Rethinking Inflammation and ASCVD

A common teaching in cardiology holds that inflammation plays a causal role in atherosclerotic cardiovascular disease (ASCVD). The recent announcement that the ZEUS cardiovascular outcomes trial of the anti-inflammatory drug ziltivekimab returned null results has seriously challenged this. John M. Mandrola, MDBefore ZEUS, the belief that inflammation had a causal role in ASCVD was strong enough for experts to have written a detailed ACC Scientific Statement on the topic. The lengthy document, published last September, includes 20 consensus recommendations. The evidence supporting the inflammatory theory stems from many sources. Pathologic specimens have shown inflammatory cells in plaques; observational studies have correlated serum levels of the inflammatory biomarker C-reactive protein (CRP) with cardiac events; and genetic studies have observed that variants of the interleukin-6 (IL-6) receptor, which mimic IL-6 blockade, predict lower ASCVD risk. The data from anti-inflammatory drug studies have been less convincing. While there have been modest reductions in CV events in two trials of colchicine (COLCOT and LoDoCo2) and one trial with the IL-1 beta blocker canakinumab, a post-MI trial of colchicine and a methotrexate study were null. ZEUS Results for IL-6 Inhibitor ZiltivekimabZEUS targeted patients deemed likely to have inflammation as an important causal pathway. Patients had to have elevated high-sensitivity CRP (hsCRP), established ASCVD, and chronic kidney disease. The more than 6000 patients were enrolled at 500 sites in 32 countries. The primary endpoint was myocardial infarction (MI), stroke, and cardiovascular death. The press release from Novo Nordisk said that ziltivekimab did not reduce the primary endpoint vs placebo (hazard ratio, 0.99; 95% CI, 0.88-1.11). The null result occurred despite the observation that the monoclonal antibody against IL-6 did exactly what it was designed to do: reduce both IL-6 and hsCRP levels. The tight confidence intervals render a false-negative finding unlikely. Overall rates of serious adverse events were similar in both groups, but a higher proportion of people treated with ziltivekimab had serious infections compared to placebo. No difference in all-cause mortality was observed.The company also wrote that the two additional ongoing cardiovascular outcomes trials investigating ziltivekimab in heart failure (HERMES) and following an acute MI (ARTEMIS) will continue and are anticipated to read out in the first half of 2027.A Caveat Before the LessonsI don’t expect that the formal trial results will alter the message of the press release, but there is a small chance we could learn something important from the full dataset. Recall that the overall primary outcome of the TOPCAT trial of spironolactone in patients with heart failure and preserved ejection fraction (HFpEF) was nonsignificant, but then we learned of a severe regional variation in effect. ZEUS was conducted in hundreds of centers in 32 countries, so such a variation is possible. The first lesson is that both atherosclerosis and inflammation are likely too complicated for drugs with one mechanism of action to counteract. Perhaps we were lucky with statins. While the LDL-cholesterol lowering from statins is partly why the drug class modifies ASCVD, it is likely that other pleiotropic effects contribute. If it were solely due to LDL-C lowering all cholesterol-lowering drugs would work equally well; that is not the case. The second lesson is that plausibility is never enough for a modern cardiology therapy. ZEUS was set up for success in that high-risk patients with documented inflammation were enrolled. IL-6 had previously been shown to be an important mediator of inflammation. In fact, in an observational substudy of CANTOS, patients who achieved low levels of IL-6 in the canakinumab group had lower levels of CV events. However, in ZEUS, the drug reduced IL-6 and hsCRP levels as designed but had no effect on outcomes. The message is clear: Surrogate markers are fine in early research, but before a new therapy is given to patients, it must be shown to reduce outcomes. In this way, ZEUS casts doubt on drugs that were approved without CV outcomes data — the oral PCSK9 inhibitor enlicitide and the small interfering RNA (siRNA) drug inclisiran, for instance. It also informs the development of lipoprotein(a) lowering drugs. Shouldn’t these drugs have to show reductions in CV outcomes before approval? The third message is that perhaps we were wrong about the causal link between inflammation and ASCVD. Consider that many diseases are known to be inflammatory: rheumatoid arthritis, inflammatory bowel disease, psoriasis, ankylosing spondylitis, and giant cell arteritis are all modified by anti-inflammatory drugs. In atherosclerotic heart disease, the story is far less clear. The fourth message is that drug development is hard. In clinical medicine, we are shielded by the challenge of making successful drugs. At meetings and in journals we see favorable phase 3 trials that are the cumulative work of decades or more. The chance of approval of a new phase 1 drug is less than 10% — and falling over time. Surely our amazing progress in treating ASCVD will make it even harder for new agents to make substantial incremental progress.My guess is that we should expect more ZEUS-like stories. 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.