APOE (Apolipoprotein E)

APOE encodes a lipoprotein that is a component of many lipid particles. The two most studied loci in the APOE gene are rs7412 and rs429358, which define three haplotypes (ε2, ε3, and ε4). In Caucasians treated with atorvastatin, individuals who are heterozygous for the rare allele of rs7412 (belonging to the ε2 haplotype) experienced a 39.9% lowering of LDL-c compared with 36.4% in individuals with the major allele (belonging to the ε3 haplotype) ( P =0.002) . Furthermore, in a study of 509 patients randomized to 10 mg atorvastatin, 20 mg simvastatin, or 10 mg pravastatin, there was an attenuated LDL-c lowering with the major ε3 haplotype compared with the minor ε2 haplotype: 30% vs 36% ( P =0.005 after adjusting for statin type) . The study also tested if the effect of ε3 haplotype could be overcome by dose escalation. Even with maximally prescribed doses (80 mg atorvastatin, 80 mg simvastatin, or 40 mg pravastatin), ε3 haplotype carriers continued to have diminished LDL-c lowering compared to ε2 carriers: 39% vs 45% ( P =0.009 after adjusting for statin type) . In a metaanalysis of over 17,000 patients treated with statins, the APOE variants emerged as the most significant (metaanalysis P -value=8×10 ⁻²⁹ ) conferring a 5% greater LDL-c lowering compared to noncarriers . Finally, a large-scale candidate gene study of 18,705 individuals revealed that rs7412 was associated with a near 3% reduction of LDL-c per allele with simvastatin therapy . Thus, the evidence is compelling for the influence of the ε3 haplotype on LDL-c lowering, and it appears that this effect cannot be overcome by simple dose escalation.

SLCO1B1 (Solute Carrier Organic Anion Transporter Family, Member 1B1)

The majority of research on the genetic contribution to statin-induced side effects has focused on SLCO1B1 (also referred to as SLC21A6 , OATP-C , or OATP1B ), which codes for a hepatic drug transporter that mediates the hepatic uptake of statins. The *5 variant is defined by the C allele of rs4149056 and encodes an alanine to valine substitution at amino acid position 174. The polymorphism interferes with the localization of the hepatic drug transporter to the plasma membrane , resulting in elevated circulating concentrations of statins, as measured by plasma area under the curve values or Cmax . As a result of reduced function of this transporter, hepatic exposure to statins is reduced in carriers thus reducing exposure to HMGCR. As a consequence, each copy of the *5 allele confers a 1%–2% smaller LDL-c lowering compared to noncarriers . Despite this statistical significance of this finding, it does not appear that carrying the *5 allele is associated with any heightened risk of cardiovascular events despite mildly higher LDL-c .

LPA (Lipoprotein(a))

LPA codes for apolipoprotein (a) which when linked with LDL particles form Lp (a)—a well studied molecule associated with coronary artery disease (CAD). The Lp (a) molecule has both atherogenic and thrombogenic effects, in vitro, but the extent to which these translate to differences in how atherothrombotic disease presents is unknown. The rs10455872 and rs3798220 single nucleotide polymorphisms (SNPs) are independently and strongly associated with the KIV-2 copy number variant in Lp (a), which encodes variability in apo (a) size and is responsible for ~30% of variance in Lp (a) levels . Since LDL-c resides (in part) in LP (a) and statins do not lower Lp (a) levels, carriers of LPA variants have a ~5% smaller LDL-c lowering in response to statin therapy in a recent metaanalysis of statin treated patients ( P =7×10 ⁻⁴⁴ ) .

SORT1/CELSR2/PSRC1

Two variants on chromosome 1p (rs646776 and rs12740374) were recently identified in a metaanalysis of 38,000 statin treated patients studied using GWAS . These variants are in proximity to three genes SORT1, CELSR2 , and PSRC1 . The rs12740374 variant is an expression quantitative trait locus (eQTL) eQTL for the three genes, resulting in increased transcription, which produces a ~1.5% greater LDL-c lowering than noncarriers. SORT1 encodes sortilin, which reduces LDL-c through a statin-independent mechanism. Carriers of these two variants have lower levels of LDL particles that are resistant to statin lowering and as a consequence higher proportion of LDL particles that are statin responsive.

In general, the pharmacogenetic associations of LDL-c lowering with statin therapy are mild (<5% differences in LDL-c) and thus are unlikely to inform clinical decision-making given that the average LDL-c lowering with statin therapy is 30%–40%.

CYP2C219

Clopidogrel is an inactive prodrug that requires hepatic bioactivation by several enzymes, including CYP2C19 . The prodrug is converted into its active metabolite in a 2-step process involving several cytochrome P450 (CYP) enzymes. This resulting active metabolite then irreversibly inhibits the platelet ADP receptor, P2Y12 .

The activity of CYP enzymes varies considerably between individuals as a result of genetic variation. CYP2C19 is involved in both steps of clopidogrel activation: (1) conversion of clopidogrel into 2-oxo-clopidogrel, and then (2) conversion of 2-oxo-clopidogrel into the active metabolite. Genetic variants that diminish the activity of the enzyme will cause shunting of the prodrug to the esterase-mediated pathway to form inactive metabolites. This will lead to decreased levels of the active metabolite and less inhibition of platelets, ultimately leading to a greater risk of cardiovascular events .

