Methotrexate and Hydroxychloroquine Split on Cardiovascular Outcomes in RA

Damian McNamara

November 13, 2020

No significant differences in major adverse cardiovascular events (MACE) emerged between methotrexate and hydroxychloroquine (HCQ) treatment in a comparison of adults 65 years or older with rheumatoid arthritis (RA). However, researchers reported some elevation in risk for stroke in the methotrexate group and for myocardial infarction (MI) and heart failure in the HCQ group.

The primary outcome, a composite of MI, stroke, or cardiovascular death, had an incidence of 23.39 per 1000 person-years in the methotrexate group vs 24.33 in the HCQ group in this observational study of nearly 60,000 people.

"These results suggest an importance of looking at different individual events of cardiovascular disease rather than the whole 'CV' disease only," Seoyoung Kim, MD, said in an interview. "The other important thing is that the mortality was not significantly different between the two groups."

For example, the researchers reported 256 cardiovascular-related deaths in the methotrexate group and 263 such deaths in the HCQ cohort.

Addressing a Recognized Risk

"It is well known that patients with rheumatoid arthritis have excessive morbidity and mortality," Kim, of the division of rheumatology at Brigham and Women's Hospital and associate professor of medicine at Harvard Medical School in Boston, said at the virtual annual meeting of the American College of Rheumatology.

Among prior studies in this area, the Cardiovascular Inflammation Reduction Trial (CIRT) found no significant reduction in cardiovascular events among people taking methotrexate vs placebo. However, the study of 4786 people was not specific to RA, Kim said. The lack of efficacy on this endpoint prompted researchers to stop CIRT early.

"So what does the conclusion of the CIRT trial mean for rheumatoid arthritis patients?" Kim asked.

To find out, she and colleagues compared risk of MACE among participants newly starting either methotrexate or HCQ. The study included 59,329 people aged 65 and older who were identified through Medicare claims data from 2008 to 2016. Mean age was 74 years, and 80% were women.

The investigators used propensity score matching to control for multiple covariates for demographics, other medications, and comorbidities. Use of other medications was similar between groups, including glucocorticoids, NSAIDs, and statins. Baseline cardiovascular morbidities likewise were well balanced, Kim said.

The hazard ratio for the primary MACE outcome was 0.96 (95% CI, 0.86 - 1.08).

Secondary Outcomes

MI was less common in the methotrexate group, for example, with an incidence of 8.49 per 1000 person-years vs 10.68 per 1000 person-years in the HCQ cohort. This finding was statically significant, Kim said, with a hazard ratio of 0.80 favoring methotrexate.

Heart failure also occurred less often in the methotrexate cohort, with an incidence rate of 8.57 per 1000 person-years vs a rate of 14.24 in the HCQ group. The hazard ratio again favored methotrexate at 0.60.

In contrast, strokes were more common with methotrexate than with (incidence of 7.94 vs 6.01 per 1000 person-years).

Another secondary outcome, all-cause mortality, was not significantly different between groups. There were 821 deaths in the methotrexate group (28.65 per 1000 person-years) and 796 deaths in the HCQ group (31.33 per 1000 person-years).

Studying Causality Next?

Session moderator Maya Buch, MD, PhD, professor of rheumatology at the University of Manchester in England, asked Kim why she found significant differences in some secondary outcomes but not the primary composite endpoint.

"When we think of cardiovascular diseases, we tend to think of them all developing through the same mechanism. But perhaps the exact mechanism might not be identical," Kim replied. The findings do not suggest causality because the study was observational, she added, "but maybe this will lead to a randomized, controlled trial."

When asked for comment, Buch said that the study was "interesting" and "suggestive of differences in type of MACE between the two drugs evaluated," but that there should be caution in interpreting the findings because of the lack of detailed information on RA disease and activity in claims databases, in addition to other factors, even though the investigators made adjustments for known differences through propensity score matching.

The division of pharmacoepidemiology and pharmacoeconomics at Brigham and Women's Hospital supported the study. Kim has received support for Brigham and Women's Hospital for unrelated research from Pfizer, AbbVie, Roche, and Bristol-Myers Squibb. Several other coauthors reported having financial relationships with pharmaceutical companies that make drugs for RA. Buch has disclosed no relevant financial relationships.

American College of Rheumatology (ACR) 2020 Annual Meeting: Abstract 1993. Presented November 9, 2020.

This article originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.