Difficulties Detecting Clinically Relevant Factor Xa Inhibitor Levels Prior to Reversal With Andexanet Alfa for Intracranial Hemorrhage

Abstract

Coagulation factor Xa (recombinant), inactivated-zhzo (andexanet alfa) is approved for reversal of life-threatening bleeding with rivaroxaban and apixaban use. Clinical decision-making to initiate reversal is reliant on dose taken and timing of last dose. In practice, timing of last dose may be unknown, and the turnaround time for drug-specific anti-factor Xa levels at some institutions may be prolonged, leaving clinicians balancing a difficult decision with limited tools. This report includes a series of 3 patients who presented to our institution with an intracranial hemorrhage and received andexanet alfa for apixaban reversal. These cases highlight the challenges clinicians are facing when using andexanet alfa for emergent rivaroxaban or apixaban reversal when the timing of last dose is unknown, or patients fall outside of the recommended timeframe for use and clinically relevant drug levels are still suspected. Based on our experiences, we encourage other institutions to evaluate their abilities to rapidly and accurately detect the presence of clinically relevant rivaroxaban and apixaban levels when utilizing andexanet alfa.

Introduction

The Food and Drug Administration approved coagulation factor Xa (recombinant), inactivated-zhzo (andexanet alfa) in 2018 for reversal of life-threatening bleeding with rivaroxaban and apixaban use. ¹ Andexanet alfa creates a decoy, which binds factor Xa inhibitors. ² In the ANNEXA-4 trial, investigators enrolled patients with a major bleed within 18 hours of last factor Xa inhibitor dose. Major bleeding was defined as bleeding with hemodynamic instability, hemoglobin decrease by > 2 g/dL, or bleeding into a critical site such as intracranial hemorrhage (ICH). ³ To be included in the efficacy analysis, it was required that baseline direct oral anticoagulant (DOAC) drug-specific anti-factor Xa levels were > 75 ng/mL. Based on this protocol, dosing depends on the dose of rivaroxaban or apixaban taken and timing of last dose.^2,3

Since approval, institutions have been tasked to decide whether to pursue formulary addition or continue use of prothrombin complex concentrate (PCC). Factors to consider include limited efficacy studies comparing andexanet alfa to PCC and high cost (average wholesale price $33,000 to $59,400). ² Based on the inclusion criteria in ANNEXA-4 and the International Society on Thrombosis and Haemostasis (ISTH) recommendations regarding anti-factor Xa level thresholds for reversal in major bleeding, it is plausible to restrict use of andexanet alfa to patients within 18 hours of their last rivaroxaban or apixaban dose or with drug-specific anti-factor Xa levels > 50 ng/mL.^3,4 It is difficult to determine this information if patients are poor historians and if drug-specific anti-factor Xa levels are not readily available. Although some institutions may have access to in-house rivaroxaban or apixaban anti-factor Xa assays, many may only offer these assays as send out tests. Some have created processes to assess for clinically relevant factor Xa levels utilizing available in-house coagulation assays, but as we demonstrate, these protocols may not be applicable at all institutions. ⁵ We present 3 cases from June 2019 to May 2021 in which andexanet alfa was administered for ICH in the setting of unclear timing of last dose or when clinically relevant drug levels were suspected, and the challenges we faced utilizing our in-house coagulation assays.

Case 1

An 89-year-old female presented from her assisted living facility (ALF) with new-onset left-sided hemiparesis and slurred speech. Her past medical history included atrial fibrillation for which she was taking apixaban 2.5 mg twice daily. A head computed tomography (CT) scan revealed a right basal ganglia intraparenchymal hemorrhage (IPH). Her last dose of apixaban was administered 3 hours prior to arrival based on the ALF medication administration record, but it was noted that she intermittently refused medications. The patient confirmed she took her medications that morning. Laboratory testing was collected (Table 1) and reversal was initiated with low dose andexanet alfa. A send out apixaban anti-factor Xa level was performed utilizing the pre-andexanet alfa blood sample and returned 2 days later. A repeat head CT was performed approximately 1 hour after reversal which demonstrated expansion of her basal ganglia hemorrhage. She was discharged back to her ALF 10 days after admission.

Table 1.

Laboratory Values Prior to Anticoagulant Reversal.

	Case 1	Case 2	Case 3
Coagulation Assays
aPTT, seconds	32.0	33.2	26.0
PT, seconds	11.7	12.0	17.0
INR	1.03	1.05	1.47
LMWH anti-factor Xa level, IU/mL	0	0.01	0.01
Apixaban anti-factor Xa level, ng/mL	19.0	51.0	82.0
Chemistry
Creatinine, mg/dL	0.71	1.60	0.65
Total bilirubin, mg/dL	1.0	0.4	0.8
AST, U/L	22	30	49
ALT, U/L	14	29	17
Alkaline phosphatase, U/L	32	135	351

Abbreviations: aPTT, activated partial thromboplastin time; PT, prothrombin time; INR, international normalized ratio; LMWH, low molecular weight heparin; AST, aspartate transaminase; ALT, alanine aminotransferase.

Case 2

An 86-year-old male presented with a headache and confusion. His past medical history included atrial fibrillation for which he was taking apixaban 2.5 mg twice daily. A head CT scan revealed a right frontotemporal subdural hemorrhage and a right temporo-occipital IPH, with a 4 mm right to left midline shift. The patient’s family confirmed his last dose of apixaban was approximately 15 hours prior. Laboratory testing was collected (Table 1) and reversal was initiated with low dose andexanet alfa. A send out apixaban anti-factor Xa level was performed utilizing the pre-andexanet alfa blood sample and returned 2 days later. Magnetic resonance imaging was performed approximately 13 hours after reversal, which demonstrated no expansion. He was discharged to acute inpatient rehabilitation 4 days after admission.

