reinforcing the need to move beyond race-based to genotype-based rosuvastatin dosing.

Plain language summary

Rosuvastatin, a commonly prescribed cholesterol-lowering drug, has differences in response between different population subgroups. Rosuvastatin may also cause muscle pains, contributing to low adherence to the medication. Asians have a significantly high frequency of genetic variation (Q141K) within ABCG2, a critical rosuvastatin-efflux pump, leading to less functional transporter and higher drug levels. This special report highlights the role of ABCG2 genotyping in prescribing rosuvastatin. Also, it describes the consequences of between-population differences in the Q141K frequency in deciding the starting dose in individuals of Asian background, using the most recent genetic-based guidelines. Among different Asian subgroups, Filipinos have the highest Q141K polymorphism frequency and are more likely to require a lower starting dose of rosuvastatin than other Asians. Knowledge of the Q141K frequency in different Asian subgroups could drive individualized rosuvastatin dosing and reduce racial disparities in drug safety and efficacy.

Tweetable abstract

The reduced ABCG2 function allele frequency is highly prevalent among Asians, especially Filipinos. Rosuvastatin is a substrate for ABCG2. Disproportionate ABCG2 allele frequency in Filipinos warrants a personalized rosuvastatin starting dose.

Statins are a group of lipid-lowering medications commonly prescribed to manage cardiovascular risk factors and prevent major cardiovascular events [1,2]. The main drugs in this family are simvastatin, lovastatin, rosuvastatin, atorvastatin, fluvastatin and pitavastatin [1,2]. Based on the most recent report by the US Agency for Healthcare Research and Quality (AHRQ) report [3], statins are one of the top prescribed drugs, 13.7% of patients living in the USA have been prescribed a statin [3–5]. Further, atorvastatin is the most commonly prescribed drug in America and is prescribed to 24.5 million (7.5%) patients in the USA [3,5]. Statins are safe and effective; however, adverse drug events, the so-called statin-associated musculoskeletal symptoms (SAMS), could occur, which could negatively affect medication adherence [6,7]. Rosuvastatin is a commonly used statin and is the 26th most prescribed drug in the USA [1,2,5]. Rosuvastatin's disposition and risk of SAMS are influenced by several factors, including polymorphisms within key transporters and phase I metabolizing enzymes (i.e., SLCO1B1, ABCG2 and CYP2C9) [1]. A genetic variant within ABCG2, a major rosuvastatin transporter in the liver and intestine, can occur at various frequencies among US populations, with the Asian subgroups having higher frequencies of decreased or nonfunctional transporter activity than non-Asians [8–11]. With the low frequency of SLCO1B1 variants in Asian subgroups, this observation suggests that genetic variants in ABCG2 among Asians are more likely to influence the disposition of rosuvastatin and the risk of SAMS. In light of the 2022 Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines for statins [1], this special report aims to highlight the importance of considering the ABCG2 genotype before prescribing rosuvastatin and the implications of interpopulation allele frequencies in determining the starting dose of the drug in Asian population subgroups (Filipino, Japanese, Korean, Hmong and Southern Han-Chinese).

Methods

PubMed was used to conduct a Medline search for our special report. Our strategy specifically involved searching PubMed for relevant articles, using the search phrases ‘pharmacogenomics’ OR ‘pharmacogenetics’ OR ‘rosuvastatin’ AND ‘Asian’ OR ‘ABCG2’. The references for chosen studies were scanned to include the most pertinent research articles. Additionally, we used the most contemporary rosuvastatin dosing recommendations from the CPIC to inform the implications of the differences in the allele frequency of rs2231142 in ABCG2 among Asian population subgroups relative to other major racial population groups.

Challenges in statins pharmacotherapy

Statin use is a cornerstone in managing lipid disorders and preventing coronary heart disease [12,13]. Numerous large clinical trials and post-marketing surveillance studies have repeatedly shown that the use of statins has a significant positive impact on cardiovascular disease-related mortality [12,13]. Despite the shared mechanism of action, various statins differentially lower cholesterol and other lipid traits. Of interest, rosuvastatin exerts its low-density lipoprotein cholesterol (LDL-C) lowering effects through the inhibition of 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase in the hepatocytes [2,14]. According to 2019 lipid treatment guidelines, moderate- or high-intensity rosuvastatin is indicated in patients with an age range of 40–75 years with an LDL-C ≥70 mg/dl and with or without diabetes mellitus (DM) and at a 10-year clinical atherosclerotic cardiovascular disease (ASCVD) risk of ≥7.5%. Additionally, rosuvastatin is indicated in patients with high LDL-C (≥190 mg/dl) or established ASCVD [2]. The recommended rosuvastatin starting doses in the general population are 10 mg and 20 mg daily [2].

