Two siRNAs cut lipoprotein(a) by more than 90%. Nobody has shown it prevents anything yet

Lipoprotein(a) is the cardiovascular risk factor that has been measurable for decades and untreatable for all of them. It is largely genetically determined, it is a causal risk factor for atherosclerotic cardiovascular disease and aortic stenosis, and its effect is thought to be mediated in part by oxidised phospholipids bound to apolipoprotein(a) and apolipoprotein B [s3]. Elevated concentrations are associated with the development of atherosclerotic cardiovascular disease [s1].

Two papers published within four days report drugs that lower it a great deal. Neither reports that lowering it helps anyone.

The first-in-human trial

Kylo-11 is described as a non-canonical, very long-acting small interfering RNA targeting lipoprotein(a) [s1]. The phase 1 trial was randomised, double-blind and placebo-controlled, conducted at a single hospital-based site in China in otherwise healthy adults aged 18–55 with elevated lipoprotein(a) [s1]. Participants were randomised 8:2 to a single subcutaneous dose of Kylo-11 or placebo across seven cohorts [s1].

Cohorts 1–6 enrolled participants with lipoprotein(a) of 75–200 nmol/L at doses of 9, 30, 75, 225, 450 and 600 mg; cohort 7 enrolled participants above 200 nmol/L at 225 mg [s1]. Between 30 May and 30 December 2024, 71 participants were randomised — 57 to Kylo-11, 14 to placebo — and 70 received a dose [s1]. Median age was 27.5 years (IQR 22.0–32.0); 46 (65%) were male [s1]. Median follow-up was 337 days (IQR 334–337) [s1].

The primary endpoint was investigator-assessed adverse events within 24 weeks [s1]. Thirty-seven of 70 participants (53%) had adverse events, mostly grade 1–2 and judged unrelated to study drug [s1]. There were no injection-site reactions, no serious adverse events, no drug-related adverse events and no deaths [s1]. Two grade 3 or higher events — transient hypertriglyceridaemia and elevated creatine phosphokinase, both at day 168 in the 225 mg cohort — were adjudicated unrelated to study drug, as was one grade 3 or higher event in the placebo group at day 84 [s1].

The pharmacodynamic results were the point. At 48 weeks after a single dose, median reductions in serum lipoprotein(a) ranged from −53% (IQR −71 to −41) and −71 nmol/L (−79 to −60) in the 9 mg cohort to −97% (−98 to −97) and −129 nmol/L (−153 to −99) in the 600 mg cohort [s1]. In cohort 7, with baseline above 200 nmol/L, the 225 mg dose produced median reductions of −96% (−98 to −91) and −208 nmol/L (−222 to −198) [s1].

The authors conclude that a single dose was well tolerated and, at 225 mg or higher, durably reduced lipoprotein(a) up to 48 weeks [s1]. The trial was funded by Kylonova Biopharma and registered as NCT06363851 [s1]. A linked Comment in the same issue is titled "Deep and durable lipoprotein(a) lowering: Kylo-11 advances RNA interference therapeutics" [s2].

The mechanism study

Three days later, a post hoc analysis of a phase 2 trial of lepodisiran, an extended-duration siRNA targeting lipoprotein(a), reported on the oxidised phospholipids [s3].

The parent phase 2 trial randomised 320 participants at 66 global centres [s3]. The analysis included 213 of them who received placebo or lepodisiran 16, 96 or 400 mg at baseline and day 180 and had oxidised phospholipid measurements at baseline, day 240 and day 360 [s3]. Median baseline levels were 122.0 nmol/L (Q1, Q3: 105.1, 137.7) for OxPL-apo(a) and 27.7 nmol/L (22.5, 36.2) for OxPL-apoB [s3].

At day 240, placebo-adjusted geometric mean changes in OxPL-apo(a) were −21.9% (95% CI −45.0 to 11.0), −65.1% (−73.8 to −53.6) and −94.2% (−95.7 to −92.2) at 16, 96 and 400 mg [s3]. For OxPL-apoB the figures were −39.5% (−51.2 to −25.0), −76.9% (−80.6 to −72.5) and −88.5% (−90.4 to −86.3) [s3]. By day 360 the effects had attenuated: OxPL-apo(a) changes were −23.7% (−47.0 to 9.8), −41.6% (−56.4 to −21.6) and −80.7% (−85.7 to −73.9) [s3]. An accompanying editorial is titled "Taking the Sting Out of Lipoprotein(a)" [s4].

What neither study is

Neither is an outcomes trial. Both measure a biomarker.

That distinction has cost the lipid field before. A drug that moves a causal biomarker in the right direction is a good bet, not a demonstrated benefit. Lipoprotein(a) is described in the lepodisiran paper as "largely genetically determined" and causal for atherosclerotic cardiovascular disease and aortic stenosis [s3] — but lifetime genetic exposure is not the same experiment as pharmacological lowering begun in adulthood, and neither of these papers tested the latter against clinical events.

Two further limits are specific. The Kylo-11 trial enrolled 70 people with a median age of 27.5 who were otherwise healthy, at a single site [s1]. That is a first-in-human safety and pharmacodynamic study and nothing more; the population is decades younger than the one that would be treated. The lepodisiran analysis is post hoc, on 213 of 320 participants, in a phase 2 trial [s3].

What to watch

The only result that will settle this is a cardiovascular outcomes trial in the class. Until one reports, the correct summary is that lipoprotein(a) has gone from unmodifiable to modifiable, and that whether modifying it prevents heart attacks remains an open question with a strong prior.

This article describes trial results. It is not medical advice, and nothing here should be used to start, stop or change treatment.

Sources

• Safety and lipoprotein(a)-lowering effects of Kylo-11, a non-canonical, long-duration small interfering RNA targeting lipoprotein(a): a first-in-human, randomised, double-blind, placebo-controlled, phase 1 trial, The Lancet, 28 August 2026
• Deep and durable lipoprotein(a) lowering: Kylo-11 advances RNA interference therapeutics, The Lancet, 28 August 2026
• Effect on Lipoprotein(a) Oxidized Phospholipids of Lepodisiran, an Extended-Duration siRNA Targeting Lipoprotein(a), Journal of the American College of Cardiology, 31 August 2026
• Taking the Sting Out of Lipoprotein(a): Oxidized Phospholipid Lowering With Lepodisiran, Journal of the American College of Cardiology, 12 August 2026