HCQ for RA: Not a Great Idea? Save
Hydroxychloroquine (HCQ) appeared less safe from a cardiovascular standpoint than methotrexate in older patients with rheumatoid arthritis (RA), analysis of Medicare data indicated.
Among RA patients 65 and older with a history of heart failure, major cardiovascular events occurred more frequently in those initiating HCQ versus methotrexate (HR 1.30, 95% CI 1.08-1.56), and all-cause mortality also was increased significantly with HCQ (HR 1.22, 95% CI 1.04-1.43), according to Elvira D'Andrea, MD, PhD, of Brigham and Women's Hospital in Boston, and colleagues.
And rates of hospitalization for heart failure were substantially greater in patients starting HCQ -- especially so for those with no previous history of the condition (HR 1.57, 95% CI 1.30-1.90), the researchers reported in the Journal of the American College of Cardiology.
HCQ was already suspected of being relatively risky for individuals with heart failure, D'Andrea and colleagues noted. In 2016, the American Heart Association put the drug on its list of those "that may cause or exacerbate heart failure," albeit on the basis of "very limited" evidence. The new study was aimed at filling that evidence gap.
D'Andrea and colleagues examined Medicare's database for individuals 65 and older diagnosed with RA from 2008 to 2016 and starting either HCQ or methotrexate for the first time. Those with nursing home stays or with cancer or lupus co-diagnoses were excluded, as were patients receiving other disease-modifying RA drugs. That left the researchers with a total of about 70,000 patients.
D'Andrea and colleagues also developed a model using patients' baseline characteristics to estimate the likelihood that an individual would be prescribed HCQ over methotrexate. Some 27,000 HCQ initiators were then matched according to this propensity score with the same number of methotrexate users for the final analysis of cardiovascular risks.
Overall, no differences in rates of major cardiovascular events (a composite of cardiovascular death and hospitalization for myocardial infarction or stroke) were seen between the HCQ and methotrexate groups (HR 1.03, 95% CI 0.79-1.35). Rates of myocardial infarction and stroke were also nearly identical, as were rates of sudden cardiac arrest and ventricular arrhythmia. Cardiovascular mortality, however, was more common with HCQ (HR 1.17, 95% CI 1.02-1.35). All-cause mortality was also increased significantly (HR 1.10, 95% CI 1.01-1.20) with HCQ.
Those findings were among all included patients irrespective of heart failure history. When stratified according to presence or absence of such history, it became apparent that only those with pre-existing heart failure were at extra risk for most cardiovascular events associated with HCQ. The sole exception was hospitalization for heart failure, which was elevated with HCQ irrespective of past heart failure history.
In an accompanying editorial, however, two rheumatologists at the University of Texas Southwestern Medical Center in Dallas argued that the case against HCQ shouldn't be taken as closed.
"The subgroup analysis findings of increased HF [heart failure] admissions and cardiovascular events in persons with preexisting HF, are exploratory and hypothesis generating and should be interpreted with caution," wrote Elizabeth Blair Solow, MD, MSc, and Bonnie L. Bermas, MD.
They pointed to a number of limitations, not least of which was the focus on patients 65 and older starting on first-line therapy; in most patients, RA is diagnosed and treatment begun far earlier.
Also, median follow-up was 180 and 228 days in the HCQ and methotrexate groups, respectively. Solow and Bermas pointed out that HCQ blood levels normally don't stabilize until about 6 months -- meaning that "one-half of the patients on HCQ had not reached steady state during the study period."
The pair questioned, too, whether cardiomyopathy from HCQ could have developed so quickly, "bringing into question as to whether HCQ itself increased HF hospitalizations."
Another way to look at the data, the editorialists suggested, is that it might not be that HCQ is particularly risky, but that methotrexate is actually cardioprotective. Some evidence points in that direction, Solow and Bermas indicated, though it is far from conclusive. (For one thing, RA itself raises cardiovascular risk and effective anti-inflammatory treatment is believed to reduce it.)
In sum, according to Solow and Bermas, the new study mainly highlights the need for more research to resolve the apparently discordant findings.