What’s your take on the Novartis Horizon trial? Their drug appeared to do a good job of reducing Lp(a), but “did not meet its primary endpoint of reducing the risk of cardiovascular events” compared to placebo.
Good question. For background, Lp(a) is a type of low density lipoprotein (LDL) that has an additional protein, apo(a), attached to it. They both carry cholesterol around the body and both generally contribute to heart disease in a dose-dependent manner, i.e. higher levels produce higher risk.
Previous trials have lowered Lp(a) levels and shown a variety of effects, but all were confounded. For example, trials using niacin dropped Lp(a) only 20 to 30% and did a dozen other things, including some harm. PCSK9 and CETP drugs lowered LDL far more than Lp(a), making it hard to identify any Lp(a) specific effects. So, HORIZON is the first large trial attempting to isolate Lp(a) reduction tested against hard outcomes.
Well, it didn’t work.
Earlier this month Novartis announced that Lp(a)HORIZON missed its primary endpoint, a composite of CV death, non-fatal MI, non-fatal stroke, and urgent revascularization. All this, despite the investigational drug, pelacarsen, substantially reducing Lp(a) in 8,323 people with established cardiovascular disease and Lp(a) ≥70 mg/dL. That’s interesting!
I still haven’t seen the details of the trial such as the hazard ratio, confidence interval, event rates, secondary endpoints, or any sub-group analyses…so it’s hard to know what exactly happened.
Still, there are a number of possible explanations for why lowering Lp(a) didn’t seem to help here:
- Five years of lowering is too short to offset decades of cumulative exposure
- Optimized background care (mean LDL ~65 mg/dL, ~78% on high-intensity statins, leaving little Lp(a)-attributable residual risk)
- Insufficient Lp(a) lowering
- Index event bias
Upcoming trials include the OCEAN(a)-Outcomes with olpasiran (>95% lowering, entry threshold ≥200 nmol/L) and the ACCLAIM-Lp(a) study with lepodisiran (for primary prevention!) amongst others. We’ll see what happens.
My take is that the guideline-recommended one-time test of Lp(a) is still reasonable to guide management. We just don’t know yet whether lowering Lp(a) itself is beneficial, either in those with established disease or for those without (primary prevention). I suspect lowering Lp(a) alone likely does help reduce risk, provided it’s reduced to a low enough level for long enough. What I’m less confident on is how much it helps. Is it incremental or large? And does it depend on whether someone already has heart disease vs. primary prevention? We’ll know more after some of these other trials.
-Jordan
3 Likes
Thank you for the detailed reply. Looking forward to more information on this trial and results of the other trials.
I prefer “Optimized background care (mean LDL ~65 mg/dL, ~78% on high-intensity statins, leaving little Lp(a)-attributable residual risk)”, since a combination of meds has lowered my LDL well below 65 mg/dL.
For a one time test, my Lp(a) has bounced around a lot on repeated tests.
Yea, there’s a decent amount of variability in the test, for sure. We’ll see where the evidence takes us!