Overview of the Questionable Drug Andexanet (ANDEXXA)

In 2018, andexanet alfa (ANDEXXA, hereafter andexanet) received accelerated approval from the Food and Drug Administration (FDA) for emergency reversal of life-threatening or uncontrolled bleeding caused by the use of two new oral anticoagulants (blood thinners): apixaban (ELIQUIS) and rivaroxaban (XARELTO).[1] These anticoagulants are approved for the following main uses: preventing stroke in patients with nonvalvular atrial fibrillation (a heart rhythm disorder), and deep vein thrombosis (blood clots in large veins, such as those in the legs) and pulmonary embolism (blood clot in the lungs) after hip or knee replacement surgery, as well as treating and reducing the recurrence risk of deep vein thrombosis and pulmonary embolism in general.[2],[3],[4]

Public Citizen’s Health Research Group has designated andexanet as a Do Not Use medication because its clinical efficacy and safety have not been adequately established. In addition, we have designated both apixaban and rivaroxaban as Do Not Use because their benefit-risk profiles are not better than that of warfarin (COUMADIN, JANTOVEN), the standard, inexpensive oral anticoagulant that has been around for decades.[5],[6] Warfarin’s blood-thinning effects have been well-established and its effects can be monitored easily using a widely available blood test and can be reversed quickly if needed with an inexpensive vitamin K injection.

Andexanet’s uncertain clinical benefits

The FDA approved andexanet based on results from two small placebo-controlled pharmacological studies and initial results from a single open-label, noncomparative, then-ongoing clinical trial called ANNEXA-4.[7] The pharmacological studies enrolled healthy volunteers, some of whom received either apixaban or rivaroxaban for four days before receiving andexanet. ANNEXA-4 enrolled subjects with sudden major bleeding who were already taking either apixaban or rivaroxaban and were subsequently given andexanet during the trial.[8] Reduction in blood thinning activity constituted the main efficacy basis for FDA approval of andexanet because its maker, Bristol-Myers Squibb, claimed that this surrogate endpoint correlates with clinical response, a hypothesis that FDA clinical reviewers doubted but was accepted by the FDA director who oversaw the review process.[9] Initial ANNEXA-4 results did not show that the drug improved hemostasis (resolution of serious bleeding).[10]

A 2019 updated analysis of ANNEXA-4 concluded that reduction in blood thinning activity was not a “robust” predictor of clinical response,[11] negating the drugmaker’s earlier claim. This analysis reported that 82% of 249 subjects given andexanet with usable data were judged to have good or excellent hemostasis after 12 hours of andexanet infusion. Because the trial did not have a comparison group, it is not known if hemostasis would have been the same in subjects treated with usual emergency supportive care (hereafter usual care).

In fact, editors of the independent drug review journal Prescrire International argued that it is highly possible that andexanet-treated subjects in ANNEXA-4 had already received usual care.[12] These editors also noted that it is uncertain whether the natural elimination of the oral anticoagulants could have played a role in the improved hemostasis seen in the trial (12 hours after end of andexanet infusion) given that it also takes 12 hours for half of apixaban or rivaroxaban to be eliminated from the body after being taken by mouth. Further, the updated analysis did not assess hemostasis relative to the time between the last use of anticoagulants and the administration of andexanet.

ANNEXA-4 appears to have been completed (with 479 enrolled subjects) but its final results have not been published as of press time for this article.[13] Regardless of any new results, the generalizability of this trial to real-world patients is limited because the trial excluded seriously ill patients who need a bleeding-reversal agent the most, including comatose patients, those with severe bleeding in the brain and those needing major surgical procedures within 12 hours of andexanet treatment. The trial also excluded subjects with a life expectancy of less than one month.

Although another ongoing trial is comparing andexanet with usual care in patients with bleeding in the brain, its results are not expected until April 2023.[14]

Andexanet’s serious risks and other issues

FDA clinical reviewers were further troubled by the initial results of ANNEXA-4 because they showed that andexanet was associated with serious and life-threatening adverse events: blood clots in arteries and veins, cardiac arrest, ischemic events (such as heart attack and ischemic stroke) and sudden death.[15] Yet the director of the FDA office responsible for reviewing andexanet downplayed this concern, arguing that these risks are “mitigated” by a black-box warning on its label — the strongest warning that the FDA can require — about these risks since approval.[16],[17]

However, the updated ANNEXA-4 results have confirmed the concerns of the FDA clinical reviewers: 54 (15%) of 352 subjects died within 30 days of receiving andexanet.[18] Even more deaths likely would have occurred had severely ill patients been permitted to participate in the trial.

Furthermore, 63 (18%) of the ANNEXA-4 subjects experienced at least one thromboembolic (blood-clot) adverse event (including deep venous thrombosis, heart attack, pulmonary embolism, stroke or transient ischemic attack) within 30 days of andexanet infusion.[19] In one-third of the 63 subjects, these adverse events occurred within the first three days of receiving andexanet. Similarly, real-world data from one U.S. hospital showed that thrombotic events occurred in 17% of 36 patients who received the drug.[20]

key logistic shortcoming of andexanet is that it is available in vials as a powder that needs to be dissolved in an accompanying solution, a process that can take up to five minutes per vial.[21] This is challenging in emergency situations because typically nine to 18 vials are needed to infuse a single dose of the drug.

