At least 55 countries have ordered large supplies of the anti-malarial drugs Hydroxychloroquine (HCQ) and Chloroquine (CQ) from India, manufacturer of 70% of the world’s HCQs, to treat COVID-19 patients.
There is a fundamental problem: evidence on the efficacy and safety of HCQ/CQ are limited at best and even stands on shaky grounds.
Evidence from laboratory works
The Wuhan Institute of Virology had carried out in-vitro studies for chloroquine and subsequently hydroxychloroquine to look at dosages needed for a substance to kill the virus. The in vitro study was conducted in a laboratory without involving real COVID-19 patients.
Their studies showed that HCQ/CQ could kill SARS-CoV-2, COVID-19 causing pathogen, in low dosages.
Although promising, an in vitro study is carried out in a lab outside the context of the human body. It cannot determine the real response of the very complex human biological systems.
Also, an in-vitro study is not designed to determine the dose needed for sufficient blood’s level to achieve the desired therapeutic effect.
Therefore, human trials on real COVID-19 patients are needed.
Methodological deficits in the first human trial
The first human trial from France claimed HCQ efficacy for COVID-19 patients. This report was published on March 20, 2020, in the International Journal of Anti-Microbial Agents.
Didier Raoult and his University of Marseille team reported their HCQ trial on 36 COVID-19 patients.
The researchers divided the patients into two groups. Group A (20 patients) received 200mg of HCQ three times a day. Group B (16 patients) received usual care without HCQ. The researchers also gave six patients in group A Azithromycin, an anti-bacterial, in addition to HCQ.
On the sixth day, all patients were RT-PCR tested. It turned out that more patients in group A were COVID-19 negative, compared to group B.
Also, all patients receiving a combination of Azithromycin and HCQ tested negative. Raoult claimed that “hydroxychloroquine treatment is significantly associated with viral load reduction/disappearance in COVID-19 patients and its effect is reinforced by azithromycin”.
Scientists later voiced out concerns over the quality of the study.
Five days after Raoult’s article, Kerstin Frie and Kome Gbinigie from The Center for Evidence-Based Medicine at Oxford University comprehensively summarised scientists’ concerns as follows:
First, the actual number of participants in group A was 26, but six patients were excluded halfway through, because of treatment termination. They were excluded because they were admitted to the intensive care unit (ICU), died, or experienced side effects. Data from these six patients were not included in the analysis.
This allows attrition bias because there was data from a portion of patients initially recruited but not analysed.
Second, with only 36 patients, this study is relatively weak. Raoult and his team calculated at least 48 patients were needed for sufficient statistical power. This means, there is a risk that this research might have produced false-positive results, i.e. the therapeutic effect is too prominent due to the small sample size.
Third, one of the patients who tested negative with RT-PCR on day six turned out to be positive again on day eight. This shows another weakness, where there was no medium or long term data on the effect HCQ.
Fourth, the experimental design is not the best trial design to determine the efficacy and safety of a drug in humans.
The best design is a randomised controlled trial, where researchers divide patients randomly into at least two groups, the HCQ/CQ recipient and the non-HCQ/CQ recipient.
Randomisation is very important to minimise allocation bias. For example, intentionally or unintentionally, patients with milder symptoms fall into the HCQ/CQ recipient group so that the result shows that more patients in this group are recovering.
Raoult divides his patients into groups A and B, without randomisation, so it is not free from the risk of allocation bias.
Does it work? Scientists are still uncertain
A group of researchers from China recently reported the first randomized controlled trial. They randomized 150 COVID-19 patients whereby 75 received HCQ on top of the standard care and another 75 received only the standard care.
Their study showed that addition of HCQ did not add to the probability of negative conversion. Negative conversion refers to change in patient’s COVID19 status based on RT-PCR swab test, from positive to negative. While this is the first report of a randomized clinical trial, a systematic review of many such trials is needed to firmly establish HCQ/CQ efficacy and safety.
Recently, another group of researchers published an observational study that analysed 1,376 COVID-19 patients in the United States. They compared clinical outcomes of patients administered with HCQ (811 patients) and without HCQ (565 patients).
The results are inconclusive. The researchers acknowledged that the results of this study could not be used to determine the efficacy and safety of HCQ.
Also, they do not recommend the usage of HCQ beyond research setting. They emphasised that randomised controlled trials are needed to decide whether or not HCQ should be used for COVID-19 patients.
Risks of adverse effects
Apart from the inconclusive results of whether HCQ/CQ are effective to treat COVID-19, risks of adverse effects are threatening.
In general, the most prominent adverse effects of HCQ/CQ lead to retinal damage (eye) and various heart disturbances.
In addition, some COVID-19 patients experience a decrease in liver and kidney function as part of their disease course. As a result, the ability of the liver and kidneys to process and remove HCQ/CQ from the body becomes very limited, hence increasing threat of poisoning.
All this shows that while HCQ/CQ efficacy remains inconclusive, their adverse effects are threatening. Countries around the world should be vigilant and avoid actively promoting its use until evidence is established.
As of May 22 2020, the Cochrane COVID-19 Study Register listed 262 human trials on the use of HCQ/CQ for COVID-19, of which 215 are designed as “randomised controlled trial”. Cochrane has also published a protocol for systematically reviewing the outcome of these studies. These may be one of the most sought after research outcome the world has been eagerly waiting.
The drama surrounding massive production and importation of HCQ/CQ throughout the world has once again served as a reminder that public policy must stand on solid scientific grounds. This is to ensure the government use its resources responsibly, especially in times of scarcity.
Editor and author note: This article is a revision from its original version entitled “Why countries should stop using anti-malarial drugs for COVID19” to respond to the recent retraction of The Lancet paper purporting significant mortality risks from using HCQ for COVID19. This article previously quoted the results of the paper.