Hydroxychloroquine for COVID-19: A Lesson in Study Quality

*Jason Gallagher is an active member of the Society of Infectious Disease Pharmacists.

Of all the issues that coronavirus disease 2019 (COVID-19) has unmasked, one of the biggest may be the power of wishful thinking. Perhaps one of the better-defined examples of this has been the different interpretations on the utility of hydroxychloroquine, an agent that has been polarized in bizarre ways.

Even before the first small study of hydroxychloroquine for COVID-19 was published, hydroxychloroquine was getting attention from clinicians facing early outbreaks. There was a reason for hope, since both hydroxychloroquine and chloroquine have in vitro activity against SARS-CoV-2. They are generally well-tolerated and hydroxychloroquine is widely-available in the US. But in vitro activity does not always predict clinical success, the agents have problematic pharmacokinetics, and the record of effectively treating respiratory viruses with antiviral therapies is unimpressive. Nevertheless, it was worth a look.

The first widely reported study started a whirlwind of effects with which that we are still dealing. It was a report by Gautret and colleagues of virologic outcomes of 26 patients who received hydroxychloroquine compared to 16 who were treated at another hospital and did not.¹ Six patients who also received azithromycin “to prevent bacterial superinfection” were included, and these had the most rapid virologic response. No clinical outcomes were reported, though of the 6 patients excluded from analysis, 3 were transferred to the ICU and another 1 died. However, this was enough to capture the interest of a scared public and desperate healthcare professionals looking for any possible therapy.

As use increased, retrospective cohort studies followed, reporting variable results. Since more severely ill patients were often more likely to receive hydroxychloroquine with or without azithromycin, the results of non-randomized studies were highly dependent on the quality of analysis performed. Some found a benefit, more found no effect, and some suggested harm.^2-5 Meanwhile, the politicization of the drug was well-underway. The announcement by the president that the Gautret results could be a “game-changer” drove prescriptions for hydroxychloroquine and azithromycin to the point of shortages and turned a 70-year-old, 18-letter drug into a polarizing household topic. The publication and of a Lancet paper that concluded that the drugs were actually harmful was cheered by one side, then derided by the other after it was retracted. Hydroxychloroquine was stuck in the middle.

Randomized, blinded clinical trials have since been performed that in my mind have put this issue to rest, at least scientifically. The RECOVERY trial in the UK found no difference in mortality between hydroxychloroquine and usual care in over 4600 hospitalized patients.⁶ Two randomized studies in less severely ill patients found no difference between groups^7,8 and perhaps most damning, a study of hydroxychloroquine as post-exposure prophylaxis did not prevent COVID-19 developing any better than placebo.⁹

As the pandemic has progressed, so has our knowledge about COVID-19 therapy. Unfortunately, the early excitement about hydroxychloroquine has not been warranted, and the time has come to abandon it and move on to explore other possible therapies. At this point the threshold for new data evaluating hydroxychloroquine must be raised to prospective, randomized, blinded studies designed to remove the bias that non-randomized studies inherently create, while we need to revert to using retrospective studies for the hypothesis-generation for which they are intended.

References:

1. ^{Gautret P, et al.} Int J Antimicrob Agents. ^{Published online March 20, 2020. doi: 10.1016/j.ijantimicag.2020.105949}
2. ^{Magagnoli et al., Outcomes of Hydroxychloroquine Usage in United States Veterans Hospitalized with COVID- 19, Med (2020), https://doi.org/10.1016/j.medj.2020.06.001}
3. ^{Geleris J, et al.} N Eng J Med. ^{Published online May 7, 2020. doi: 10.1056/NEJMoa2012410}
4. ^Rosenberg JAMA^{. doi:10.1001/jama.2020.8630}
5. ^{Arshad S et al. IJID: https://doi.org/10.1016/j.ijid.2020.06.099}
6. ^{Horby P et al.} https://doi.org/10.1101/2020.07.15.20151852
7. ^{Mitja O et al} Clinical Infectious Diseases^{, ciaa1009,} https://doi.org/10.1093/cid/ciaa1009
8. ^{Skipper CP, Pastick KA, Engen NW, et al. Hydroxychloroquine in nonhospitalized adults with early COVID-19. A randomized trial. Ann Intern Med. 16 July 2020. [Epub ahead of print]. doi:10.7326/M20-4207}
9. ^{Boulware D et al. NEJM. DOI: 10.1056/NEJMoa2016638}