2864Growth differentiation factor 15 and bleeding risk with ticagrelor in patients with prior myocardial infarction: insights from PEGASUS-TIMI 54

Abstract

Background: Long‐term ticagrelor reduces thrombotic risk in patients with prior MI but increases bleeding. Growth Differentiation Factor 15 (GDF‐15) predicts bleeding risk in patients with atrial fibrillation and acute coronary syndromes. It is unknown whether GDF‐15 predicts bleeding risk in high risk patients with prior MI. Purpose: To test the hypothesis that GDF‐15 is associated with bleeding risk in stable patients with prior MI treated with potent antiplatelet therapy. Methods: GDF‐15 (Roche Diagnostics) was measured at baseline in 8,620 patients with prior MI randomized to ticagrelor 90 mg twice daily, ticagrelor 60 mg twice daily or placebo in PEGASUS‐TIMI 54. TIMI major bleeding was adjudicated by a blinded CEC. Reported hazard ratios were adjusted for age, sex, race, region, BMI, smoking, COPD, hypercholesterolemia, hypertension, diabetes mellitus, eGFR<60, prior stroke or TIA, PAD, atrial fibrillation, heart failure, angina, >1 prior MI, prior PCI, prior CABG, multivessel CAD, qualifying MI type, and time from MI. Results: The median GDF‐15 concentration was 1241 pg/mL (IQR 913 ‐ 1775). Higher GDF‐15 levels were associated with age, BMI, female sex, smoking, cardiovascular risk factors (e.g. hypertension, diabetes mellitus, renal dysfunction), PAD, more extensive CAD (multivessel disease, prior revascularization), and index NSTEMI vs STEMI. There was a strong graded relationship between baseline GDF‐15 levels and the risk of TIMI major bleeding at 3 years in patients randomized to either dose of ticagrelor but not placebo (Figure). After adjustment for baseline differences, TIMI major bleeding increased 49% per 1‐SD logtransformed GDF‐15 in patients randomized to ticagrelor 60 mg (Adj‐HR 1.49, 95% CI 1.05‐2.13, p=0.027) and 72% in those allocated to ticagrelor 90 mg (Adj‐ HR 1.72, 95% CI 1.28‐2.31, p=0.003) but was not associated with bleeding with placebo (Adj‐HR 1.06, 95% CI 0.61‐1.86, p=0.83). Similarly, analyzed by quartile of GDF‐15, the adjusted risk of TIMI major bleeding was nearly 4‐fold higher in highest vs. lowest quartile of GDF‐15 among patients randomized to ticagrelor (Figure). Conclusions: In stable patients with prior MI, baseline GDF‐15 is associated with the risk of TIMI major bleeding with ticagrelor. This biomarker may be useful to profile bleeding risk in patients being considered for long‐term treatment with ticagrelor. (Figure presented).