Introduction: A drug-specific chromogenic assay is not immediately available, so it hampers the treatment of patients who present in a clinical emergency. In this pilot study, we aimed to create a formula to predict a plasma edoxaban level based on the international normalized ratio (INR) and heparin-calibrated anti-Xa activity and derive a novel workflow for routine laboratory diagnosis. Method: Forty-two patients prescribed edoxaban were recruited and randomized to a testing or validation cohort. Plasma levels from the testing cohort were used to create a prediction formula that was then validated in a validation cohort and real-world clinical requests. Results: The INR-derived formula had high sensitivity (95.8–100%) to predict the plasma edoxaban level > 50 ng/mL and >100 ng/mL but with low specificity. However, the specificity of predicting the plasma edoxaban level of ≥100 ng/mL was 100% by using an INR ≥ 1.5 as cut-off. Heparin-calibrated anti-Xa-derived formula had a high sensitivity (90.9–100%) and specificity (93.8–100%) in real clinical situations. A two-tier approach of combining INR-derived and heparin-calibrated anti-Xa-derived formulae can overcome the low specificity and utilize the advantages of wide availability and a short turnaround time of the INR-derived formula. Conclusions: Both INR-derived and heparin-calibrated anti-Xa-derived formulae can be applied to calculate the plasma edoxaban level. A two-tier workflow of combining these two formulae greatly helps streamline the treatment of patients prescribed edoxaban who present in a clinical emergency. Adoption of this framework is feasible for routine diagnostic laboratories.
Read full abstract