In Seniors With Lupus, No Flare Worries When Stopping HCQ

Withdrawal of hydroxychloroquine (HCQ) in older patients with stable systemic lupus erythematosus (SLE) — primarily out of concern for ocular toxicity — did not increase the risk of disease flares, a retrospective study found.

Among 26 patients who discontinued HCQ after at least 5 years of use, five patients experienced a lupus flare within a year, as did five out of 32 matched controls who continued on the drug, according to Peter M. Izmirly, MD, of New York University School of Medicine in New York City, and colleagues.

This represented a nonsignificant difference in risk between the groups, with an odds ratio of 1.28 (95% CI 0.31-5.30, P=0.73), the researchers reported online in Arthritis Research & Therapy.

The benefits of HCQ in SLE are numerous and well established, ranging from protection against organ damage and limiting the risk of thrombosis to improving survival.

Earlier reports suggested that the development of ocular toxicity was a rare event, but refinements in ophthalmologic techniques recently have led to estimates that the prevalence of maculopathy may reach 20% after 20 years at daily dosages of 4 to 5 mg/kg and 40% with doses above 5 mg.

There also have been reports of cardiomyopathies with antimalarial treatment, including severe irreversible damage, generally with high cumulative doses.

As treatments for SLE have improved in recent years, there is likely to be substantially increased longevity, which may be associated with elevated risks for adverse events such as maculopathy and cardiomyopathy with long-term drug treatment. Therefore, to investigate the effect of HCQ withdrawal on disease control, the researchers conducted a chart review of patients in three major SLE databases in New York City.

The 26 patients in the withdrawal group were required to have used HCQ at daily dosages of 200 to 400 mg for 5 years or more, to have a disease activity index of 4 or lower, and to have discontinued the drug after age 55.

Almost all patients were female, and the mean age was 60. Disease duration was 24.3 years in the withdrawal group and 17.8 years among controls. Fewer patients in the withdrawal group were on prednisone (7.7% vs 15.6%) or other immunosuppressants (15.4% vs 25%). Disease activity scores were slightly higher among controls (1.8 vs 0.9, P=0.08), and significantly more controls had low levels of C3 or C4 (43.8% vs 12%, P=0.02).

After adjustment for those baseline differences, the risk of disease flare remained nonsignificant, with an odds ratio of 1.31 (95% CI 0.18-9.49, P=0.78).

Moreover, in a propensity score analysis that adjusted for other potential confounders, the risk of flare was still lower, with an odds ratio of 1.18 (95% CI 0.23-6.16, P=0.84).

None of the disease flares in either group were severe, and the number of moderate flares was lower in the withdrawal group (7.7% vs 15.6%, OR 0.45, 95% CI 0.10-2.72, P=0.37).

The majority of flares involved cutaneous manifestations and/or arthritis, and one patient in the control group developed pericarditis.

Among the withdrawal group, the most common reasons for stopping the drug were retinal toxicity, in 42.3%, and patient preference in 34.6%. One patient discontinued because of biopsy-confirmed cardiac toxicity.

No flares were observed among patients whose discontinuation related to ocular or cardiac toxicity.

During the yearlong follow-up, there were no new cases of diabetes, strokes, or acute coronary syndromes in either group, although one patient in the withdrawal group had a lower extremity deep venous thrombosis and pulmonary embolism following surgery; this was managed with a 3-month period of anticoagulation. This single event did not permit the researchers to draw firm conclusions about thrombotic risks associated with treatment withdrawal, they noted.

There were no deaths in either group, but larger populations and longer follow-up will be needed to assess mortality in this context.

A Canadian study from 3 decades ago that was considered seminal and paradigm-changing reported that patients with SLE who stopped taking HCQ were 2.5 times more likely to experience a flare. But the patients in that study had taken the drug for shorter periods and were younger and the results may not be applicable to an elderly population, where disease quiescence is more common.

“Further prospective studies will be needed to confirm these reassuring observations and to assess the potential differences in metabolic, thrombotic, and mortality outcomes from HCQ withdrawal in the elderly lupus population,” Izmirly and colleagues concluded, adding that research should also focus on identifying potential biomarkers of disease flares in older patients.

Limitations of the study included its retrospective design and small sample size.