Although statin therapy is highly effective in reducing morbidity and mortality associated with heart diseases, it is alarming that as many as half of the patients discontinue statin therapy within the first year after commencing lipid-lowering treatment. With the compounded benefits of taking statins, adherence to statin therapy becomes crucial for optimal treatment outcomes [2,15]. While there are multiple reasons for stopping statin therapy, drug adverse events, especially myalgia, were one of the most common reasons among former statin users [16]. SAMS risk factors could include female sex, liver or renal dysfunction, hypothyroidism, advanced age, concomitant use of interacting medications [6,7] and specific genetic polymorphisms in transporters known to affect blood levels of statins [1]. In a meta-analysis of cohort and randomized clinical trials, self-reported Asian ancestry was significantly associated with a higher incidence of statin intolerance (OR: 1.3; p < 0.05) compared with European (EUR) [17]. The disproportionate incidence of statin intolerance among individuals of Asian ancestry highlights the need for further investigations to explain the basis of this observation. With the high burden of dyslipidemia among select Asian subgroups, tools to optimize selecting the proper statin at the optimal dose are crucially needed to reduce the health disparities of cardiovascular diseases among Asian populations [18,19].

Rosuvastatin & Asian ancestry

Unlike other statins, rosuvastatin is minimally metabolized by cytochrome P450 2C9 (CYP2C9) (∼10%) and primarily excreted unchanged into the bile [20]. CYP2C9 forms the active metabolite, the N-desmethyl rosuvastatin, which accounts for one-sixth to one-half of the inhibitory activity of HMG-CoA reductase of the parent drug [14]. After intravenous administration, the drug is eliminated through the liver (72%) and kidneys (28%) [4]. Overall, greater than 90% of active plasma HMG-CoA reductase inhibitory activity is accounted for by rosuvastatin [14]. As a hydrophilic drug, rosuvastatin is highly dependent on drug transporters to pass through cell membranes to reach its site of action [20].

The effects of hepatic uptake and efflux transporters on the pharmacokinetics (PK) and pharmacodynamics (PD) of rosuvastatin have been thoroughly documented in the literature [1,21]. Efflux transporters, like the apical ATP-binding cassette transporter G2 (ABCG2) (i.e., breast cancer resistance protein), eliminate rosuvastatin into the bile, while the uptake transporters, like the organic anion transporting polypeptide (OATP) family 1B1 (OAT1B1), facilitate rosuvastatin intake into hepatocytes where the drug inhibits the rate-limiting step in the biosynthesis of cholesterol [1].

Nevertheless, it is critical to consider the interindividual variabilities in the exposure and response to rosuvastatin to mitigate the risk of intolerable side effects and suboptimal drug response. Additionally, initiating a lower dose of rosuvastatin in the population of Asian descent has been widely recommended by the US FDA and current lipid guidelines due to increased drug exposure and high risk for adverse drug events [14]. However, the basis of such a recommendation remains elusive. In light of the 2022 CPIC guidelines on dosing statins [1], this special report aims to highlight the implications of the ABCG2 genotype when initiating rosuvastatin and the ramifications of interpopulation subgroup allele frequencies in determining the starting dose of the drug in Asian population subgroups. Furthermore, the Asian categorization is a broad descriptor of multiple population subgroups of historical, cultural and genetic differences. Therefore, aggregating Asian population subgroups becomes a barrier to personalized patient care and a mask for granular inter-subgroup differences. Thus, highlighting inter-subgroup differences in ABCG2 allele and genotype frequencies could move us beyond race-based patient care recommendations. Also, this will provide a new perspective on personalized rosuvastatin dosing among patients of Asian ancestry who otherwise would be prescribed the lowest rosuvastatin dose due to their self-reported ancestral background.