Case 3

A 70-year-old female hospitalized with an acute pulmonary embolism taking apixaban 5 mg twice daily developed slurred speech. A head CT angiography revealed a new acute thrombus in the left middle cerebral artery, and the patient was taken for thrombectomy. The following morning, she developed an acute neurological decline and was found to have a diffuse subarachnoid hemorrhage and ventriculomegaly on head CT. Her last dose of apixaban was approximately 24 hours prior. Laboratory testing was collected (Table 1). Reversal was initiated with low dose andexanet alfa. A send out apixaban anti-factor Xa level was performed utilizing the pre-andexanet alfa blood sample and returned 2 days later. Repeat head imaging was not completed after reversal. She was transitioned to comfort measures and expired 3 days after admission.

Discussion

Our institution added andexanet alfa to formulary in June 2018. Use is restricted to ICH with last dose of apixaban or rivaroxaban within the previous 18 hours and requires approval by a neurocritical care attending. Since formulary addition, we have administered an average of nine doses annually. One of the difficulties we have faced is accurately determining timing of last dose and presence of clinically relevant factor Xa inhibitor drug levels requiring reversal. Given use in ICH, patients may be unreliable historians. Additionally, patient factors may impact drug clearance resulting in the presence of clinically relevant drug levels outside of the 18-hour timeframe.

Per the ISTH, a drug-specific anti-factor Xa level > 50 ng/mL warrants reversal in patients with major bleeding. ⁴ The ANNEXA-4 trial utilized a higher threshold for the efficacy outcome and included patients with a drug-specific anti-factor Xa level ≥ 75 ng/mL. ³ In our patients, reversal was initiated with apixaban anti-factor Xa levels of 19.0 ng/mL, 51.0 ng/mL, and 82.0 ng/mL; however, this data was not available at the time of reversal administration. Given the significant cost associated with andexanet alfa, the ability to rapidly and accurately assess for the presence of clinically relevant anti-factor Xa levels prior to initiating reversal could be a valuable tool.

Available coagulation tests such as PT, INR, aPTT, and LMWH or UFH anti-factor Xa levels are generally widely available but have variation in ability to detect clinically relevant factor Xa inhibitor levels.^6-12 The American College of Chest Physicians and ISTH have conflicting recommendations regarding which commonly available coagulation tests may be useful for quantifying rivaroxaban and apixaban anti-factor Xa levels.^6,7 The International Council for Standardization in Hematology recommendations state that the PT and/or aPTT may not be reliable to detect the presence of clinically relevant anti-factor Xa levels. ⁸ However, in a patient with known factor Xa inhibitor exposure, a prolonged PT or aPTT may be considered secondary to drug effect until proven otherwise.

When examining the relationship between drug-specific anti-factor Xa levels and PT, Barrett et al. reported a linear relationship with PT and apixaban anti-factor Xa levels; however, an analysis in a larger cohort concluded that PT was inadequately sensitive to apixaban anti-factor Xa levels > 50 ng/mL.^9,10 These findings are similar across studies analyzing aPTT values. ¹⁰ In our 2 patients with apixaban anti-factor Xa levels > 50 ng/mL, only one demonstrated an elevated PT and INR, and aPTT was within normal limits.

When examining the relationship between drug-specific anti-factor Xa levels and UFH or LMWH anti-factor Xa levels, Gosselin et al. demonstrated a linear, concentration-dependent relationship between rivaroxaban anti-factor Xa levels and both LMWH and UFH anti-factor Xa levels. ¹¹ Billoir et al. confirmed these results and found a LMWH anti-factor Xa level < 0.50 IU/mL corresponds to a rivaroxaban or apixaban anti-factor Xa level < 30 ng/mL. ¹² It is important to note that these relationships described vary by instrument and assay utilized, which limits the relationships reported by other investigators to be universally appropriate for all institutions. ¹¹ We utilize the ACL Tops instrument with the HemosIL Heparin reagent for our LMWH and UFH anti-factor Xa levels, which as we demonstrated in our 3 cases, has a low sensitivity for the detection of clinically relevant apixaban levels. Internal calibrations are required for our LMWH or UFH anti-factor Xa assays to be useful when determining if reversal with andexanet alfa is indicated.

The ability to rapidly and accurately detect clinically meaningful rivaroxaban and apixaban levels is also essential to minimize potential risk of thrombosis due to unnecessary administration. In ANNEXA-4, 10% of patients experienced a thrombotic event up to 30-days after andexanet alfa administration. ³ Although a direct comparison is not available, only 7.8% of patients who received PCC and 6.4% of patients who received plasma for warfarin reversal experienced a thrombotic event up to 45-days after administration. ¹³ This additional tool could assist clinicians and more clearly balance the potential benefits with risks.

Beyond considering cost and lack of randomized controlled trials comparing clinical efficacy outcomes between andexanent alfa and PCC, we also encourage institutions to evaluate their abilities to rapidly and accurately detect clinically relevant presence of rivaroxaban and apixaban prior to pursuing formulary addition of andexanet alfa. If available for administration, ensure clinicians are educated regarding limitations of their institutional laboratory assays.