Randomized evaluation of the loss‐of‐function carboxylesterase 1 (CES1) G143E variant on clopidogrel and ticagrelor pharmacodynamics Authors: Joshua P Lewis, Kathleen A Ryan, Elizabeth A Streeten, Hilary B Whitlatch, Melanie Daue, Keith Tanner, James A Perry, Jeffrey R O'Connell, Alan R Shuldiner, Braxton D Mitchell Published: 2024 ABSTRACT AbstractAntiplatelet therapy with a P2Y12 receptor inhibitor, in combination with aspirin, is standard of care for medical management of patients with coronary artery disease, and flexibility in prescribing options among these medications offers great potential for individualizing patient care. Previously, we showed that a loss‐of‐function missense mutation (G143E) in carboxylesterase 1 (CES1), the primary enzyme responsible for clopidogrel degradation, significantly impacts on‐clopidogrel platelet aggregation and recurrent cardiovascular event risk. In the current investigation, we conducted a prospective randomized crossover study of clopidogrel (75 mg/day for 7 days) and ticagrelor (180 mg/day for 7 days) in 50 individuals stratified by CES1 G143E genotype (N = 34 143GG and 16 143GE) to determine the effect of drug choice on inhibition of platelet aggregation (IPA). Consistent with prior reports, we observed strong association between G143E and adenosine diphosphate‐stimulated platelet aggregation following clopidogrel administration (IPA = 71.6 vs. 48.0% in 143E‐allele carriers vs. non‐carriers, respectively, p = 3.8 × 10−5). Similar significant effects on platelet aggregation were also noted between 143E‐allele carriers versus non‐carriers in response to stimulation with arachidonic acid (45.8 vs. 25.8%, p = 0.04), epinephrine (44.4 vs. 18.8%, p = 0.03), and collagen (5 μg/mL, 25.8 vs. 11.4%, p = 3.7 × 10−3). In contrast, no relationship between CES1 G143E and IPA was observed following ticagrelor administration regardless of the platelet agonist used. Collectively, these data suggest that on‐clopidogrel platelet aggregation is substantially modified by CES1 G143E genotype, that this variant does not modify ticagrelor pharmacodynamics, and that more consistent inhibition of platelet aggregation may be achieved by using ticagrelor in patients who carry clopidogrel response‐modifying alleles in CES1. Study Highlights WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC? WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC? Variability in on‐treatment platelet aggregation after administration of antiplatelet agents such as clopidogrel and ticagrelor is well known and impacts patient outcomes. A loss‐of‐function missense variant (G143E) in carboxylesterase 1 (CES1) has previously been implicated in clopidogrel response, and it has been postulated that providing ticagrelor to individuals who have clopidogrel response‐modifying alleles in CES1 may lead to more consistent inhibition of platelet aggregation. WHAT QUESTION DID THIS STUDY ADDRESS? WHAT QUESTION DID THIS STUDY ADDRESS? In this investigation, we perform a prospective, randomized crossover trial of clopidogrel and ticagrelor in healthy individuals stratified by CES1 G143E genotype to evaluate the influence of this variant on agonist‐induced platelet aggregation at baseline and after each of these medications. WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE? WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE? This study convincingly shows that platelet aggregation in response to several agonists is substantially altered in clopidogrel‐treated individuals carrying the CES1 G143E genotype while no effect on these phenotypes was observed in the same participants when ticagrelor was administered. HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE? HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE? Understanding the genetic and non‐genetic determinants of platelet aggregation in response to antiplatelet medications is critical to establish more individualized treatment options for patients with clinical indications for these drugs. Better management of platelet function in these individuals can result in a reduction in rates of recurrent cardiovascular events while simultaneously minimizing adverse reactions such as pathological bleeding. INTRODUCTION Secondary prevention of atherothrombotic events in acute coronary syndrome patients with or without percutaneous coronary intervention (PCI) is medically managed using antiplatelet therapy with aspirin and a P2Y12 receptor inhibitor (i.e., clopidogrel, ticagrelor, and prasugrel). While these medications are generally effective, variability in response to these agents, defined by inhibition of platelet function, recurrence of cardiovascular outcomes, or pathological bleeding risk, occurs in 5%–30% of patients and presents a significant challenge to physicians in determining optimal individualized patient care. 1 , 2 Currently, no substitute for aspirin therapy exists; however, choice regarding which P2Y12 receptor inhibitor to use offers clinicians flexibility in treatment options. Therefore, a better understanding of the determinants of response to these medications is critical for developing individualized medicine strategies to reduce recurrent thrombotic events. Carboxylesterase 1 (CES1) is a well‐described hepatic esterase that accounts for approximately 80%–95% of hydrolytic activity in the liver. 