In addition to the serious efficacy and safety issues associated with andexanet, many hospitals have not added it to their formularies due to its prohibitive cost:[22] approximately $29,000 to $48,100 for a single administration, depending on the needed dose.[23]

What You Can Do

If you have a condition that requires you to use an oral anticoagulant, do not take rivaroxaban, apixaban or any other new oral anticoagulant that would require the use of andexanet or another new antidote in case of bleeding. Instead, talk to your doctor about choosing warfarin.

Never stop any anticoagulant abruptly without consulting your doctor first because doing so increases your risk of blood clots and stroke.
 

References

[1] Food and Drug Administration. Accelerated approval letter for coagulation factor Xa (recombinant), inactivated-zhzo (BL125586/0). May 3, 2018. https://www.fda.gov/media/113285/download. Accessed December 4, 2020.

[2] Food and Drug Administration. Summary basis for regulatory action for ANDEXXA/coagulation factor Xa (recombinant), inactivated-zhzo. May 3, 2018. https://www.fda.gov/media/113954/download. Accessed December 4, 2020.

[3] Bristol-Myers Squibb. Label: apixaban (ELIQUIS). November 2019. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/202155s024lbl.pdf. Accessed December 4, 2020.

[4] Janssen Pharmaceuticals, Inc. Label: rivaroxaban (XARELTO). March 2020. https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/202439s031,022406s035lbl.pdf. Accessed December 4, 2020.

[5] Is XARELTO really the 'right move' for patients with blood clots or risk for stroke? Worst Pills, Best Pills News. April 2016. /newsletters/view/1026. Accessed December 4, 2020.

[6] Do Not Use the new oral blood thinner apixaban (ELIQUIS). Worst Pills, Best Pills News. June 2018. /newsletters/view/1202. Accessed December 4, 2020.

[7] Food and Drug Administration. Summary basis for regulatory action for ANDEXXA/coagulation factor Xa (recombinant), inactivated-zhzo. May 3, 2018. https://www.fda.gov/media/113954/download. Accessed December 4, 2020.

[8] 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):1326-1335.

[9] Food and Drug Administration. Summary basis for regulatory action for ANDEXXA/coagulation factor Xa (recombinant), inactivated-zhzo. May 3, 2018. https://www.fda.gov/media/113954/download. Accessed December 4, 2020.

[10] Portola Pharmaceuticals, Inc. Label: andexanet alfa (ANDEXXA). September 2020. https://dailymed.nlm.nih.gov/dailymed/fda/fdaDrugXsl.cfm?setid=ae7f0c50-ff2d-49e5-8e10-4efa861556e6&type=display#S5. Accessed December 4, 2020.

[11] 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):1326-1335.

[12] andexanet alfa (ondexxya) to xabans as an antidote. Prescrire Int. 2020;29(217):173-176.

[13] A study in patients with acute major bleeding to evaluate the ability of andexanet alfa to reverse the anticoagulation effect of direct and indirect oral anticoagulants (extension study). Last updated October 8, 2020. https://clinicaltrials.gov/ct2/show/NCT02329327. Accessed December 4, 2020.

[14] Food and Drug Administration. Accelerated approval letter for coagulation factor Xa (recombinant), inactivated-zhzo (BL125586/0). May 3, 2018. https://www.fda.gov/media/113285/download. Accessed December 4, 2020.

[15] Food and Drug Administration. Summary basis for regulatory action for ANDEXXA/coagulation factor Xa (recombinant), inactivated-zhzo. May 3, 2018. https://www.fda.gov/media/113954/download. Accessed December 4, 2020.

[16] Ibid.

[17] Portola Pharmaceuticals, Inc. Label: andexanet alfa (ANDEXXA). September 2020. https://dailymed.nlm.nih.gov/dailymed/fda/fdaDrugXsl.cfm?setid=ae7f0c50-ff2d-49e5-8e10-4efa861556e6&type=display#S5. Accessed December 4, 2020.

[18] Ibid.

[19] Ibid.

[20] Parsels KA, Seabury RW, Darko W, et al. Is it truly “alpha”? Incidence of thrombotic events with andexanet alfa at a single academic medical center. Ann Emerg Med. 2020;75(5):675-676

[21] Bernard C, Villeneuve E, Genest M, Gosselin S. Real-world utilization of andexanet alfa. Am J Emerg Med. 2020;38(4):827.

[22] Parsels KA, Seabury RW, Darko W, et al. Is it truly “alpha”? Incidence of thrombotic events with andexanet alfa at a single academic medical center. Ann Emerg Med. 2020;75(5):675-676.

[23] Mujer MTP, Rai MP, Atti V, et al. An update on the reversal of non-vitamin k antagonist oral anticoagulants. Adv Hemato. 2020;7636104(January 27).