You may have been reading and hearing lately about Hydroxychloroquine and its use in either prevention or treatment of COVID-19. An early breakout candidate for treating the novel coronavirus, evidence began to mount against its use as multiple studies showed no benefit to the medication but instead safety concerns. So it has not been surprising to see health authorities all around the world, as well as the World Health Organization, halting hydroxychloroquine studies and recommending against its use. What is surprising is a new vigorous debate about it. Foremost among such, is a curious piece in Newsweek from Yale Professor of Epidemiology Dr. Harvey Risch who characterizes the drug as a “classic example of how extra-scientific factors overrode clear-cut medical evidence.” Dr. Risch, who notes an interest in “epidemiologic methods” in his biography, must have fallen asleep during his methods class. But let me instead tell you my experience with the drug.
In March, I thought Hydroxychloroquine (brand name Plaquenil) was worth additional study in the treatment of COVID-19. As such, I began to refer newly diagnosed coronavirus patients as well as recently exposed patients to one of two clinical trials run by Dr. David Boulware at the University of Minnesota. In the first study, early treatment of COVID-19 with hydroxychloroquine was compared to placebo. In the second study, the efficacy of hydroxychloroquine in preventing disease soon after a high-risk exposure (e.g. a household member with COVID) was evaluated – again with a placebo comparison group. We will return to the result of these trials (and others) later in the post.
I did also receive a number (and by a number I mean a lot) of requests for Hydroxychloroquine to be prescribed as a treatment, a preventive agent and a “just in case.” I did not prescribe the medication in any of these instances for there was not yet evidence that the drug would be beneficial (we have this “first do no harm” motto at work). It is not to say that these requests were without merit. In fact, hydroxychloroquine has been shown to have antiviral activity in the laboratory against both SARS and COVID-19. This occurs by interfering with the angiotensin-converting–enzyme 2 (ACE2) receptor, which is the key binding site for viral “spike protein” (remember the spike protein for future reference – it is the target site for therapeutics and vaccines). It is through this spike protein that COVID-19 enters and infects human cells. A recent study has suggested that receptor-binding domain (RBD) of the spike protein may be a particularly effective vaccine target. While this is all theoretically interesting, there were no actual clinical data to support its use.
So why Bouleware’s study? Two reasons. Firstly, the study was designed as a double blinded, randomized, placebo controlled trial. In this study design, recruited patients were equally likely to receive hydroxychloroquine or placebo – and they would not know. Similarly, the researchers and data analysts would not know to which group a patient belonged. This ‘blinding’ and random assignment are critical study design pieces to ensure validity of patient reporting, data collection and in the statistical evaluation of the results. Only at the very end are investigators ‘unblinded’ to which group received the treatment. These elements make the double blind, randomized, placebo controlled trial the gold standard of clinical investigations – or as Tolkien might say, “One Study Design to rule them all.”
With an additional arm for close contacts, Dr. Bouleware’s research became unique opportunity for those infected with the virus as well as their close contacts to advance medical research. And, as I said above, I thought that hydroxychloroquine (although unproven) was worth studying as both a treatment and a preventive agent (note that Tamiflu for influenza is used similarly – there is a treatment dose as well as a prophlactic dose). The potential for side effects worried me greatly, but the dosing levels in Boulware’s study was lower than that in a parallel Brazilian study. In the Brazilian study 32% of participants suffered an adverse event, 12% had worrisome EKG changes (QT prolongation), 1.1% had an arrhythmia and 0.3% of the study population died.
In July, Dr. Bouleware published the results of his study. There was no benefit in terms of decreased severity of COVID-19 symptoms, in faster resolution of illness or in preventing hospitalization among those taking hydroxychloroquine. A similar negative result was obtained when using Hydroxychloroquine as a preventive agent. After high-risk or moderate-risk exposure to Covid-19, hydroxychloroquine did not prevent illness compatible with Covid-19 or confirmed infection when used as postexposure prophylaxis.
Anybody who relies on a single study, no matter how well designed, as sole support for an argument is misguided. But similar to the Bouleware trial, no benefit has been observed to date in other randomized, controlled trials evaluating hydroxychloroquine for the treatment of Covid-19 (RECOVERY trial, SOLIDARITY trial, Tang et al., BMJ., Cavalcanti NEJM). All of these high-quality studies had been stopped because hydroxychloroquine was providing no benefit at all for patients.
While the studies cited above are not perfect and do warrant critique – from low numbers of enrolled patients, to gaps in diagnostic testing, and self reported patient data – none suffer when compared to those frequently cited by Hydroxychloroquine proponents such as Dr. Risch. These include a study by Gautret et al., which has been criticized for statistical flaws and poor research design. Ninety-five percent of the patients in a study by Million et al. study had a low degree of clinical illness, and authors had no control group or even attempted a comparison. Another widely mentioned study is by Dr. Vladimir Zelenko of Monroe, NY and, not joking, is not even in a peer reviewed journal – it’s a Google Document.
Again, it is not as if Hydroxychloroquine isn’t a reasonable target therapy – I certainly thought so in March. But with mounting evidence, the FDA has revoked the Emergency Use Authorization that allowed hydroxychloroquine and chloroquine to be used for hospitalized patients with COVID-19, and the World Health Organization (WHO) and the National Institutes for Health have ceased trials of its use in this setting on the grounds of lack of benefit.
Succinctly summed up by Dr. Steven Nissen, a veteran clinical trialist at the Cleveland Clinic, “it is probably time to move on and start testing other therapies.”