3 In addition to endogenous compounds such as fatty acid esters, CES1 plays a critical role in the metabolism of multiple classes of therapeutic agents including angiotensin‐converting enzyme inhibitors, anticancer medications, antiplatelet agents, neuraminidase inhibitors, opioids, and more. 4 Furthermore, a growing evidence base shows that genetic variation in CES1 (e.g., rs2244613, rs115629050) significantly impacts drug metabolism and therapeutic response. 5 , 6 , 7 Furthermore, CES1 is the primary enzyme responsible for degradation of the clopidogrel prodrug, its intermediate metabolite 2‐oxo‐clopidogrel, and the active metabolite into biologically inactive carboxylic acid derivatives. Indeed, up to 85% of ingested clopidogrel is metabolized by this enzyme, thus reducing availability of the bioactive thiol metabolite (SR26334). 8 , 9 We previously observed that a loss‐of‐function single nucleotide polymorphism (SNP) in exon 4 of CES1 (rs71647871), which results in a glycine (G)‐to‐glutamic acid (E) amino acid change at position 143 (G143E) in the catalytic site, leads to significantly increased formation of the clopidogrel active metabolite and a substantial reduction in on‐clopidogrel ADP‐stimulated platelet aggregation. 10 These data are consistent with results generated in over 3300 patients who participated in the International Clopidogrel Pharmacogenomics Consortium (ICPC) where it was shown that, apart from the well‐described CYP2C19*2 variant, CES1 G143E was the strongest genetic determinant of clopidogrel response. 11 Like clopidogrel, prasugrel pharmacokinetics is dependent on carboxylesterases, although currently available data suggests that CES1 has a minor role in its metabolism compared to CES2. 12 In contrast, however, CES1 is not known to impact ticagrelor response as this agent is biologically active in its native state and not subject to CES1‐mediated degradation. Therefore, we hypothesized that treatment of individuals who carry response‐modifying alleles in CES1 such as G143E would have significantly altered genotype‐specific differences in platelet aggregation when administered clopidogrel while no such differences would be observed when given ticagrelor. If correct, more consistent platelet inhibition could be achieved in these individuals with ticagrelor resulting in more predictable platelet function and presumably better clinical outcomes. In the current investigation, we test this hypothesis by conducting a prospective, randomized crossover study of clopidogrel (75 mg daily for 7 days) and ticagrelor (90 mg twice daily for 7 days) in healthy participants stratified by CES1 G143E genotype. Extensive phenotyping including ex vivo platelet aggregometry was performed pre‐ and post‐drug administration in order to assess the interaction of genotype and drug choice on on‐treatment platelet function. METHODS Study population In this investigation, 51 apparently healthy Amish individuals were recruited based on CES1 G143E genotype at the Amish Research Clinic in Lancaster, PA between September 2017 and July 2022. Eligible participants included healthy men and women age 20–74 years of Amish descent. Participants were excluded if they were pregnant or less than 6 months had passed since delivery, breast feeding, history of a bleeding disorder or major spontaneous bleed (e.g., peptic ulcer, epistasis, or intracranial bleed), history of gastrointestinal bleed, severe hypertension (blood pressure > 160/95 mm Hg), coexisting malignancies, creatinine levels >2.0 mg/dL, ALT or AST >2 × the upper limit of normal, hematocrit <28%, TSH <0.4 or >5.5 mLU/L, polycythemia (platelet count > 500,000), thrombocytopenia (platelet count < 75,000), surgery within the last 6 months, or aspirin allergy. Participants were also excluded if they were currently taking aspirin, clopidogrel, prasugrel, ticagrelor, or other anti‐coagulant, such as warfarin, heparin, or GPIIb/IIIa antagonist and had conditions that might place them at increased risk from withdrawal of these medications 7 days prior to protocol initiation, including history of unstable angina, MI, angioplasty (including stent placement), coronary artery bypass surgery, atrial fibrillation, stroke or transient ischemic attack, diabetes, deep vein thrombosis or other thrombosis, or if they were taking vitamins or other supplements and unwilling to discontinue their use at least 1 week prior to protocol initiation, and finally if they were taking medications that would affect outcome(s) to be measured and could not willingly and safely, in