Good cholesterol control is vital to keep cardiovascular diseases at bay and ensure normal functioning of essential cellular processes in the body. In the human body, cholesterol is either taken up from dietary sources or predominantly synthesized de novo in the liver. (Lecerf and de Lorgeril, 2011). Excess cholesterol is subsequently removed via conversion into bile acids, where the process is regulated by the rate-limiting enzyme, cholesterol 7α-hydroxylase (CYP7A1). (Pandak et al., 2001). In conditions where cholesterol levels require drug interventions to achieve control, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, i.e., statins, are the cornerstone therapy. (Crismaru et al., 2020).

However, inter-individual differences in serum lipid responses to statin treatment have been frequently reported, of which genetics is one important intrinsic factor that contribute to these differences. There is increasing evidence to show that CYP7A1 is highly polymorphic and patients harbouring CYP7A1 polymorphisms, such as −204 A/C (rs3808607), −278 A/C (rs3808607) and rs8192875), experienced differential lipid lowering responses to statins. (Kajinami et al., 2005; Aruna Poduri et al., 2010; Jiang et al., 2012; Liu et al., 2017; Na Liu et al., 2020).

Multiple studies have shown that there is an association between CYP7A1 polymorphisms and a reduced response to statin therapy, where homozygous carriers of the variant allele that encode for the reduced expression of CYP7A1 experienced less than desirable cholesterol control when administered the same dose of statin as non-carriers. (Kajinami et al., 2005; Aruna Poduri et al., 2010; Jiang et al., 2012; Liu et al., 2017; Na Liu et al., 2020). The implication is that over time, such patients who are resistant to therapy may be subjected to a higher statin dose or polypharmacy, or experience poor cholesterol control, consequently leading to intolerable toxicities, risk of drug-drug interactions and disease progression. Because of this, there has been increasing research interest in identifying and studying functional CYP7A1 single nucleotide polymorphisms (SNPs) and their impact across different populations.

While there is an increasing number of studies looking into this, the sample sizes of the individual studies were small, and results were inconsistent. Hence, this systematic review and meta-analysis aimed to pool the relevant studies together to provide a larger sample size for enhanced statistical power so as to ascertain any small but relevant effects of CYP7A1 SNPs on the cholesterol lowering response to statins.

In this review, we aimed to answer the following research question: In individuals receiving statin treatment, do CYP7A1 SNPs affect the serum lipid responses to the treatment? Specifically, we are interested in comparing the lipid response outcomes to statin treatment between carriers of the variant allele versus the non-variant allele of CYP7A1 SNPs.

This meta-analysis serves as an update to a previously published review that meta-analysed the results from 3 studies to evaluate the impact of one specific CYP7A1 SNP, i.e., −204 A/C, on the treatment effects of statins on total cholesterol, LDL-C, HDL-C and triglycerides (Hofman MK et al., 2005; Kajinami et al., 2005; Aruna Poduri et al., 2010; Wei et al., 2011; Li et al., 2014). It is an important verification of earlier trends by analysing a more recent and larger data set. To the best of our knowledge, this is the first meta-analysis that pooled relevant studies without restricting to a specific CYP7A1 SNP. This not only allows us to amass a larger sample size for better statistical power and more precise point estimates, but also provides us with insights on how these CYP7A1 SNPs can collectively affect lipid outcomes in subjects with or without the variant allele who were administered a statin. Additionally, we expanded our scope of review to look into the effect of treatment duration on the lipid parameters through a stratified analysis, which was not explored in an existing review (Li et al., 2014).

A total of 6 studies comprising 1,686 subjects for the assessment of total cholesterol, LDL-C and HDL-C, and 1,156 subjects for the assessment of triglycerides were included in this meta-analysis. The study findings showed that subjects who were non-carriers of a CYP7A1 SNP had a statistically significantly greater reduction in total cholesterol and LDL-C levels as compared to their counterparts who borne the variant allele of CYP7A1 SNPs when administered a statin, irrespective of the duration of therapy. This implied that patients who carry the variant allele of CYP7A1 SNPs are at risk of a reduced treatment efficacy when administered an equivalent dose of statin as compared to patients who do not carry the variant allele. Consequentially, these patients might require a higher than necessary statin dose to achieve a similar reduction in total cholesterol and LDL-C levels, and be predisposed to the detrimental side effects of statins such as rhabdomyolysis. (Noyes et al., 2021). One possible explanation for the observed differences in reduction of total cholesterol and LDL-C levels could be that there is a reduced expression of CYP7A1 in subjects who borne a variant allele of a CYP7A1 SNP, which in turn reduces the metabolism of cholesterol to bile acids in the liver thereby leading to accumulation of cholesterol. (Wang et al., 2018; Chiang and Ferrell, 2020). A reduced expression of CYP7A1 may also downregulate LDL-receptor activity leading to reduced catabolism of LDL-C. (Pullinger et al., 2002). Our study results are also congruent with the previously published meta-analysis where subjects who were non-carriers of a CYP7A1 SNP were shown to have a greater reduction in total cholesterol (MD -2.56, 95% CI -4.49, −0.63, p = 0.009; n = 3) and LDL-C levels (MD -0.33, 95% CI -0.26, 1.41, p = 0.71; n = 3) as compared to their counterparts who borne a variant allele of CYP7A1 -204 A>C (Li et al., 2014). While the previously published meta-analysis did not establish a significant difference in LDL-C levels between carriers and non-carriers of CYP7A1 -204A/C when administered a statin, our study reported a significant difference in reduction of LDL-C levels in favour of non-carriers of CYP7A1 -204A/C.

