Pharmacogenomics of response to statin treatment and susceptibility to statin-induced adverse drug reactions in Asians: a scoping review

The literature search was conducted in two databases, Medline and Embase, using free-text terms. The Boolean operator “OR” was used to group search terms into three main subjects. The first subject included terms relating to pharmacogenetics – “pharmacogen*”, or “genetic*”, or “genomic*”, or “genotype”; secondly, terms related to statins – “statins”, or “atorvastatin”, or “rosuvastatin” or “simvastatin” or “pravastatin”; and lastly terms relating to the treatment responses – “treatment response” or “efficacy” or “response”. In terms of adverse effects, the first two subjects used were similar, the subject used included the terms “adverse effect” or “adverse drug reaction”. Each of the groups of search terms was searched simultaneously with the “AND” Boolean operator to identify original research articles published up to December 2021 on pharmacogenomics of statin treatment responses, which focused on atorvastatin, rosuvastatin, simvastatin, and pravastatin. The search was limited to articles published in the English language with human subjects. Abstracts or conference proceedings were included if the content was not published before.

A total of 1907 papers from Medline and 1,382 papers from Embase were identified for treatment response to statins. After removing duplicates, 1939 original papers remained (Figure 1). For adverse effects, 56 papers from Medline and 231 papers from Embase were identified equating to 284 individual articles after removing duplicates. Compilation and elimination of duplicates was conducted using the referencing software “Mendeley”. The title and abstract of each article were reviewed by researchers YHY and VS and verified by SH and NAR. Only papers from 2008 to December 2021 and studies carried out in Asian populations were included. Reference lists of past reviews were checked for other papers of interest to ensure that the search was all inclusive.

Figure 1.

Articles that did not answer the research question were excluded. Articles were excluded if they contained: pharmacogenomic studies of statins other than atorvastatin, rosuvastatin, simvastatin, and pravastatin; pharmacogenomic studies with subjects of unreported ethnicities; non-pharmacogenomic studies (such as pharmacokinetic studies, gene expression studies, or cell culture studies); and population genotyping studies not involving statins. The study protocol was prospectively registered with Open Science Framework (OSF) registries (https://osf.io/sqp6x).

Due to the nature of the scoping review, the quality of the individual studies was not assessed in the present study. Extracted data included information on authors, publication year, study population, statin(s) studied, gene(s) studied, study findings, and data of significant statistical value. Four reviewers (YHY, SH, NAR, and VS) independently extracted data into the data extraction form. Any conflicts were resolved within the team.

The data on (i) genetic variants studied in statin treatment response and (ii) genetic variants studied in statin-induced ADRs were summarized. The Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) was used as a checklist in documenting the rationale, methodology, and findings of the scoping review [7].

Among the 35 papers investigating treatment response, 28 papers were studies investigating a single statin (rosuvastatin, n = 9; simvastatin, n = 10; atorvastatin, n = 8; pravastatin, n = 1), 6 papers investigated multiple statins, and 1 paper did not specify the statin(s). The parameters used to determine the treatment response in the included studies were the changes in total cholesterol (TC), triglycerides (TG), LDL-C, and high-density lipoprotein cholesterol (HDL-C) levels. The data extracted from these papers are summarized in Table 1. Figure 2 depicts the important genes that have been significantly associated with response to statin treatment [8].

Figure 2.

Transporters are membrane proteins present in cells to regulate the influx of essential nutrients and ions and the efflux of cellular waste and xenobiotics, such as toxins and drugs. The two major superfamilies of transporters that play an essential role in drug disposition are solute carrier (SLC) and ATP-binding cassette (ABC) transporters.