the opinion of the study physician, discontinue these medications for 1 week prior to protocol initiation. All study protocols were approved by the Institutional Review Board at the University of Maryland and adhered to the principles of the Declaration of Helsinki. Written informed consent was obtained from each participant. Individuals were compensated for their participation. This trial has been registered at clinicaltrials.gov and assigned the identifier NCT03161678. Intervention One week prior to the start of the protocol, participants discontinued the use of all prescription medications, supplements, and vitamins. If they were taking antiplatelet agents such as aspirin or P2Y12 inhibitor, these medications were discontinued at least 2 weeks prior to the initiation of the study. Prior to the first clinic visit, participants were randomized to undergo either the clopidogrel (75 mg/day for 7 days) or ticagrelor (180 mg/day for 7 days) intervention first. Clinic staff, blinded to CES1 genotype, were notified of the order of drug intervention through the delivery of a sealed, opaque envelope from an independent research team member who was also blinded to participant genotype. Assignment envelopes were shuffled prior to study initiation and delivered sequentially. Participants who first received clopidogrel later underwent the ticagrelor intervention after a washout period of at least 2 weeks and vice versa. A schematic outlining the overall study design is shown in Figure S1. At the first clinic visit, physical examinations, information regarding medical and family histories, anthropometric measures, blood samples, and other phenotype data were collected following an overnight fast. Baseline platelet aggregation was recorded as described below. Participants were then observed while taking their first dose of clopidogrel or ticagrelor (based on randomization procedure), discharged to home with additional tablets, and instructed to take the appropriate dose (75 mg/daily for clopidogrel or 180 mg/day for ticagrelor) at the same time each day. After 7 days, they returned for the 2nd clinic visit, were observed taking their last dose, and fasting blood samples were drawn 2 h post‐dose for measurement of agonist‐stimulated platelet aggregation and sample banking. Drug adherence was assessed by participant logbook and pill count. Each participant then underwent at least a 14‐day washout period to restore basal platelet function, consistent with the average lifespan of a platelet (7–10 days). 13 Following the washout period, individuals returned for their 3rd clinic visit in the fasting state. Blood samples were collected for repeat measures of baseline platelet function and sample banking. Participants were then observed taking their first dose of clopidogrel or ticagrelor (whichever one was not administered during the first intervention period), discharged to home with additional tablets, and instructed to take the appropriate dose at the same time each day. After 7 days, they returned to the clinic for their final study visit, were observed taking their last dose, and fasting blood samples were drawn 2 h post‐dose for measurement of platelet aggregation and sample banking. Drug adherence was again assessed as described above. Platelet aggregation testing Platelet‐rich plasma (PRP) was isolated from fasting blood samples drawn into 3.2% citrate‐anticoagulated tubes (Becton‐Dickinson, Franklin Lakes, NJ) and platelet counts were adjusted to 200,000 platelets/μL using platelet‐poor plasma. Platelet function was assessed by optical aggregometry using a PAP8E Aggregometer (Bio/Data Corporation, Horsham, Pennsylvania) according to the manufacturer's instructions after stimulation with ADP (2, 5, and 20 μM), collagen (1, 2, and 5 μg/mL), epinephrine (10 μM), and arachidonic acid (1.6 mM) and was expressed as the maximal percentage change in light transmittance using platelet‐poor plasma as a referent as described previously. 14 Genotyping Given the low minor allele frequency (MAF) of CES1 G143E (~0.01–0.03 depending on race/ethnicity), we leveraged pre‐existing whole‐exome sequencing data generated in over 7000 Amish individuals to facilitate an efficient genotype‐directed recruitment strategy. Details regarding the methods used for exome sequencing have been described previously. 15 Repeat genotyping to confirm CES1 G143E genotype in all participants was performed using a TaqMan SNP genotyping assay (Applied Biosystems, Foster City, California) according to the manufacturer's instructions. The mean genotype concordance rate for this polymorphism between exome sequencing and Taqman genotyping was 100% and the Taqman genotype call rate was 98.3%. Statistical analysis Summary statistics, distributions, and frequencies were calculated using SAS v9.4 (SAS Institute Inc., Cary, North Carolina). Comparison of agonist‐specific platelet aggregation between baseline clinic visits (i.e., prior to administration of clopidogrel and ticagrelor) was performed in SAS v9.4 using a paired sample t‐test. The impact of CES1 G143E on agonist‐stimulated platelet aggregation pre‐ and post‐clopidogrel or ticagrelor administration was calculated using a variance component method under an additive model that simultaneously adjusted for age, sex, body mass index (BMI), and relatedness among study participants. Relatedness among participants was accounted for by including a polygenic component as a random effect as previously described. 