Interindividual variabilities in the exposure to rosuvastatin

According to the FDA-approved rosuvastatin drug label, individuals of Asian descent are recommended to start with a lower dose (5 mg) of rosuvastatin than non-Asians (10 mg) as they are at risk of a twofold increase in median drug levels compared with Whites [14]. According to multiple PK studies, several Asian population subgroups, including Filipino, Japanese, Korean and Chinese, appear to have higher rosuvastatin exposure than EUR [22,23]. However, the reason for this effect needs to be further elucidated. It is proposed that the interethnic disparities in rosuvastatin exposure have not been linked to extrinsic factors such as the environment, diet, or body weight, according to previous studies [23,24]. The transport of rosuvastatin into and out of liver cells by drug transporters may impact the observed interethnic differences more than CYP450 enzymes that minimally metabolize the drug. Also, a growing body of evidence has linked the reduced function genetic polymorphisms in both transporters encoding genes of OATP1B1 (SLCO1B1) and BCRP (ABCG2) to significantly influence the exposure of rosuvastatin and to be a probable cause of interethnic variations in rosuvastatin PK [23].

ABCG2 is expressed in various tissues and organs, including the liver, kidney, brain and intestine [1]. The missense variant rs2231142 C>A (Q141K) within ABCG2 results in an amino acid change in the encoding sequence that further changes the tertiary structure of ABCG2 protein (p.Gln141Lys), making it more susceptible to breakdown [25]. This change can result in a 30–40% decrease in BCRP expression and a reduction in the transport function of rosuvastatin and other endogenous compounds (e.g., uric acid) [1,8]. Different allele and genotype frequencies have been detected in several racially and ethnically diverse population subgroups [1,10,11,26]. Importantly, the frequency of Q141K was significantly more than double in all Asian subgroups (46%, 28%, and 26% in Filipino, Korean and Japanese, respectively) than non-Asians (11% and 3% in EUR and African Americans, respectively) (p < 0.001) [10,11]. The high prevalence of the same variant was also found in other Asian subgroups, such as the Minnesota Hmong, generally ascribed as Han-Chinese, (36%) and Southern Han-Chinese (26%) [9,26]. Similarly, the respective frequencies of the decreased and poor function phenotype of BCRP were 51% and 21% in Filipino, 35% and 8% in Japanese, 32% and 12% in Korean and were significantly more than triple the prevalence in EUR (20% and 1%) and African Americans (6% and 0.1%) [10,11].

Among Asian subgroups, Filipino had the highest prevalence of Q141K relative to other Asians (46% vs ∼30%) [10,11,26]. Likewise, Filipinos are most likely to have decreased and poor function BCRP (51% and 21%, respectively) than non-Filipino Asians (∼37% and ∼11%, respectively) (Table 1) [10,26]. The prevalence of Q141K within ABCG2 in Filipinos reported by both Alghubayshi et al. [8] and Roman et al. [11] is the highest reported thus far for the same variant. Consequently, Asian population subgroups, predominantly Filipino, may be more susceptible to decreased function BCRP and higher exposure to rosuvastatin versus non-Filipinos.

Table 1. . Implications of ABCG2 (c.421C>A) genotype frequencies in rosuvastatin starting dose among Asian subgroups, Europeans and African–Americans.

ABCG2 (c.421C>A) genotype	Population groups †							Transporter phenotype‡	Predicted treatment outcomes	2022 CPIC recommended starting dosing	FDA recommended starting dosing
	Filipino (%) (n = 179)	Japanese (%) (n = 187)	Korean (%) (n = 92)	Hmong (%) (n = 230)	CHS (%) (n = 105)	EUR (%) (n = 113)	AA (%) (n = 2203)
Homozygous wild-type (CC)	28	57	57	42	56	79	94	NF	Regular SAMS risk and rosuvastatin exposure	Standard dose and adjust using population-specific and disease-specific guidelines	Starting dose of ≥10 mg in the general population and 5 mg in Asian patients
Heterozygous variant (CA)	51	35	32	44	36	20	6	DF	Unknown SAMS risk, elevated rosuvastatin exposure versus NF, and improved lipid-lowering effects	Standard dose and adjust using population-specific and disease-specific guidelines	Starting dose of ≥10 mg in the general population and 5 mg in Asian patients
Homozygous variant (AA)	21	8	12	14	7.6	1	0.1	PF	Unknown SAMS risk, elevated rosuvastatin exposure versus NF and DF, and improved lipid-lowering effects	Starting dose of ≤20 mg and adjust using population-specific and disease-specific guidelines	Starting dose of ≥10 mg in the general population and 5 mg in Asian patients

^†All genotype frequencies are obtained from references [8–11,31].

