By George Delgado, M.D.
Hydroxychloroquine (HCQ) and its chemical cousin, chloroquine, have been at the heart of a heated debate amongst physicians, scientists, policymakers and politicians. Does HCQ help heal COVID-19 or is it just snake oil?
There are some basic science studies and clinical trials in humans—mostly small and non-randomized—that do support the rational use of HCQ in battling SARS-CoV-2, the virus that causes COVID-19. The science starts with the realization that areas in the world that have endemic malaria, such as Africa and southeast Asia have been relatively spared of COVID-19 compared to the more developed areas of the world such as Europe and North America. Scientists do not think that malaria itself prevents COVID-19, although that is a possibility. Rather, the hypothesis is that the HCQ or chloroquine taken by many people in malaria-endemic areas to prevent malaria, might actually stop the infection with SARS-CoV-2. 
Next, scientists have been busy in the basic laboratory. There, they have demonstrated that HCQ blocks SARS-CoV-2 infections in animal cells in cultures. The versatile molecule appears to block several necessary steps in viral takeover of cells. 
Finally, we know that HCQ modulates the immune system. That’s why it is such a great drug for rheumatoid arthritis and lupus. Wouldn’t those immune-adjusting properties be perfect to ward off the dreaded “cytokine storm” that hits severely ill COVID-19 patients?
The most important study to date that demonstrated HCQ effectiveness in treating COVID-19 was conducted in Michigan by the Henry Ford Health System.  This was a retrospective review—in a sense a natural experiment—whereby the researchers looked back in time and compared patients who had received HCQ and azithromycin (a common antibiotic), HCQ alone or neither.
Although it was not randomized, prospective or blinded, it was a well conducted study. The researchers looked at over 2,000 COVID-19 patients that came through their hospital system. The bottom line was that the patients who received HCQ alone or HCQ with azithromycin did much better that the patients who had azithromycin alone or neither. The doctors who conducted the study felt that the medications worked because the patients were treated relatively early in the course of their illnesses when the HCQ (and possibly the azithromycin) could stop the virus from entering into and hijacking cells. Also, the immune-modulating effects of HCQ would theoretically be much more effective at preventing cytokine storm that stopping it once it was established, just like dousing a small campfire is much easier that battling a raging forest fire. Interestingly, the HCQ group did better than the HCQ plus azithromycin group.
As for safety, the HCQ did very well. Many experts have expressed great concern about the heart side effects (arrhythmias) of HCQ, as well as azithromycin. In the Henry Ford study, the doctors were very careful to monitor the patients closely and they kept the patients’ electrolytes in tip-top shape.Their diligence paid off; there were no cases of serious arrhythmias in the treated patients.
The other often cited study took place in France. This study, by Gautret and colleagues looked at patients who came to a medical facility, some symptomatic, some without symptoms but with positive PCR SARS-CoV-2 tests.  The study showed that the group that received HCQ had a conversion of their tests to negative at a faster rate compared to those who did not receive the therapy. Six patients also received azithromycin and they did even better.
The Gautret study has been hit with criticism because of its small size and because six out of the 26 who were in the HCQ group were “lost during follow-up” mostly because they were transferred to the ICU (i.e. they become sicker). The main end point of conversion of the nasal PCR test from positive to negative, apparently could not be verified in those six patients. However, the fact that the patients worsened suggests that the HCQ failed in those patients and the data certainly were affected by them being dropped from the analysis.
A follow-up study by the same French group, authored by Million, included over a thousand patients. Again, it was a look back in time and did not include a placebo group. The study reaffirmed that, with close monitoring of EKG and electrolytes, HCQ in combination with azithromycin is safe. The patients who did the best in this study, not surprisingly, were those who were younger, had fewer chronic conditions and were less sick with COVID-19, when they started treatment. There was no comparator group that did not receive HCQ.
A high-profile study in New York State, sometimes called the Zelenko study, was authored by Derwand, Scholz and Zelenko, and has not been published in a peer-reviewed journal, although it was purported to be submitted.  This was an interesting experiment where patients with COVID-19 were given treatment consisting of HCQ, zinc and azithromycin for five days if they were older than age 60, had certain chronic medical conditions or were short of breath. They were compared to patients outside of Dr. Zelenko’s practice who were not treated with the triple-therapy. The results were that the 141 patients receiving therapy required hospitalization much less than the ones who did not get the triple-therapy. (2.8% vs. 15.4%) The patients tolerated the treatment well and there were no cardiac therapeutic misadventures.