16 To account for the effect of baseline platelet function on post‐drug maximal platelet aggregation (MPA) measurements, Inhibition of Platelet Aggregation (IPA) was calculated and similar analyses were performed. Specifically, IPA was defined as [(MPAbaseline – MPAday 7)/MPAbaseline] × 100 where MPAbaseline represents maximal platelet aggregation values obtained prior to drug administration and MPAday 7 represents the same value obtained after 7 days of either clopidogrel or ticagrelor treatment. IPA analyses were performed using the variance component method described above and included identical covariates. The prespecified primary outcome of this investigation was ADP‐stimulated IPA in response to clopidogrel and ticagrelor on day 7 by CES1 G143E genotype. Secondary outcomes included genotype‐specific IPA analyses in response to other platelet agonists (i.e., collagen, epinephrine, and arachidonic acid) as well as evaluation of maximal agonist‐stimulated platelet aggregation pre‐ and post‐drug treatment. Based on our prior investigation, 10 our planned sample size of 30 CES1 143GG homozygotes and 30 143GE heterozygotes was chosen to provide 90% power to detect differences in IPA between groups, assuming α = 0.05. Of note, no CES1 143EE homozygote has been identified to date in the Lancaster Amish or in other large genomic databases such as the UK Biobank, NHLBI Exome Sequencing Project, and Trans‐Omics for Precision Medicine (TOPMed) Program. All statistical analyses were two‐sided. RESULTS After screening for eligibility, 63 participants (41 143GG and 22 143GE) began the intervention, 12 of whom (6 143GG and 6 143GE) withdrew during the course of the protocol. Data from 1 individual with the CES1 143GG genotype was removed from analysis due to platelet aggregometer failure. The final sample size for analysis was 50 individuals including 34 143GG homozygotes and 16 143GE heterozygotes (Figure 1). Enrollment into the trial ended when no more prospective participants carrying the CES1 143E‐allele could be identified. No harms, serious adverse events, or unintended effects were observed regardless of intervention or participant genotype. CONSORT participant flow chart showing number of recruited individuals through each stage of the randomized crossover trial. Clinical characteristics of the analyzed cohort are shown in Table 1. Briefly, participants were generally healthy, middle‐aged, drug‐naïve, and had normal measures of blood pressure, lipids, glucose, and blood cell counts. When stratifying the sample by CES1 G143E genotype, individuals with the 143GE genotype were, on average, younger than those with the 143GG genotype (44.3 ± 14.6 vs. 49.1 ± 10.5 years, respectively, p = 0.001). No other significant differences in clinical characteristics were observed by genotype (Table 1). TABLE 1Clinical characteristics of CES1 G143E crossover study participants.Characteristic (units)Total143GG143GE p‐ValueNumber (n)503416Age ± SD (yr)47.6 ± 12.049.1 ± 10.544.3 ± 14.6 0.001 Female n (%)22 (43.1)17 (48.6)5 (31.3)0.18BMI ± SD (kg/m2)27.5 ± 4.427.4 ± 4.327.6 ± 4.80.25Systolic blood pressure ± SD (mmHg)117.9 ± 12.5118.0 ± 12.6117.6 ± 12.50.56Diastolic blood pressure ± SD (mmHg)72.7 ± 7.773.7 ± 7.970.6 ± 7.00.06Total cholesterol ± SD (mg/dL)209.8 ± 49.7211.1 ± 46.3206.9 ± 57.90.43LDL‐cholesterol ± SD (mg/dL)136.2 ± 42.0135.3 ± 37.9138.1 ± 51.20.15HDL‐cholesterol ± SD (mg/dL)57.8 ± 17.159.8 ± 18.153.6 ± 14.40.63Triglycerides ± SD (mg/dL)65.1 ± 31.266.4 ± 34.062.2 ± 24.80.95Glucose ± SD (mg/dL)89.3 ± 16.888.6 ± 13.590.8 ± 22.80.50Hematocrit ± SD (%)41.3 ± 3.441.0 ± 3.041.9 ± 4.10.87Red blood cell count ± SD (n × 1,000,000/μL)4.6 ± 0.44.6 ± 0.34.6 ± 0.40.51White blood cell count ± SD (n × 1000/μL)5.6 ± 1.45.6 ± 1.55.5 ± 1.10.82Platelet count ± SD (n × 100,000/μL)242.2 ± 54.0251.5 ± 59.1222.1 ± 34.40.63Mean platelet volume ± SD (fL)10.4 ± 0.810.4 ± 0.810.5 ± 0.80.83Abbreviations: BMI, body mass index; CES1, carboxylesterase 1; dL, deciliter; fL, femtoliter; HDL, high‐density lipoprotein; kg, kilogram; LDL, low‐density lipoprotein; m, meter; mg, milligram; mm, milimeter; SD, standard deviation; yr, year. Bold values denote statistical significance at p < 0.05. Clinical characteristics of CES1 G143E crossover study participants. Abbreviations: BMI, body mass index; CES1, carboxylesterase 1; dL, deciliter; fL, femtoliter; HDL, high‐density lipoprotein; kg, kilogram; LDL, low‐density