Statins are transported into the liver by the organic anion transporting polypeptide 1B1 (OATP1B1) encoded by SLCO1B1, which controls the systemic exposure of many statins and is associated with statin-induced myopathy. Out of the many single nucleotide polymorphisms (SNPs) in SLCO1B1, the two most studied are A388G and T521C [9]. Even though the T521C SNP was not associated with the lipid-lowering response of simvastatin and rosuvastatin [10,11,12], Lee et al. [11] reported an association between increased plasma concentrations of rosuvastatin and impaired N-demethylation of rosuvastatin. This was validated in a clinical pharmacokinetic study that reported higher plasma exposure of rosuvastatin and its metabolites among Asian populations in the United States when compared with Caucasians [13]. A study on 177 hypercholesterolemic Indian patients revealed that there was significantly higher LDL-C reduction in patients who were homozygous for the A allele in the A388G SNP when treated with 10 mg of atorvastatin compared to G homozygotes [14]. In contrast, studies among Thai and Chinese hypercholesterolemic patients found no association between A388G SNP and lipid-lowering response [10,11,12]. Interestingly, this SNP has a higher variant allele frequency (57%–75%) compared to wild-type allele frequency (25%–43%) in Chinese, Indian, and Thai populations [10,11,12, 14] than in Caucasians (38.2%) [15]. An intronic SNP, rs4149081 G>A, identified previously in a genome-wide association studies (GWAS) study was studied by Hu et al. [12] among Chinese patients and they reported that rs4149081 G>A SNP was significantly associated with a 4.6% and 4.0% greater reduction of LDL-C compared with those with wild-type alleles in response to rosuvastatin and simvastatin treatment, respectively. However, T89595C and C463A SNPs were not associated with treatment response in Indian and Thai patients treated with simvastatin and atorvastatin, respectively [10, 14].

Multidrug resistance-associated protein 2 (MRP2) also known as ABC sub-family C member 2 (ABCC2) encoded by the ABCC2 gene, is an efflux transporter expressed on the apical domain of epithelial cells. This transporter has also been associated with hepatobiliary excretion of statins, and SNPs in this gene may have a role in the response to statins [16]. A study on 318 simvastatin-treated Chinese patients revealed that the variant allele of the-24 C>T SNP and female gender were significantly associated with low response of HDL-C elevation upon simvastatin treatment [17]. However, the genotypes were not associated with the differences in TC, TG, LDL-C, or HDL-C among genotypes before and after treatment [17]. On the other hand, in another study, the change in LDL-C was more pronounced among patients who carried both the variant allele for rs717620 in ABCC2 gene and wild-type allele for rs1128503 in ABCB1 gene compared to those who carried wild-type genotypes for both of these SNPs [18].

The ABCB1 gene encodes for P-glycoprotein, which is another efflux pump that transports statins and metabolites from hepatocytes to bile and from renal cells to urine. The two most studied SNPs in ABCB1 are G2677T (rs2032582) and C3435T (rs1045642). Among Indian atorvastatin-treated patients, carriers of variant allele for G2677T had 3 times more significant reduction of LDL-C as compared to those who carried the wild-type allele (P < 0.05) [14]. This study also reported significant association between C3435T polymorphism and LDL-C reduction where carriers of variant allele had greater LDL-C reduction treated with atorvastatin compared to wild-type allele carriers [14].

The ABCG2 gene that encodes for ABC, sub-family G2 protein or alternatively known as breast cancer resistant protein (BCRP), is mostly found on the plasma membrane and functions as a direct drug efflux pump [19]. A study on 291 Chinese patients reported that C421A*AA homozygotes exhibited 63% and 41% higher mean plasma concentrations of rosuvastatin and its metabolite as compared to heterozygous carriers (CA), while 120% and 99% greater than in those with the homozygous wild-type (CC) genotype when treated with rosuvastatin [11]. Significant correlation was observed between plasma concentrations of rosuvastatin and the percentage reduction in LDL-C (r =−0.194; P = 0.001), which disappeared when adjusted for the C421A polymorphism [11]. Hu et al. [20] and Tomlinson et al. [21] reported that this polymorphism is significantly associated with reduction of LDL-C level in a gene-dose dependent manner among Chinese patients, where carriers of homozygous variant genotype (AA) exhibited higher reduction in LDL-C level. On the other hand, for ABCG5 gene, another study in Indian patients treated with atorvastatin demonstrated no association between ABCG5 rs6720173 SNP and LDL-C levels [14].

