UPDATED: A Combination Therapy that Eliminates COVID-19 Virus
Subsequent studies cast doubt on the results of small French Trial that showed exciting results in a treatment for COVID-19.
[UPDATED 25 March 2020 to include more details about the trial and its limitations]
[UPDATED 21 April 2020 to include results from a retrospective analysis of SARS-CoV-2 patients in the US and another from France.]
[UPDATED 9 May 2020 to include results from an observations study of COVID-19 patients in hospitalized in New York City]
9 May 2020 Update: An observational study of 1376 patients hospitalized in New York City found no significant benefit of hydroxychloroquine treatment. The patients receiving hydroxychloroquine were sicker than the other patients. The authors analyzed the data in two ways. First, they compared the 811 patients who received hydroxychloroquine with all 565 who did not. Second, they compared the 811 patients receiving hydroxychloroquine with those patients with similar characteristics (age, sex, race and ethnic group, body mass index, smoking status, various co-occurring conditions (like chronic lung disease, diabetes, cancer, chronic kidney disease, among others), medications, initial vital signs, and initial laboratory test results. This second analysis is referred to as the “propensity-score — matched” analysis, because it matches patients in each group according to how sick they were and what other conditions they had. The endpoints used were either death or intubation.
By just comparing all of the hydroxychloroquine-receiving patients with all of the other patients, the hydroxychloroquine-receiving patients fared worse (32.3% died or were intubated); whereas only 14.9% of the patients that did not received hydroxychloroquine died or were intubated. However, the propensity-score — matched analysis, which accounted for the differences among patients within the groups, found not significant difference between the two groups. Based on these results, the clinical center where this was done removed the suggestion of using hydroxychloroquine to treat COVID-19 patients.
J. Geleris, et al. Observational Study of Hydroxychloroquine in Hospitalized Patients with Covid-19. N. Engl. J. Med. 10.1056/NEJMoa2012410 (2020). DOI: 10.1056/NEJMoa2012410
21 April 2020 Update: A retrospective study of 368 patients in the United States Veterans Health Administration medical centers found that hydroxychloroquine alone significantly increased the risk of death compared with patients who received hydroxychloroquine + azithromycin or neither drug. Indeed, 28% of patients receiving hydroxchloroquine alone died, 22% receiving both drugs died, and only 11% of those who did not receive either drug died. A caveat of this retrospective study is that the patients receiving hydrochloroquine with or without azithromycin tended to be sicker with more severe disease based on clinical measures.
J. Magagnoli, et al., Outcomes of hydroxychloroquine usage in United States veterans hospitalized with Covid-19. medRxiv 2020.04.16.20065920 (20 April 2020) DOI: 10.1101/2020.04.16.20065920 [Non-peer reviewed preprint]
A retrospective study of 181 patients in 4 French hospitals found no benefit to hydroxychloroquine treatment when comparing the risk of developing acute respiratory distress syndrome or death. Of the 84 patients who initially started hydroxychloroquine treatment, 8 had to discontinue due to heart complications (a known hazard of this drug). A caveat to this study is that the number of patients who died was very low in both groups, making the number of events not statistically informative. A confounding issue is that 8 patients in the group that did not receive hydroxychloroquine within 48 hours of admission did subsequently receive hydroxychloroquine later. Some patients in the hydroxychloroquine group also received azithromycin or other antibiotics.
M. Mahevas, et al., No evidence of clinical efficacy of hydroxychloroquine in patients hospitalized for COVID-19 infection with oxygen requirement: results of a study using routinely collected data to emulate a target trial. medRxiv 2020.04.10.20060699 (14 April 2020) DOI: 10.1101/2020.04.10.20060699 [Non-peer reviewed preprint]
Collectively, these studies show that hydroxychloroquine can be dangerous to patients hospitalized with COVID-19 and that widespread adoption of such a treatment is not appropriate.
A vaccine is not going to be available for many months, long after many people have died in the absence of treatment or suppression of the spread of the new coronavirus SARS-CoV-2. But there is light starting to shine on this horrible situation. Researchers have begun clinical trials for treatments that reduce the viral load in patients and will hopefully speed recovery and reduce lethality.