lipoprotein; m, meter; mg, milligram; mm, milimeter; SD, standard deviation; yr, year. Bold values denote statistical significance at p < 0.05. Prior to performing CES1 genotypic‐specific analyses, we compared baseline platelet aggregation in response to multiple agonists (i.e., ADP, collagen, epinephrine, and arachidonic acid) at clinic visit #1 (protocol initiation) and clinic visit #3 (post‐washout visit). These analyses were performed to confirm that platelet function returned to the basal state prior to participants beginning their 2nd drug intervention. Indeed, we observed no significant difference in baseline platelet aggregation between clinic visits #1 and #3 for any agonist or concentration tested (2 μmol/L ADP 31.4 vs. 31.8, p = 0.95; 5 μmol/L ADP 55.5 vs. 56.4, p = 0.80; 20 μmol/L ADP 69.1 vs. 69.2, p = 0.93; 1 μg/mL collagen 62.2 vs. 62.0, p = 0.96; 2 μg/mL collagen 66.9 vs. 67.7, p = 0.74; 5 μg/mL collagen 72.5 vs. 73.2, p = 0.58; 10 μM epinephrine 66.0 vs. 65.0, p = 0.79; 1.6 mM arachidonic acid 64.0 vs. 64.0, p = 0.99). In addition, visual inspection of histograms of baseline platelet aggregation at protocol initiation and following the washout period showed a high degree of consistency (Figure S2). To evaluate whether CES1 G143E was a determinant of baseline platelet aggregation, we performed association analyses of agonist‐stimulated maximal platelet aggregation by CES1 G143E genotype prior to clopidogrel or ticagrelor administration. Based on these analyses, we observed no evidence of association between CES1 G143E and maximal platelet aggregation prior to clopidogrel or ticagrelor administration for any of the agonists and concentrations that were tested (Table 2). Following clopidogrel treatment, however, we observed strong association between this variant and ADP‐stimulated platelet aggregation whereby individuals who carried the CES1 143E‐allele had significantly lower MPA compared to those who were homozygous for the CES1 143G‐allele (Table 2); an effect that was more pronounced with increasing concentrations of ADP (2 μmol/L ADP, 5.6 vs. 14.3%, p = 0.04; 5 μmol/L ADP, 14.5 vs. 25.3%, p = 1.7 × 10−3; 20 μmol/L ADP, 20.6 vs. 35.4%, p = 2.4 × 10−4). A nominal association was also observed between CES1 G143E and on‐clopidogrel platelet aggregation in response to 5 μg/mL collagen (55.0 vs. 63.3%, p = 0.02). In contrast, little evidence of association was observed between CES1 G143E and platelet aggregation regardless of the agonist used following ticagrelor exposure (Table 2). TABLE 2Association between CES1 G143E and maximal platelet aggregation (MPA).Maximum aggregation phenotypeBaseline #1 (prior to clopidogrel)Baseline #2 (prior to ticagrelor)143GG143GE p‐Value143GG143GE p‐ValueADP (2 μmol/L)32.8 ± 4.328.5 ± 5.80.8733.8 ± 4.827.5 ± 5.20.18ADP (5 μmol/L)54.0 ± 2.858.7 ± 4.30.3856.6 ± 3.355.9 ± 4.10.97ADP (20 μmol/L)68.0 ± 1.374.1 ± 2.10.1069.2 ± 1.869.4 ± 1.90.93Collagen (1 μg/mL)59.9 ± 2.766.9 ± 4.00.0963.3 ± 2.959.3 ± 3.90.57Collagen (2 μg/mL)66.2 ± 2.068.5 ± 3.20.5868.0 ± 2.367.1 ± 2.10.83Collagen (5 μg/mL)71.6 ± 0.974.4 ± 1.60.1773.7 ± 1.572.2 ± 1.20.48Epinephrine (10 μM)66.5 ± 2.364.9 ± 5.50.9366.1 ± 3.562.4 ± 5.30.82Arachidonic acid (1.6 mM)62.3 ± 2.267.6 ± 2.50.2363.0 ± 2.766.2 ± 2.00.44Maximum aggregation phenotypeOn‐clopidogrelOn‐ticagrelor143GG143GE p‐Value143GG143GE p‐ValueADP (2 μmol/L)14.3 ± 2.35.6 ± 0.9 0.04 7.9 ± 1.34.9 ± 0.9 0.02 ADP (5 μmol/L)25.3 ± 2.414.5 ± 1.4 1.7 × 10 −3 15.2 ± 1.712.3 ± 1.30.06ADP (20 μmol/L)35.4 ± 2.820.6 ± 2.4 2.4 × 10 −4 23.0 ± 1.621.6 ± 1.10.10Collagen (1 μg/mL)35.8 ± 3.027.6 ± 3.90.3225.2 ± 2.823.2 ± 3.70.85Collagen (2 μg/mL)47.6 ± 2.835.7 ± 4.00.0635.7 ± 3.135.5 ± 3.40.73Collagen (5 μg/mL)63.3 ± 1.655.0 ± 2.7 0.02 56.3 ± 2.456.2 ± 2.30.76Epinephrine (10 μM)52.5 ± 3.735.7 ± 6.00.0639.4 ± 4.132.7 ± 6.40.79Arachidonic acid (1.6 mM)45.3 ± 3.837.7 ± 4.90.1940.2 ± 3.631.8 ± 4.80.34 Note: Data are expressed as mean maximal platelet aggregation values ± standard deviation.Abbreviations: ADP, adenosine diphosphate; CES1, carboxylesterase 1. Bold values denote statistical significance at p < 0.05. Association between CES1 G143E and maximal platelet aggregation (MPA). Note: Data are expressed as mean maximal platelet aggregation values ± standard deviation. Abbreviations: ADP, adenosine diphosphate; CES1, carboxylesterase 1. Bold values denote statistical significance at p < 0.05. In order to account for variability in baseline platelet aggregation among participants and to more accurately assess the potential impact of CES1 G143E on clopidogrel and ticagrelor response, we utilized the Inhibition of Platelet Aggregation (IPA) variable, which represents a percentage change of post‐drug platelet aggregation compared to baseline. When performing these analyses, we observed a strong effect of CES1 G143E on platelet aggregation in response to multiple agonists post‐clopidogrel exposure; however, no association was observed