Prof Carl Heneghan & Tom Jefferson
Can antivirals defeat Covid-19?
Can antivirals help us defeat coronavirus? The odds don’t look good. The use of antiviral compounds against respiratory viruses has a chequered history. Hundreds have been tested; very few have made it to market. And even fewer make a difference. What’s more, the evidence of their impact on mortality rates – the most important outcome of all – is thin.
The race for a magic bullet to overcome Covid-19 has been going for months now. At least 254 treatments are currently undergoing development, ranging from antibodies to cell-mediated treatments. And there are over 3,500 trials underway. Twenty-five different antivirals are in various stages of development, with a variety of funders, from industry to academia to governments and private donors, often in cooperation, sometimes in competition.
In drug development, getting to the market first matters. Regardless of who funds the race, the winners (those compounds ultimately ending up with a licensed indication for use in Covid) are likely to be used in massive quantities at great expense. Like some of their predecessors, they may end up in government stockpiles. In 2014, the UK government increased its antiviral stockpile for influenza to over £500m, despite remaining on the shelf – unused – for over a decade.
But amidst this race, it’s important that these trials are done properly, with the right amount of scrutiny. When it comes to the hunt for a suitable coronavirus antiviral, there is evidence that this isn’t always happening.
In testing the effects of antivirals, trials should be registered in a publicly accessible database before the first subject to test them on is signed up. Journals and funders insist on such registration, and the WHO considers it essential for research transparency that strengthens the validity and value of the evidence. Yet nine compounds under investigation do not currently have entries in the clinical trials registers. Their existence and progress are almost invisible. Some compounds may, in the end, be found to provide significant clinical benefit with side effects that are seen to be widely acceptable. However, it is equally important that data on the ‘failed’ compounds are made publicly available.
Problems are further exacerbated if a trial switches the reporting of its main outcomes. A recent trial of Remdesivir reported that it helped patients to recover more quickly. But was this really so? The original plans in the trial were apparently to assess death rates, and the need for mechanical ventilation for patients afflicted with Covid-19. But at the last minute, days before the press release was issued, researchers switched to ‘time to recovery’. In the Oval Office at the White House, Anthony Fauci, director of the US National Institute of Allergy and Infectious Diseases that ran the trial was able to announce ‘a clear-cut, significant positive effect’. Outcome switching, akin to changing the rules of a football match at half-time, is a major problem. Over a third of studies show a discrepancy between what is pre-specified and what is reported.
As a result of this, people are likely to be losing trust in the reporting of medical research. But it gets worse when you consider that journals often overplay benefits and underplay harms in an effort to present a rosy version of reality to those who prescribe the treatment. In one analysis, over 80 per cent of Cochrane systematic reviews did not report data on the main harm outcome of interest; and in the five highest impact medical journals, one in five treatment trials show discrepancies relating to the primary outcome.
In Europe, Remdesivir has now been granted a conditional marketing authorisation based on a patient's 'time to recovery'. The conditional authorisation allows early access to the drug with less complete data than normally expected. The US study underpinning authorisation was stopped early. A significant number of patients were therefore not assessed. And by stopping the trial early, we can’t currently know whether Remdesivir reduces mortality or not. Further studies are promised. But it would be wise to not expect a miracle cure.
