Questions over WHO decision to pause global hydroxychloroquine trial

Anastasia Tsirtsakis

4/06/2020 5:18:08 PM

Studies drawing on data collected by a little-known US company have been called into question by the medical community.

Tedros Adhanom Ghebreyesus
WHO Director General Dr Tedros Adhanom Ghebreyesus. (Image: AAP)

Reports questioning the reliability of data that informed recent COVID-19 studies published in two leading medical journals have been called ‘disturbing’.
One of the studies, published in The Lancet, showed coronavirus patients taking hydroxychloroquine were at higher risk of heart problems and death – results that led the World Health Organization (WHO) to halt its global Solidarity trial on use of the anti-malaria drug as a treatment for COVID-19. That decision was overturned this week.
‘On the basis of the available mortality data, the members of the committee recommended that there are no reasons to modify the trial protocol,’ WHO director general Dr Tedros Adhanom Ghebreyesus said during a press conference on Wednesday.
‘The executive group received this recommendation and endorsed continuation of all arms of the Solidarity trial, including hydroxychloroquine.’
The study, led by vascular surgeon Dr Mandeep Mehra, claims to have analysed Surgisphere data collected from nearly 15,000 patients with COVID-19 who received hydroxychloroquine alone or in combination with antibiotics from 1200 hospitals around the world.
Among them were five Australian hospitals, which had recorded 600 positive cases and 73 deaths as of 21 April, according to Surgisphere – contrary to data from Johns Hopkins University that shows only 67 deaths by 21 April.
Surgisphere’s CEO and study co-author Dr Sapan Desai attributed the discrepancy, which was identified by The Guardian, to the accidental inclusion of a hospital in Asia.
However, further investigation by The Guardian found that Surgisphere did not work with several Australian hospitals, and that a number of its employees have little or no background in science or data analysis.
‘It appears that they collected some data on Australian cases without having informed any Australian researchers, and there’s no known repositories of data for the Australian cases,’ Dr Gaetan Burgio, head of the transgenesis facility at the John Curtin School of Medical Research at the Australian National University, told newsGP.
‘So this all casts doubt on the validity of the study, and it’s particularly concerning in the middle of the pandemic.’
To date, the peer-reviewed study has been disputed by 120 doctors.
Meanwhile, questions have also been raised about a study published in the New England Journal of Medicine (NEJM) using the Surgisphere database, which found angiotensin-converting-enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) were not associated with a higher risk of harm in COVID-19 patients.
It is unclear how Surgisphere, initially established as an education company publishing medical textbooks, came to own an international database with access to data for tens of thousands of patients around the world. Its methodology for the collection and analysis of that data in such a short period is also not clear.
‘It is not something that any hospital could realistically do,’ Peter Ellis, chief data scientist of international management consultancy Nous Group, told The Guardian.
‘De-identifying is not just a matter of knocking off the patients’ names; it is a big and difficult process. I doubt hospitals even have capability to do it appropriately.
‘It is the sort of thing national statistics agencies have whole teams working on, for years.’
Given the possible implications, Dr Ian Musgrave, a molecular pharmacologist and senior lecturer at the University of Adelaide, told newsGP the WHO’s initial decision to postpone its trial was ‘totally sensible’.
But more than anything, he says, the Lancet case highlights the importance of data transparency, particularly in the pandemic environment where research is being published at a rapid pace.
‘We’re living in an age of big data where we rely on access to very large databases, and the question here is mostly around, is it credible? Are the databases they claim to have access to really available? How are they curated? Do they have the information they claim they have?’ Dr Musgrave said.  
‘Most journals, I believe [including] The Lancet itself, have stipulations that data should be placed in a data repository so other people can check. But in this case apparently there are confidentiality and commercial competence agreements between lawyers and these data are not available.
‘It might be 100% legitimate; they might have all this data there. Or the data may be there and they’ve made a simple error in doing the data analysis.
‘But we have no way of checking.’
Dr Musgrave said journals must ensure data requirements are stringently applied to all studies that come through, and doctors and scientists must continue to point out inconsistencies.
Hydroxychloroquine has been touted as a possible treatment for COVID-19, with a number of trials taking place around the world, including one looking at the drug as a possible prophylactic against COVID-19 for Australian healthcare workers on the frontline. 
But, to date, there is no evidence proving the drug is effective, with the National COVID-19 Clinical Evidence Taskforce cautioning against its use outside of a clinical trial.
More than 5000 articles have been published on COVID-19 since the pandemic started, with more than 1100 clinical trials underway.
‘There’s a lot of hype and research is rapidly being published. The Lancet case shows the important role rigorous processes play,’ Associate Professor Julian Elliot, Chair of the National COVID-19 Clinical Evidence Taskforce, told newsGP.

‘We do have to move quickly, but while following rigorous processes – otherwise there could be dire consequences.’
Log in below to join the conversation.

coronavirus COVID-19 hydroxychloroquine WHO

newsGP weekly poll Would you be willing to provide a firearms health assessment for your patient?

newsGP weekly poll Would you be willing to provide a firearms health assessment for your patient?



Login to comment