Kevin A Friede1,2, Rachel A Myers1,2, Jordan Gales3, Ilya Zhbannikov1,2, Thomas L Ortel4, Svati H Shah2,5, William E Kraus2,5, Geoffrey S Ginsburg6, Deepak Voora1,2. 1. Center for Applied Genomics & Precision Medicine, Duke University, Durham, NC, USA. 2. Division of Cardiology, Duke University, Durham, NC, USA. 3. Department of Cardiology, Texas Heart Institute, Houston, TX, USA. 4. Division of Hematology, Duke University, Durham, NC, USA. 5. Duke Molecular Physiology Institute, Duke University, Durham, NC, USA. 6. All of Us Research Program, National Institutes of Health, Bethesda, MD, USA.
Abstract
AIMS: Gene expression biosignatures may hold promise to individualize antiplatelet therapy in conjunction with current guidelines and risk scores. The Aspirin Response Signature (ARS) score is comprised of a weighted sum of correlated, prothrombotic gene transcripts measured in whole blood. In prior work where volunteers were exposed to aspirin 325 mg daily, higher ARS score was associated with lower platelet function; separately, in a clinical cohort of patients, higher ARS scores were associated with increased risk of adverse cardiovascular events. To better understand this apparent paradox, we measured ARS gene expression and score in volunteers to determine aspirin dose-response and ticagrelor relationships with ARS score and separately in patients to assess whether ARS is associated with incident bleeding. METHODS AND RESULTS: Blood samples were collected from volunteers (N = 188) who were exposed to 4 weeks of daily aspirin 81 mg, daily aspirin 325 mg, and/or twice-daily ticagrelor 90 mg. ARS scores were calculated from whole blood RNA qPCR, and platelet function and protein expression were assessed in platelet-rich plasma. In mixed linear regression models, aspirin 81 mg exposure was not associated with changes in ARS gene expression or score. Aspirin 325 mg exposure resulted in a 6.0% increase in ARS gene expression (p = 7.5 × 10-9 vs. baseline, p = 2.1 × 10-4 vs. aspirin 81 mg) and an increase in expression of platelet proteins corresponding to ARS genes. Ticagrelor exposure resulted in a 30.7% increase in ARS gene expression (p < 1 × 10-10 vs. baseline and each aspirin dose) and ARS score (p = 7.0 × 10-7 vs. baseline, p = 3.6 × 10-6 and 5.59 × 10-4 vs. aspirin 81 mg and 325 mg, respectively). Increases in ARS gene expression or score were associated with the magnitude of platelet inhibition across agents. To assess the association between ARS scores and incident bleeding, ARS scores were calculated in patients undergoing cardiac catheterization (N = 1421), of whom 25.4% experienced bleeding events over a median 6.2 years of follow-up. In a Cox model adjusting for demographics and baseline antithrombotic medication use, patients with ARS scores above the median had a higher risk of incident bleeding (HR 1.26 [95% CI 1.01-1.56], p = 0.038). CONCLUSIONS: The ARS is an Antiplatelet Response Signature which increases in response to greater platelet inhibition due to antiplatelet therapy and may represent a homeostatic mechanism to prevent bleeding. ARS scores could inform future strategies to prevent bleeding while maintaining antiplatelet therapy's benefit of ischemic cardiovascular event protection.
AIMS: Gene expression biosignatures may hold promise to individualize antiplatelet therapy in conjunction with current guidelines and risk scores. The Aspirin Response Signature (ARS) score is comprised of a weighted sum of correlated, prothrombotic gene transcripts measured in whole blood. In prior work where volunteers were exposed to aspirin 325 mg daily, higher ARS score was associated with lower platelet function; separately, in a clinical cohort of patients, higher ARS scores were associated with increased risk of adverse cardiovascular events. To better understand this apparent paradox, we measured ARS gene expression and score in volunteers to determine aspirin dose-response and ticagrelor relationships with ARS score and separately in patients to assess whether ARS is associated with incident bleeding. METHODS AND RESULTS: Blood samples were collected from volunteers (N = 188) who were exposed to 4 weeks of daily aspirin 81 mg, daily aspirin 325 mg, and/or twice-daily ticagrelor 90 mg. ARS scores were calculated from whole blood RNA qPCR, and platelet function and protein expression were assessed in platelet-rich plasma. In mixed linear regression models, aspirin 81 mg exposure was not associated with changes in ARS gene expression or score. Aspirin 325 mg exposure resulted in a 6.0% increase in ARS gene expression (p = 7.5 × 10-9 vs. baseline, p = 2.1 × 10-4 vs. aspirin 81 mg) and an increase in expression of platelet proteins corresponding to ARS genes. Ticagrelor exposure resulted in a 30.7% increase in ARS gene expression (p < 1 × 10-10 vs. baseline and each aspirin dose) and ARS score (p = 7.0 × 10-7 vs. baseline, p = 3.6 × 10-6 and 5.59 × 10-4 vs. aspirin 81 mg and 325 mg, respectively). Increases in ARS gene expression or score were associated with the magnitude of platelet inhibition across agents. To assess the association between ARS scores and incident bleeding, ARS scores were calculated in patients undergoing cardiac catheterization (N = 1421), of whom 25.4% experienced bleeding events over a median 6.2 years of follow-up. In a Cox model adjusting for demographics and baseline antithrombotic medication use, patients with ARS scores above the median had a higher risk of incident bleeding (HR 1.26 [95% CI 1.01-1.56], p = 0.038). CONCLUSIONS: The ARS is an Antiplatelet Response Signature which increases in response to greater platelet inhibition due to antiplatelet therapy and may represent a homeostatic mechanism to prevent bleeding. ARS scores could inform future strategies to prevent bleeding while maintaining antiplatelet therapy's benefit of ischemic cardiovascular event protection.