Two for the heart: Dutch Pharmacogenetic Working Group prescribing guidance on statins and sulfonylureas to reduce cardiometabolic risk

Wolthuis and colleagues recently reported [1] new prescribing guidance from the Dutch Pharmacogenetic Working Group (DPWG) informed by a systematic review to optimize pharmacotherapy of dyslipidemia (statins) and hyperglycemia (sulfonylureas). Novel to this report is the recommendation to conduct SLCO1B1 genotyping on patients prior to prescribing high-dose simvastatin, proactive risk factor consideration to guide selection and dose of other statins, and the first prescribing guidance pertaining to CYP2C9-sulfonylurea interactions. The DPWG combined these two classes of medications because both are known to decrease cardiovascular risk in high-risk patients, albeit through different pathways.

Cardiovascular disease (CVD) (ischemic heart disease and stroke) remains the number one cause of death in the world [2]. Statins act as inhibitors of the 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) to block cholesterol synthesis and are a mainstay of treatment for reducing low density lipoprotein (LDL) cholesterol and triglycerides to reduce the risk of major cardiovascular events and for primary prevention of cardiovascular disease [3]. Although their efficacy for lipid reduction and CVD prevention are well established, of particular concern with statins are myopathy or a range of statin-associated muscle symptoms (SAMS). SAMS are reported by 10-25% of patients taking statins [4], and although rare (occurs in ~ 1/100,000 patients), there is significantly higher risk of potentially life-threatening rhabdomyolysis for patients prescribed combination therapies that include cytochrome P450 (CYP)3A4 inhibitors.

A polymorphism in the solute carrier organic anion family member 1B1 (SLCO1B1) gene (c.521 T > C) that is present in two alleles (*5 and *15) reduces the activity of this transporter that facilitates passage of statins from blood into the liver, which may lead to higher blood levels of statins and higher risk of myopathy [5]. Although not addressed by this review, the same transporter contributes to the transport of the sulfonylureas. Indeed, variants in the SLCO1B1 gene have been linked to the efficacy of these glucose-lowering medications [6]. The systematic review identified individual studies and meta-analyses that evaluated associations between SLCO1B1 c.521 C > T and endpoints of efficacy, side effects including SAMS, and hepatic bioavailability when available for atorvastatin, fluvastatin, rosuvastatin, and simvastatin. The body of evidence and strength of studies were evaluated as seen in Supplementary Table 1 of the DPWG guideline paper [1], with 0 indicating the lowest quality evidence to 4 indicating high-quality clinical trials or meta-analyses. The Table 1 below summarizes the DPWG prescribing recommendations regarding SLCO1B1 genotype and statins.

Table 1.

Summary of Therapeutic Recommendations.

Genotype	Drug	Recommendation	Clinical consequence
SLCO1B1
521 TC or CC	Atorvastatin	Use as an alternative to simvastatin for patients with no additional risk factors for statin-induced myopathy. Atorvastatin is also affected by CYP3A4 inhibitors such as amiodarone, verapamil, and diltiazem; consider pravastatin or rosuvastatin if the patient is taking a CYP3A4-inhibiting drug.	Increased theoretical risk of myopathy if patients are taking a CYP3A4 inhibitor because the enzyme is also involved in atorvastatin metabolism.
	Pravastatin	Consider as an alternative to simvastatin because of less influence of SLCO1B1 gene variation. NO action is required for this gene-drug interaction.	Pravastatin is influenced to a lesser extent by genetic variation in SLCO1B1 and is also not influenced by CYP3A4 inhibitors such as amiodarone, verapamil, and diltiazem.
	Rosuvastatin	If patient has additional significant risk factorsa for statin-induced myopathy, keep the rosuvastatin dose as low as possible (e.g., by adding ezetimibe), and advise the patient to contact healthcare providers in case of muscle complaints. If patient has NO significant additional risk factors for statin-induced myopathy, advise the patient to contact healthcare providers in case of muscle complaints.	Rosuvastatin is influenced to a lesser extent by genetic variation in SLCO1B1 and is also not influenced by CYP3A4 inhibitors such as amiodarone, verapamil, and diltiazem.
	Simvastatin	Choose an alternative such as a statin less severely affected by SLCOB1 gene variation such as pravastatin or rosuvastatin, both of which are not influenced by CYP3A4 inhibitors. Fluvastatin is not influenced by genetic variation in SLCO1B but there was not sufficient evidence to make a clinical recommendation for fluvastatin. The DPWG considers genotyping for SLCO1B1 before starting simvastatin at a dose of 80 mg/day to be “essential” for drug safety. If an alternative to simvastatin is not an option, consider risk factorsa for statin-induced myopathy before starting a lower dose (40 mg/day or less) simvastatin and advise the patient to report muscle symptoms.	Dose-dependent increase in the risk of statin-induced myopathy. No difference in LDL cholesterol lowering by genotype.
CYP2C9
*1/*2; *1/*3; intermediate metabolizer, other genotype; *2/*2; *2/*3; *3/*3; and poor metabolizer, other genotype	Glibenclamide,b gliclazide, glimepiride, or tolbutamide	NO therapy adjustments are required for these gene-drug interactions.	Increased bioavailability of these drugs. Either an increase in hypoglycemia or no negative clinical effect to outweigh the clinical importance of the increased effectiveness it signals.

