Intensive LDL Cholesterol Targeting in Atherosclerotic Cardiovascular Disease

Authors, Journal, Affiliations, Type, DOI

• Authors: Yong-Joon Lee, Seung-Jun Lee (co-first authors), Byeong-Keuk Kim (corresponding), et al., for the Ez-PAVE Investigators
• Journal: New England Journal of Medicine, 2026;394:1365–1375. Published online March 28, 2026; print April 9, 2026
• Affiliations: Severance Hospital, Yonsei University College of Medicine, Seoul (lead); 17 sites across South Korea
• Type: Investigator-initiated, multicenter, open-label, randomised superiority trial
• Funding: Cardiovascular Research Center (South Korea) + Yuhan (Korean pharmaceutical company); investigator-initiated grant; funders had no role in design, data collection, analysis, or manuscript writing
• DOI: 10.1056/NEJMoa2600283
• Trial Registration: NCT04626973

Overview

The Ez-PAVE trial (Effects of Ezetimibe Combination Therapy for Patients with Atherosclerotic Cardiovascular Disease — Randomized Comparison of LDL Cholesterol Targeting <70 vs. <55 mg/dL) enrolled 3,048 patients with documented ASCVD across 17 South Korean sites, randomly assigned 1:1 to an LDL-C target of <55 mg/dL (intensive) vs <70 mg/dL (conventional). This is the first large RCT to directly compare these two specific LDL-C targets — the precise evidence gap underlying the recent guideline upgrade from <70 to <55 mg/dL for secondary prevention. At median 3.0 years, the intensive-targeting group achieved a significant 33% reduction in the primary composite endpoint (CV death, nonfatal MI, nonfatal stroke, any revascularisation, or UA hospitalisation; HR 0.67; P=0.002), driven primarily by revascularisation and nonfatal MI reductions. Safety was equivalent between groups, with lower creatinine elevation unexpectedly observed in the intensive arm.

Keywords

LDL cholesterol targeting, secondary prevention, atherosclerotic cardiovascular disease, ezetimibe, statins, PCSK9 inhibitors, intensive lipid lowering, cardiovascular events, dyslipidemia, LDL-C goals

Key Takeaways

Background

• Despite guideline recommendations lowering the ASCVD secondary prevention LDL-C target from <70 to <55 mg/dL (ESC/EAS 2019; ACC expert consensus), the evidence base for this stricter target derived from drug trials (high-intensity statins, IMPROVE-IT ezetimibe, FOURIER/ODYSSEY evolocumab/alirocumab) — not trials designed around specific LDL-C targets
• The TST trial compared <70 vs 90–110 mg/dL in ischaemic stroke — does not address the current clinical question
• Real-world implementation gap: <25% of ASCVD patients achieve <55 mg/dL; <10% receive ezetimibe; ~1% receive PCSK9 inhibitors

Methods — Trial Design

• Open-label (unblinded to patients and physicians); event adjudication committee blinded to group assignment and LDL-C levels
• 17 sites, South Korea; January 2021 – July 2022 enrollment
• No interim analyses; missing data not imputed
• Additional 2×2 secondary randomisation: statin monotherapy (rosuvastatin or atorvastatin) vs statin + ezetimibe — to balance drug distribution and enable target achievement
• LDL-lowering escalation sequence: increase statin dose → add ezetimibe → add PCSK9 inhibitor (PCSK9i use constrained by South Korean reimbursement policy)

Methods — Trial Population

• Inclusion: Age 19–80 years; documented ASCVD (≥1 of: prior ACS, stable angina with imaging/functional confirmation, coronary or arterial revascularisation, stroke/TIA, PAD)
• Key exclusion: LDL-C <70 mg/dL WITHOUT statin therapy (already at target without treatment)
• Baseline characteristics (balanced): Mean age 64.4±9.0 years; 20.9% women; median baseline LDL-C 76 mg/dL (IQR 61–96)
• Prior ACS: 55.6%; stable angina: 48.4%; revascularisation: 67.2%; stroke/TIA: 3.8%; PAD: 8.7%
• East Asian (Korean) population only

Methods — Endpoints

• Primary: Composite of CV death + nonfatal MI + nonfatal stroke + any revascularisation + hospitalisation for UA at 3 years (Kaplan-Meier cumulative incidence; log-rank test; Cox proportional hazards)
• Secondary (efficacy): Individual primary components; composite CV death/nonfatal MI/nonfatal stroke
• Safety: New-onset diabetes; worsening glycaemic control; statin-associated muscle symptoms → therapy change; cancer; cataract surgery; aminotransferase elevation (≥3× ULN); creatinine elevation (>1.5× baseline); CK elevation (≥4× ULN)
• Per-protocol and competing-risk sensitivity analyses performed

Results — LDL-C Achieved

• Intensive group: Median 56 mg/dL (IQR 48–67); 1.4 mmol/L — maintained consistently throughout
• Conventional group: Median 66 mg/dL (IQR 58–76); 1.7 mmol/L
• Between-group LDL-C difference sustained throughout follow-up (~10 mg/dL separation)
• Target achievement at 3 years: 60.8% reached <55 mg/dL (intensive); 68.1% reached <70 mg/dL (conventional)
• 85.2% of intensive group reached <70 mg/dL by 3 years
• Drug intensity at 3 years (intensive vs conventional): high-intensity statin 48.4% vs 32.3%; ezetimibe 66.6% vs 56.7%; PCSK9i 2.3% vs 0.9%
• 39% of intensive group failed to reach <55 mg/dL target at 3 years (limited PCSK9i use; no inclisiran/bempedoic acid available in South Korea during trial)