In an ongoing Chinese trial, researchers found benefit in giving patients chloroquine, a drug used for decades to treat malaria. With results from >100 patients so far reported, chloroquine reduced symptoms of pneumonia and viral load, improved lung morphology on images, and shortened the duration of the disease. These data have not been published, just a summary of the findings has been reported.
A second promising trial just published results, and it doesn’t take a statistician to see the effects. Not only is the trial good news because it seems to work to reduce viral load, but the drugs tested are already approved for use in humans for other conditions and they are not expensive. So, repurposing these for COVID-19 therapy should be easy if additional tests support the results.
French clinician scientists tested hydroxychloroquine, which is a safer version of chloroquine that has been used to treat malaria for decades, and azithromycin, an antibiotic approved by the FDA in 2002, and the combination of these 2 drugs in COVID-19 patients. The combination reduced detectable virus in some patients tested within 6 days. Even hydroxychloroquine alone was effective in eliminating detectable virus in 70% of the subjects. Adding azithromycin to the remaining virus-positive patients eliminated the detectable virus.
The patients in the trial ranged from 10 years old to 60 years old, included males and females, and represented severe disease, mild disease, and asymptomatic disease. However, the trial only included 36 total subjects, so the number of subjects in each patient group is low. Additionally, not everyone can take hydroxychloroquine or azithromycin, but these results are extremely encouraging.
With social distancing and other practices in place to slow the spread, there is a real chance that treatments will become available before the death toll becomes high throughout the world.
NEW SECTION [25 March 2020]: Caveats and Limitations
This trial has been criticized by many scientists and clinicians, and there are many caveats that need to be understood about the trial design and the data.
The method and design of this trial is open-label, non-random. This means that the doctors and patients knew who was receiving the treatment and who was in the control standard of care group. It also means that patients were intentionally selected for treatment. For example, children (< 18 years old) were only included in the standard of care group; they were excluded from the treatment group based on the information about reduced risk of severe symptoms in patients of this age group.
The treatment group was comprised of patients at a single center in Marseille, where the researchers with the experience with this treatment were located. The patients receiving standard of care only were in 4 different locations throughout the south of France. The patients in the control group from the center in Marseille included those who were excluded from participation because they did not meet the study criteria or refused the experimental treatment.
Additionally, the data are only presented through day 6, even though the study design and methods indicate that the study lasted 14 days. However, the methods clearly state that the primary endpoint of this part of the study is “virological clearance at day-6 post-inclusion.”
In the graph presented above, the data from some patients are missing. One patient still had detectable virus at day 6 after receiving hydroxychloroquine and was given azithromycin on day 8. The virus was undetectable at day 9. Other missing patient data includes 6 hydroxychloroquine-treated patients: 3 patients had symptoms that became worse and were transferred into intensive care, 1 patient died, 1 patient left the hospital, 1 patient stopped treatment because of side effects (nausea). So, of the original 26 patients in the treatment group, the graph only shows data for at most 19 patients.
- Study only evaluated detectable virus in nasal and throat swabs.
- Study did not include data about patient outcomes (survival, length of hospital stay, and need for assisted respiration).
- Even some patients in the treatment group got worse or died.
Gautret et al. Hydroxychloroquine and azithromycin as a treatment of COVID‐19: results of an open‐label non‐randomized clinical trial. International Journal of Antimicrobial Agents — In Press 17 March 2020 (2020) DOI : 10.1016/j.ijantimicag.2020.105949 Article
Gao J, Tian Z, Yang X. Breakthrough: Chloroquine phosphate has shown apparent efficacy in treatment of COVID-19 associated pneumonia in clinical studies. Biosci. Trends. 14, 72–73 (2020). DOI:10.5582/bst.2020.01047
Hydroxychloroquine. DrugBank (accessed 19 March 2020) https://www.drugbank.ca/drugs/DB01611
Azithroymcin. DrugBank (accessed 19 March 2020) https://www.drugbank.ca/drugs/DB00207
Update with information about WHO Trials (23 March 2020)
Update with information about chloroquine and hydroxychloroquine (25 March 2020)