"That's why I wish the inclusion criteria would have tightened up to something like within 10 hours of taking FXa inhibitor rather than 15 hours."
I agree that the higher the concentration of X or E, the comparatively better A will look versus KCentra. For perspective, here is the actual indication for A. "ANDEXXA is indicated for patients treated with rivaroxaban or apixaban, when reversal of anticoagulation is needed due to life-threatening or uncontrolled bleeding." Annexa-4 used an 18 hour from last dose cutoff. The criteria for determining high/low dose A uses an 8 hour inflection point. The healthy volunteer studies on A used 3 and 4 hour delays prior to A administration. In a perfect world I think the best way to administer A is to do a rapid diagnostic to determine the actual FXA concentration to determine the need for and the dose of A to be used. Serum concentration of X and E can vary widely from patient to patient and also relying on patient or caregiver memory to determine the time of the last dose doesn't strike me as the best clinical practice.
Preclinical work does establish that A more effectively and rapidly reverses 10X inhibition that KCentra. That paper's abstract was just released and its conclusion is below. TF=Tissue factor and TG=Thrombin generation
"Conclusions: Specific reversal agents are effective for restoration of TF-TG by either direct sequestration of the FXa inhibitor and restoration of FXa activity (in the case of AnXa with FXa inhibitors) or by replacement of inactive factors due to warfarin treatment (in the case of 4F-PCCs). Addition of coagulation factors or use of 4F-PCCs had limited effect on restoration of normal TG in the presence of a FXa inhibitor. These data demonstrate that AnXa can normalize TF-TG over a wide range of FXa inhibitor concentrations (0-2000 ng/mL), in particular over the range observed in bleeding patients in the ANNEXA-4 study, where 77% of bleeding patients had inhibitor levels ≥75 ng/mL at baseline. In contrast, 4F-PCCs were only able to restore normal TF-TG at low inhibitor levels (<75 ng/mL). These in vitro data are consistent with ANNEXA-4, where AnXa demonstrated >90% immediate reversal of anti-FXa activity and >80% hemostatic efficacy with a wide range of FXa inhibitor levels and multiple bleed types."
There is this from Annexa-4.
"Rapid specific reversal of factor Xa inhibition to hasten hemostatic control should improve clinical outcomes. The hemostatic efficacy of 82% in our trial compares well with the hemostatic efficacy of 72% observed in a previous study of prothrombin complex concentrate involving patients with major bleeding who were treated with vitamin K antagonists, which used similar criteria for assessment of anticoagulation reversal.11
Here is that KCentra study in Warfarin reversal.
So the big question is will KCentra do better against Rivaroxiban and Apixaban than it did against Warfarin?? There is also this retrospective comparison of KCentra to Andexanet that tried to address this question. It's a smallish sample size and the larger sample (n=44) would have been ineligible for A-4. The smaller sample (n=31) that was A-4 eligible did achieve 80% "effective" hemostasis. The n=44 sample ineligible for A-4 however only achieved 55% "good/excellent" hemostasis.
So to cut to the chase, the preclinical data is my touchstone for expecting A to do better than KCentra in the phase 4 trial. There just is not enough clinical data on which to base a guess. I don't think the 15 hour limit will hurt. If the Portola data from ASH on PCC 4 reversal is accurate, then the large majority of patients in the ph 4 will have X and E concentrations above the inflection point at which KCentra is largely ineffective at restoring thrombin generation.