^aRisk factors identified by authors for statin-associated muscular symptoms include female gender, age greater than 65 years, hypothyroidism, diabetes, and trauma.

^bGlibenclamide is also known as "glyburide".

These guidelines differ somewhat from the most recent Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines published on this gene-drug interaction [7]. The CPIC guideline also recommends prescribing of an alternative statin for patients with SLCO1B1 decreased function or SLCO1B1 possible decreased function phenotypes depending on the clinically desired potency needed for each patient. CPIC designed a prescribing algorithm to be used in conjunction with a 2018 prescribing guideline published by the American College of Cardiology (ACC) and the American Heart Association (AHA) [8] such that statin recommendations including statin intensity and dose were stratified by SLCO1B1 phenotype based on risk for SAMS with black box warnings for statins and doses at the highest risk for SAMS indicated separately for low-intensity, moderate-intensity, and high-intensity statin clinical indications. For example, high-intensity statins (known to reduce low density lipoprotein cholesterol (LDL-C) by ≥50%) are recommended by the ACC/AHA guidelines for patients with existing atherosclerotic cardiovascular disease (ASCVD), or patients at very high risk for a CVD event. Very high risk status applies to patients with a history of ASCVD events (recent or historical acute coronary syndrome or myocardial infarction, ischemic stroke, or peripheral arterial disease), age ≥65 years, history of familial hypercholesterolemia, prior coronary artery bypass surgery or percutaneous coronary intervention outside of the major ASCVD event, hypertension, chronic kidney disease, current smoking, or who have persistently elevated LDL-C despite maximally tolerated statin therapy and ezetimibe, or a history of congestive heart failure. Like the recent CPIC guideline, no recommendations were made regarding CYP3A4 genotype and statin prescribing; yet unlike the CPIC guideline, the DPWG guideline provides recommendations to avoid concomitant use of atorvastatin with CYP3A4-inhibiting drugs (e.g., amiodarone, verapamil, and diltiazem) and suggests alternative statins (pravastatin or rosuvastatin) that are not significantly affected by CYP3A4. There are some other subtle differences between the present guidelines and the aforementioned CPIC guidelines, but they are not contradictory in directionality or substantial. Both guidelines only made recommendations for the c.521 C > T polymorphism and did not support evidence for prescribing recommendations based on the SLCO1B1 *14 allele (contains both the c.388 A > C and c.463 C > A polymorphisms).

