Hydroxychloroquine
Rachel Balick
For a while there, hydroxychloroquine (HCQ) became the new toilet paper: Everyone thought they needed it to stay safe during COVID-19, and resulting demand placed stress on supply.
HCQ had support from high places. President Trump called it “a game changer” at a White House press briefing on March 19. “What do you have to lose? Take it. I really think they should take it,” he said at another briefing on April 4. But in late April, FDA cautioned against its use for COVID-19 outside of a hospital setting. By June 15, FDA had put the kibosh on the whole thing—it revoked the drug’s emergency use authorization (EUA).
How did we get here?
Promise
Researchers had been exploring the use of antimalarials like HCQ and chloroquine (CQ) against viral infections for decades, so the recommendation wasn’t totally implausible. CQ had shown promise against the 2002 outbreak of sudden acute respiratory syndrome (SARS), another coronavirus, by inhibiting viral replication.
In addition, some hypothesized that CQ or HCQ could inhibit immune response that could lead to lung and other organ damage in patients with COVID-19.
Trump’s initial endorsements came from two sources. A letter in BioScience Trends summarized the experience of a Chinese medical group, reporting that “results from more than 100 patients have demonstrated that chloroquine phosphate is superior to the control treatment in inhibiting the exacerbation of pneumonia, improving lung imaging findings, promoting a virus negative conversion, and shortening the disease course.” The letter, however, included no additional detail about study design, patient selection criteria, outcomes, or safety. Read the letter at https://apha.us/BioSciTrends.
An International Journal of Antimicrobial Agents study out of France included more detail but still had significant limitations. Read that study at https://apha.us/HCQFrance.
The French study had a very small sample size—42 patients were enrolled, of whom 16 were controls. Still, it found that patients who were positive for COVID-19 and were treated with HCQ, sometimes in combination with azithromycin depending on a patient’s clinical presentation, had a significantly reduced viral load compared with controls.
HCQ hype was born. Though evidence supporting their effectiveness was scant, FDA approved HCQ and CQ for COVID-19 treatment by EUA on March 28. EUA was granted based on SARS-CoV-2’s potential to cause life-threatening disease; a reasonable belief that HCQ/CQ could effectively treat COVID-19 based on the limited in vitro and anecdotal clinical data in case series, and that the drugs’ known and potential benefits outweighed their known and potential risks; and the lack of an adequate, approved, and available alternative treatment.
Literature about HCQ and COVID-19 began to proliferate, but most consisted entirely of speculation (https://apha.us/Speculation), hypothesis (https://apha.us/Hypothesis), calls for further study (https://apha.us/FurtherStudy), or trial design proposals (https://apha.us/Proposal).
Value as a prophylactic
When President Trump announced in May he’d been taking the drug as prophylaxis, controversy boiled over again.
Signs that HCQ had some value in preventing COVID-19 came from early reports and passing comments out of China indicating that patients taking HCQ for autoimmune disorders and as an antimalarial had not developed COVID-19 (https://apha.us/EarlyReports). But these reports were based on survey or observational data, not a clinical trial, and soon there were case studies of HCQ patients contracting the disease (https://apha.us/HCQpatient).
One open-label, nonrandomized, noncontrolled trial out of South Korea did offer some degree of useful information on HCQ’s efficacy as postexposure COVID-19 prophylaxis. Read the study at https://apha.us/PEprophylaxis.
A long-term care hospital social worker (the index case) was diagnosed with COVID-19 following 3 days of fever, malaise, myalgia, and cough. They had been at work for 2 of those 3 days, coming into contact with many patients and staff.
The study considered all 193 inpatients to have been exposed to the virus, as the social worker had visited every floor of the facility, including the ICU. Twenty-four employees were also considered to be exposed and were sent home to quarantine for 14 days. All patients and workers were tested, and one worker (the second case) received a positive result although they were not known to have encountered the index case. That worker had a headache during a 6-hour shift but wore a mask the entire time. Still, 14 additional workers were sent into quarantine based on contact with the second case.
The intervention was 400 mg of HCQ once a day for 14 days as postexposure prophylaxis (PEP)—all patients and 29 workers were to receive the intervention. Two patients died of non-COVID-19 causes and were excluded. Two patients and one worker refused to participate, and six workers quit their jobs; they were also excluded.
One to two days before the end of the HCQ regimen, no participating patients or workers beyond the two original cases had tested positive for COVID-19 by polymerase chain reaction (PCR) test. Though this could indicate HCQ PEP efficacy, there’s no way to know what would have happened without intervention. The study had other limitations. The study design has inherent flaws, and there’s no way to know if staff quarantines, patient transfers, masks, or other interventions affected the outcomes. And, if HCQ does indeed inhibit viral replication, PCR tests may not have been sensitive enough to detect viral loads.
Perils
There was scant evidence in favor of HCQ and CQ even when President Trump touted it. But then multiple studies cast doubt on HCQ’s effectiveness and suggested some patients suffered negative consequences or experienced hastened deaths.
By April 24, FDA had issued a warning that HCQ and CQ should not be used for COVID-19 outside of a hospital or clinical trial. The drugs, especially in combination with QT-prolonging drugs like azithromycin, were reportedly associated with serious heart rhythm problems. Read FDA’s statement at https://apha.us/FDA_HCQ_caution.
The HCQ story demonstrates the significant real-world impact of inappropriately applying conclusions from data and misunderstanding a study design’s limitations.
The most recent and most dramatic example is a May 2 Lancet article by Mehra and colleagues that was subsequently retracted. The retrospective, multinational registry analysis found a higher risk of mortality as well as a higher risk of ventricular arrhythmia for patients who received either HCQ or CQ with and without an added macrolide antibiotic. Read the study at https://apha.us/LancetStudy.
Lancet authors focused on in-hospital mortality, with secondary outcomes of clinically significant ventricular arrhythmia during hospitalization—defined as a first occurrence of a nonsustained but at least 6 seconds long or sustained ventricular tachycardia or ventricular fibrillation; progression to mechanical ventilation; total hospital length of stay; and total ICU length of stay.
The study subsequently found statistically significant hazard ratios for both outcomes in all four treatment groups.
It seemed like a final answer to the HCQ/CQ question—it was well-designed with a large sample size, inclusion of multiple institutions across multiple continents, well-matched control groups, and minimized confounds. But much like everything we think we know about HCQ/CQ, all wasn’t quite as it seemed.
Upon its publication, media seized on the study, the World Health Organization halted other trials, and patients and providers altered therapeutic choices. But soon after came data integrity concerns, and 200 scientists and researchers have called for the release of its full methodology and data set to allow for peer review; this had not been done at press time. Data from Australia doesn’t match information provided by the Australian government. It is customary in a big data study like the one in Lancet to include machine learning code and full data set, but the authors did not do that in this case. Even more suspiciously to some, the authors refused to provide information about contributing institutions and countries. They cited privacy concerns, but the data was already de-identified, and the large sample size should make such concerns negligible.
On June 5, the authors retracted the study. Read the retraction statement at https://apha.us/LancetRetraction.
FDA calls it quits
Finally, on June 15, FDA revoked HCQ’s EUA for COVID-19. “Based on its ongoing analysis of the EUA and emerging scientific data, [FDA] determined that chloroquine and hydroxychloroquine are unlikely to be effective in treating COVID-19 for the authorized uses in the EUA,” the agency said. “The totality of scientific evidence currently available indicate a lack of benefit.” Read the agency’s statement at https://apha.us/RevokedEUA.
For the latest on all things COVID-19, take advantage of the resources at www.pharmacist.com/coronavirus.