Validation of a Modified Anti-FXa Assay on STA-Compact Analyzer for Measuring FXa Inhibitor Levels in the Presence of Andexanet Alfa

Introduction: The anticoagulants rivaroxaban and apixaban inhibit Factor Xa (FXa) activity and are effective therapeutic agents for the prevention and treatment of thromboembolism. Andexanet alfa (andexanet) is the only specific reversal agent approved for anticoagulation reversal in patients presenting with life-threatening bleeding treated with rivaroxaban or apixaban.

Anti-FXa assays are functional assays used for measuring direct and indirect FXa inhibitor levels in plasma. Current commercially available anti-FXa assays are not suitable for accurately measuring anti-FXa activity in patient samples in the presence of andexanet due to the high sample dilutions (e.g.,1:44 with STA-Liquid Anti-Xa), which causes dissociation of the inhibitor from andexanet (due to reversible binding). This results in substantial underestimation of the reversal activity of andexanet, and in some cases with erroneously elevated anti-FXa levels in patient samples following andexanet treatment. Therefore, we modified the current anti-FXa assays to measure apixaban and rivaroxaban anti-FXa levels in the presence of andexanet.

Methods: The modified anti-FXa assays were performed on STA®-Compact analyzer using reagents from STA-Liquid Anti-Xa assay kits (Stago). Andexanet was provided as a 10 mg/mL frozen stock. Plasma samples with FXa inhibitor in the presence or absence of andexanet were prepared using Stago Calibrator 1 and pooled normal human plasma (Precision Biologic). Previous anti-FXa assays (calibrator range: 0-500 ng/mL) employed 1:4 dilution of test plasma in Owren-Koller buffer followed by addition of FXa substrate and bovine FXa with an overall 1:44 sample dilution. In the modified assays, calibration curves for the anti-FXa assay were generated using apixaban or rivaroxaban calibrators in the range of 0 to 80 ng/mL; test samples were analyzed neat with an overall 1:2.6 sample dilution. Rivaroxaban and apixaban concentrations in the test samples were interpolated from the respective assay specific calibration curves. Validation of the modified anti-FXa assays was performed by assessing a) linearity and precision of the calibration curve and controls, b) lower limit of quantitation (LLOQ), c) inter-assay precision and d) sample stability.

Results: The calibration curves for both apixaban and rivaroxaban assays demonstrated good correlation with a mean “r” value of 0.997 and 0.998, respectively (n=6). The percent recovery and precision of the modified anti-FXa assay are summarized in Table 1.

The LLOQ for apixaban and rivaroxaban using the modified assays were <18.8 and <20.6 ng/mL, respectively, based on the preliminary results. Both apixaban and rivaroxaban assays demonstrated potent reversal of anti-FXa activity in presence of andexanet. Samples containing 232 ng/mL apixaban (0.5 µM) treated with equimolar andexanet (0.5 µM) resulted in 77.18 ng/mL of quantifiable apixaban (~66.7% reversal). Similarly, samples containing 242 ng/mL rivaroxaban (0.5 µM) treated with equimolar andexanet (0.5 µM) resulted in 29.44 ng/mL quantifiable rivaroxaban (~87.8% reversal). The short-term stability study with the modified anti-FXa assay included storage of plasma samples at 2-8^oC up to 24 hours and under frozen conditions at -20^oC up to 2 weeks.

Conclusions: The modified anti-FXa assays intended for measuring FXa inhibitor levels in the presence of andexanet produced acceptable analytical performance characteristics. Application of the modified anti-FXa assays in plasma samples from andexanet-treated human subjects has yet to be studied.