Insight into pill-in-pocket oral anticoagulation therapy with continuous monitoring of atrial fibrillation using a patient-level computational model

Abstract Introduction Atrial fibrillation (AF) is associated with increased risk of stroke. Continuous oral anticoagulation therapy (OAC) significantly reduces the incidence of stroke while potentially promoting bleeding events. Pill-in-pocket OAC guided by cardiovascular implantable electronic devices has been proposed as an alternative OAC strategy, but its benefits for reducing major bleeding while maintaining effective stroke prevention remain uncertain. Purpose To explore the efficacy and safety of pill-in-pocket OAC guided by continuous rhythm monitoring using a patient-level computational model of AF. Methods The patient-level modeling framework included two parallel Markov-like models; one with 7 clinical states (sinus rhythm, symptomatic/asymptomatic AF each with/without stroke and death) and one with 2 states (non-bleeding and major bleeding). AF-related atrial remodeling was incorporated and transition probabilities in both models were influenced by age, sex, and/or atrial remodeling (Figure 1A). Model calibration/validation was performed by reproducing a series of epidemiological and clinical data. Continuous OAC was simulated after clinical AF diagnosis (symptomatic AF episode of ≥3 hours) in the control group and pill-in-pocket OAC was simulated based on protocols described in the ongoing REACT-AF trial. Results The model simulates individual AF episodes and clinical events during the entire lifetime of a virtual patient. Simulated stroke incidence, non-fatal major bleeding incidence in the general population, and major bleeding incidence in AF patients with or without OAC were consistent with the observed data (Figure 1B-E). Virtual RCTs were simulated based on real-world trial design (Figure 2A), and the Kaplan-Meier curves showed no significant difference in the incidence of stroke and all-cause mortality between continuous and pill-in-pocket OAC strategies (Figure 2B, P = 0.88). However, major bleeding risk was significantly lower in the group with pill-in-pocket OAC guided by continuous monitoring (Figure 2C, P = 0.011). Conclusion Our novel patient-level AF model can simulate clinical outcomes such as stroke and major bleeding in large cohorts of virtual patients. By simulating the study protocol of an ongoing clinical trial, the model predicted that on-demand OAC is non-inferior in stroke and death, and superior in major bleeding compared with the standard continuous OAC. The model is a useful tool for the preliminary efficacy and safety assessment of future randomized clinical trials.Figure 1 Figure 2