^‡Phenotype was assigned using CPIC allele functionality and diplotype to phenotype translation tables [51].

AA: African Americans; ABCG2: Apical ATP-binding cassette transporter G2; CHS: Southern Han-Chinese; CPIC: Clinical Pharmacogenetics Implementation Consortium; DF: Decreased function; EUR: Europeans; NF: Normal function; PF: Poor function; SAMS: Statin-associated musculoskeletal symptoms; SNP: Single nucleotide polymorphism.

The variant Q141K is a well-characterized polymorphism with disease relevance where a reduced secretory function of BCRP of uric acid (UA) in the kidney and intestine is associated with elevated serum UA, leading to an increased risk of developing hyperuricemia and gout, particularly in population subgroups known to have a high prevalence of this variant [8]. The missense Q141K (C>A) polymorphism does not only have disease relevance but also drug relevance, where individuals with homozygous for the variant allele (AA) showed 100% greater area under the plasma rosuvastatin concentration–time curve (AUC_0–∞) after a single 20 mg of oral dose than heterozygous carriers (CA) (152 ng.h/ml and 76 ng.h/ml, respectively) and over 140% greater than in those with the wild-type genotype (CC) (152 ng.h/ml and 62 ng.h/ml, respectively) (p ≤ 0.01) [27].

Besides, the peak plasma rosuvastatin concentration was the highest in the individuals carrying AA relative to CA and CC (C_max: 16 ng/ml, 7.8 ng/ml and 7.1 ng/ml, respectively; p < 0.03) [11]. Likewise, a recent meta-analysis suggested that the variant allele carriers (CA/AA) had a significantly 1.5-fold higher rosuvastatin AUC_0–∞ (mean difference [MD]: 0.43; p < 0.0001) and C_max (MD: 0.42; p < 0.0001) compared with CC [28]. These PK changes were observed in both Asian and EUR individuals carrying the Q141K variant; however, exposure to rosuvastatin was still greater in Asian versus EUR individuals with the same ABCG2 genotype, possibly due to increased absorption of statin in this population [23].

SLCO1B1 is a liver-specific transporter-encoding gene, another significant statin transporter. Genetic variations in SLCO1B1, specifically, the missense rs4149056 T>C (c.521T>C), the defining variant for SLCO1B1*5, can result in lower levels of OATP1B1 protein on the basolateral surface of human hepatocytes and reduced function, leading to diminished hepatic uptake of statins [24,29]. The same genetic variant was also significantly associated with SAMS [29]. Different genotype–phenotype frequencies between Asian and non-Asian population subgroups were also observed [30–32]. The respective frequencies of the decreased and poor function SLCO1B1 were 18% and 2.9% in Japanese, 20.5% and 1.5% in Korean, and 23% and 2% in Chinese. Overall, the defected SLCO1B1 (CT and CC) frequencies in Asian were similar or slightly lower than in EUR (25.5% and 2.8%). Besides, African Americans had markedly lower heterozygous and homozygous SLCO1B1 genotype variants frequencies (6.9- 0.1%) than EUR (25.5- 2.8%) (Table 2) [30–32]. In Filipino, the frequency of SLCO1B1*5 is unknown, at the time of this publication, indicating the need for studying under-represented subgroups to uncover possible allele frequency differences between the Asian subgroups, ultimately, to better characterize statin exposure. SLCO1B1*5 is broadly associated with increased exposure to different statins, including rosuvastatin, atorvastatin, simvastatin, lovastatin and fluvastatin [1]. Specifically, the AUC_0–48 of rosuvastatin metabolite was found to be 100% greater in homozygous variant carriers (CC) than those with the wild-type genotype (TT) [33]. Although SLCO1B1*5 can influence exposure to rosuvastatin, its effect is reduced compared with the ABCG2 variant Q141K, which is significantly more common in Asians than non-Asians [8–11,29]. This observation implies that ABCG2 is an important predictor of rosuvastatin exposure among Asian subgroups [29].

