However, two experts caution that, because of confounding, the trial is unable to answer convincingly the question of whether HCQ can benefit COVID-19 patients.
Wei Tang, with the Departments of Pulmonology and Critical Care Medicine at Ruijin Hospital, in Shanghai, China, and colleagues enrolled patients with COVID-19 from 16 treatment centers in China in February. They posted their findings on the medRxiv preprint server, but their paper has not been peer reviewed. A coauthor told Medscape Medical News the work has been submitted to a journal.
The overall 28-day negative conversion rate of SARS-CoV-2, which was the primary endpoint, was similar in the two 75-patient treatment groups. The Kaplan-Meier estimate for negative conversion rate was 85.4% in the HCQ plus standard of care (SOC) arm, vs 81.3% in the SOC-only group (P = .341). Negative conversion rates for the two groups were similar at days 4, 7, 10, 14, and 21.
Adverse events were reported in 8.8% of patients in the control group compared with 30% in the HCQ group. Diarrhea was the most common side effect, occurring in 10% of patients in the HCQ group vs none in the control group. Two patients in the HCQ arm had serious adverse events; one experienced disease progression, and the other experienced upper respiratory tract infection.
Patients in the HCQ group received a high loading dose of 1200 mg daily for 3 days followed by a maintenance dose of 800 mg daily for the remaining days. Total duration was 2 weeks for patients with mild or moderate disease and 3 weeks for those with severe disease.
No Difference in Relief of Symptoms
The two arms were similar in alleviation of symptoms by day 28: 59.9% with HCQ plus SOC vs 66.6% with SOC alone.
However, the researchers said that in a post hoc analysis, they found a significant reduction of symptoms after adjusting for the confounding effects of antiviral agents (hazard ratio, 8.83; 95% confidence interval, 1.09 – 71.3).
In addition, Tang and colleagues report a significantly greater reduction of C-reactive protein (CRP), a biomarker for inflammation, from baseline to day 28 in the HCQ group in comparison with the control group (6.986 vs 2.723 mg/L).
The authors suggest the alleviation of symptoms may come from HCQ’s anti-inflammatory effects.
The mean age of the patients was 46 years, and 55% were male. Almost all patients had mild or moderate disease; two had severe disease.
Experts Say Study Arms May Not Have Been Comparable
J. Michelle Kahlenberg, MD, PhD, research professor of rheumatology at the University of Michigan in Ann Arbor, told Medscape Medical News that it’s important to note that in the post hoc analysis, 89% of the patients in this trial were receiving other therapy in addition to HCQ.
“When [the researchers] say they saw improvement in symptoms when they removed the confounders, what they actually did was remove the patients from the analysis that got antivirals, and that left 14 patients in each arm,” Kahlenberg said.
Moreover, Kahlenberg noted, 20% of patients who received HCQ had mild symptoms, whereas only 9% of those in the SOC group did.
“We don’t know how those patients played out in the post hoc analysis — whether it was the patients who were really mild that didn’t get the antivirals that were left in the hydroxychloroquine group and that’s why they had a slightly faster resolution of symptoms,” she said.
She said that in this study, the researchers calculated CRP in milligrams per liter, whereas in the United States, it is measured in milligrams per deciliter. The conversion highlights the fact that the reduction in CRP was not terribly noteworthy, she said.
“The patients with COVID who tend to tank and have cytokine storms ― their CRP is much higher,” she said. “So the small improvement in CRP wasn’t that exciting.
“I don’t think this gets us anywhere closer to an answer. It’s another muddy study,” she said.
Similarly, Christopher V. Plowe, MD, MPH, director of the Global Health Institute at Duke University in Durham, North Carolina, told Medscape Medical News he sees no convincing answers in this study.
Plowe, professor of medicine, molecular genetics, microbiology, and global health at Duke, also noted differences between the two groups at enrollment.
For example, the HCQ group had more than three times the number of patients with shortness of breath (22.1% vs 5.9%); more with sputum production (16.2 vs 5.9%); and more with cough (51.5% vs 38.2%). In addition, the average age was 4 years higher in the HCQ group.
“It makes me wonder whether the randomization was truly random,” Plowe said.
Plowe also questioned the authors’ statement that they didn’t see cardiac arrhythmia events, such as prolonged QT intervals. “I can’t see any evidence that they did an EKG on anybody,” he said.
“This study leaves the door open to the possibility that hydroxychloroquine may have a clinical benefit. If there is a benefit, it seems to be related to the drug’s anti-inflammatory properties. If that’s the case, I’m not sure this particular drug, as opposed to others, would be the way to go,” Plowe said.