A trial emulation study indirectly comparing tafamidis and acoramidis in transthyretin amyloid cardiomyopathy: the ReplicATTR study

Scritto il 14/09/2026
da Philippe Debonnaire

J Comp Eff Res. 2026 Sep 14:e260114. doi: 10.57264/cer-2026-0114. Online ahead of print.

ABSTRACT

Aim: Tafamidis and acoramidis are transthyretin stabilizers approved for the treatment of transthyretin amyloid cardiomyopathy. We compared the effectiveness of acoramidis versus tafamidis regarding all-cause mortality (ACM). Materials & methods: This study used a hybrid design to compare 42-month ACM with acoramidis versus tafamidis using trial emulation, calibration, and indirect treatment comparison. Two real-world cohorts of participants initiating tafamidis were identified from TriNetX (2019-2024) using observational analogues to emulate ATTR-ACT and ATTRibute-CM trial eligibility criteria and baseline characteristics. A calibration factor was derived by comparing 42-month ACM in the ATTR-ACT tafamidis arm with the ATTR-ACT emulation cohort to quantify the net effect of systematic differences between the populations. This factor was applied to the ATTRibute-CM emulation tafamidis cohort and compared with the ATTRibute-CM acoramidis arm. Using these values and corresponding variance, Monte Carlo simulation (MCS)-estimated medians and percentile-based 95% CIs were derived. Results: In the ATTRibute-CM emulation tafamidis cohort (n = 617), 42-month ACM risk was 35.4% (95% CI: 29.6-42.0), and decreased to a MCS-estimated median of 31.2% (95% CI: 25.3-39.0) after application of the calibration factor (MCS-estimated median: 1.14 [95% CI: 0.91-1.43]). Comparison with 42-month ACM risk observed in the ATTRibute-CM acoramidis arm (24.0%, 95% CI: 20.0-28.7) yielded a relative risk of 0.78 (95% CI: 0.60-0.99), consistent with lower observed ACM risk in the acoramidis arm. Conclusion: Acoramidis was linked to a lower estimated ACM risk in this emulated comparative analysis. These findings should be considered as hypothesis-generating, requiring confirmation in head-to-head studies.

PMID:42734113 | DOI:10.57264/cer-2026-0114