Tafamidis with versus without subsequent vutrisiran initiation in transthyretin amyloid cardiomyopathy: a two-database propensity-matched landmark study

Transthyretin cardiac amyloidosis (ATTR-CM) is a progressive infiltrative cardiomyopathy associated with substantial morbidity and mortality despite treatment with tafamidis. Real-world comparative data evaluating the combination of transthyretin stabilization and gene-silencing therapy remain limited. We conducted retrospective, active-comparator, new-user analyses using the TriNetX and Epic Cosmos electronic health record databases. Adults initiating tafamidis from March 20 through July 9, 2025, were classified during a 180-day treatment-assignment window into a tafamidis-plus-vutrisiran group if they newly initiated vutrisiran or a tafamidis-alone group if they did not initiate a transthyretin gene-silencing therapy. A common day-180 landmark aligned treatment assignment and outcome time zero and was chosen to reduce immortal-time bias that would arise if patients were classified according to a later vutrisiran initiation while follow-up began at tafamidis initiation. This initiation-based classification did not require confirmation of continuous concurrent use of both agents. Follow-up began after the landmark and continued for 180 days. After 1:1 propensity-score matching, each TriNetX treatment group included 268 patients and each Epic Cosmos treatment group included 414 patients. The primary outcome was a composite of all-cause mortality or an acute heart-failure event. In TriNetX, the primary composite outcome occurred in 36 of 268 patients (13.4%) in the tafamidis-plus-vutrisiran group and 47 of 268 patients (17.5%) in the tafamidis-alone group (risk ratio [RR], 0.77; 95% CI, 0.51–1.14; hazard ratio [HR], 0.78; 95% CI, 0.51–1.18). In the separate Epic Cosmos replication cohort, the corresponding risks were 52 of 414 (12.6%) and 66 of 414 (15.9%), respectively (RR, 0.79; 95% CI, 0.56–1.10; HR, 0.80; 95% CI, 0.56–1.13). Neither primary-outcome comparison was statistically significant. No statistically significant between-group differences were observed for all-cause mortality, acute heart-failure events, arrhythmic outcomes, or cardiogenic shock in either database. In post hoc matched-cohort phenotype comparisons, documented baseline peripheral/ATTR neuropathy remained more common in the tafamidis-plus-vutrisiran group than in the tafamidis-alone group in TriNetX (27.6% vs. 15.3%) and Epic Cosmos (29.2% vs. 16.4%). In two retrospective, propensity-matched cohorts of patients with ATTR-CM, there was no statistically significant difference between the tafamidis-plus-vutrisiran and tafamidis-alone groups in the 180-day composite of all-cause mortality or an acute heart-failure event. No statistically significant differences were observed in secondary outcomes. Although point estimates numerically favored tafamidis plus vutrisiran, the confidence intervals included no difference. Residual phenotype imbalance, particularly the higher prevalence of documented neuropathy in the combination group, and the short follow-up limit causal interpretation. These hypothesis-generating findings do not demonstrate an incremental treatment benefit and should not be interpreted as evidence of equivalence.