CHICAGO — Low-dose methotrexate is no better than placebo at reducing key inflammatory markers or cardiovascular (CV) events in high-risk patients on statin therapy with stable atherosclerosis, according to results of the prematurely stopped Cardiovascular Inflammation Reduction Trial (CIRT).
Observational data suggesting fewer CV events in patients with arthritis raised hopes that the generic, anti-inflammatory drug might deliver similar effects as the more costly monoclonal antibody, canakinumab (Ilaris, Novartis).
Yet when the same trialists randomly assigned a similar set of more than 4000 patients in CIRT, 170 patients on low-dose methotrexate and 167 patients on placebo had a nonfatal myocardial infarction (MI), nonfatal stroke, or cardiovascular death (P = .91).
Adding hospitalization for unstable angina leading to urgent revascularization to MACE failed to change the primary outcome (201 vs 207 patients; P = .67).
“To find out at the end of this study that low-dose methotrexate neither lowered what we believe to be a critical pathway in atherosclerosis, the IL-1 to IL-6 pathway, but also did not lower cardiovascular event rates for me is a very informative negative control,” principal investigator Paul Ridker, MD, Brigham and Women’s Hospital, Harvard University, Boston, Massachusetts, told theheart.org | Medscape Cardiology.
“When you take the two trials together, what’s exciting is that we now have a very good sense of where we need to go.”
The results were presented in a late-breaking science session here at the American Heart Association (AHA) 2018 Scientific Sessions and published online simultaneously in the New England Journal of Medicine.
“The inflammatory pathway is not dead. This doesn’t kill it; this hones it down to where you need it to be,” Gervasio Lamas, MD, chief of cardiology, Mount Sinai Medical Center, Miami Beach, Florida, who was not involved in the study, told theheart.org | Medscape Cardiology. “And it’s certainly not dead in the individuals who have the highest degree of cardiovascular inflammation, which is diabetic patients. Obesity is rampant and growing in the US and worldwide.”
CIRT investigators at 417 sites enrolled 6158 patients with a prior MI or multivessel coronary disease and either type 2 diabetes (68%) or metabolic syndrome (32%). The trial was stopped early this spring after a median follow-up of 2.3 years in 4786 patients. The target dose of methotrexate was 15 to 20 mg weekly and titrated via a computerized algorithm. All patients received 1 mg of folate daily.
At 8 months, patients assigned methotrexate rather than placebo had significantly higher alanine aminotransferase and aspartate aminotransferase levels and greater reductions in leukocyte counts and hematocrit levels (P < .001 for all comparisons). There was no clinically relevant change in lipids.
Unlike in CANTOS, however, methotrexate had absolutely no impact on the inflammatory markers of IL-β, IL-6, or high-sensitivity C-reactive protein (hs-CRP).
Although patients in both trials received aggressive lipid-lowering therapies, CANTOS enrolled only patients with residual inflammation, whereas CIRT did not screen for hs-CRP. As a result, the median CRP at randomization was 4.2 mg/L vs only 1.6 mg/L in CIRT, emphasized Ridker.
IL-β to IL-6 signaling, initiated at the NLRP3 inflammasome level, drives the inflammatory process and higher CRP levels, he said. In contrast, methotrexate’s mechanisms of action remain uncertain but likely reflect adenosine-mediated anti-inflammatory effects.
“These two trials — CANTOS positive, CIRT effectively a neutral control — thus point directly toward future work targeting upstream inhibition of the NLRP3 inflammasome or downstream inhibition of IL-6 as potential targets for novel cardiovascular therapeutics,” said Ridker. “Remember this unique pathway has genetic data, vascular outcome data, and clinical data all supporting a future direction.”
Study discussant Sidney C. Smith Jr, MD, University of North Carolina, Chapel Hill, said patients in CANTOS were arguably at higher risk and that more information is needed on the relationship between the efficacy of anti-inflammatory agents at varying low-density lipoprotein cholesterol (LDL-C) levels and statin dose.
He pointed out that mean LDL-C was reduced to well within the 2018 cholesterol guideline levels for intensive statin therapy in CIRT (68 mg/dL) but was higher in CANTOS (mean, 82 mg/dL; 75th percentile, 107 mg/dL).
There was no significant difference between the methotrexate and placebo groups in the individual components of CIRT’s primary endpoint or in the secondary outcomes of all-cause mortality (96 vs 83 patients), hospitalization for heart failure (48 vs 53 patients), and MACE or revascularization (278 vs 288). No subgroup appeared to benefit from treatment with methotrexate.
“Taken together, CIRT and CANTOS provide potentially helpful mechanistic observations about selecting drug effects in targeting inflammation,” Smith noted. “Two trials, LoDoCO and COLCOT, involving colchicine, which also acts on IL-β, should provide valuable information in this regard.”
Oral NLRP3 inhibitors that may affect the IL-β to IL-6 to CRP pathway are also under investigation, according to Ridker.
Commenting to theheart.org | Medscape Cardiology, John T. Wilkins, MD, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, said, “We really don’t have a complete handle on why methotrexate is so effective as an anti-inflammatory agent, but the hypothesis that this would be efficacious was reasonable. So the results are a little surprising, but highly informative.”
Multiple pathways can lead to inflammation and the development of atherosclerotic plaques, but “canakinumab is clearly a much, much more targeted agent,” he said.
“The biggest takeaway is that inflammation is becoming a very reasonable target for the reduction of atherosclerotic disease risk,” Wilkins said. “We need more trials.”
Despite the positive findings from CANTOS, a CV indication for canakinumab remains hung up with the US Food and Drug Administration. In a recent complete response letter, the agency requested more information from Novartis on the subgroup of responders with lower-than-median levels of hsCRP who had a 27% reduction in MACE, compared with an overall reduction of 15% to 17%.
“I don’t know why the FDA is balking,” but it may be because “it was only one study and it’s a new mechanism,” Lamas said in an interview. Cardiologists may be slow to warm to canakinumab without a major advertising push, he suggested, in part because the drug is not well understood and the effect somewhat modest.
Of special note, the sodium glucose cotransporter type 2 (SGLT2) canagliflozin (Invokana, Janssen) was recently approved to prevent MACE in adults with type 2 diabetes and established CV disease, while the SGLT2 inhibitor empagliflozin (Jardiance, Boehringer Ingelheim/Lilly) has carried a CV mortality claim for years.
When surveyed on whether they feel comfortable prescribing SGLT2 inhibitors to their patients with diabetes and high-risk coronary artery disease risk, 84% of attendees responded “yes” following the presentation of the DECLARE-TIMI 58 trial results during the same session.
The study was funded by the National Heart, Lung, and Blood Institute (NHLBI). The trial drug and matching placebo were purchased from Teva Pharmaceuticals, which donated costs for packaging and shipping. Ridker reports receiving grants from NHLBI, Novartis, Pfizer, Kowa; personal fees from Inflazome, Novartis, Pfizer, Corvidia, and CiviBio; and a patent with royalties paid to Siemens and AstraZeneca. Lamas and Smith reported no relevant conflicts of interest.
American Heart Association (AHA) Scientific Sessions 2018. LBS.02. Presented November 10, 2018.
N Engl J Med. Published online November 10, 2018. Full text.