A post-exposure prophylaxis study was published in the New England Journal of Medicine.  Its aim was to prove or disprove that HQC can prevent illness if taken by people exposed to the coronavirus. It was a cleverly designed, randomized, double-blinded, placebo-controlled study using social media to recruit patients. It did not show any difference in COVID-19-like illness in those who took HCQ compared to those who took placebo. A flaw in the study was an inability to test those who were exposed. Because of the poor availability of testing, they went by patients’ reports of symptoms. Another potential flaw was the young age of those who entered the study (median age 40); since younger patients tend to not get severe disease (or may be more likely to develop asymptomatic disease).
There was a high profile study that did not show any benefit with HCQ, published in the prestigious Lancet journal, that eventually was retracted.  The company that provided the data for the study was called into question. When its leaders refused to allow third parties, including the study’s authors to evaluate the raw data, there was no other option other than retracting the article. This marred the reputation of the authors as well as the Lancet and exposed a business of medical data being treated as a commodity.
Let’s look at some of the other studies examining if HCQ might help people already sick with COVID-19. One of the biggest negative studies took place in England, the RECOVERY trial.  It was a well designed, randomized, controlled trial that was not blinded. About 4700 patients were included. There was no statistically significant difference in the percentage of patients who received HCQ and died compared to those who received usual care. In both groups, about 25% died. There was no increase in heart arrhythmias in those who were given HCQ. Those who advocate for HCQ argue that this study looked at the wrong group of patients — they may have been too sick to respond to HCQ.
Another randomized, controlled study in Brazil, published in the New England Journal of Medicine, compared three groups of hospitalized patients.  One group received usual care, another HCQ and a third HCQ and azithromycin. The authors found no differences in the clinical outcomes after 15 days of starting treatment. There were no heart side effects.
Where does the COVID-19 truth meter lead us? I think we can draw at least a few conclusions and generate a slew of questions.
First, HCQ is not the wonder drug that some have proclaimed it to be. It obviously does not work in very sick patients. Good studies showing benefit are lacking. Second, it does not cause cardiac arrhythmias in the vast majority of those who take it.
The question remains, does HCQ have a role in treating patients before they get very sick? There are flawed studies that certainly suggest that it can help. The task remains for researchers to test the drug, with or without azithromycin and zinc in the group of patients where it has shown the most promise. There are several studies that have been planned to further investigate HCQ. Many have been suspended because of the negative hospital studies.
If HCQ proves to be helpful for early cases of COVID-19, it would not be an unprecedented situation. This might be analogous to using an oral antibiotic to treat a pneumonia in a patient who is not hospitalized. You would not expect that same oral antibiotic to do the trick in a patient who is critically ill with pneumonia in the hospital.
For now, I believe that, like any area of medicine, physicians should be allowed to prudently prescribe HCQ when they feel that their patients might benefit. The potential patients who might benefit include those older than 60 and those with other serious illnesses. Clearly, it is not the right drug for those who are already very ill, in the ICU.
As more data emerge, those who are HCQ advocates or opponents should be willing to accept the proof or disproof of the utility of HCQ. Let the data fall where they may. If it is proven to work, let’s use it. If it is disproven by good studies, continuing to use it would be like prescribing snake oil.
- MF, Bangdiwala AS, et al. A Randomized Trial of Hydroxychloroquine as Postexposure Prophylaxis for Covid-19. N Engl J Med. Published online June 3, 2020. doi:10.1056/NEJMoa2016638. https://www.nejm.org/doi/full/10.1056/NEJMoa2016638
- Mehra MR, Ruschitzka F, Patel AN. Retraction—Hydroxychloroquine or chloroquine with or without a macrolide for treatment of COVID-19: a multinational registry analysis. The Lancet doi: 10.1016/S0140- 6736(20)31324-6 18. https://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(20)31324-6.pdf