Abstract
Background Prior analyses suggest nucleoside reverse transcriptase inhibitor (NRTI) abacavir (ABC), but not tenofovir (TFV), is associated with a 2-fold increase in the hazard of myocardial infarction. REPRIEVE is ideally suited to evaluate the role of ABC and TFV backbones, tenofovir alafenamide (TAF) and tenofovir disoproxil fumarate (TDF), in major adverse cardiovascular events (MACE).
Methods We compared hazard of first MACE among people with HIV (PWH) at low-to-moderate cardiovascular risk using ABC (n=883), TAF (n=957), and TDF (n=4274) at entry. Overlap weights balanced biasing factors, including age, sex at birth, ASCVD risk, CD4, eGFR, and anchor ART. Associations between entry NRTI and MACE were estimated using a marginal Cox proportional hazards model. Change of NRTI, or “switching,” was common during follow-up. Additional associations were estimated by further censoring at first switch and applying time-updated inverse probability of censoring weighting (IPCW). REPRIEVE was registered with ClinicalTrials.gov , NCT02344290 .
Results Baseline-adjusted associations suggest clinically relevant increases in hazard of first MACE for ABC vs TAF (HR=1.5; 95% CI: 0.9, 2.3), ABC vs TDF (HR=1.4; 95% CI: 0.9, 2.1), but not TAF vs TDF (HR=0.9; 95% CI: 0.6, 1.5). With censoring at switch, HRs increased to 1.6 (95% CI: 0.9, 2.7) for ABC vs TAF, 2.0 (95% CI: 1.2, 3.4) for ABC vs TDF, and 1.2 (95% CI: 0.7, 2.2) for TAF vs TDF. The largest HR observed was for ABC vs TDF and MI (IPCW HR=3.5; 95% CI: 1.3, 9.4).
Conclusions Antiretroviral therapies with ABC backbones are associated an increase in MACE compared to TFV backbones among PWH at low-to-moderate cardiovascular risk.