The protein encoded by ABCG8 functions as a half-transporter to limit intestinal absorption and promote biliary excretion of sterols [19]. An intron 11 SNP rs4148222 was shown to be significantly associated (P = 0.046) with baseline levels of HDL-C among Chinese patients treated with atorvastatin [22]. Another study among Chinese patients treated with atorvastatin found no association between ABCG8 C1199A polymorphism and the lipid-lowering response of atorvastatin, yet the response appeared to be affected by synergistic effect of CYP7A1 and ABCG8 polymorphisms [23]. Greater TG and TC reduction (P = 0.001 and P = 0.009, respectively) were observed in patients with the CYP7A1 A-204C homozygous wild-type genotype (AA) compared to the homozygous variant genotype (CC) [23]. A Taiwanese study found that patients carrying the ATTATCGAC haplotype demonstrated greater reduction of LDL-C [24]. However, an Indian study did not show any association between the D18H variant and LDL-C levels [25].

The ABCA1 gene encodes for a human transporter known as cholesterol efflux regulatory protein (CERP). The KK homozygotes of the R219K SNP in ABCA1 exhibited significantly higher increase in HDL-C level compared to wild-type homozygotes RR in Chinese patients treated with pravastatin [26].

Genetic polymorphisms in cytochrome P450 (CYP) genes have been associated with the efficacy of selected statins by influencing their hepatic metabolism. The CYP450 enzymes catalyze the metabolism of most statins, except pravastatin and rosuvastatin (Figure 2). Atorvastatin and simvastatin are metabolized by the CYP3A4 isoenzyme and fluvastatin via CYP2C9 [27]. In addition, metabolism of statins may be mediated by other CYP450 isoenzymes, including CYP2D6 [28]. A meta-analysis of population-scale sequencing project showed that the distribution of CYP alleles, including CYP3A4, CYP3A5, CYP2C9, CYP2C19, and CYP2D6, differs considerably between different ethnic groups (Europeans, Africans, East Asians, South Asians, and admixed Americans) [29].

Hu et al. [30] found no significant association between CYP3A4*1G and CYP3A4*22, along with other gene polymorphisms (CYP3A5*3 and PPARA rs4823613 A>G), with response to 40 mg simvastatin in 273 Chinese patients with hypercholesterolemia. However, a prospective study by Gao et al. [31] reported a CYP3A4*1G-dose-dependent reduction of TC in hyperlipidemic patients treated with atorvastatin but not simvastatin. Peng et al. [32] also showed that Chinese patients with ischemic stroke who carried the rs2242480*C allele had better lipid-lowering effect with atorvastatin treatment compared to those who were TT homozygotes. However, this association did not survive Bonferroni correction [32]. Hypercholesterolemic Indian patients with the wild-type genotype of rs2740574 (CYP3A4) demonstrated significant reductions in LDL-C in response to atorvastatin therapy [14]. However, results have been inconsistent with the CYP2C9 polymorphisms. Lin et al. [33] reported that Chinese patients carrying the CYP2C9*1/*3 or CYP2C9*3/*3 genotypes showed a significant lipid-lowering efficacy when treated with rosuvastatin, with higher TC and LDL-C reduction compared to patients with the wild-type genotype. However, other CYP2C9 SNPs (rs1934967; rs1057910) were not associated with statin treatment response in Chinese patients [11, 24].

Studies found no significant associations between statins response in LDL-C reduction and CYP2C19 polymorphisms (rs10786172; rs4244285; rs4986893) among Chinese patients [11, 24, 32]. However, Finkelman et al. [34] reported that changes in lipid profiles from baseline for poor metabolizers (PMs: *2*2, *2*3, *3*3) and extensive metabolizers (EMs: *1*1, *1*2, *1*3) of CYP2C19 were comparable except for TG among healthy Taiwanese receiving multiple doses of 20 mg rosuvastatin. The PMs demonstrated a greater TG reduction compared to EMs.

Li et al. [35] investigated the effects of the lipid-lowering effect of simvastatin and CYP2D6*10 in 200 Chinese Ningxia Hui patients with hyperlipidemia. The lipid-lowering effects of simvastatin 20 mg daily, after 8 weeks of treatment, was increased significantly with the genotype CC, leading to the authors recommending that patients carrying the CC genotype in this population should be prescribed lower doses of simvastatin compared to those with CT and TT genotypes. Additionally, Chinese patients carrying the rs1065852*G allele showed better lipid-lowering effect with atorvastatin therapy [32].

