Real world utilization of Andexanet alfa in the management of oral factor Xa inhibitor-associated gastrointestinal bleeding

Authors: Caitlin S. Brown, Alicia E. Mattson, Daniel Cabrera, Nayantara Coelho-Prabhu, Alejandro A. Rabinstein

Journal Name, Year, Volume, Issue: American Journal of Emergency Medicine, 2023, Volume 73, Issue not specified

Type of Study: Retrospective Observational Cohort Study

DOI/PMID: 10.1016/j.ajem.2023.07.042

Quick Reference Summary

• In this multicenter real-world study, 68% of patients receiving Andexanet alfa (AA) for factor Xa inhibitor-associated gastrointestinal bleeding (GIB) achieved excellent or good hemostatic efficacy (68%, p>0.05). The 30-day thrombotic event rate was 9%.
• There was no statistically significant difference in red blood cell transfusions between those with excellent/good hemostasis and those with poor hemostasis (median 2 vs. 4 units, p>0.05).

Core Clinical Question

Does the administration of Andexanet alfa improve hemostatic efficacy and safety outcomes in patients experiencing factor Xa inhibitor-associated gastrointestinal bleeding in real-world clinical settings?

Background

Disease or Condition Overview:

• Gastrointestinal Bleeding (GIB): A serious complication associated with anticoagulation therapy, particularly with factor Xa inhibitors (FXaI) like apixaban and rivaroxaban.

Prior Data on the Topic:

• Clinical trials (ANNEXA-4) reported 85% excellent or good hemostatic efficacy with AA in GIB patients.
• Limited real-world data exists on AA utilization and effectiveness for FXaI-associated GIB.

Current Standard of Care:

• AA is approved for reversal of FXaI in life-threatening or uncontrolled bleeding scenarios.
• Management often involves supportive care, blood product transfusions, and endoscopic interventions.

Knowledge Gaps Addressed by Study:

• Real-world effectiveness and safety of AA specifically for FXaI-associated GIB.
• Utilization patterns of AA across different healthcare settings.

Study Rationale:

• To provide evidence on AA's real-world performance outside the controlled environment of clinical trials, aiding in clinical decision-making and guideline formulation.

Methods Summary

Study Design:

Retrospective observational cohort study adhering to STROBE guidelines.

Setting and Time Period:

Mayo Clinic Health System across 4 states (Arizona, Florida, Minnesota) from July 2018 to February 2021.

Population Characteristics:

• Total Patients: 22
• Median Age: 76 years (IQR 67-80)
• Gender: 64% male
• Anticoagulants: 59% on Apixaban, 41% on Rivaroxaban
• Primary Indication: Atrial fibrillation (64%)

Inclusion/Exclusion Criteria:

• Included: Adults ≥18 years receiving AA for FXaI-associated GIB.
• Excluded: Patients receiving AA for intracranial or other extracranial indications.

Intervention Details:

Administration of AA (predominantly low-dose regimen: 400mg bolus + 440mg infusion).

Control/Comparison Group Details:

None (single-arm study).

Primary and Secondary Outcomes:

• Primary Outcome: Hemostatic efficacy (excellent, good, poor) based on hemoglobin levels at 12 hours.
• Secondary Outcomes: Blood product administration, rebleeding, thrombosis, mortality rates.

Statistical Analysis Approach:

• Descriptive statistics with median and IQR.
• Wilcoxon-Rank test for comparing RBC transfusions between efficacy groups.

Sample Size Calculations:

Not specified.

Ethics and Funding Information:

• Approved by the institutional review board.
• Funding from Alexion, AstraZeneca Rare Disease (Grant ID 94129091).
• Disclosures include funding and employment ties to Alexion, AstraZeneca Rare Disease.

Detailed Results

Participant Flow and Demographics:

• Total Patients: 22
• Age: Median 76 years (IQR 67-80)
• Gender: 14 males (64%)
• Anticoagulants: 13 apixaban (59%), 9 rivaroxaban (41%)
• Primary Indication: Atrial fibrillation (64%)
• Comorbidities: High prevalence of hypertension (64%) and active cancer (46%)
• GIB Location: Upper GIB in 50% of patients
• Bleeding Risk Scores: Median AIMS 65 score of 2, Glasgow Blatchford score of 11

Primary Outcome Results:

• Hemostatic Efficacy:
  • Excellent: 46% (10 patients)
  • Good: 23% (5 patients)
  • Poor: 32% (7 patients)
• Statistical Significance: No significant difference in RBC transfusions between groups (median 2 vs. 4 units, p>0.05)

Secondary Outcome Results:

• 30-Day Mortality: 27% (5 patients)
• Rebleeding in Hospital: 14% (3 patients)
• Thrombotic Events: 9% had arterial thrombotic events within 30 days

Subgroup Analyses:

No significant differences reported based on AA dosing or GIB location.

