LAST Friday, hard on the heels of the retraction by The Lancet of its now-notorious paper purporting to show that hydroxychloroquine not only did not help Covid-19 patients, but actually made them worse, came the termination of the hydroxychloroquine ‘arm’ of the UK’s RECOVERY (Randomised Evaluation of COVid-19 thERapY) clinical trials. A huge embarrassment was conveniently overlain by news from Oxford University that sorry, hydroxychloroquine really isn’t any good. So even if The Lancet paper was fake, ‘a political hit job’ as one American doctor had it, Oxford’s clinical trial showed the same result.
Professor Martin Landray, MB ChB, PhD, FRCP, FHEA, FASN, FBPhS, FESC, Professor of Medicine and Epidemiology at Oxford, and deputy chief investigator of the RECOVERY trial, was interviewed by the increasingly astute France Soir. It began well: the trial termination was triggered by a request from the Medicines and Healthcare products Regulatory Agency who were bothered about safety (reasons not explained). In consequence the study was ‘unblinded’, that is, results were revealed. Finding no significant difference between treated and untreated patients, they called off the trial, with Oxford triumphantly announcing that hydroxychloroquine was no good for Covid-19, at least for hospitalised patients.
Since those following the issue already knew that no anti-viral was likely to help very sick late-stage Covid-19 patients, this wasn’t news. The trial design had already been savaged within days of launch; the results were no surprise. Internet sleuths also got to work on the very heavy doses of the drug that were given – 2400 mg in the first 24 hours, a ‘dose fit for a gorilla’ as one critic had it. Quizzed about this, Landray defended the dosage, twice, as being usual for other diseases such as amoebic dysentery. Say again? Hydroxychloroquine is used for lupus and arthritis as well as malaria, but dysentery? As a footnote in medical history, the older chloroquine was used half a century ago in attempts to control dysentery, but Professor Christian Perronne, head of infectious diseases at Garches, France, told France Soir that it had been abandoned before 1976. Was Landray confusing hydroxychloroquine with the hydroxyquinolines, which are used for dysentery? Perhaps all those post-nominals had induced some medical dyslexia.
Landray explained there was no approved dosing for Covid because it was a new disease. Well, yes, but the toxic dose won’t change depending on the illness. Asked whether the UK had a maximum dose for hydroxychloroquine, Landray wasn’t sure, but opined it would be much larger, say six to ten times the trial’s dose. That makes 24 whole grams. NICE says about 490 mg per day for a 75kg adult. In France 1800 mg in a day mandates hospitalisation as a poisoning. Twenty-four grams at one go would be almost certainly lethal, possibly even to a gorilla. So Landray has had notice of some hard questions on dose, which will no doubt be explained in the full report, not yet released.
Others were shocked by the overall mortality: about a quarter of patients had died after 28 days. These is very poor compared with hospitals and regions elsewhere. In New York and the Bouches-du-Rhône department of France, hospital mortality is around 13 per cent. At the University Hospital in Marseille it is 15.6 per cent for the sickest patients, those not just hospitalised, but in intensive care. Either the RECOVERY patients were very ill indeed, or something is badly wrong with the ‘standard of care’ in the 175 NHS hospitals involved.
RECOVERY was scattered over multiple hospitals, a network, and Prof Landray admitted that he did not treat any of the patients himself. Professor Didier Raoult, director of the University Hospital in Marseille, which has diagnosed more than 6,000 cases, described the mortality as ‘effroyable’ – appalling. Raoult published his hydroxychloroquine/azithromycin dual therapy in early March and has since used it in well over 3,000 patients with 18 deaths – about 0.5 per cent. He has, though, always been clear that ‘quand il est trop tard, il est trop tard’. Like other successful clinicians, he insists that early treatment is key. This is because, as Dr Harvey Risch of Yale explains: ‘Early outpatient illness is very different from later hospitalised florid disease, and the treatments differ.’ It is clear from world experience that the early infectious stage (‘just a mild illness’) is very different from the ‘inflammatory’ stage when patients start drowning in their own fluids. You can find a chart of the clinical course here.
What is killing patients is their own body’s immune response in overdrive. Treating this with anti-viral medications misses the target; the infection is no longer their really big problem.
In an interview on Tuesday, Raoult scathingly called RECOVERY ‘the Marx Brothers doing science’. He runs through the defects: lack of positive diagnosis, failure to discriminate different stages of the disease, the huge dosage, lack of virological follow-up, and the shocking mortality – which is when his simile ceases to be funny. The one upside is that no cardiac deaths occurred even with the thumping initial dose, showing that talk of the drug’s cardiac risk is massively overblown. But we did not need to experiment on more than 4,000 very ill people to find that. So more hard questions for RECOVERY’s report; we shall see.
Landray wrapped up his interview with France Soir: ‘We have demonstrated that this drug is no good for this disease whatever one wants to believe.’ Thousands of Professor Raoult’s patients, successfully treated with his hydroxychloroquine combination therapy, probably will go on wanting to believe something different. The cognitive dissonance needs resolving. A clue may lie in one of Landray’s papers published in February and entitled The Magic of Randomisation versus the Myth of Real-World Evidence which I first thought was irony. Only in today’s dreaming spires would it be possible for someone to call ‘real-world evidence’ a ‘myth’, revoking centuries of Oxford’s fabled reliance on primary sources. Rattling dice now determines the answer.
Professor Peter Horby, the chief investigator of the trial, doubled up: ‘The RECOVERY trial has shown that hydroxychloroquine is not an effective treatment in patients hospitalised with COVID-19. Although it is disappointing that this treatment has been shown to be ineffective, it does allow us to focus care and research on more promising drugs.’
His first statement is broadly true, and may well follow from the evidence (though we havn’t seen it all yet). Hydroxychloroquine, not in combination with adjuncts potentially important to its anti-viral effect, isn’t likely to work in very ill hospitalised patients with late stage disease. But plenty of doctors using the drug successfully knew that already. Their goal was keeping patients out of hospital.
Horby’s second statement isn’t so qualified and open to easy misrepresentation, so let’s be accurate now: RECOVERY has shown nothing whatever about the effectiveness of hydroxychloroquine-based combinations in the clinically very different early-stage infections, before the ‘inflammatory’ stage has set in. Horby does however declare his future recommendations, ‘to focus . . . on more promising drugs’. Good to have that made clear, but is this ‘following the science’, or what some major sponsors wish to hear?
It is well that the hydroxychloroquine ‘arm’ of RECOVERY has been cancelled. It was always looking in the wrong place. Will Oxford’s PRINCIPLE trials, based in GP surgeries, looking to treat early-stage disease, start to take note of successful empirical experience around the world, and arrange a trial of the Marseille combination, or the related protocol with zinc supplementation pioneered by Dr Zelenko in New York? Or will Oxford’s latter-day denial of ‘real-world evidence’ continue to blind them to what successful clinicians elsewhere – and sometimes whole countries – are reporting?
With a destructive lockdown ongoing, and the worst deaths per million in the world after Belgium, the UK needs far better than clinical trials run by the Marx Brothers. Wake up, Oxford, and rediscover your empirical inheritance. With a trail of deadly failure behind us, taking an objective look at what others have already done is a pretty small ask.
