Laboratory Medicine in the Clinical Decision Support for Treatment of Hypercholesterolemia: Pharmacogenetics of Statins

Abstract

Statin responsiveness is an area of great research interest given the success of the drug class in the treatment of hypercholesterolemia and in primary and secondary prevention of cardiovascular disease. Interrogation of the patient's genome for gene variants will eventually guide anti-hyperlipidemic intervention. In this review, we discuss methodological approaches to discover genetic markers predictive of class-wide and drug-specific statin efficacy and safety. Notable pharmacogenetic findings are summarized from hypothesis-free genome wide and hypothesis-led candidate gene association studies. Physiogenomic models and clinical decision support systems will be required for DNA-guided statin therapy to reach practical use in medicine.

Trial registration: ClinicalTrials.gov NCT00767130.

Keywords: Coenzyme Q(10); Lipid metabolism; Myalgia; Myopathy; PCSK9 inhibitors; Pharmacogenetics; Physiogenomics; Statins.

Copyright © 2016 Elsevier Inc. All rights reserved.

Figures

Figure 1

Associations to LDL cholesterol lowering through for ACACB rs34274 (left panel) and ACACB rs2241220 (right). Each circle represents a subject (genotype), with the horizontal axis specifying the low density lipoprotein cholesterol, and the vertical axis the genotype: bottom circles – homozygous for major allele, middle circles – heterozygous, top circles – homozygous for minor allele. A LOESS fit of the allele frequency as a function of LDLC (thick line) is shown. LOESS: LOcally-wEighted Scatter plot Smooth. (From Ruaño G, Thompson PD, Kane JP, et al. Physiogenomic analysis of statin-treated patients: domain-specific counter effects within the ACACB gene on low-density lipoprotein cholesterol? Pharmacogenomics. 2010 Jul;11(7):959–71; with permission.)

Figure 2

Manhattan plot for Myalgia index. The scale on the ordinate represents statistical significance as the log-score given by s = −log10p, and the abscissa represents the chromosomal location, with the chromosome boundaries indicated by the coloring of the data points. The strongest SNP associations (log-score > 5) are indicated by a filled circle (●), and the top six markers are labeled with the name of the associated gene for 865,483 SNPs.

Figure 3

Genomic locus and effect on myalgia for four of the genes identified in the whole-genome screen. The panels show log-scores of association for all SNPs within 200 kb of the index SNP along the chromosome, and on the right show the effect of the variant allele on the probability of myalgia in subgroups of patients taking one of the three major statins. (Top Panel from Ruaño G, Windemuth A, Wu AH, et al. Mechanisms of statin-induced myalgia assessed by physiogenomic associations. Atherosclerosis, 2011. 218(2): p. 451–6; with permission)

Figure 4

Left panel: Frequency distribution showing the numbers of patients carrying from 0 to 6 risk alleles, respectively, at the 3 polymorphic sites rs4693570 (COQ2, para-hydroxybenzoate-polyprenyltransferase), rs6732348 (DMPK, myotonin, protein kinase), and rs17381194 (Plasma membrane calcium-transporting ATPase 1). All patients were treated with statins and ~40% were diagnosed with statin myalgia. Right panel: Statin myalgia risk index curve based on the same SNP markers. According to the function, a patient with 0 or 1 risk allele has less than 1% chance of experiencing myalgia on statin. A patient with 5 risk alleles has a 54% chance and a patient with 6 risk alleles, a 70% chance. As the number of predictive SNP markers in the model is increased, it may be possible to refine further the prediction of statin myalgia beyond 70%.

Figure 4

Left panel: Frequency distribution showing the numbers of patients carrying from 0 to 6 risk alleles, respectively, at the 3 polymorphic sites rs4693570 (COQ2, para-hydroxybenzoate-polyprenyltransferase), rs6732348 (DMPK, myotonin, protein kinase), and rs17381194 (Plasma membrane calcium-transporting ATPase 1). All patients were treated with statins and ~40% were diagnosed with statin myalgia. Right panel: Statin myalgia risk index curve based on the same SNP markers. According to the function, a patient with 0 or 1 risk allele has less than 1% chance of experiencing myalgia on statin. A patient with 5 risk alleles has a 54% chance and a patient with 6 risk alleles, a 70% chance. As the number of predictive SNP markers in the model is increased, it may be possible to refine further the prediction of statin myalgia beyond 70%.

Figure 5

SINM PhyzioType results and drug recommendations for 3 patients (Patients #1, #2, #3). The PhyzioType Result is a matrix defined by predictions for 3 phenotypes (Myalgia Index, locCK, LDL) in response to 3 drugs (rosuvastatin, atorvastatin, simvastatin). The values on the left of the cells are predicted responses, with percentile ranks in parentheses on the right. The color coding of the cells with the predictions for each patient represents their quartile rank in a population distribution: RED, worst quartile with unfavorable response; GREEN, best quartile with favorable response; YELLOW, second and third quartiles with intermediate response.