May 26, 2020

expert reaction to WHO temporarily suspending a trial into hydroxychloroquine

The World Health Organisation (WHO) have announced the temporary suspension of a trial into hydroxychloroquine.

Prof Peter Horby, Professor of Emerging Infectious Diseases and Global Health in the Nuffield Department of Medicine, University of Oxford, said:

“The WHO decision to temporarily suspend hydroxychloroquine in their trial is based on data from a study that has looked at routinely collected data from hospitalised patients, and which reports an increased risk of death in patients who were given hydroxychloroquine.  However, these types of studies are difficult to interpret because the decision to give the drug will be based on the severity of disease in the patient. It’s a bit like giving diabetic drugs to diabetics: the drug doesn’t cause the disease, you are given the drug because of the disease. The authors have tried to control for this “indication bias” in the study but it is very hard to do it fully.  In response to that paper we looked very carefully at our data over the weekend, to make sure we are not putting patients at risk. Since RECOVERY patients are randomised, our data are much less vulnerable to the biases that plague studies that use routine health care data. An independent committee has looked at our data and did not see any safety concerns. We discussed our findings with Medicines and Healthcare products Regulatory Authority, who have agreed with our interpretation that the data provide reassurance that continued enrolment into the hydroxychloroquine arm is safe and that we should press ahead with getting a reliable answer on hydroxychloroquine through the RECOVERY trial.”

Prof Robin Ferner, Honorary Professor of Clinical Pharmacology, University of Birmingham & Honorary Consultant Physician, City Hospital Birmingham, said:

“On 22nd May 2020 the medical journal The Lancet published a study by Professor Mehra and colleagues on chloroquine and hydroxychloroquine in patients with COVID-19. The study looked at data from 671 hospitals worldwide on more than 96,000 patients who had tested positive for the SARS-CoV-2 virus, 16% of whom had been treated with chloroquine or hydroxychloroquine. They excluded patients who were on mechanical ventilators, or those whose treatment was delayed by more than 48 hours from diagnosis—which meant that information on the sickest patients was not analysed.

“Since the choice of treatments was not random, the authors of the study set out to correct the raw data by allowing for a wide range of factors that could affect a patient’s outcome. These factors included obesity, sex, ethnic group, co-existing heart disease, smoking, and two measures of disease severity. They did not correct for some other factors likely to influence the outcome, such as the country or health system in which treatment took place. The treatments were not allocated at random. This makes it impossible to even out the effects of unknown factors that might affect the result, and which are allowed for by assigning treatments at random.

“The mortality in in the patients who received neither chloroquine nor hydroxychloroquine was 9%, and in those who received either of the drugs it was 21%. Even when the authors allowed for all the known factors influencing outcome, the mortality was higher in the patients treated with chloroquine or hydroxychloroquine; and higher still if they also received azithromycin or a related (macrolide) antibiotic.

“The authors correctly concluded that in their study they were ‘unable to confirm a benefit of hydroxychloroquine or chloroquine, when used alone or with a macrolide, on in-hospital outcomes for COVID-19. Each of these drug regimens was associated with decreased in-hospital survival and an increased frequency of ventricular arrhythmias when used for treatment of COVID-19.’

“These and other data[1] offer no encouragement to those who are promoting hydroxychloroquine as a treatment for COVID-19. However, they do not provide definitive evidence, because the treatments were not allocated at random, so cannot allow for unknown influences. 

“WHO was in a difficult position, since this study might mean that (hydroxy)chloroquine does more harm than good. In the circumstances, suspending its trial was reasonable. But we really won’t know the answer to the question ‘Does (hydroxy)chloroquine help patients with COVID-19 at all’ until we have the answers from randomized controlled trials. One such trial, RECOVERY, has recruited several thousand patients, and its lead investigator, Professor Peter Horby, said today that a preliminary analysis had shown no evidence that hydroxychloroquine increased mortality.

“So let’s hope that the full data from that trial are available soon and give a clear answer to the question.”

[1] https://www.bmj.com/content/369/bmj.m1432

Dr Reecha Sofat, clinical pharmacologist, University College Hospital, said:

“I think you do have to use caution when interpreting observational versus randomised studies. This decision brings to the fore that granting off label use without evidence can lead to harms. This is particularly true when the biological rationale for the treatment is not that convincing – it is better to wait for the results of a properly designed trial before using potentially harmful interventions. Incidentally, the national RECOVERY trial running at Oxford is looking at this very issue and is likely to report soon – this should help inform decision-making.”