CYP3A5 is another CYP3A family enzyme that is involved in metabolism of simvastatin and atorvastatin. However, no studies found any significant associations between CYP3A5 polymorphisms (rs776746, rs4646450, rs3800959, and rs776746) with the lipid-lowering effects of simvastatin or atorvastatin among Chinese or Indian patients [14, 22, 30, 36].

CYP3AP1 is a pseudogene in the human CYP3A gene, which can play a vital role in gene conversion and recombination with a functional gene. Li et al. [37] investigated the lipid-lowering response of atorvastatin and simvastatin with CYP3AP1*3 (−44G>A) polymorphisms in 379 Chinese Han hyperlipidemic patients. There was a greater LDL-C reduction in response to simvastatin therapy in women who were CYP3API*3/*3 carriers compared to CYP3API*1/*3 carriers. In the atorvastatin group, the TC level was significantly reduced in women who were CYP3API*3/*3 carriers than in CYP3API*1/*3 carriers [37].

P450 oxidoreductase (POR) plays an essential role for CYP450 activity by transferring electrons from nicotinamide adenine dinucleotide phosphate (NADPH) to CYP3A enzymes, which metabolize various statins, including atorvastatin [38]. Wei and Zhang [36] investigated the association of POR*28 polymorphisms and the lipid-lowering effects of atorvastatin (20 mg/d for 4 weeks) in 179 Chinese patients with hyperlipidemia. Homozygotes for the T allele showed significantly lower LDL-C levels than homozygotes for the C allele (wild-type) after atorvastatin treatment. After adjustment for age, sex, and body mass index, CYP3A5 non-expressors who were POR*28 wild-type homozygotes (CC) presented with significantly higher mean TC and LDL-C levels than those who were POR*28 variant homozygotes (TT) and heterozygotes (CT) at baseline and after atorvastatin treatment [36]. This indicates that POR*28 SNPs are associated with significant increases in plasma lipids in non-expressors of CYP3A5.

Apolipoproteins have a role in regulating lipoprotein metabolism [39]. Polymorphisms in several apolipoproteins, such as APOE, APOA5, and APOB, have been reported to affect statin response. Apolipoprotein A-V (APOA5) is particularly involved in the metabolism of TG-rich lipoproteins [40], and has previously been shown to be associated with a high degree of interindividual variation in serum TG [41,42,43]. A study by Hua et al. [44] in 195 Chinese patients with treatment-naïve dyslipidemia explored the association of APO5 rs662799 (T-1131C) SNP following a 12-week randomized treatment of three statins: rosuvastatin, atorvastatin, or simvastatin. Patients carrying the rs662799*TT genotype exhibited greater LDL-C reduction regardless of whichever statin they were being treated with. However, this genotype was only correlated with HDL-C and TG in atorvastatin- and simvastatin-treated groups [44]. Interestingly, this lack of association of rosuvastatin with this SNP was validated by Hu et al. [45] in their cohort of Chinese patients with hyperlipidemia.

Apolipoprotein E (APOE) is expressed in the brain and the liver and is a ligand for the LDL receptor (LDLR) [46]. Similar to APOA5 polymorphism, Hu et al. [45] reported that there was no significant association between APOE E2/E3/E4 polymorphisms and response to rosuvastatin among Chinese patients with hyperlipidemia. Thai hypercholesterolemic patients who were APOE2 and APOE3 carriers had significantly greater TC and LDL-C reduction compared to APOE4 carriers after 3 months of simvastatin treatment [47]. On the other hand, a study by Hu et al. [20] found that the LDL-C reduction after at least 4 weeks of rosuvastatin 10 mg daily treatment was significantly higher among carriers of the rs4420638*A allele within the APOE/C-I/C-IV/C-II gene cluster.

Liu et al. [48] found that there was a significant reduction in TC, LDL-C, and APOB levels after 12 weeks of pravastatin therapy only among Chinese hyperlipidemia patients expressing the APOA1 (G75A) AA and GA genotypes.

3-Hydroxy-3-methyl-glutaryl-coenzyme A reductase (HMGCR) is the rate-controlling enzyme for cholesterol synthesis via the mevalonate pathway, which is the target enzyme of statins [49].

