Association of ABCB1 (C3435T) and ABCC1 (G2012T) Polymorphisms with Clinical Response to Atorvastatin in Iranian Patients with Primary Hyperlipidemia

Background: Atorvastatin is prescribed for the primary and the secondary prevention of coronary artery diseases. A wide variation in inter-individual statin response suggests that genetic differences may contribute to this variation. This study investigated the association of ABCB1 (C3435T) and ABCC1 (G2012T) polymorphisms with clinical response to atorvastatin in Iranian primary hyperlipidemic patients. Methods: Individuals (n=179) with primary hypercholesterolemia were enrolled, and peripheral blood samples were collected. Genotyping of two polymorphisms were performed by amplification refractory mutation system PCR. Results: Following four weeks of treatment, a significant reduction of LDL-C was observed in variant groups (CT+TT) of ABCB1 (P=0.018) and wild-type group (GG) of ABCC1 genes (P=0.029). Logistic regression analysis revealed a significant difference between male and female responses to 10 mg/day atorvastatin (P=0.004, odds ratio=0.2, CI 95%=0.06-0.6). Conclusion: Our finding indicated that these polymorphisms may be attributed to LDL-C serum levels in the primary hypercholesterolemia patients receiving atorvastatin.


INTRODUCTION
oronary artery disease (CAD) was responsible for approximately 300,000 deaths in 2010 [1] . It has also been indicated to have a strong association with dyslipidemia [2] . The third report of the National Cholesterol Education Program [3] emphasizes the necessity of treatment of lipoprotein abnormalities in patients at high risk for coronary heart disease (CHD). The discovery of 3-hydroxy-3-methylglutarylcoenzyme reductase inhibitors (statins) introduced a highly effective therapeutic approach for the reduction of both morbidity and mortality of patients with established CHD [4] . Serum lipid concentration and individual's responsiveness to statin are influenced by environmental and genetic factors over 50% [5] .
Pharmacogenetic studies have already identified more than 30 genes affecting statin. P-glycoprotein (P-gp) and multidrug resistance protein 1 (ABCC1) play an important role in drug transport through the cell membrane, which may contribute to drug disposition and response [6] . P-gp is encoded by a polymorphic gene named multidrug resistance 1 (ABCB1), located on chromosome 7. Until now, more than 20 polymorphisms have been recognized; some of which have been associated with altered P-gp expression and activity [6][7][8] .
Several investigations have focused on a synonymous polymorphism in the ABCB1 gene, C3435T and reported that this polymorphism is associated with variation in atorvastatin efficacy and safety [9][10][11] . The C3435T (rs1045642) is located in the C exon 26 and has a relationship with differences in total serum cholesterol, LDL-C, and HDL-C in response to atorvastatin in American populations [12] . ABCC1 appears to be ubiquitously expressed in many human tissues and transports a wide spectrum of substrates and toxicants [13] . A non-synonymous G2012T (rs 45511401) polymorphism, located in exon 16, is common among Caucasian populations [14] . Recently, one study has found an association between polymorphisms in ABCB1 and ABCC1 genes and differences in serum LDL-C and HDL-C levels, which indicates the potential role of these polymorphisms in the pharmacogenetics of the atorvastatin [4,15] . Regarding these points, this study, for the first time, examined the association of ABCB1 (C3435T) and ABCC1 (G2012T) polymorphisms with para-clinical and clinical responses to atorvastatin in Iranian patients with hyperlipidimia.

