Effect of NOAC vs. VKA on clinical outcomes in AF patients independent of burden of comorbidities

Introduction and methods

Patients with AF often have many comorbidities [1]. Comorbidities can increase bleeding risk in patients on anticoagulants [2,3]. Also, risk of stroke increases with many comorbidities and increasing age.

Treatment with warfarin or other VKAs can effectively reduce the risk of stroke and systemic embolic events (SEE) in AF patients [4]. However, anticoagulant therapy is often withheld in patients with a high bleeding risk or other perceived contraindications [5,6]. The decision to prescribe antithrombotic agents can be influenced by the presence of multiple comorbidities [7]. This study investigated the relationship between the burden of comorbidities and relative efficacy and safety of edoxaban vs. warfarin.

The current study was a post-hoc analysis of the ENGAGE AF-TIMI 48 trial. The ENGAGE AF-TIMI 48 trial was a three-group, randomized, double-blind, double-dummy, multinational trial in 21105 AF patients with moderate-to-high-risk (CHADS2 score ≥2). Patients with severe comorbidities such as active malignancies, advanced liver disease, class IV or V chronic kidney disease, and hepatitis B or C were excluded. Patients were randomized to receive warfarin (n=7036), higher dose edoxaban regimen (HDER) (60 mg, n=7035), or lower dose edoxaban regimen (LDER) (30 mg, n=7034). The median follow-up was 2.8 years. The updated Charlson Comorbidity index (CCI) was used to classify the burden of comorbidities [8]. Data were analyzed by prespecified CCI categories: 0 (32.0% of patients), 1 (7.3%), 2 (42.1%), 3 (12.7%) or ≥4 (6.0%).

The primary efficacy outcome was stroke or SEE. Key secondary efficacy endpoint was major adverse cardiac events (MACE), defined as MI, stroke, SEE or CV death. Primary safety outcome was major bleeding. The net clinical outcome was a composite of stroke/systemic embolism, major bleeding or death.

Main results

• Annualized event rates for the primary efficacy endpoint of stroke or SEE were 1.85%, 1.83%, 1.67%, 1.77% and 2.54% for CCI= 0, 1, 2, 3 and ≥4, respectively (P-trend=0.55). The incidence of stroke or SEE was significantly lower in patients with CCI=0 compared to those with CCI ≥4 (P=0.01).
• Prevalence of MACE was higher in patients with higher CCI. Rates increased from 3.6% for CCI=0 to 8.33% for CCI ≥4 (P-trend <0.001).
• Incidence of major bleeding was significantly lower in patients with CCI=0 (2.4%/year) compared to patients with CCI≥4 (4.7%/year). Event rates increased with increasing CCI group (P-trend<0.001).
• Rates of the net clinical outcome were significantly higher in patients with higher CCI (P-trend<0.001).
• No significant interactions between CCI score and the relative efficacy of HDER vs. warfarin were seen. HR’s for the primary efficacy endpoint of stroke/SEE were 0.88, 0.73, 0.90, 0.73, and 1.02 for CCI = 0, 1, 2, 3, ≥4, respectively (P-interaction=0.85). Also, among the secondary efficacy endpoints, no significant treatment interactions by CCI group were observed (all P-interaction >0.20).
• For the primary safety endpoint of major bleeding, no significant interactions were observed between CCI score and HDER vs warfarin (P-interaction >0.64). HR’s for the primary safety endpoint were 0.83, 0.79, 0.71, 0.96, 0.81 for CCI = 0, 1, 2, 3, ≥4, respectively.
• There was no interaction for CCI score with treatment effect of HDER vs. warfarin on net clinical outcome (P-interaction 0.31).
• No significant interaction between CCI and LDER vs warfarin was observed for efficacy outcomes. For safety outcomes, LDER significantly reduced major bleeding vs. warfarin across CCI groups, with a more pronounced reduction in bleeding in CCI=0 group (HR:0.33, P-interaction=0.016).

Conclusion

References

Find this article online at Eur Heart J Cardiovasc Pharmacother