Table 2. . Implications of SLCO1B1*5 (c.521T>C) genotype frequencies in rosuvastatin starting dose among Asian subgroups, Europeans and African–Americans.

SLCO1B1*5 (c.521T>C) genotype	Population groups †					Transporter phenotype‡	Predicted treatment outcomes	2022 CPIC recommended starting dosing
	Japanese (%) (n = 104)	Korean (%) (n = 200)	CHS (%) (n = 103)	EUR (%) (n = 4300)	AA (%) (n = 2203)
Homozygous wild-type (*1/*1)	79	78	75	72	93	NF	Typical SAMS risk and rosuvastatin exposure	Standard dose and adjust using population-specific and disease-specific guidelines
Heterozygous variant (*1/*5)	18	20.5	23	25.5	6.9	DF	Typical SAMS risk§ and elevated rosuvastatin exposure versus NF	Standard dose and adjust using population-specific and disease-specific guidelines
Homozygous variant (*5/*5)	2.9	1.5	2	2.8	0.1	PF	Typical SAMS risk§ and elevated rosuvastatin exposure versus NF and DF	Starting dose of ≤20 mg and adjust using population-specific and disease-specific guidelines

^†All genotype frequencies are obtained from references [30–32]. Allele and genotype frequencies are unknown for the Filipino American population subgroup.

^‡Phenotype was assigned using CPIC Allele Functionality and Diplotype to Phenotype Translation tables [51].

^§Typical myopathy risk with doses ≤20 mg.

CHS: Southern Han-Chinese; AA: African Americans; CPIC: Clinical Pharmacogenetics Implementation Consortium; DF: Decreased function; EUR: Europeans; NF: Normal function; PF: Poor function; SAMS: Statin-associated musculoskeletal symptoms; SLCO1B1: Solute carrier organic anion transporter family member 1B1.

In addition to genetic factors, concomitant medications and certain herbals can modulate the activity of transporters, ultimately affecting exposure to rosuvastatin [34,35]. The coadministration of other drugs such as cyclosporine, gemfibrozil and anti-HIVs (e.g., lopinavir) [35] can potentially inhibit both transporters. Particularly, cyclosporine is an immunosuppressant and a potent inhibitor of ABCG2 and SLCO1B1 that can markedly influence rosuvastatin disposition [14,35,36]. Indeed, rosuvastatin dosing should be limited to 5 mg or 10 mg daily in patients using cyclosporine or lopinavir, respectively. Also, rosuvastatin use should be avoided among gemfibrozil users [14,35]. Additionally, the coadministration of the antihypertensive drug, telmisartan 80 mg, daily for 6 days markedly increased the exposure to rosuvastatin (101% and 18% in C_max and AUC, respectively) in healthy Korean volunteers, compared with rosuvastatin 20 mg alone [37]. The interaction was confirmed in Chinese patients [38] and in an in vitro study suggesting that this interaction was mediated by telmisartan inhibiting ABCG2 [38]. While drugs can modulate the activity of rosuvastatin transporters, flavonoids (i.e., catechins) that are contained in a typical cup of green tea, commonly consumed by individuals of Asian descent, can inhibit ABCG2 activity [39]. Conversely, grapefruit juice might upregulate ABCG2 in the intestine in an animal model, which was denoted by promoting UA efflux [40].

Hepatic expression of ABCG2 differs between sexes; however, the data remain inconclusive [41,42]. On the one hand, a male mouse model had a higher Bcrp1 expression than a female mouse model [42]. Analysis of hepatic expression of human BCRP also indicated a higher expression in men versus women [42]. Estrogen hormones, on the other hand, had direct effects on major kidney transporters, including ABCG2 [43]. For example, estrogen response binding sites have been identified in the promoter region of ABCG2 [43], implying that ABCG2 can be transcriptionally upregulated by estrogen, reducing the risk of gout. In contrast, other reports have demonstrated that ABCG2 protein expression is downregulated by estrogen, increasing the risk of gout [44]. Similar conflicting results on the role of estrogen in ABCG2 expression have persisted. For example, an observational study demonstrated that shorter exposure to endogenous estrogen was associated with a high risk of gout [45]. Conversely, exposure to exogenous estrogens such as oral contraception and hormone replacement therapy was associated with an increased risk of gout [45]. In contrast, in vitro and in vivo investigations demonstrated that administration of estradiol, an exogenous hormone replacement therapy, was associated with ABCG2 upregulation and increased uric acid clearance from the intestine [46]. Sex hormones may affect the expression of ABCG2; however, the direction and the overall effect remain inconclusive. Furthermore, there are no studies that have investigated the effects of circulating sex hormones on statin dosing requirements.