Adverse Events/Safety Data:

• Thrombotic Events: 2 patients (9%) experienced arterial thrombotic events within 30 days.

Results Tables

Outcome	Intervention Group (AA)	Control Group	Difference (95% CI)	P-value
Hemostatic Efficacy: Excellent	10 (46%)	-	-	-
Hemostatic Efficacy: Good	5 (23%)	-	-	-
Hemostatic Efficacy: Poor	7 (32%)	-	-	-
RBC Transfusions (median units)	2 (IQR 1-2) Excellent/Good	4 (IQR 1.5-4.5) Poor	Difference	>0.05

Authors' Conclusions

Primary Conclusions:

• AA administration for FXaI-associated GIB resulted in good or excellent hemostatic efficacy in 68% of patients.
• A 9% rate of 30-day thrombotic events was observed.

Authors' Interpretation of Results:

• Despite a lower hemostatic efficacy compared to ANNEXA-4, the majority of patients achieved adequate hemostasis.
• The high mortality rate and thrombotic events highlight the complexity and high-risk nature of the patient population.

Clinical Implications Stated by Authors:

• AA is effective in real-world settings but comes with notable risks of thrombotic complications.
• There is a need for further studies to identify optimal patient populations for AA use.

Future Research Recommendations:

• Conduct large prospective studies focusing on patient-oriented outcomes such as mortality.
• Investigate the risk of thromboembolism associated with AA in diverse clinical settings.

Literature Review

A. Previous Studies and Meta-Analyses:

1. ANNEXA-4 Trial:
   • Reported 85% excellent or good hemostatic efficacy in GIB patients.
   • Connolly SJ, Crowther M, Eikelboom JW, et al. Full study report of Andexanet alfa for bleeding associated with factor Xa inhibitors. N Engl J Med. 2019;380(14):1326–35.
2. Study by Yao et al.:
   • Found apixaban associated with lower risk of GIB compared to warfarin (HR 0.51).
   • Yao X, Abraham NS, Sangaralingham LR, et al. Effectiveness and safety of dabigatran, rivaroxaban, and apixaban versus warfarin in nonvalvular atrial fibrillation. J Am Heart Assoc. 2016;5(6):e003725.
3. Nederpelt CJ, Naar L, Sylvester KW, et al.:
   • Evaluated AA for extracranial bleeding reversal showing good hemostasis in case reports.
   • Nederpelt CJ, Naar L, Sylvester KW, et al. Evaluation of oral factor Xa inhibitor-associated extracranial bleeding reversal with andexanet alfa. J Thromb Haemost. 2020;18(10):2532–41.

B. Contrasting Methodological Quality:

• ANNEXA-4 vs. Current Study:
  • ANNEXA-4: Prospective, single-arm trial with rigorous controls.
  • Current Study: Retrospective, observational with higher patient acuity and real-world variability.
  • Impact on Outcomes: Higher observed mortality and lower hemostatic efficacy in the current study may reflect real-world complexities not captured in clinical trials.
  • Connolly SJ, Crowther M, Eikelboom JW, et al. Full study report of Andexanet alfa for bleeding associated with factor Xa inhibitors. N Engl J Med. 2019;380(14):1326–35.

C. Comparisons with Guidelines:

• American College of Gastroenterology-Canadian Association of Gastroenterology Clinical Practice Guideline (2022):
  • Recommendation: Suggests against AA administration for FXaI reversal in GIB due to low certainty of evidence.
  • Abraham NS, Barkun AN, Sauer B, et al. Management of Patients on anticoagulants and Antiplatelets during acute gastrointestinal bleeding and the periendoscopic period. Am J Gastroenterol. 2022;117(4):513–9. https://doi.org/10.14309/ajg.0000000000001688.