While the *1 allele of CYP2C19 has full enzymatic activity, the *2 (rs4244385) variant is the most common of the reduced-function variants and produces a complete loss of enzymatic activity. Carriers of *2 produce a lower amount of active metabolite and therefore have attenuated clopidogrel-induced platelet inhibition . With the *2 allele, a gene–dose effect is seen, where an increasing number of reduced-function alleles results in a decreasing amount of platelet inhibition . In a multicenter, randomized clinical trial, CYP2C19 genotype was associated with on-treatment reactivity (OTR) while receiving clopidogrel after PCI, at 12–24 hours ( P =2.2×10 ⁻¹⁵ ), 30 days ( P =1.3×10 ⁻⁷ ), and 6 months ( P =1.6×10 ⁻¹¹ ) after PCI . Furthermore, the risk of high OTR 12–24 hours after PCI was 2.5-fold greater for carriers of one reduced-function allele and 4.5-greater for carriers of two reduced-function alleles.

Apart from *2, other loss-of-function variants exist [*3 (rs4986893), *4 (rs28399504), *5 (rs56337013)]. These variants are rare but produce similar enzymatic defects as the *2 allele .

For carriers of *2, one treatment strategy is to consider higher doses of clopidogrel. A multicenter, randomized, double-blind trial showed that in patients with stable cardiovascular disease, tripling the maintenance dose of clopidogrel to 225 mg daily in CYP2C19*2 heterozygotes achieved levels of platelet reactivity similar to that seen with the standard 75 mg dose in noncarriers; in contrast, for CYP2C19*2 homozygotes, doses as high as 300 mg daily did not result in comparable degrees of platelet inhibition . Alternative P2Y12 inhibitors are also available. Ticlopidine is a first generation thienopyridine but does not seem to be impacted by the *2 or *3 CYP2C19 polymorphisms . Prasugrel (third generation), like clopidogrel, is also a prodrug but is unique in that its bioactivation appears to be less dependent on CYP2C19 . In contrast to the metabolism of clopidogrel, esterases are part of the activation pathway and work in series with CYPs in the oxidation steps to produce the active metabolite of prasugrel . Carriers of the *2 allele produce equivalent concentrations of active metabolite and achieve similar degrees of platelet inhibition compared to noncarriers . Ticagrelor, a nonthienopyridine P2Y12 antagonist, is orally active and thus is not influenced by genetic variation at CYP2C19 , thus providing another option for carriers .

CYP2C19

Since there is significant evidence for the influence of CYP2C19 variants on clopidogrel-induced laboratory outcomes, there has been considerable investigation in extending these observations to clinical outcomes. In patients who received PCI after ACS and were treated with clopidogrel, carriers of at least one *2 allele had a 1.5-fold increased risk of death, MI, and stroke in the subsequent year of follow-up compared to noncarriers . In patients with ST-segment elevation myocardial infarction (MI) who received clopidogrel, carriers of any two loss-of-function alleles (*2, *3, *4, or *5) had a twofold increase in the risk of death from any cause, nonfatal stroke, or MI in the year of follow-up . In addition to these composite outcomes, the incidence of stent thrombosis was also found to be increased threefold in carriers of at least one *2 allele and up to sixfold in carriers of two alleles . Therefore, the association between CYPC219 loss-of-function alleles and the clinical response to clopidogrel is consistent with that seen in the associations between the alleles and platelet function.

The CYP2C19 associations with clinical response are also context specific. For example, in the ACTIVE-A trial, which contained atrial fibrillation patients ineligible to receive warfarin that received a modest benefit on 75 mg clopidogrel over placebo, there was no influence of CYP2C19 loss-of-function alleles on the primary efficacy outcomes (stroke, MI, noncentral nervous system embolism, cardiovascular death) or safety outcomes (major bleed) . Furthermore, in non-PCI populations, such as the CURE trial of medically-managed ACS, there was also no effect of CYP2C19 on outcomes. Therefore, outside of the ACS/PCI window, it is unlikely that CYP2C19 polymorphisms play an important role in influencing outcomes in these conditions.

Another allele, *17, is often reported as a “gain-of-function” allele. Many initial reports found an association with improved clinical outcomes compared to noncarriers (10.0% vs 11.9%, meta analysis P =0.005) in patients with CAD. In four of these six studies, *17 carriers had an increased risk of bleeding (8.0% vs 6.5%, metaanalysis P =0.006) . However, what is often not appreciated is that the *17 is in near perfect linkage disequilibrium with the *2 allele ( D′ =1.0). Put another way, in the vast majority of patients, carrying the *17 allele implies the absence of the *2 allele and vice versa. Therefore, any analysis of *17 must account for the effects of *2 in order to assess its independent contribution. After adjustment for the *2 allele, the *17 allele confers no difference in the platelet response to clopidogrel .

In contrast to clopidogrel, prasugrel is not associated with an increased risk of cardiovascular death, MI, stroke, or stent thrombosis in carriers of the *2 allele, thus providing the rationale for substituting prasugrel for clopidogrel in carriers of *2 . Because ticagrelor is administered as an orally active drug, we should not expect genetic variation in CYP2C19 to influence the response to ticagrelor. To confirm this, investigators have shown that CYP2C19*2 does not influence platelet aggregation or clinical outcomes in patients treated with ticagrelor, unlike clopidogrel. In an effort to identify genetic variants beyond CYP2C19 that are associated with the response to ticagrelor, a recent GWAS was conducted on levels of ticagrelor and its active metabolite from the PLATO clinical trial. Genetic variants in SLCO1B1 that were in linkage disequilibrium with the SLCO1B1*5 , loss of function variants were associated with ticagrelor and active metabolite levels. None of the variants in SLCO1B1 , however, were associated with bleeding or ischemic events in the ticagrelor treated arm .