Andexanet Improves Hemostasis in ICH Tied to Factor Xa Inhibitors

In patients with intracerebral hemorrhage (ICH) who are taking factor Xa inhibitor anticoagulants, andexanet alfa (Andexxa/Ondexxya, AstraZeneca) rapidly reduced antifactor Xa (anti-Xa) activity and increased the rate of hemostatic efficacy, but was also associated with an increase in thrombotic events in the ANNEXa-I study.

"Andexanet can be considered for patients with ICH associated with factor Xa inhibitors but, as with many treatments, we will need to balance the benefits and risks and use careful risk stratification for patient selection," lead investigator of the study, Stuart Connolly, MD, professor of medicine, McMaster University, Hamilton, Ontario, Canada, concluded.

Connolly presented the ANNEXa-I study at the recent World Stroke Congress (WSC) 2023 held in Toronto, Canada.

Andexanet, which is conditionally approved in Canada, the US, and Europe, is specifically designed to rapidly reverse the anticoagulation effects of direct oral factor-Xa inhibitors, such as apixaban, rivaroxaban, and edoxaban, due to life-threatening or uncontrolled bleeding.

In a previous nonrandomized open-label study (ANNEXA-4) in patients with acute major bleeding linked to the use of factor Xa inhibitors, the agent was associated with a 92% reduction from baseline in anti-Xa activity, with 82% of patients showing excellent or good clinical hemostasis 12 hours after andexanet infusion.

That was the precursor for the current ANNEXa-I study, which investigated the use of andexanet in patients on a factor Xa inhibitor who are experiencing an ICH.

Connolly noted that patients who experience an ICH while on a factor Xa inhibitor have a high rate of hematoma expansion, leading to an increased risk of death and disability.

ANNEXa-I was a randomized, open-label, blinded endpoint trial that aimed to enroll 900 patients within 6 hours of symptom onset of an acute ICH with a hematoma volume of 0.5 mL to 60 mL and who had received treatment with an oral factor Xa inhibitor within 15 hours.

Eligible patients were randomly assigned to andexanet alfa or usual care. The primary efficacy endpoint was hemostatic efficacy, defined as an increase in baseline National Institutes of Health Stroke Scale (NIHSS) score of less than 7 points at 12 hours, together with an increase in hematoma volume on repeat neuroimaging at 12 hours of less than 35%, and with no rescue therapies — surgery or prothrombin complex concentrate (PCC) — between 3 and 12 hours after randomization.

Trial Stopped Early

The trial was stopped earlier this year by the data and safety monitoring board after an interim analysis when 450 patients had been randomized because of superior efficacy in the andexanet group based on a P value of less than 0.031 for the primary endpoint.

Baseline characteristics of patients in the trial showed a mean age of 79 years, a heavy burden of underlying vascular disease with a history of myocardial infarction (MI) in 11%, and stroke in 20% to 22%. Additionally, atrial fibrillation was very common (85% - 90%), the median NIHSS score was 9, and the median hematoma volume was 10 mL.

The low dose of andexanet was used in 75% of patients and the high dose in the remaining 25%. The dose was prespecified based on the type and dose of factor Xa inhibitor that the patient had received and the timing of enrollment. PCC was given to 87% of patients in the usual care arm.

Median time from symptoms to baseline scan was 2.3 hours, and median time from scan to randomization was 1.1 hours.

Results showed a sharp reduction in anti-Xa activity after 1 hour in the andexanet group. The reduction in anti-Xa activity was over 90% for patients on apixaban or rivaroxaban and around 72% for those on edoxaban, in line with previously reported observations.

Good or excellent hemostatic efficacy was achieved in 63.9% of patients in the andexanet arm vs 52.4% in the usual care arm, giving an increase of 11 patients per 100 treated, a highly significant result (P = .008), Connolly reported.

Excellent hemostatic efficacy (hematoma expansion less than 20%) occurred in 55.9% of andexanet patients vs 45.3% of those given usual care, and good hemostatic efficacy occurred in 8% vs 7.1%, respectively.

An absolute increase in hematoma volume of 12.5 mL or more occurred in 11.6% of andexanet patients vs 19% of the usual care group, a difference of 7.4 patients per 100 patients treated.