between this variant and agonist‐induced platelet aggregation following ticagrelor exposure (Table 3). Indeed, clopidogrel‐treated participants who carried the 143E‐allele had much stronger inhibition of platelet aggregation compared to non‐carriers for all concentrations of ADP tested (2 μmol/L ADP, 73.1 vs. 53.0%, p = 7.7 × 10−3; 5 μmol/L ADP, 74.7 vs. 53.3%, p = 4.0 × 10−5; 20 μmol/L ADP, 71.6 vs. 48.0%, p = 3.8 × 10−5). Furthermore, clopidogrel‐treated CES1 143E‐allele carriers also displayed higher inhibition of platelet aggregation in response to arachidonic acid (45.8 vs. 25.8%, p = 0.04), epinephrine (44.4 vs. 18.8%, p = 0.03), and collagen (5 μg/mL, 25.8 vs. 11.4%, p = 3.7 × 10−3). While statistical thresholds of association were not met for analyses between this variant and IPA in response to 1 or 2 μg/mL collagen, point estimates suggest a similar direction of effect (57.9 vs. 40.7%, p = 0.08 and 48.6 vs. 19.7%, p = 0.12, respectively). No evidence of association was observed between CES1 G143E and IPA for any platelet agonist or concentration when participants were exposed to ticagrelor (Table 3). TABLE 3Association between CES1 G143E and inhibition of platelet aggregation (IPA).IPA phenotype (%)On‐clopidogrelOn‐ticagrelor143GG143GE p‐Value143GG143GE p‐ValueADP (2 μmol/L)53.0 ± 4.673.1 ± 4.6 7.7 × 10 −3 74.3 ± 2.875.0 ± 5.10.93ADP (5 μmol/L)53.3 ± 3.374.7 ± 2.0 4.0 × 10 −5 73.6 ± 2.275.5 ± 3.70.95ADP (20 μmol/L)48.0 ± 3.971.6 ± 3.0 3.8 × 10 −5 67.2 ± 1.968.7 ± 1.60.23Collagen (1 μg/mL)40.7 ± 4.657.9 ± 5.10.0852.8 ± 7.859.5 ± 6.30.75Collagen (2 μg/mL)19.7 ± 10.548.6 ± 4.80.1247.5 ± 4.046.6 ± 5.10.59Collagen (5 μg/mL)11.4 ± 2.225.8 ± 3.6 3.7 × 10 −3 24.1 ± 2.622.0 ± 3.30.37Epinephrine (10 μM)18.8 ± 5.844.4 ± 7.2 0.03 33.1 ± 9.344.2 ± 9.00.63Arachidonic acid (1.6 mM)25.8 ± 5.845.8 ± 6.4 0.04 25.4 ± 13.752.1 ± 7.20.27 Note: Data are expressed as mean inhibition of platelet aggregation percentages ± standard deviation and represents a percentage change of post‐drug platelet aggregation compared to baseline.Abbreviations: ADP, adenosine diphosphate; CES1, carboxylesterase 1. Bold values denote statistical significance at p < 0.05. Association between CES1 G143E and inhibition of platelet aggregation (IPA). Note: Data are expressed as mean inhibition of platelet aggregation percentages ± standard deviation and represents a percentage change of post‐drug platelet aggregation compared to baseline. Abbreviations: ADP, adenosine diphosphate; CES1, carboxylesterase 1. Bold values denote statistical significance at p < 0.05. DISCUSSION Carboxylesterase 1 is an abundant hepatic serine esterase involved in the hydrolysis of multiple endogenous compounds (e.g., lipids, cholesterol esters) and xenobiotic agents (environmental pollutants, insecticides, etc.). 3 It is also responsible for the metabolism of numerous therapeutic agents spanning different drug classes including clopidogrel, methylphenidate, oseltamivir, ACE inhibitors, and several anti‐cancer drugs. 17 , 18 , 19 , 20 , 21 , 22 Therefore, understanding the determinants of variation in CES1 activity has potentially important clinical implications. Indeed, prior investigations have shown that CES1 expression and/or catalytic function is influenced by many clinical and anthropometric variables include age, sex, disease state, nutrition, as well as several drug interactions. 3 , 23 , 24 , 25 , 26 , 27 Furthermore, it is becoming increasingly apparent that genetic variation in CES1 has an important role in regulating enzyme expression and activity. We and others have previously shown that a loss‐of‐function single nucleotide polymorphism (G143E, rs71647871) in CES1 strongly impacts clopidogrel pharmacokinetics, resulting in increased formation of the active metabolite and enhanced inhibition of on‐clopidogrel platelet reactivity. 10 , 11 , 28 , 29 , 30 , 31 Mechanism‐based pharmacokinetic/pharmacodynamic modeling data revealed similar findings showing that CES1 G143E led to an 83% reduction in CES1 activity and severely influenced clopidogrel active metabolite formation and platelet aggregation. 32 Consistent with these investigations, in vitro kinetic studies of this variant with other drugs (i.e., methylphenidate, oseltamivir) show that the 143E minor allele results in an approximate 85%–100% reduction in CES1 activity. 33 Based on currently available x‐ray crystallography data, it is presumed that CES1 G143E leads to a near‐complete loss of enzyme function through disruption of an oxyanion hole located adjacent to active site catalytic triad and thus destabilizes substrate‐enzyme intermediates. 