Study population
Patients with primary hypercholesterolemia were selected from the outpatients, who were evaluated for the presence of CAD, at the cardiovascular disease clinics affiliated with Shiraz University of medical sciences (Shiraz, Iran) between January 2012 and October 2013. The study protocol was approved by the Ethics Committees of Shiraz University of Medical Sciences (code number 91-01-36-5018). Pregnant women as well as individuals with serum triglycerides above 400 mg/dl or with thyroid disease (TSH≥5 or ≤0.4 µu/dl), renal failure (SCr≥2 mg/dl for women, SCr≥2.5 mg/dl for men) [16] , liver diseases (ALT≥40 mg/dl), diabetes mellitus (who had two consecutive FBS≥126 mg/dl) and individuals treated with oral contraceptives and lipid-lowering drugs were not enrolled in the study [16] . All the patients signed an informed consent form. Four weeks before the study, all the subjects visited a dietitian and were instructed to consume a low-cholesterol diet (total daily fat intake 25 to 35% of total calories, trans fats to less than 7% of calories) [16] . Information on age, gender, body mass index, blood pressure or anti-hypertensive drug history, physical activity (20 minutes walking per day) [16] , family history and past medical history of each patient were recorded. Alcoholic and smoker patients were excluded from the study due to the limited numbers. Individuals were categorized into CHD and non-CHD groups according to National Cholesterol Education Program ATP-III guidelines [3] . For non-CHD subjects, the risk factor was calculated, and global risk factor assessment was carried out to determine the initial treatment goals. All cases had two or more risk factors and were treated with 10, 20 or 40 mg atorvastatin (Sobhan Darou, Tehran, Iran) orally once daily for four weeks. The patients received an appropriate dose of atorvastatin according to their ideal LDL-C goal (patients with CHD, LDL-C≤70 mg/dl and LDL-C≤130 mg/dl for non-CHD patients).

Lipid measurement
One week before and four weeks after atorvastatin administration, peripheral blood samples were collected from the individuals after an overnight fasting (8-12 h) [16] . Total serum cholesterol, HDL-C, LDL-C and triglyceride concentrations were measured. Before and after the administration of atrovastatin, ALT and creatinine kinase were determined to detect the possible adverse drug reactions.

Statistical analysis
Continuous variables were presented as mean±SD. Categorical variables were reported as counts (percentage). To find any potential association between genotypes in candidate genes and demographic phenotypes and response to atorvastatin, we first divided the patients into two different dosage groups, individuals treated with 10 mg atorvastatin per day and those who were treated with 20 or 40 mg. To evaluate the association between ABCB1 and ABCC1 polymorphisms and serum lipid concentration, individuals carrying the homozygous form of the less common allele were grouped with the heterozygous carriers (variant group), and those carrying the homozygous form of the more common allele were put in the wild-type group. General linear regression model was employed to evaluate the association of C3435T and G2012T genotypes with LDL-C reduction. Logistic regression analysis was carried out to compare the effect of demographic factors and C3435T and G2012T polymorphisms on clinical response to atorvastatin (or achievement of LDL-C to therapeutic goal). The allele frequency of these variants was assessed for deviation from the Hardy-Weinberg equilibrium using Chi-square test. A P<0.05 was considered statistically significant. SPSS 17 software (SPSS Inc., Chicago, Illinois) was used for statistical analysis.

Subjects' characteristics and the frequency distribution of the polymorphisms
Patients (n=179) with primary hypercholesterolemia were enrolled in this study. Based on the laboratory data, the mean±SD of LDL-C, triglyceride, total cholesterol and HDL-C were139.8±27.2, 163.4±60.8, 217.7±37.9 and 42.6±8.6 mg/dl, respectively. As shown in Table 1, environmental factors have similar frequencies between the two dosage groups (P>0.05). The genotype and allele distribution of two polymorphisms are shown in Table 1. The success rate for obtaining the LDL-C level was 76%.

Effect of ABCC1 and ABCB1 polymorphisms on serum lipid level
The percentage change in lipid parameters by genotype is shown in Table 2. Multivariate logistic regression analysis showed significant differences between males and females' responses to atorvastatin following four weeks of treatment with 10 mg/day atorvastatin (P=0.004, odds ratio=0.2, CI 95%= 0.06-0.6). The result indicated that women are better responders to the therapy by five-fold. Furthermore, no significant association was detected between the ABCB1 (C3435T) and ABCC1 (G2012T) polymorphisms and clinical response with different dosages of atorvastatin in Iranian hyperlipidemic patients. Linkage disequilibrium between C3435T and G2012T polymorphisms was not observed in our study population (P=0.56).