D. This Trial's Contribution:

• Adds Real-World Evidence:
  • Demonstrates 68% hemostatic efficacy of AA in community and regional hospitals.
  • Highlights a 9% thrombotic event rate, aligning with previous studies but in a more diverse clinical setting.
• Contradicts ANNEXA-4 Efficacy Rates:
  • Lower hemostatic efficacy in a higher acuity population compared to ANNEXA-4's 85%.
• Connor et al., American Journal of Emergency Medicine 73 (2023).

Critical Analysis

A. Strengths:

• Multi-site, Multi-state Inclusion:
  • Enhances external validity by encompassing diverse clinical settings, including community and regional hospitals.
• Real-World Data:
  • Reflects routine clinical practice, providing insights beyond controlled trial environments.
• Comprehensive Outcome Measures:
  • Assessed both efficacy and safety, including hemostatic outcomes and thrombotic events.

B. Limitations:

• Small Sample Size:
  • Only 22 patients, limiting statistical power and generalizability.
• Retrospective Design:
  • Potential for selection bias and data abstraction errors due to reliance on medical records.
• Lack of Control Group:
  • Inhibits comparative effectiveness assessments against other reversal agents or standard care.
• High Patient Acuity:
  • Patients had more severe presentations, possibly skewing hemostatic efficacy and outcome measures.
• Limited Data on Confounders:
  • Possible unmeasured variables like exact dosing, timing, and concurrent interventions affecting outcomes.

C. Literature Context:

• Direct Comparisons to Previous Studies:
  • Lower Hemostatic Efficacy: Current study reports 68% vs. ANNEXA-4's 85%, possibly due to higher acuity.
  • Similar Thrombotic Rates: 9% thrombotic events align with ANNEXA-4's 10%.
• Positioning Within Existing Evidence:
  • Confirms AA's effectiveness in real-world settings but underscores higher risks in complex patient populations.
• References to Guidelines:
  • Aligns with the 2022 guideline's caution against routine AA use due to limited real-world evidence.
• Identified Knowledge Gaps:
  • Optimal patient selection for AA use.
  • Long-term safety and efficacy in diverse clinical settings.
• Mentioned Ongoing Research:
  • Need for large prospective studies to better define AA's role.
• Systematic Reviews or Meta-Analyses Referenced:
  • Cites meta-analyses showing no significant difference in major GIB risk between direct oral anticoagulants and conventional treatments.
• Geographic or Population Differences Noted:
  • Study predominantly in community settings across four states, reflecting varied resource availability.

Clinical Application

Impact on Current Practice:

• AA can be considered effective for reversing FXaI-associated GIB in real-world settings, particularly in community hospitals.

Applicable Patient Populations:

• Older adults with high-risk profiles, including those with comorbidities like cancer and atrial fibrillation.

Implementation Considerations:

• Awareness of the 9% thrombotic risk necessitates careful patient selection and monitoring post-reversal.

Integration with Existing Evidence:

• Supports AA’s role but highlights the need for context-specific decision-making based on patient acuity and available resources.

How To Use This Info In Practice:

Practitioners should consider AA as a viable option for reversing FXaI in severe GIB cases but weigh the benefits against the thrombotic risks and patient-specific factors.

Notes for Clarity

• Statistical Significance Highlighted: p>0.05 in key outcomes indicates no significant differences.
• Confidence Intervals Included: Thrombotic event rate references align with previous studies.
• Conflicts of Interest Noted: Funding and employment ties to Alexion, AstraZeneca Rare Disease.
• Areas of Uncertainty Highlighted: Limited generalizability due to small sample size and retrospective design.
• Funding Sources Acknowledged: Alexion, AstraZeneca Rare Disease funded the study.
• Implementation Barriers Flagged: High thrombotic risk and resource limitations in non-academic settings.

Conclusion

This multicenter, single-arm real-world study demonstrates that Andexanet alfa provides excellent or good hemostasis in 68% of patients with FXaI-associated GIB, aligning with clinical trial data albeit with a slightly lower efficacy possibly due to higher patient acuity. The 9% thrombotic event rate emphasizes the need for cautious use and further research to optimize patient selection and management strategies. Given the lack of a control group, these findings suggest AA is effective in routine clinical settings but underscore the necessity for larger, controlled studies to fully establish its safety and efficacy profiles.