Add-On Treatment With Zilebesiran for Inadequately Controlled Hypertension: The KARDIA-2 Randomized Clinical Trial

Abstract

Importance: In prior monotherapy studies of patients with hypertension, single subcutaneous doses of zilebesiran, an investigational RNA interference therapeutic, reduced serum angiotensinogen levels and systolic blood pressure (SBP) at 3 and 6 months.

Objective: To evaluate the efficacy and safety of zilebesiran vs placebo when added to a standard antihypertensive medication.

Design, setting, and participants: This phase 2, randomized, prospective, double-blinded trial enrolled adults with uncontrolled hypertension from 150 sites across 8 countries between January 2022 and June 2023. The final follow-up date was December 11, 2023, and analyses were conducted on March 1, 2024.

Interventions: Eligible patients were initially randomized in cohorts to receive open-label run-in treatment for at least 4 weeks with indapamide 2.5 mg, amlodipine 5 mg, or olmesartan 40 mg (4:7:10 randomization), each administered once daily. Within cohorts, adherent patients with 24-hour mean ambulatory SBP of 130 mm Hg to 160 mm Hg were subsequently randomized (1:1) to additional blinded treatment to receive single subcutaneous doses of zilebesiran 600 mg or matching placebo.

Main outcomes and measures: The primary end point in each cohort was the difference between zilebesiran and placebo in change from baseline in 24-hour mean ambulatory SBP at 3 months.

Results: Of 1491 patients entering the run-in phase, 663 (130 receiving indapamide, 240 receiving amlodipine, and 293 receiving olmesartan) were randomized to receive zilebesiran (n = 332) or placebo (n = 331). The least-squares mean difference between zilebesiran and placebo in change from baseline to 3 months in 24-hour mean ambulatory SBP was -12.1 mm Hg (95% CI, -16.5 to -7.6; P < .001) for indapamide, -9.7 mm Hg (95% CI, -12.9 to -6.6; P < .001) for amlodipine, and -4.5 mm Hg (95% CI, -8.2 to -0.8; P = .02) for olmesartan. Across cohorts, more patients who received zilebesiran than placebo experienced hyperkalemia (18 [5.5%] vs 6 [1.8%]), hypotension (14 [4.3%] vs 7 [2.1%]), and acute kidney failure (16 [4.9%] vs 5 [1.5%]) events, but most episodes were mild and resolved without medical intervention.

Conclusions and relevance: In patients with uncontrolled hypertension despite treatment with indapamide, amlodipine, or olmesartan, the addition of single-dose zilebesiran resulted in significant SBP reductions compared with placebo at 3 months, with low rates of serious adverse events.

Trial registration: ClinicalTrials.gov Identifier: NCT05103332.

Figure 1.. Patient Flow in the KARDIA-2 Study of Add-On Treatment With Zilebesiran for Hypertension

Data are shown for the full analysis set. Owing to multiple patient enrollments, which were discovered after initial database lock, 4 patients accounting for 9 patient IDs were excluded from all analyses. ^aA patient could have multiple reasons for exclusion or run-in failure and was counted separately for each reason. ^bOne patient was excluded owing to Good Clinical Practice Guidelines violation. ^cOne patient assigned to the placebo group accidentally received a partial dose of zilebesiran (100 mg), and is therefore included in the placebo group of the full analysis set and the zilebesiran group of the safety analysis set and excluded from the pharmacodynamic analysis set. ^dNo patients who stopped participating did so because of an adverse event.

Figure 2.. Change From Baseline at 3 Months in Systolic Blood Pressure (SBP) by Background Therapy Cohort and Treatment Group

Analyses are presented for the full analysis set. Box plots demonstrate median (horizontal line), mean (circle), IQR (box upper and lower boundary), highest and lowest values within 1.5 × the IQR (whiskers), and more extreme values (diamonds). Least-squares mean (LSM) values were derived from mixed model for repeated measures analysis including treatment, visit, treatment × visit, and race (Black; all other races) as fixed factors and corresponding baseline SBP and baseline estimated glomerular filtration rate as covariates; assessments from patients who received rescue therapy within 2 weeks of an assessment were censored.