Prof Trudie Lang, Director of The Global Health Network, Nuffield Department of Medicine, University of Oxford, said:

“The WHO temporarily halting the use of Chloroquine in COVID-19 clinical trials highlights why we need to run carefully designed clinical trials during outbreaks. This enables us to learn as quickly as possible about whether potential therapies can tackle the virus and are safe. Properly designed and managed Clinical trials are the only way we can see whether drugs might also cause harm. They are designed to assess the safety of the drug relative to the ability to bring any benefit.

“We have long known that Chloroquine can cause harmful cardiac related side effects from the use of Chloroquine in the treatment of malaria. Recent trials carried out in China of a lower Chloroquine dose didn’t show efficacy in relation to Covid 19. A higher Chloroquine dose could bring increased risk of harmful side effects.

“Most countries have restricted the use of chloroquine (and hydroxychloroquine) to be used in the treatment of COVID-19 only within a clinical trial. This is so the use and safety can be carefully monitored and data captured in comparison to other treatments, or standard care, and answers on safety and efficacy accurately assessed. Using chloroquine ‘off license’ or outside a trial does not help answer these questions and can also present a risk to the patient and the risks to the individual remain unknown furthermore each individual might not be adequately monitored to prevent any possible harm.

“The WHO have stopped the use of Chloroquine because they have had independent safety review committees assessing the clinical trial data in real time as these studies have progressed. These review panels are standard in clinical trials and this process would have identified accurately that more side effects were being reported in the patients taking Chloroquine. Therefore, it was completely appropriate to recommend this drug is withdrawn from these studies as soon as this signal was observed and until this data is fully analysed and reviewed.”

Dr Stephen Griffin, Associate Professor in the School of Medicine, University of Leeds, said:

“The WHO has taken the wise precaution of halting arms of their wide-ranging therapeutics trial relating to Chloroquine (CQ) and Hydroxychloroquine (HCQ). This is largely based upon a study published last week that retrospectively analysed thousands of patients receiving these drugs as part of their COVID19 treatment, either alone or in combination with antibiotics. The study could find no evidence for a beneficial effect in patients taking these drugs. However, worryingly, patients appeared to be at heightened risk of cardiac complications, especially when taking CQ or HCQ alongside macrolide antibiotics.

“Whilst not the gold standard of a phase three, randomised, placebo-controlled trial, such a broad retrospective analysis of real-world patient data is certainly one of the next best options available. Previous COVID19 studies on CQ and HCQ have yielded conflicting outcomes, many of which may have been attributable to their relatively small size. It must be remembered that undertaking large-scale trials is an investment both in clinical resource and financially, but patient safety must remain absolutely paramount. If it is not obvious from preliminary studies that a clear patient benefit is likely, which out-weighs any risk, then it is difficult to justify moving to the next stages under normal circumstances. Clinical investigation of CQ and HCQ was fully justified at the outset due to the urgency and severity of the COVID19 crisis, yet it is equally important to recognise new data and to act quickly in the best interests of patients.

“We should remember that the mechanism by which CQ and HCQ are thought to act against SARS-CoV2 is not specific to this virus, instead acting against the cellular processes that mediate the internalisation of infectious virus particles. Thus, these drugs could conceivably act against a broad range of viruses that also hijack the same cellular pathway, yet this has not previously been extensively explored. However, it is not well understood whether such outcomes are achievable using these drugs in the context of human tissue, which may explain a potential lack of antiviral potency in patients compared to cell culture. By contrast, increasing evidence points to adverse physiological effects relating to a dysregulated heartbeat in COVID19 patients. Thus, in my view it is pertinent to pause ongoing trials to fully evaluate whether harmful effects of these drugs in COVID19 patients are indeed more prevalent than any potential benefit. Moreover, these drugs have existing applications for autoimmune conditions for which they are well suited and essential for many patients’ wellbeing; they should be preserved for such use. Going forward, it would seem wise at present to avoid the use of these medicines “off-label” for the treatment and/or prevention of SARS-CoV2 infection, and promotion of such use by public figures ought to be rigorously challenged until such time that supportive, well-controlled, clinical data are available for genuine patient benefit.”