DISCUSSION
The present investigation showed that the frequencies of the ABCB1 genotype (CC: 2.80%, TT: 11.70% and CT: 85.50%) were not similar to those found in Iranian population in previous studies (P>0.05) [17][18][19] . Azarpira et al. [17] have found ABCB1 wild-type genotype (3435CC) in 19% of Iranian renal transplant patients, whereas 51% and 30% were heterozygous (3435CT) and homozygous (3435 TT), respectively. The genetic risk variants such as APOE, PCSK9,CDKN2A and CDKN2B are associated with conventional risk factors for CAD and very commonly occur in 50% of the population [20] . In this light, we suggest that ABCB1 polymorphism might be considered as a genetic risk factor for CAD development, and it may affect the genotype profile of these patients. However, no data were found regarding ABCB1 genotype distribution in healthy control group to compare with our patient group.
In the present study, logistic regression analysis revealed significant differences between males and females' response to atorvastatin regarding the achievement of LDL-C goal after four weeks of treatment with 10 mg/day atorvastatin (P=0.004, odds ratio=0.2, CI95%=0.06-0.6). Women in the first group showed better response than men. One study indicated lower mean area under the concentration-time curve and shorter half-life of atorvastatin in women than in men [22] . In a previous study on Japanese women with primary hypercholesterolemia LDL-C level was decreased to a greater extent in women than men after three months of treatment with atorvastatin [23] , which was in accordance with our findings. The current study revealed that four weeks of treatment with 10 mg/day atorvastatin significantly reduced LDL-C level in various groups of ABCB1 gene polymorphism (CT+TT) and wild-type group of ABCC1 gene (GG).
The LDL-C reduction demonstrated that the effect of these polymorphisms on atorvastatin efficacy was more clear in low-dose atorvastatin. Taking dose dependency of atorvastatin into consideration [16] , we suggest that in higher doses of atorvastatin, the effect of dose is more significant than the polymorphism effects.
In a recent study, Kajinami et al. [10] found no association between CC genotype of ABCB1 and smaller reductions in LDL-C, but larger increases were observed in HDL-C after treatment with atorvastatin (10 mg/day). Furthermore, our study confirmed the investigation by Hoenig et al. [24] on Australian patients treated with 80 mg atorvastatin. The result indicated that the CC genotype at the C3435T polymorphism in ABCB1 was associated with reduced atorvastatin efficacy regardless of the cholesterol metabolism following a six-week administration of atorvastatin. In Alzoubi et al. [25] study, both the TT genotype of G2677T (ABCB1) and the TT genotype of the C3435T (ABCB1) polymorphisms were associated with lower levels of LDL-C after atorvastatin treatment in Jordanians hypercholesterolemic patients. However, the effects of atorvastatin on the levels of total cholesterol, triglyceride and HDL-C had no relationship with any of the genotypes in both polymorphisms. In Rebecchi et al. [15] study on Brazilian population, no relationship was found between LDL-C reduction and ABCC1 (G2012T) polymorphism, but ABCB1 and ABCC1 mRNA levels were modulated by atorvastatin and ABCB1 G2677T/A polymorphism. Also, ABCB1 baseline expression was related to the differences in the serum LDL-C and ApoB in response to atorvastatin. Unequal group sizes and the risk of selection bias with the groups are the main limitations of this study. A short-term follow-up and the lack of control group for comparison may be considered as other important limitations. However, we concluded that in a low dose of atorvastatin, ABCB1 and ABCC1 polymorphisms may have an effective impact on atorvastatin efficacy in reduction of LDL-C serum level in Iranian patients with primary hypercholesterolemia.