Interindividual variabilities in response to rosuvastatin

Based on the patient's risk of ASCVD, the most current lipid management guideline suggests particular statin intensities [2]. The goal of the recommendation was to harness the plethora of evidence now available to make it easier for healthcare providers to manage dyslipidemia in patients who are at risk for ASCVD and to ensure that patients who will benefit from different statin intensities will achieve it. This recommendation emphasizes the necessity for moderate- and high-intensity statins in patients at risk of ASCVD. Nonetheless, the same approach does not account for the interpatient variability in response to statin therapy, which should be evaluated when treating patients with statins for dyslipidemia. Variability in response to rosuvastatin has been well documented in clinical studies [1,2,28]. The proportion of patients treated with rosuvastatin and achieving a <50% reduction in LDL-C from baseline was only 23% at the 5 mg dose and 57% at the 20–40 mg dose range [28]. Factors associated with variable rosuvastatin response after adjusting for rosuvastatin dose were age and sex, baseline LDL-C and comorbid diseases (Figure 1) [28]. Specifically, suboptimal response to rosuvastatin was associated with young age, male sex, low baseline LDL-C (0–140 mg/dl) and diabetes [28].

Figure 1. . A proposed model integrating factors affecting the response to rosuvastatin.

Interindividual variability in response is often associated with genetic polymorphisms in rosuvastatin transporters, especially in ABCG2 [1,28,47]. While the polymorphism Q141K is a strong predictor of rosuvastatin plasma levels in Asian individuals [23], and it is associated with enhanced response to the drug [27,29], individuals with the variant genotype have a better response to rosuvastatin (7% reduction in LDL-C) than the wild-type genotype, which is equivalent to the impact of doubling the dose of the drug [29,48]. According to Tomlinson et al. [48], the percent change in the mean LDL-C levels was significantly different across the three genotypes of Q141K(C>A) in Chinese patients treated with rosuvastatin (CC, CA and AA, least-squares means: –50.2, –54.3, –57, respectively; p = 0.0006) [29]. To illustrate this observation, Q141K within ABCG2 reduces the levels of expression of BCRP leading to reduced transporter efflux activity in individuals carrying the variant allele. Poor function BCRP increases rosuvastatin absorption in the intestine while reducing drug efflux in the biliary tract [27]. Hence, the dual impact of improved absorption and decreased hepatic elimination explains rosuvastatin accumulation in the blood and liver. Indeed, this variant is likely to be a significant contributor, at least partly, to the ethnicity-related differences in rosuvastatin PK, given the markedly high frequency of Q141K polymorphism in ABCG2 among Asian and Chinese population subgroups. Besides, these PK changes could contribute to the LDL-C reduction variations associated with rosuvastatin.