Chien et al. [24] explored the association of two HMGCR SNPs (rs12916 and rs3931914) and efficacy after 9 months of treatment with various statins, including atorvastatin, simvastatin, and pravastatin, in 386 Chinese patients with hyperlipidemia. Patients who carried the rs12916*CT and TT genotypes showed a significant greater reduction in LDL-C than the CC homozygotes after correction for multiple tests in an additive model [24]. Additionally, the CCGTCCA haplotype was associated with a significant increase in LDL-C. However, there was no significant association with rs3931914. Among 24 healthy Koreans, Chung et al. [50] noticed that the HMGCR rs3846662 GG genotype demonstrated significantly higher levels of LDL-C at baseline and all observation times (7–28 d after atorvastatin therapy), with mean differences of 18%–33%.

Hu et al. [20] reported that LDLR rs1433099 polymorphism was one of the predictors for the LDL-C changes in response to rosuvastatin in 386 Chinese patients. Individuals with the TT genotype demonstrated a significantly lower reduction in LDL-C level compared to those with the CC and CT genotypes after adjustment for age, sex, and having familial hypercholesterolemia, but not significant after Bonferroni correction [20].

In addition to LDL receptors, proprotein convertase subtilisin/kexin type 9 (PCSK9) regulates levels of plasma LDL by binding to the LDLRs and directing it to endosomes or lysosomes for destruction [51]. Reduction in LDLRs decreases clearance of LDL-C and ultimately leads to accumulation of LDL-C in the blood circulation. Wanmasae et al. [47] investigated the variation in three PCSK9 SNPs (R46L, I474V, and E670G) on the lipid-lowering effects of simvastatin in 223 Thai hypercholesterolemic patients. There was a significant reduction of LDL-C in PCSK9 474IV carriers compared to 474II carriers, but no significant difference for TC, TG, and HDL-C. However, the association between PCSK9 R46L and E670G and simvastatin response was not analyzed due to low frequencies of minor alleles [47].

Sterol regulatory element-binding transcription factor 1 (SREBF1) is the transcription factor that regulates the expression of hepatic LDLR [24]. A study explored the association of SREBF1 rs9902941 polymorphisms and response to various statins among 386 Chinese patients with hyperlipidemia. Those subjects carrying minor T alleles had significantly higher reduction of LDL-C levels compared to those who were CC homozygous [24].

Cholesterol 7α-hydroxylase (CYP7A1) is the rate-limiting enzyme in bile acid production, which is essential for cholesterol elimination from the body. Hence, genetic variants in the gene may indirectly affect an individual's sensitivity to the lipid-lowering effects of statins [52]. Among the Chinese Han hypercholesterolemia patients whose LDL levels were recorded in pursuance of atorvastatin treatment, a significantly greater reduction was, among all the 7 variants studied, demonstrated only for rs8192870*T homozygotes [22]. A similar effect was also seen with Indian and Chinese patients treated with atorvastatin who were homozygous for the rs3808607*A allele [14, 23]. However, the study by Chien et al. [24] did not find any associations with the two SNPs, rs4738687 and rs2162459, in their cohort of Chinese patients. Recently, Liu et al. [53] reported that CYP7A1 rs3824260 is associated to different responses to simvastatin among those Chinese hypercholesterolemia patients. They revealed that AA homozygotes tend to have low response to LDL-C reduction, whereas GG homozygotes were at a higher risk of HDL-C reduction after 12 weeks of simvastatin therapy [53].

Cholesterol ester transfer protein (CETP) encoded by the CETP gene is a plasma protein that facilitates the transfer of lipids between HDL and TG-rich lipoproteins. The most studied SNP in CETP is the TaqIB variant, which has been associated with statin response [16]. Previous studies reported that the B1 allele is associated with increased CETP concentrations as well as activity and thus low HDL-C levels while the opposite was reported for the B2 allele [54,55,56]. In 225 hypercholesterolemic Thai patients treated with simvastatin, better lipid-lowering effect with higher reductions in TC and LDL-C was observed among B1 carriers (B1B1 + B1B2) compared to the B2 homozygotes [47].

Lipoprotein Lipase (LPL) plays a role in lipid metabolism by catalyzing the hydrolysis of triacylglycerol in chylomicrons and very low-density lipoproteins (VLDL) for tissue utilization [57]. Hu et al. [20] reported that Chinese patients with the C1421G CC genotype demonstrated a significantly lower reduction in LDL-C levels compared to those with the CG or GG genotypes in response to rosuvastatin treatment.