Results — Primary and Secondary Endpoints

• Primary composite (3-year KM incidence): 6.6% (intensive) vs 9.7% (conventional)
• Primary HR: 0.67 (95% CI 0.52–0.86; P=0.002) — 33% relative risk reduction
• Per-protocol analysis consistent with ITT
• Nonfatal MI: 0.8% vs 1.7%; HR 0.46 (95% CI 0.23–0.91)
• Any revascularisation: 4.8% vs 7.5%; HR 0.63 (95% CI 0.47–0.84) — primary driver of composite
• CV death: 31 (2.0%) vs 29 (1.9%) all-cause deaths — not significantly different
• Nonfatal stroke: Included in composite; individual component not significantly different
• Hospitalisation for UA: Included in composite; individual component not significant
• Composite CV death/nonfatal MI/nonfatal stroke: reported in Table 2; not achieving significance individually
• Competing-risk sensitivity analyses consistent with primary results

Results — Subgroup Analyses

• Hazard ratios for primary endpoint consistent with direction of benefit across prespecified subgroups: age, sex, BMI, prior ACS, revascularisation, stroke/TIA, PAD, hypertension, diabetes, CKD, baseline LDL-C
• No formal interaction term reported as significant

Results — Safety

• New-onset diabetes: No significant difference
• Worsening glycaemic control: No significant difference
• Statin-associated muscle symptoms → therapy change: No significant difference
• Cancer: No significant difference
• Cataract surgery: No significant difference
• Aminotransferase elevation (≥3× ULN): No significant difference
• CK elevation (≥4× ULN): No significant difference
• Creatinine elevation (>1.5× baseline): 1.2% (intensive) vs 2.7% (conventional); P=0.004 — lower in intensive group (unexpected protective signal; mechanism unclear)

Discussion Key Points

• First large RCT to validate the guideline-recommended shift from <70 to <55 mg/dL as a target for secondary prevention — provides direct target-based RCT evidence that was previously lacking
• Mechanism: intensive LDL lowering slows atherosclerosis progression and promotes plaque regression (intravascular imaging studies using high-intensity statins, ezetimibe, PCSK9i)
• Revascularisation accounts for the majority of primary endpoint events and most of the benefit — a softer endpoint than CV death or MI; raises the question of whether the composite is driven by a clinical guideline for intervention rather than a pure event
• 39% of intensive group failed target → real-world PCSK9i use, inclisiran, and bempedoic acid could increase target attainment and potentially amplify benefit
• Real-world context: <25% ASCVD patients currently achieve <55 mg/dL; <10% receive ezetimibe; ~1% PCSK9i — aggressive intensification is needed
• Creatinine protection signal: several studies suggest statins are renoprotective; intensive LDL lowering may slow CKD progression, but mechanism and long-term significance require further study
• Current guidelines mention <40 mg/dL for extreme-risk patients — this lower threshold not tested in Ez-PAVE

Limitations of the Document

• Open-label — physician and patient knowledge of target could influence revascularisation decisions (the main driver of benefit), potentially inflating the primary endpoint difference; event committee was blinded to group and LDL-C
• Primary endpoint event rate lower than assumed: Anticipated 15% 3-year conventional-group rate; actual 9.7%. Trial was still powered to detect the observed difference but with less precision than planned
• 39% of intensive group did not reach <55 mg/dL target — limited PCSK9i reimbursement; no inclisiran/bempedoic acid available in South Korea during trial; true "on-target" comparison is diluted
• Benefit driven predominantly by revascularisation (soft endpoint): No significant reduction in individual hard endpoints (CV death, stroke); HR for MI (0.46) significant but based on small absolute numbers (0.8% vs 1.7%)
• East Asian (Korean) only: May not generalise to other ethnicities with different ASCVD risk profiles, statin pharmacokinetics, or drug access
• Short 3-year follow-up: LDL-lowering trials typically show greater separation of mortality/hard endpoint curves over longer periods (e.g., FOURIER median 2.2 years showed less CV death reduction than ODYSSEY OUTCOMES at 2.8 years)
• No imputation for missing LDL-C values — potential downward bias with underestimation of true LDL-C levels
• Revascularisation endpoint includes any revascularisation — heterogeneous endpoint combining elective and urgent procedures; open-label design raises concern about differential threshold for referral
• PCSK9i use minimal (2.3% intensive vs 0.9% conventional at 3 years) — does not reflect current practice in high-resource settings where PCSK9i are more widely accessible

Key Concepts Mentioned

• concepts/Dyslipidemia-Management — direct evidence for <55 vs <70 mg/dL target comparison in secondary prevention; first RCT validating the stricter guideline target
• concepts/ASCVD-Risk-Assessment — secondary prevention population; ASCVD definition used for enrolment

Key Entities Mentioned

• entities/Chronic-Coronary-Disease — 48.4% stable angina; 67.2% revascularisation history; most patients represent CCD population
• entities/Acute-Coronary-Syndrome — 55.6% prior ACS; directly relevant to post-ACS LDL management

Wiki Pages Updated

• wiki/sources/intensive-ldl-ezpave-nejm-2026.md — created
• wiki/concepts/Dyslipidemia-Management.md — Ez-PAVE RCT evidence added; contradiction added; source added; source_count 4→5
• wiki/sourceindex.md — new entry added
• wiki/wikiindex.md — Dyslipidemia-Management concept entry updated