There is, however, insufficient evidence to suggest that the presence of variant allele of CYP7A1 SNPs had a significant effect on the reduction of triglycerides levels or improvement in HDL-C levels when subjects were administered a statin. This is in contrast to the previously published meta-analysis where subjects who were non-carriers of a CYP7A1 SNP had a greater reduction in triglycerides (MD -1.92, 95% CI -7.48, 3.63, p = 0.64; n = 2) and a greater increase in HDL-C levels (MD -0.39, 95% CI -1.13, 0.35, p = 0.30; n = 3) as compared to their counterparts who borne a variant allele of a CYP7A1 SNP (Li et al., 2014). One plausible explanation for the lack of a significant effect on HDL-C could be that statins generally have a mild effect on HDL-C, which resulted in minimal change from baseline across all subjects and consequentially, negligible differences between subjects with the homozygous dominant genotype and variant allele genotype. This is not surprising as statins are known to have a modest effect on HDL-C and other pharmacologic agents (e.g., niacin and fibrates) in place of statins are routinely used in practice to improve HDL-C levels (Chapman, 2004; Schachter, 2004; RobertOh, 2007; Amit and Khera, 2013; Parhofer KG and Laufs, 2019). Further, CYP7A1 is not a key enzyme involved in the regulation of triglycerides levels, and therefore perturbation of its bioactivity via genetic polymorphism may not exhibit direct impact on triglycerides metabolism.

We were also able to learn from the study results that a longer treatment duration with statins (at least 12 weeks), at this juncture, has a favourable impact on total cholesterol and LDL-C levels in subjects who were non-carriers of a CYP7A1 SNP as compared to subjects with the variant allele of the CYP7A1 SNPs. This also reflects the clinical significance of CYP7A1 genotype in the long term management of cholesterol, which is a more plausible scenario than as an acute treatment. However, duration of therapy did not have a significant impact on HDL-C and triglycerides levels. Future studies, if available, can be conducted to further elucidate the association between study duration and treatment effect of statins on all four lipid parameters in subjects with or without the variant allele of the CYP7A1 SNPs.

We highlighted earlier that a previously published meta-analysis focused on one specific CYP7A1 SNP, i.e., −204 A/C, while our meta-analysis was conducted on all relevant studies without restricting to a specific CYP7A1 SNP. Specifically, the studies included in our meta-analysis looked into the CYP7A1 SNPs rs3808607 and rs8192875. The CYP7A1 SNP, rs3808607, is located in the promoter regions, while rs8192875 is located on the fourth exon of the gene and is associated with an amino acid change from aspartic acid to asparagine. It is known that the promoter region of the CYP7A1 gene contains multiple functional binding sites for liver-enriched transcription factors. (Molowa DT et al., 1992; Cooper et al., 1997; Deng et al., 2017). It is therefore possible that a polymorphism in this region can inhibit the binding of transcription factors thereby decreasing transcription and resulting in a reduced expression of the CYP7A1 enzyme. It is also suggested in literature that CYP7A1 expression can possibly be modulated by two interacting regulatory SNPs, specifically an enhancer SNP and promoter SNP, adding weight to our perspective that it is meaningful to examine the pooled impact of CYP7A1 SNPs on lipid outcomes. (Wang et al., 2018). As for exonal SNPs, they have the ability to replace the amino acid that encodes a protein and potentially alter the amino acid sequence such that protein structures and functions are impacted. (Deng et al., 2017). Therefore, by including studies without restricting to a specific CYP7A1 SNP to the meta-analysis, we not only can increase the sample size for a meaningful statistical analysis but can also better appreciate how these CYP7A1 SNPs can collectively affect the lipid outcomes in subjects with or without the variant allele of the CYP7A1 SNPs who were administered a statin.

Nonetheless, our study has the following potential limitations that should be taken into consideration when interpreting the study results. Firstly, we were not able to assess the impact of statin intensity on the lipid outcomes between the two groups of subjects as a majority of the included studies used a moderate-intensity statin. Future studies can be conducted using either low or high doses of statins to allow a more holistic assessment of the impact of a differential statin dose. Secondly, we were also not able to assess the impact of different statins on the lipid outcomes. A total of 5 out of 6 studies ((Kajinami et al., 2005; Aruna Poduri et al., 2010; Wei et al., 2011; Liu et al., 2017; Abed et al., 2021) used atorvastatin, while the remaining study ((Hofman MK et al., 2005)) used pravastatin. Considering different statins have different chemical structures which in turn affect their pharmacokinetic-pharmacodynamic properties and potentially efficacy and safety profiles, conducting future studies with a diverse range of statins will aid in providing more comprehensive and meaningful results (Schachter, 2004). Future research should be conducted to shed light on the impact of different statins and statin doses on the lipid outcomes in subjects with or without variant allele of CYP7A1 SNPs. Thirdly, many non-genetic extrinsic factors, such as comorbidities and comedications, can interact with the genotypes resulting in a phenomenon known as phenoconversion. (Shah and Smith, 2015). Phenoconversion converts genotypic extensive metabolisers (EMs) into phenotypic poor metabolizers (PMs) of drugs, thereby modifying patients’ clinical response and can significantly impact the interpretation of genotype-focused studies. (Shah and Smith, 2015). It will be prudent to consider patients’ clinical features and medical history when considering the impact of CYP7A1 SNPs on their lipid outcomes with the administration of statins.

Our meta-analysis showed that the presence of variant alleles of CYP7A1 SNPs can result in a smaller reduction in total cholesterol and LDL-C levels as compared with individuals who do not carry the variant allele, when administered an equivalent dose of statin. There was, however, no significant impact of variant alleles of CYP7A1 SNPs on triglycerides and HDL-C levels when subjects were administered an equivalent dose of statin. Taken together, the exact pharmacogenetic relevance of CYP7A1 SNPs to statin therapy awaits wider clinical association and mechanistic investigation.