Prof Stephen Evans, Professor of Pharmacoepidemiology, London School of Hygiene & Tropical Medicine, said:

“The Solidarity randomised trial was designed to compare several treatments and the actual treatments used could vary from country to country:

1. lopinavir with ritonavir

2. hydroxychloroquine (the original protocol allowed for chloroquine also but this was not used)

3. remdesivir

4. lopinavir with ritonavir plus Interferon beta-1a..

5. Control: optimized support care

“WHO has announced that the Executive Committee (not the Data Monitoring Committee) has suspended recruitment into (and possibly continuing treatment in but that is not clear) the hydroxychloroquine arm. 

“This decision is largely because of concerns raised by regulators and investigators following The Lancet observational study. There are data from smaller studies and some randomised evidence but none so far contradicts the large observational study with any reliability.

“It is always difficult to decide to suspend a trial or one arm of a trial, and it will usually be done on the basis of data from that trial, combined if necessary with information from other studies, especially randomised trials. It is not ideal, but it is difficult in a pandemic to be sure that it is correct, but the first requirement is to “do no harm”. This seems a sensible decision to review as much randomised data as possible to allow the trial treatment to be continued (if either benefit or at least no harm was seen). It will depend on co-operation from the other trials, but it could be argued that they have a public health duty to provide confidential data.

“WHO acknowledge that large randomised trials are necessary and is asking for data from these other large trials. They are aware of at least 7 other trials, including the UK “RECOVERY” trial, which include hydroxychloroquine as one of the treatments. The SOLIDARITY trial data monitoring committee will then look at the unblinded data from its trial and from any other trials who contribute data before making a decision to either lift the suspension of the hydroxychloroquine arm or to terminate it.

“The SOLIDARITY trial has randomised in total 3500 patients though not all have been followed to obtain an outcome.

“The COPCOV trial is rather different; it is using hydroxychloroquine to try and prevent Covid-19 or reduce its severity in those infected with SARA Cov-2, and so is given to currently healthy people. The WHO, RECOVERY and similar trials are using hydroxychloroquine as treatment for diagnosed disease, generally in hospital patients. It is theoretically possible that hydroxychloroquine could be dangerous in treatment but effective in prevention, but this seems rather unlikely.

“In trials of a treatment intended for prevention of disease, it is absolutely vital that healthy people are not harmed. It is known that hydroxychloroquine is effective in malaria and some auto-immune diseases and so the benefit/risk balance is positive in such conditions, in spite of the known harms that are not necessarily frequent. The benefit/risk balance in healthy people, but who are at risk of Covid-19 disease is unknown and that is why the COPCOV trial is being done. It is a difficult decision for them. Suspending recruitment or treatment could damage the trial temporarily or permanently and to do so prior to the outcome of the SOLIDARITY trial investigation of the data could be regarded as premature, but it is for the COPCOV trial committees to decide and it won’t be easy.

“We do not have totally reliable evidence that hydroxychloroquine is beneficial for treatment or prevention of Covid-19. There is some, non-randomised and thereby less certain, evidence that it increases mortality. There is good evidence that, especially combined with azithromycin or similar antibiotic that relatively rare heart rhythm abnormalities are notably increased, as expected from what is known about these drugs. I argued in comments on The Lancet observational study (22 May 2020) that it may be only ethical to continue to give patients hydroxychloroquine in the context of a randomised trial. That view seems not only to be supported by WHO, but their actions to obtain as much randomised data as possible before continuing with hydroxychloroquine for treatment, reflect those concerns that hypothesised benefits are not seen in practice.”

All our previous output on this subject can be seen at this weblink: www.sciencemediacentre.org/tag/covid-19

Declared interests

Prof Peter Horby: Chief Investigator of the RECOVERY trial

Prof Stephen Griffin: “No conflicts”

Prof Trudie Lang: “No COIs”

Prof Stephen Evans: “No conflicts of interest. I am funded (1 day/week) by LSHTM.  They get funding from various companies, including Astra Zeneca and GSK but I am not funded by them, I have no involvement in obtaining funding from them and I am not an investigator or any grants obtained from them.  I am the statistician to the “meta-Data Safety and Monitoring Board” for CEPI [https://cepi.net/].  I will probably be paid for my attendance at meetings and expenses for travel.”

None others received.