34 In the current investigation, we, for the first time, prospectively evaluate the effect of CES1 G143E on platelet aggregation at baseline and in clopidogrel‐treated individuals. Furthermore, we assess whether treatment with an alternative P2Y12 receptor inhibitor that is not dependent on CES1‐mediated metabolism (i.e., ticagrelor) results in more consistent inhibition of platelet function. Our evidence suggests that CES1 G143E does not meaningfully alter baseline platelet function, consistent with the hypothesis that this variant exerts its effect on platelet aggregation by affecting antiplatelet drug metabolism. We do show, however, that this variant does substantially alter agonist‐induced platelet aggregation in the presence of clopidogrel. While genotype‐dependent effects were most apparent when ADP was used as an agonist, which isn't surprising given that clopidogrel irreversibly inhibits the ADP receptor (P2Y12), it is interesting to note that we also observed significant associations between this variant and platelet aggregation in response to other agonists (i.e., collagen, epinephrine, and arachidonic acid). We speculate that the latter associations are indirectly caused by genotype‐specific differences in inhibition of ADP‐related platelet aggregation pathways, which, in turn could influence platelet aggregation stimulated by other platelet agonists via secondary platelet signaling; however, we cannot rule out the possibility that CES1 and variants in this gene more directly impact other agonist‐specific platelet aggregation pathways. We observed little to no evidence of association between CES1 G143E and platelet aggregation in ticagrelor‐treated individuals. Unlike clopidogrel, which is subject to CES1‐mediated degradation in the liver, ticagrelor is bioactive in its ingested form as well as after the formation of an active metabolite following O‐deethylation by CYP3A4. 35 In our initial analyses of raw maximal platelet aggregation values, nominal evidence of association was shown between CES1 G143E and platelet aggregation in response to 2 μmol/L ADP (but not 5 or 20 μmol/L ADP) in ticagrelor‐treated participants (Table 2). However, when performing analyses using the IPA variable, no association was found between this variant and on‐ticagrelor IPA regardless of agonist or concentration used, suggesting that nominally significant associations observed in analyses of maximal platelet aggregation data were caused by differences in baseline platelet aggregation and not due to a pharmacogenetic relationship between CES1 and ticagrelor. Again, this is consistent with the fact that, based on currently available information, CES1 does contribute to ticagrelor metabolism. While CES1 is the primary enzyme responsible for clopidogrel degradation, several other enzymes, most notably CYP2C19, contribute to its metabolism. In fact, multiple prior investigations have shown that genetic variation in CYP2C19, particularly the loss‐of‐function CYP2C19*2 variant, results in significantly decreased clopidogrel active metabolite concentration, altered on‐treatment platelet function, and adverse patient outcomes. 9 Therefore, it is important to note that such genetic variation could influence the results of the current investigation. While de novo genotyping of CYP2C19*2 was not performed as part of this study, we did have prior CYP2C19*2 genotype data in 43 of our participants (30 CES1 GG homozygotes and 13 CES1 GE heterozygotes) using exome‐wide sequencing. Data generated from CYP2C19*2‐adjusted analyses in this subset of participants were consistent with our primary findings (data not shown), suggesting that CES1 G143E is an important genetic determinant of clopidogrel response independently of CYP2C19 genotype. These results may have important implications for medical management and personalized treatment of patients who have a clinical indication for antiplatelet agents. Today, physicians have flexibility in prescribing options for dual antiplatelet therapy regimens. More specifically, aspirin can be used in combination with clopidogrel, ticagrelor, or prasugrel. While each of these medications performs generally well in inhibiting platelet function in most patients, drug‐specific variability does exist in some subsets of patients. This can lead to an increased risk of experiencing a recurrent major adverse cardiovascular event in cases where platelet inhibition is inadequate or, on the other hand, can lead to pathological bleeding if the effect of the drug is enhanced. As mentioned above, bioactivation of clopidogrel is also dependent on the cytochrome P450 system and patients who carry the CYP2C19*2 allele are poor metabolizers of clopidogrel, experience increased platelet reactivity, and exhibit higher cardiovascular event risk. 9 While strong genetic modifiers of ticagrelor and prasugrel have yet to be firmly established, comprehensive genome‐wide investigations in large samples are still limited. Given this information, it is critical to identify drug‐specific sources of variation, both genetic and non‐genetic, in order to better individualize antiplatelet treatment to manage risk of recurrent cardiovascular events while simultaneously reducing adverse drug events such as bleeding. In the current investigation, we observed that clopidogrel‐treated individuals who carried the CES1 143E‐allele had greater inhibition of platelet aggregation compared to 143G‐allele homozygotes; yet, no difference was noted by genotype when ticagrelor was administered. Prior studies have postulated that reduced platelet activity in patients who carry the 143E‐allele may result in increased risk of adverse bleeding events. 