Assessment of the compatibility of the real-world nonvalvular atrial fibrillation patients in Turkey with the study population of Phase 3 novel oral anticoagulant trials: an auxiliary study of NOAC-TR

Introduction: Although the indication of novel oral anticoagulant (NOAC) treatment in atrial fibrillation (AF) is comparatively similar, Phase 3 NOAC trials have variable inclusion criteria that differentiate each other and also from the real-world population. Aim: We aim to investigate the similarity between real-world nonvalvular AF patients and the population of Phase 3 NOAC trials in terms of eligibility. Methods: A total of 2802 patients using rivaroxaban, dabigatran, and apixaban were retrospectively evaluated. All the patients met the exclusion criteria of NOAC Phase 3 trials. These patient population were compared with the population of Phase 3 rivaroxaban (ROCKET-AF), dabigatran (RELY), apixaban (ARISTOTLE), and edoxaban (ENGAGE) trials in terms of inclusion criteria. Furthermore, the patients were stratified on the basis of CHA 2 -DS 2 -VASCc is enaogh score. Results: The proportion of population who met the eligible criteria for ARISTOTLE trial (91%) was different from that of RELY (78%), ROCKET-AF (50%), and ENGAGE (61%) trials (P < 0,001). For the population at intermediate risk (CHA 2 DS 2 -VASc score ≥1), the proportion which met the inclusion criteria for RE-LY trial (99%) was different from that of ARISTOTLE (91.2%), ROCKET-AF (50%), and ENGAGE trials (61%) (P < 0.001). For the population at high risk (CHA 2 DS 2 -VASc score ≥2), the proportion which met the inclusion criteria was as follows: 94% for ARISTOTLE, 83% for RELY, 65% for ENGAGE, and 53% for ROCKET-AF trials (P < 0.001). In this population, 38% of patients using rivaroxaban, 46% of patients using dabigatran, and 12% patients of using apixaban did not meet the inclusion criteria for the ROCKET-AF, RE-LY, and ARISTOTLE trials, respectively. Conclusion: Eligibility of the real-world population for NOAC trials is variable. A considerable number of real-world patients using NOAC do not meet the inclusion criteria of the corresponding drug.