Three Lessons From the 'Negative' ZEUS and HORIZON Trials

Makers of the interleukin-6 (IL-6) blocker, ziltivekimab, and the lipoprotein(a) (Lp[a])-lowering drug pelacarsen announced that their drugs failed to reduce cardiac outcomes in large phase 3 randomized trials. At the upcoming American Heart Association meeting, we will learn the details of the ZEUS and HORIZON trials, but the announcements are clearly a disappointment for proponents of the drugs. John Mandrola, MDI see three lessons for cardiologists and researchers alike. But first let's lay out why experts thought these drugs had promise. Lowering Lp(a) and IL-6 Passes the Plausibility TestIt's remarkable how much sense these drugs made — on a whiteboard. First, both pathways are tied to atherosclerotic cardiovascular disease (ASCVD). In the case of IL-6 blockade, we know that inflammation has a role in atherosclerosis. We also know that two anti-inflammatory drugs have modestly lowered ASCVD events. Canakinumab, a blocker of IL-1-beta, which is upstream from IL-6, modestly reduced major adverse cardiac events in the CANTOS trial. The less specific anti-inflammatory drug colchicine reduced cardiac outcomes in the LoDoCo2 and COLCOT trials, though it failed in sicker myocardial infarction (MI) patients in the CLEAR SYNERGY trial. Plausibility for Lp(a) lowering seems stronger than any anti-inflammatory agent. The low-density lipoprotein (LDL)-like particle has been shown to promote clotting, inflammation, and fatty plaque. A 2009 meta-analysis of 127,000 people in 36 prospective studies found that higher Lp(a) predicts coronary disease and ischemic stroke. The relationship was continuous and linear. The second causal connection of these two pathways is genetics. In the case of IL-6, a 2012 Mendelian randomization study found that people who carry a variant in the gene for the IL-6 receptor, one that dampens IL-6 signaling (similar to the arthritis drug tocilizumab), had about a 5% reduction in the odds of coronary disease events for each copy of the variant they inherited. And gene variants largely determine Lp(a) levels, which correlate in a continuous way with ASCVD risk. ZEUS and HORIZON Trials and Preliminary Results ZEUS randomized more than 6300 patients with established ASCVD, chronic kidney disease, and a high-sensitivity C-reactive protein (hsCRP) ≥ 2 mg/L to ziltivekimab 15 mg per month subcutaneously or placebo. The press release announcing the null findings told us that ziltivekimab had no effect on the primary endpoint of cardiovascular (CV) death, MI, or stroke (hazard ratio [HR], 0.99; 95% CI, 0.88-1.11). This, despite previous observations from a phase 2 trial that the drug substantially lowers the inflammatory marker hsCRP.Even worse than a null primary endpoint with tight confidence intervals was that serious infections were higher in the ziltivekimab arm. In early September, Novo Nordisk announced early termination of the HERMES and ATHENA trials, which were studying the drug in patients with heart failure. We know a little less about the HORIZON trial of pelacarsen vs placebo. The trial enrolled slightly more than 8300 patients with an elevated Lp(a) and established ASCVD. That press release noted that the trial "missed" its primary endpoint, a composite of CV death, MI, stroke, and urgent coronary revascularization. This was despite the drug lowering Lp(a) levels, though it's not yet clear by how much, but a phase 2 study described an 80% reduction with a similar dose to that used in HORIZON. Three Lessons Learned The first lesson applies mostly to clinicians. The key message is that no matter how much sense an intervention makes, you need trials to prove a benefit in hard outcomes. Antiarrhythmic drugs after MI made great sense in the 1980s until the CAST trial found them to be harmful. Now over three decades later, ZEUS and HORIZON reconfirm the necessity of evidence in the practice of medicine. Plausibility is surely part of clinical appraisal; you would be suspicious of a treatment with no plausible mechanism. But plausibility is not enough to accept a treatment. For this you need positive outcomes trials. The second lesson applies to both clinicians and researchers: Incremental progress is hard. While ASCVD remains a large public health problem, two forces have made new progress in therapeutics challenging. First is the decline of ASCVD events over time — due to measures including smoking cessation, improvement in air quality, and removal of trans fats from the food supply. Second is the huge reduction in cardiac events from drug therapy, primarily from statins, but more recently, GLP-1 receptor agonists. Statins reliably reduce ASCVD events by 20%-25%, and in the SELECT trial, semaglutide further reduced cardiac events by 20% in patients with cardiovascular disease and overweight/obesity. People age and aging brings atherosclerosis, so there is a ceiling to how much more event reduction is possible. And even if anti-inflammatory or Lp(a)–lowering drugs (or others) were successful, the price per event reduction would likely be too high for widespread acceptance. The third lesson applies mostly to researchers (but also to clinicians): Atherosclerosis is complex. For instance, with statins, we make the simple correlation between LDL-C reduction and benefit, but surely statins have non-LDL-C-lowering mechanisms of benefit. If it were simply lipid lowering, all drugs that reduced lipid levels would reduce events — but they don't. ConclusionThe null trials don't rule out benefit from other types of inflammation blockade or different Lp(a)–lowering drugs. More trials are planned in these two pathways. But ZEUS and HORIZON underscore the need for humility when it comes to finding additional therapeutic benefit in the complex disease atherosclerosis. I will keep an open mind, but for now I'm skeptical that another game changer for chronic atherosclerosis lies hidden and yet unfound. John Mandrola practices cardiac electrophysiology in Louisville, Kentucky, and is a writer and podcaster for Medscape. He espouses a conservative approach to medical practice. He participates in clinical research and writes often about the state of medical evidence.