Asian-specific genotype-based dosing recommendation for rosuvastatin

In 2012, the CPIC conducted a comprehensive review of the literature and published the recommendations for SAMS for SLCO1B1 and simvastatin only [33]. Later in 2022, the guidelines were updated to include recommendations for additional statins and CYP2C9 and ABCG2 [1]. Although the CPIC quantified the need for reducing rosuvastatin dose, based on PK studies involving ABCG2 and dose–response relationship models, there is no contemporary evidence as to whether genetic variations in the aforementioned genes directly predict SAMS. The CPIC recommends a rosuvastatin dose of ≤20 mg as the initial dose for patients with poor functioning ABCG2 due to the high drug exposure; additionally, if a higher dose is warranted, an alternative statin is recommended. While the FDA recommends an initial dose of 5 mg in all Asians [14], regardless of ABCG2 genotype, CPIC recommends considering Asian ancestry when deciding on the dose based on individual genotype–phenotype (Table 1) [1]. The high prevalence of poor functioning ABCG2 phenotype in Asian subgroups coupled with genotype-based dosing recommendation by the CPIC indicates that Filipinos (21%) are most likely to benefit from receiving a lower dose of rosuvastatin relative to other Asian subgroups (8–14%) and non-Asians (0.1–1%) (Table 1). This highlights the importance of acknowledging the interindividual differences in exposure to rosuvastatin due to ABCG2 genetic polymorphism differences among different Asian subgroups. Further, Filipinos have been greatly enriched with the Q141K polymorphism than other Asian, EUR and African Americans subgroups, thereby having higher odds of exposure to the drug. Previous preliminary evidence suggested that Filipino have one of the highest global frequencies of the pleiotropic genetic polymorphism (Q141K) in ABCG2 [8,10,19,49]. Given this observation, it is conceivable that when exposed to rosuvastatin, Filipinos may be negatively impacted and disproportionately predisposed to SAMS compared with other ancestral population subgroups. Nevertheless, future pharmacogenetic or genetic studies are warranted to replicate and validate the former results with improved sampling and population demographics. Nonetheless, clinicians treating patients of Asian ancestry, especially Filipinos, are strongly encouraged to consider pre-emptive genotyping of ABCG2 before prescribing rosuvastatin to Filipino patients with dyslipidemia. Further, prescribers could consider alternative statins less impacted by this transporter when poor transporter function exists. Alternative statins could include pravastatin, fluvastatin (highly affected by CYP2C9 function), lovastatin, pitavastatin and simvastatin [1]. Given the polygenic nature in response to statin therapy, these recommendations should be made while considering the combined genetic variations in other genes (Figure 2).

Figure 2. . Implications of combined ABCG2 and SLCO1B1 phenotype in rosuvastatin starting dose using the 2022 CPIC guidelines.

ABCG2: Apical ATP-binding cassette transporter G2; SLCO1B1: Solute carrier organic anion transporter family member 1B1; CPIC: Clinical Pharmacogenetics Implementation Consortium; DF: Decreased function (yellow); NF: Normal function (green); PF: Poor function (gray); SAMS: Statin-associated musculoskeletal symptoms.

* In all cases, the starting dose is recommended in the general population with the consideration of population-specific and disease-specific adjustment guidelines.

Combined genotype-based dosing recommendations for rosuvastatin among Asian

The co-occurrence of genetic variations within other statin-transporter or -metabolizing encoding genes (e.g., ABCG2, SLCO1B1, CYP2C9 and CYP3A4/5) may potentially occur. Indeed, this co-occurrence could further impact the exposure to several statins, including simvastatin, rosuvastatin, atorvastatin, lovastatin, pitavastatin, pravastatin and fluvastatin [1,21,29]. Genetic variations within both transporters ABCG2 and SLCO1B1 also impact atorvastatin, simvastatin and fluvastatin [21]. However, the available data are insufficient to make dosing recommendations [1]. Nevertheless, the association studies evaluating the impact of combinatorial genetic variations within rosuvastatin transporters on drug exposure or response are limited. A study by Lee et al. [29] suggested that Chinese patients with the combined genetic variations within ABCG2 and SLCO1B1 had significantly higher rosuvastatin plasma concentration compared with wild-type carriers (p < 0.001), suggesting that the higher the number of genetic variants, the higher the plasma concentration and risks of SAMS [1]. The combined genetic variations were significantly associated with plasma levels of rosuvastatin. Indeed, including the combined genetic effect as a covariate improved the prediction model where it explained 16.6% (vs 14.4% for ABCG2 alone) of the variance in rosuvastatin exposure (p < 0.001) [29]. It is worth noting that genetic polymorphism within ABCG2 alone was still a strong predictor for rosuvastatin plasma concentration. The latter observation is supported by the significantly high Q141K frequency in select Asian subgroups [8–11] and the evidence of higher rosuvastatin exposure in Q141K carriers versus SCLO1B1*5 [23,29]. Further, there was no discussion of the influence of genetic variations in CYP2C9 as their frequencies are very low or undetected in Asians [50], and it mainly affects the PK of fluvastatin [1].