Lecithin-cholesterol acyltransferase (LCAT) plays a key role in cholesterol transport and removal by catalyzing the conversion of cholesterol and lecithin to cholesteryl esters and lysophosphatidylcholines on the surface of HDLs, which is a crucial step in the maturation of HDLs [58]. A study investigated the association of LCAT rs255052 with the lipid-lowering effect of rosuvastatin in 386 Chinese patients and found that the GG genotype carriers demonstrated a greater reduction in LDL-C levels than those carrying the A allele [20].

In addition to the aforementioned gene variations affecting pharmacokinetics and pharmacodynamics of statins, other gene polymorphisms were also studied among Asian populations. Flavin-containing monooxygenase 3 (FMO3) catalyzes the oxygenation of various compounds and plays a regulatory role in cholesterol metabolism [59]. A study revealed that the FMO3 V257M variation was one of the predictors of LDL-C response in Chinese patients on rosuvastatin therapy. Patients with the V257M GG genotype showed higher reductions in LDL-C levels in comparison to those with the GA and AA genotypes [20].

Leptin may also be involved in regulating cholesterol metabolism via other mechanisms, such as downregulating activity of HMG-CoA reductase and upregulating activity of sterol and cholesterol hydroxylases to decease VLDL levels [60]. Li et al. [61] investigated the association between LEP G2548A and LEPR Q223R polymorphisms and efficacy of simvastatin in 212 Chinese patients with primary hyperlipidemia. Patients carrying the Q223R RR genotypes had significantly higher reductions in TG and TC levels than those with the QR genotypes among poor respondents. However, this significance did not survive Bonferroni correction [61]. Sun et al. [62] observed that the increment of HDL-C levels was significantly lesser among patients who were AA homozygotes than those with the GG genotype. However, there was no significant association between LEP G2548A SNP with treatment response [61].

Paraoxonase 1 (PON1) is an HDL-associated enzyme involved in reducing lipid peroxide accumulation on LDL, suggestive of a protective role against cardiovascular disease [63]. Fu et al. [64] noticed that there was significant association between plasma HDL-C changes with PON1 Q192R polymorphism among 236 Chinese patients with coronary heart disease receiving simvastatin therapy. Patients with the RR genotype presented with greater increment in HDL-C levels in response to simvastatin therapy, than patients with the QR or QQ genotypes [64].

Farnesoid X receptor (FXR), the bile acid-activated nuclear receptor with an essential role in lipid and carbohydrate metabolism, is a regulator of drug transporters involved in statin disposition, including SLCO1B1 [65]. A study investigated the impact of FXR G-1T polymorphism on the lipid-lowering response of rosuvastatin among Chinese patients with hyperlipidemia. The authors observed that individuals who carried the T allele (GT or TT) had greater reductions in LDL-C and TC compared to those who were homozygous for the G allele [66]. Sterol regulatory element-binding factors cleavage activating protein (SCAP) has a regulatory role in lipid homeostasis. A study showed that Indian patients carrying the SCAP A2386G*G allele presented with significant reductions in their TC and LDL-C levels in comparison to those homozygous for the *A allele [67].

T-cell immunoglobulin and mucin domain 4 gene (TIMD4) and hepatitis A virus cellular receptor 1 gene (HAVCR1; also known as TIMD1) are genes essential in regulating immune responses. HAVCR1 is expressed on T-helper cells and co-stimulates T-cell activation, whereas TIMD4 is expressed on antigen-presenting cells and mediates phagocytosis of apoptotic cells [68]. Zhang et al. [69] investigated the effect of three TIMD4-HAVCR1 SNPs (rs12522248, rs1501908, and rs2036402) on the lipid-lowering effect of atorvastatin in a population of 724 individuals of southern Chinese Han ethnicity. They demonstrated that patients who carried the rs1501908*G and rs12522248*C alleles had lower TC and LDL-C levels than those carrying the rs1501908*C and rs12522248*T alleles, respectively. Additionally, patients with the rs2036402*TC genotype presented with lower TC and LDL-C levels than those with the TT genotype. Interestingly, patients with the rs1501908*CC, rs12522248*TT, or rs2036402*TT genotypes presented with lower APOA1 levels than other respective genotypes after atorvastatin treatment [69]. These associations highlighted that atorvastatin had significantly higher efficacy in reducing TC and LDL-C levels in patients who carried the TIMD4-HAVCR1 rs1501908*G and rs12522248*C alleles. On the other hand, Takahashi et al. [70] reported that the C-857T polymorphism in the promoter region of the TNF-α gene was associated with significantly higher LDL-C levels and lower cholesterol-lowering effects in patients who were carriers of the T allele in comparison to non-T carrier when treated with statins.