A modified Rankin Scale (mRS) score of 3 or less (suggesting a good functional outcome) at 30 days was achieved by 28% of andexanet patients and 30.9% of the usual care group, a nonstatistically significant difference.

Safety results showed an increase in thrombotic events in the andexanet group, with 10.3% of andexanet patients experiencing at least one thrombotic event compared with 5.6% of those on usual care (P = .048).

Connolly noted that there was an excess of ischemic strokes (6.5% vs 1.5%) and of MI (4.2% vs 1.5%) in the andexanet group, but no difference in deaths (27.8% vs 25.5%).

Subgroup analysis did not show any significant interactions for any particular patient group in terms of treatment benefit, which was fairly uniform across all patient groups.

With regard to thrombotic events, interaction P values were not calculated for these as the event rates were small and the interaction P values would not be significant, Connolly noted. But there were some suggestions of interactions. For example, patients with prior history of stroke or TIA seemed to have a higher incidence of a thrombotic events.

Balancing the Risks and Benefits

Asked for comment on these results for theheart.org | Medscape Cardiology, experts said it will be important to see further data from the trial to understand more fully the balance between risks and benefits.

Rustam Al-Shahi Salman, MD, professor of clinical neurology at the University of Edinburgh, Edinburgh, Scotland, said: "ANNEXa-I is a landmark trial in that it has used a randomized design. Most previous studies of these antidotes just simply looked at coagulation parameters in treated patients showing that coagulation measurements are corrected in an observational design, so this is a big step forward.

"While the trial clearly shows that this drug works in re-establishing effective hemostasis, this has to be balanced by the increased risk of thrombotic events," he said. "From the results so far, I think it difficult to judge whether there is a definite net benefit."

Salman said more data on the clinical outcomes is needed to know what the overall effects are. "Also, because it is such an expensive drug, there is going to have to be a health economics evaluation to work out whether it is cost-effective," he added.

"The ANNEXa-I results indicate that the number needed to treat to increase hemostatic efficacy was less than the number needed to treat to see harm from an increase in thrombotic events," co-chair of the WSC session where the study was presented, Ashkan Shoamanesh, MD, associate professor of medicine at McMaster University. "This would suggest overall benefit from andexanet alfa in this population."

He also noted that patients without prior history of stroke or MI did not seem to be at a higher risk of thrombotic events with andexanet.

"Unfortunately, however, the measure of functional outcome (mRS at 1 month) was performed too early and was also underpowered to provide meaningful insights in the interpretation of results," he added. "For instance, in recent years we have learned that the benefits of acute ICH treatments generally require 6 to 12 months to manifest themselves."

Shoamanesh said that several uncertainties remain, and further analysis is required for optimization of patient selection for direct reversal of factor Xa inhibitors with andexanet.

"These will ultimately improve our understanding of the modification of treatment effects on efficacy and safety outcomes according to time from symptom onset to treatment, baseline hematoma volumes, antifactor-Xa levels, and dose/type of PCC used in the usual care arm. Further detail on the clinical impact of failing to achieve hemostatic efficacy and thrombotic events is also required to best interpret these results," he commented.

"In the interim, the main results of ANNEXa-I indicate that andexanet alfa should be made available to physicians and can be considered for the reversal of factor Xa-associated ICH on an individualized basis weighing its potential benefits against the risk of thrombotic events on a case-by-case basis," he suggested.

Shoamanesh also pointed out that in subgroup analyses there was no suggestion of a reduced effect of andexanet vs usual care for hemostatic efficacy in the 87% of patients who were destined to receive PCC as part of usual care vs those who did not receive PCC in the control arm.

"This data, although limited, implies that PCC has little if any benefit in mitigating ICH growth in patients treated with factor Xa inhibitors. This would be consistent with the already established lack of effect of PCC on antifactor-Xa activity and the reported long time to normalization of endogenous thrombin potential with PCC in these patients."

The ANNEXa-I trial was funded by Alexion-Astra Zeneca. Connolly reported receiving consulting fees and research funding from Astra Zeneca.

15th World Stroke Congress (WSC) 2023. Presented October 11, 2023.

For more news, follow Medscape on Facebook, X (formerly known as Twitter), Instagram, and YouTube.