Assuming the effect is additive or synergistic, an individual with poor functions for both SLCO1B1 and ABCG2 (i.e., high-risk phenotype individuals) could have higher rosuvastatin exposure compared with individuals with poor function (PF) in a single transporter (Figure 2) [1]. The findings from Lee et al. [29] need further validation and replication in larger and multiethnic studies as their results did not provide information on how to adjust doses based on combined genotype information. To address the polygenic nature in the response to statins, the CPIC guidelines have provided combinatorial genotype-based dosing recommendations for rosuvastatin and classified them as optional [1]. Further, the combinatorial genotype-based rosuvastatin dosing recommendations were made possible by extrapolating the studies supporting single gene associations [1].

Using the 2022 CPIC rosuvastatin starting dose recommendations, individuals with combined genetic information for SLCO1B1 and ABCG2 could garner additional benefits by closely estimating the starting dose (Figure 2). However, the limited knowledge about the effect of combinatorial genetic variations within statin pharmacogenes limits the strength of the same recommendations. Studies identifying the prevalence of high-risk phenotype population subgroups, investigating their association with statin exposure and providing information on dose adjustment, are needed.

Conclusion

Recognizing the role of genetic polymorphisms in major transporters such as SLCO1B1*5 (rs4149056 T>C, c.521T>C) on rosuvastatin PK and PD, the Q141K genetic variant within ABCG2 has a greater impact on rosuvastatin dosing in individuals of Asian ancestry. With markedly high frequency, especially among Asian population subgroups (e.g., Filipino), it is plausible that the Q141K polymorphism in ABCG2 would warrant a lower initial dose of rosuvastatin compared with White [9]. Further, a holistic understanding of the underlying mechanism and risk factors contributing to the development of SAMS is necessary to minimize their incidence and achieve the ideal dose regimen for improved statin adherence in each individual. Furthermore, involving population subgroups in genetic research uncovers meaningful genetic information that could influence treatment decisions regarding dosing. Alternatively, aggregating Asian subgroups may expose some patients to greater risk for harm or preclude them from harnessing the benefits of taking the most appropriate statin dose.

Future perspective

The constraints and ramifications of aggregating different Asian subgroups (e.g., Filipino, Japanese, Korean) under a broader racial classification are underlined by this special report. Additionally, it highlights how diverse Asian population subgroups differ markedly in the frequency of the clinically actionable genetic polymorphism Q141K within ABCG2, suggesting that Filipinos have the highest frequency of this polymorphism compared with other Asian subgroups. After replication and validation of the preliminary findings, these results, in conjunction with the 2022 Statins CPIC recommendations, might be used to advocate for pharmacogenetic testing among specific Asian subgroups with a high frequency of decreased or poor ABCG2 transporter function. Patients' genetic information may improve our prediction model and move us beyond race-based patient care to more individualized care. Analysis of well-characterized population subgroups is needed to identify population-specific variants and distinct allele frequencies. Nevertheless, including unique ancestral subgroups in genetic and pharmacogenetic studies would improve evaluations of polygenic risk scores to precisely predict the response and exposure to a specific therapy and estimate disease risk. Furthermore, accounting for population subgroups differences when conducting biomedical and genetic research is needed.

Executive summary.

• Statins are widely used medications for the primary and secondary prevention of cardiovascular diseases. However, adherence to statin therapy could be greatly reduced due to adverse drug events.

• Statins are generally safe and effective medications; however, statin-associated musculoskeletal symptoms could disproportionately affect certain patient population groups with specific risk factors profiles.

• Rosuvastatin, a frequently prescribed statin, is characterized by interindividual variability in systemic exposure among various patient population subgroups.

• Asian subgroups are more likely to have the missense genetic polymorphism Q141K within ABCG2, a gene encoding for a key efflux transporter, which results in reduced transport function and increased systemic exposure to rosuvastatin.

• Q141K genetic polymorphism has a stronger influence on rosuvastatin's pharmacokinetic and pharmacodynamic in patients of Asian ancestry, despite the importance of other genetic polymorphisms within other transporters like SLCO1B1.

• Compared with other Asian subgroups, Filipinos will likely need a lower initial dose of rosuvastatin, based on the frequency of the Q141K variant in ABCG2 and the most recent CPIC recommendations on statin dosing.

• The discovery of substantial interpopulation disparities, determined by the Q141K genotype frequencies of aggregated Asian subgroups, supports the need to move from a race-based to a genotype-based approach for rosuvastatin dosing.