A particular strength of this guideline is the application of a clinical implication score that categorizes pre-therapeutic pharmacogenetic (PGx) results as “essential”, “beneficial”, or “potentially beneficial” based on four criteria: the clinical effect associated with the gene-drug interaction, the level of evidence supporting the clinical effect, the effectiveness of the intervention in preventing the clinical effect (number needed to genotype), and the PGx information in the drug-label. The DPWG considers genotyping for SLCO1B1 before starting simvastatin at a dose of 80 mg/day to be “essential” for drug safety. Patients who test positive for 521 C carrier status should be prescribed an alternative statin that is less severely affected by SLCOB1 gene variation such as pravastatin or rosuvastatin, both of which are not influenced by CYP3A4 inhibitors. If an alternative to simvastatin is not an option, consider risk factors^§ for statin-induced myopathy before starting a lower dose simvastatin and advise the patient to report muscle symptoms. The risk factors identified by authors for statin-associated muscular symptoms include female gender, age greater than 65 years, hypothyroidism, diabetes, and trauma.

As with CVD, the prevalence of type 2 diabetes mellitus (T2DM) is increasing worldwide [2], and T2DM is a major risk factor for CVD. Evidence supports the efficacy of tight glucose control to reduce the risk of major adverse cardiovascular events [9]. Sulfonylurea medications are insulin secretagogues – that is, they stimulate pancreatic beta cells to release insulin by binding to their receptors on the adenosine triphosphate (ATP)-sensitive potassium channel, igniting a cascade of potassium efflux, calcium influx, increasing insulin exocytosis and release into the bloodstream [1].

Although no longer considered first-line monotherapy for the treatment of type 2 diabetes, sulfonylureas remain widely used in combination with metformin and are a mainstay of treatment in low and middle-income countries. Although statins are associated with a higher risk of hypoglycemia, there are not statistically significant differences in treatment-associated mortality between second and third-generation sulfonylureas and other, newer hypoglycemic agents including glucagon-like peptide 1 (GLP-1) analogues, dipeptidyl-peptidase 4 inhibitors, thiazolidinediones, or sodium-glucose co-transporter 2 inhibitors when used in combination with metformin [10].

Genetic variants in the cytochrome P450 family 2 subfamily C member 9 (CYP2C9) gene result in absent or reduced metabolic activity of the CYP2C9 enzyme. The DPWG found that although CYP2C9 variants may lead to increased blood levels of glibenclamide (also known as “glyburide”), gliclazide, glimepiride, and tolbutamide, no therapy adjustments based on CYP2C9 metabolic phenotype are required in patients with these variants because there was either no negative clinical effect or an increase in hypoglycemia that they considered to be less important than the increase in effectiveness of the drug that it signals.

This systematic review provides many clinically actionable resources in the tables and supplementary materials. These include recommendations stratified by genotype and drug for both SLCO1B1-statin and CYP2C9-sulfonylurea gene-drug interactions, annotations of genetic variants and their frequencies across ethnic/racial groups, a structured literature review with ratings of the level of evidence of each included study, the clinical implementation score and the criteria by which it is based, such as look-up texts for electronic prescribing systems designed for pharmacists and physicians. These can be used to program clinical decision support systems.

The review was limited by not addressing the clinical implications of genotyping SLCO1B1 for sulfonylurea prescribing, possibly because of limited evidence and one could consider a “null” recommendation for the CYP2C9-sulfonylurea interactions as a limitation. However, the contribution of a clinical implementation score could actually be empowering to prescribers in both clinical contexts to identify patients and gene-drug pairs for preemptive genotyping. In low- and middle-income countries where resources are relatively scarce and the prevalence of noncommunicable diseases such as CVD and T2D are increasing, it is reassuring to know that preemptive genotyping is not essential to optimize glycemic status as part of a multiple CVD risk factor reduction strategy. Moreover, if preemptive genotyping is not essential for prescribing statins other than simvastatin, a SAMS risk-based prescribing approach can provide the precision needed to minimize the risk of myopathy. As such, this review offers clinical value to a diverse spectrum of ASCVD risk patients relevant to the global community with or without the resource of pharmacogenetic genotyping.

Author contributions

Dr. David conceived of and wrote the commentary.

Funding

Dr. David receives funding from the National Human Genome Research Institute Project Number R21HG013413 (PI: S. David) and the National Center for Advancing Translational Science Project Number UL1TR002389 (MPIs: J. Solway, K. Kim, J. Jacobs), National Institutes of Health.

Competing interests

The author declares no competing interests.