Abstract Background The dynamic changes and stability of blood biomarkers over time and after treatment are not well known. In this study, we describe changes in 12 centrally quantified known and novel cardiovascular biomarkers, prior to and 3 months after ablation for atrial fibrillation (AF). Purpose In patients enrolled in the AXAFA-AFNET5 trial, we 1) characterised changes in 12 biomarker levels pre and post-ablation, 2) ascertained if biomarker changes are consistent between males and females, and 3) identified biomarkers which predict recurrent AF post-ablation. Methods and results Of the 674 patients who were recruited and randomised, 633 received the study drug and underwent ablation. Peripheral blood samples were available for 488 patients at baseline and 434 at 3 months follow-up (median age [Q1, Q3] 64 [58, 70] years; 34% female). Between baseline (BL) and follow-up (FU), paired comparisons revealed that 3 biomarkers decreased, ANG2 (median [Q1, Q3] BL 2.185 [1.711, 3.115], FU 1.827 [1.457, 2.297] ng/mL, p<0.001), BMP10 (BL 2.056 [1.810, 2.380], FU 1.986 [1.757, 2.260] ng/mL, p<0.001), and NTproBNP (BL 2.219 [0.858, 5.731] per 100pg/mL, p<0.001), while 1 biomarker increased, FABP3 (BL 2.911 [2.425, 3.508], FU 2.911 [2.462, 3.521], p=0.005). The remaining 8 biomarkers remained unchanged. Significant differences in ANG2, BMP10, NTproBNP and FABP3 were driven by patients who remained arrhythmia free at follow-up whereas biomarker levels remained unchanged in 121 patients who experienced recurrent AF (39%; Figure). Change scores were mainly consistent between males and females, however, CRP decreased significantly more in females. Recurrent AF episodes were not different between males and females (p=0.319). Cox proportional hazards model assessed the relationship of individual biomarkers at baseline for predicting recurrent AF. Elevated ANG2 (hazard ratio, HR per ng/mL [95% confidence interval] 1.214 [1.113, 1.325]), BMP10 (HR per ng/mL 1.516 [1.039, 2.214]), and NTproBNP (HR per 100 pg/mL 1.050 [1.025, 1.076]) significantly predicted increased risk for recurrent AF, after adjustment for age, sex, body mass index, hypertension, diabetes, chronic obstructive pulmonary disease, stroke, heart failure, ablation type (PVI, PVI and other, other), ablation energy (radiofrequency, cryoablation, other), and treatment arm. Conclusion In this study, most cardiovascular biomarkers are unchanged after ablation for AF, however, ANG2, BMP10, and NTproBNP decreased at follow-up. These effects are driven by patients who remained arrhythmia free and could potentially reflect improvement in vascular (ANG2), endothelial (BMP10), and myocardial load (NTproBNP) parameters post-ablation. This outcome corresponds with the observation that elevated levels of these biomarkers at baseline predict recurrent AF at 3 months. Both males and females demonstrate similar changes in biomarker profiles and benefit equally from ablation for AF. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): DZHK (German Centre for Cardiovascular Research) and BMBF (German Ministry of Education and Research) to AFNET.Additional support from European Union [grant agreement No. 633196 (CATCH ME)]. Biomarker changes