Matrix metalloproteinase 9 (MMP9) has been shown to be associated in the pathogenesis and progression of cardiovascular diseases [71]. Xu et al. [71] investigated the association of MMP9 C1562T polymorphism on the lipid-lowering effect of simvastatin in 264 Chinese patients with coronary artery disease. Their results showed that the reduction of LDL-C with simvastatin therapy was significantly greater among patients who were TT homozygotes compared to CC homozygotes [71]. Polymorphism on neurocan/cartilage intermediate layer protein 2/pre-B-cell leukemia homeobox 4 (NCAN/CILP2/PBX4) gene located on chromosome 19 has been shown to be associated with different serum lipid phenotypes among Europeans and Asians [72,73,74]. Hu et al. [20] investigated the association of rs16996148 with the lipid-lowering effect of rosuvastatin in Chinese patients and observed that patients who were homozygous for the rs16996148*G allele had greater LDL-C reduction than the GT heterozygotes.

Endothelial lipase (EL) has a role in modulating lipid metabolism by lowering HDL-C levels [75]. However, Cai et al. [76] reported that there was no significant effect of EL 2037 T/C and 2237 G/A polymorphisms on the lipid-lowering effects of rosuvastatin (10 mg/d for 4–8 weeks) in 121 Chinese with coronary artery disease.

Five papers investigated associations between various genetic variants with statin-induced ADRs, studying the effects of either just one statin [33, 77] or multiple statins [78,79,80] (Table 2). As the studies used different methods in determining statin-induced ADRs and studied different genes, direct comparisons to derive a definitive conclusion were not possible. The most robust and well-defined method of defining statin-induced ADR was adopted in the research of Sai et al. [78]. In this study, statin-related myopathy was presented as three distinct categories: myalgia (defined by creatine kinase [CK] value <3-fold of the upper limit of normal with muscle symptoms), myositis (defined by CK value between 3- and 10-fold of the upper limit of normal (ULN) with muscle symptoms), and rhabdomyolysis (defined by CK value >10-fold of the ULN with muscle symptoms) [78]. The remaining four studies depended on either patient self-reported ADR symptoms or checking for ADRs in medical records.

Sai et al. [78] compared genetic variants in three candidate genes (SLCO1B1, RYR2, and GATM) and four HLA genes in 52 patients who had experienced statin-induced myopathies and 2878 healthy controls. Of the variants tested, only HLA-DRB1*04:06 was statistically significant after Bonferroni correction. However, HLA-DRB1*11:01 was previously identified to be significantly associated with antibody positive myopathy in Caucasians and African patients [81]. These differences could have been due to the differences in allele frequencies between the Japanese and Caucasian/African populations.

The study with the largest cohort classified statin-induced myopathy based on CK levels and geographic location in China (392 patients from Dongzhi and 195 patients from Beijing) over two time points, specifically after 4 weeks’ and 8 weeks’ treatment with simvastatin [77]. Jiang et al. [77] only found significant elevation of CK levels among patients in Dongzhi with either the AA genotype in the LEP G2548A variant and RR or QQ genotypes of the LEPR Q233R variant. However, this finding was not replicated in patients from Beijing.

On the other hand, Liu et al. [79] reported that the risk for myotoxicity was significantly higher among patients who were C allele carriers of SLCO1B1 T521C taking rosuvastatin. However, this effect was statin-dependent and was not observed in patients taking either atorvastatin or simvastatin [79]. Controversially, Zhong et al. [80] reported that healthy Chinese Hakka subjects with the CC genotype (n = 47) did not experience any statin-induced myopathy.