11 However, given the relatively low minor allele frequency of this variant (approximately 0.0001–0.03 depending on race/ethnicity 36 , 37 ), well‐powered investigations attempting to answer this hypothesis have yet to be performed. Though we could not assess risk of bleeding in the current study given the participant population, it is interesting to note that when we compared both maximal platelet aggregation and IPA in CES1 143E‐allele carriers treated with clopidogrel versus ticagrelor, we observed no statistical difference between these groups for any agonist or concentration tested (Table S1). While platelet aggregation is not the sole driver of bleeding complications, these results may suggest that risk of adverse bleeding events in CES1 143E‐allele carriers might not be appreciably different when treated with clopidogrel versus ticagrelor. However, additional studies more formally evaluating the effect of this variant on pathological bleeding are warranted, both in the context of clopidogrel and other P2Y12 receptor inhibitors, before such claims could be made. There are several limitations to this investigation. First, this trial was conducted in relatively healthy individuals. While this is advantageous in pharmacogenetic studies to minimize confounding variables that inhibit our ability to detect genetic associations and estimate effect sizes (e.g., polypharmacy, existing comorbidities, etc.), additional characterization in patients with more severe cardiovascular burden is required to better determine clinical impact and generalizability. In addition, all participants included in this study were of European descent (i.e., Amish). While there are currently no data to suggest that the effect of CES1 G143E may be different in other populations, the allele frequency of this variant does vary between groups and further evaluation in larger, more ancestrally diverse groups of people are warranted. Finally, another limitation was the relatively short intervention periods (7 days each for both clopidogrel and ticagrelor). In clinical settings, patients who are prescribed these medications are typically required to take them for at least 6–12 months depending on clinical indication; thus, examination of more persistent effects of CES1 G143E over longer on‐treatment periods may allow for additional insights. In conclusion, we prospectively show that genetic variation in CES1 significantly impacts on‐clopidogrel inhibition of platelet aggregation while little to no effect is observed in the same individuals when administered ticagrelor. As the field of antiplatelet interventional cardiology continues to expand with the development of new medications and treatment paradigms that consider more tailored approaches based on specific patient characteristics, identification of genetic and non‐genetic determinants of drug response will be critical to prescribe the optimal medication to each patient. In doing so, clinicians will be better equipped to reduce rates of recurrent cardiovascular events while simultaneously lessening the risk of adverse drug events such as pathological bleeding. AUTHOR CONTRIBUTIONS J.P.L. wrote the manuscript, and all co‐authors had the opportunity to review and edit all parts of it. J.P.L., A.R.S., and B.D.M. contributed to the design of the research. J.P.L., E.A.S., H.B.W., and K.T. performed the research. J.P.L., K.A.R., M.D., and J.A.P. analyzed the data. J.R.O. contributed new analytical tools. FUNDING INFORMATION Research reported in this publication was supported by National Heart, Lung, and Blood Institute of the National Institutes of Health under award number R01HL137922 awarded to JPL. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. CONFLICT OF INTEREST STATEMENT ARS is an employee of Regeneron Pharmaceuticals, Inc. and receives salary, stock, and stock options as compensation. ARS is also a part‐time faculty member at the University of Maryland School of Medicine, in addition to his employment at Regeneron. All other authors declared no competing interests for this work. Supporting information Appendix S1 ACKNOWLEDGMENTS We gratefully acknowledge our Amish liaisons, field workers, and the Amish community, without whom these studies would not have been possible. Lewis JP, Ryan KA, Streeten EA, et al. Randomized evaluation of the loss‐of‐function carboxylesterase 1 (CES1) G143E variant on clopidogrel and ticagrelor pharmacodynamics. Clin Transl Sci. 2024;17:e70079. doi: 10.1111/cts.70079 REFERENCES Associated Data Supplementary Materials Appendix S1