Safety of low lipoprotein(a) levels: the FOURIER trial.
Baris Gencer, Robert P Giugliano, Xinhui Ran et al.
✦ AI-curated · Sources linked
In 30 seconds
This analysis of the randomized, multicentre FOURIER trial examined associations between lipoprotein(a) [Lp(a)] levels and safety outcomes in patients with stable atherosclerotic cardiovascular disease receiving evolocumab or placebo alongside statin therapy. Among 25,090 participants with baseline Lp(a) measurements, lower Lp(a) was associated with higher prevalent and incident diabetes risk: for every 50 nmol/L lower Lp(a), the adjusted hazard ratio for incident diabetes was 1.05 (95% CI 1.02-1.08; P = .002). Low Lp(a), including levels ≤13 nmol/L, was not associated with increased risk of the other specified adverse outcomes.
Key findings
- For every 50 nmol/L lower baseline Lp(a), incident diabetes risk was higher (adjusted HR 1.05, 95% CI 1.02-1.08; P = .002).
- Lower Lp(a) was associated with prevalent diabetes at baseline (adjusted OR 1.03, 95% CI 1.02-1.04; P < .001, per 50 nmol/L lower Lp(a)).
- Lp(a) concentration was not associated with incident haemorrhagic stroke, serious bleeding, neurocognitive events, malignancy, or atrial fibrillation, including at levels ≤13 nmol/L.
- Evolocumab was not associated with increased diabetes risk irrespective of baseline Lp(a); in the top Lp(a) decile, adjusted HR was .72 (95% CI .38-1.35).
Why it matters
As Lp(a)-lowering therapies enter clinical testing, safety concerns about very low Lp(a) are relevant. These findings are reassuring for several specified adverse outcomes, while identifying an association between lower Lp(a) and diabetes that warrants attention when interpreting safety data. The diabetes association does not establish that low Lp(a) causes diabetes.
What to keep in mind
The analysis examined associations within a trial population with stable ASCVD; the relationship between Lp(a) and diabetes was not established as causal.