Should we test for SLCO1B1 genotype before prescribing statins?—a discussion of clinical trial results

Genetic biomarkers that allow prediction of drug response or toxicity hold promise for the personalization of pharmacological treatment. By now, more than 120 drugs have received pharmacogenetic drug labels by FDA or EMA that provide guidance regarding indications, contraindications or dosage recommendations in relation to patient genotype (1,2). One of those pharmacogenetic associations regards statins, which are widely used as cholesterol-lowering drugs for the prevention and treatment of cardiovascular disease in high-risk patients. Specifically, there is strong genetic as well as mechanistic evidence for a link between statin-induced myopathy and rhabdomyolysis with the presence of the variant rs4149056 (c.521T>C; p.V174A) in SLCO1B1 (3), the gene encoding the hepatic statin transporter OATP1B1. Rs4149056 has been shown experimentally to result in reduced transporter function (4) and has been repeatedly associated with elevated statin plasma concentrations due to impaired hepatic uptake (5,6). As a consequence of this available evidence, SLCO1B1*5 and *15, the main alleles containing rs4149056, are included in guidelines by the Clinical Pharmacogenetics Implementation Consortium (CPIC) to mitigate statin-related myotoxicity (7). Of note however, this guideline only refers to the interpretation and implementation of test results if available, not whether pharmacogenetic testing should be conducted.

The decision of whether preemptive pharmacogenetic testing should be conducted is complex and requires the evaluation of a multitude of factors, including the prevalence and predictive power of the pharmacogenetic biomarker, costs of genetic testing, the severity and costs associated with the adverse event, as well as safety, cost, and efficacy of alternative therapies. Moreover, there are important concerns regarding the effects of genotyping on prescribing practices and patient adherence. Specifically, the presence of a risk genotype might discourage prescription of recommended statin levels or adversely affect adherence, which might result in a worsening of hypercholesterolemia and, accordingly, negative impacts on public health. This concern is of particular importance as previous evaluations of the clinical validity, utility and cost-effectiveness of preemptive SLCO1B1 genotyping for statin risk have remained controversial (8-10).

Recently, Vassy and colleagues presented the results of a multicenter randomized non-inferiority trial to test whether preemptive genotyping of statin-naive patients at high risk of atherosclerotic cardiovascular disease would not have negative effects on low density lipoprotein cholesterol (LDL-C) reduction (11). The study was conducted over the course of 3.5 years across eight primary care practices in the Veterans Affairs Boston Healthcare System, randomizing a total of 408 patients. Importantly, patients in the intervention group experienced LDL-C reductions at 12 months that was not significantly lower compared to the control group (mean change in LDL-C in intervention vs. control group, −1.1 vs. −2.2 mg/dL; difference, −1.1 mg/dL; 90% CI, −4.1 to 1.8 mg/dL; P<0.001). Furthermore, while the fraction of patients receiving statin therapy in agreement with the guidelines by the American College of Cardiology-American Heart Association (ACC-AHA) was overall low in both groups, the number in the intervention group was noninferior compared to the control group [12 patients (6.2%) in the intervention group vs. 14 patients (6.5%) in the control group; 90% CI, −0.038 to 0.032; P<0.001]. Only few muscle-related adverse events were registered in any of the groups [2 cases (1%) in the intervention group vs. 3 cases (1.4%) in controls]. Notably, none of these were related to CPIC guideline-discordant prescriptions, neither in the control nor the intervention group, suggesting that the slightly higher adverse event rates in the control group would not have been prevented by genetically-guided prescribing.

Besides variants in SLCO1B1, also variability in ABCG2 (rs2231142), encoding the export transporter protein BCRP, has been implicated in statin pharmacokinetics, specifically of rosuvastatin (12-14). While impacts of rs2231142 in ABCG2 are overall markedly lower than for rs4149056 in SLCO1B1, it remains to be investigated whether incorporation of this variant in pharmacogenetic guidance might contribute to avoiding the remaining safety events in individuals with normal SLCO1B1 phenotypes. Furthermore, both SLCO1B1 as well as ABCG2 harbor a plethora of other, mostly rare variations, which have been estimated to account for a total of 5–10% of the overall genetically encoded functional variability of OATP1B1 and BCRP (15,16). Analogously to the incorporation of additional genes, consideration of these additional variations using sequencing approaches might provide further refinements for the optimization of personalized prescribing.

A main limitation of the study is the considerable heterogeneity in statin prescriptions with fewer patients with decreased or poor SLCO1B1 function genotypes in the intervention group than in the control group combined with a relatively small cohort size. Specifically, only 24 and 26 patients received statin prescriptions in the control and intervention groups, respectively, and of these only less than one-third were heterozygous or homozygous for rs4149056. Moreover, among these patients with decreased or poor transporter genotype only one patient in the control group received a prescription discordant with the CPIC guidelines for the respective genotype (20 mg simvastatin); however, the patient was not among those experiencing an adverse event. Despite those limitations, the presented results are encouraging and ameliorate concerns that preemptive genotyping for risk variations of statin-related myopathy might reduce adherence or worsen patient outcomes. While we believe that the current data is not sufficient to endorse the implementation of SLCO1B1 testing to guide statin therapy in primary care, such a strategy does at the very least not seem to negatively impact atherosclerotic cardiovascular disease prevention, thus paving the way for further evaluations of the cost-effectiveness and patient benefits of genotype-guided prescribing in larger dedicated trials.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.