Remdesivir for COVID-19: randomised study shows similar antiviral effect to placebo

14 May 2020. Related: COVID-19: investigational drugs, COVID-19.

Simon Collins, HIV i-Base

Leaked results from a study of the antiviral drug remdesivir, have reported no benefit when used in late-stage COVID-19. The top-line results were posted online on the WHO website and have since been taken down, but a screen shot is still available, included in a report from STAT news. [1]

The summary results showed that 237 participants hospitalised with severe COVID-19 were randomised to remdesivir (n=158) or control (n=79). There were no differences in clinical outcomes: clinical improvement [HR 1.23 (95%CI: 0.87 to 1.75)] or mortality at 28 days [13.9% vs 12.8%, difference 1.1 (95%CI: –8,1 to +10.3)]. The summary results also include no difference in viral load (PCR) although further details are not included. Both groups reported side effects in about 65% of participants. More people stopped treatment due to side effects in the remdesivir group: 18 (11%) vs 4 (5%).

That is all though. A press release from Gilead Sciences, the company developing remdesivir, confirmed that the results were released before they had been peer-reviewed, and that the study, being run in China, had been stopped early due to low recruitment. [2]

The results are important because they provide randomised data where remdesivir use is compared to a control group not getting this drug. The only other published data on remdesivir is from open-label use in an expanded programme. These results were difficut to interpret because some people recovered (as they would anyway) and some people still died (just showing that remdevisir cannot help everyone in late-stage COVID-19). [3]

However, remdesivir was being used in late stage COVID-19 when it might be more effective much earlier. Remdesivir is an antiviral drug that was previously studied to treat Ebola virus. In-vitro studies showed activity against a range of viruses, including SARS-1, SARS-2 and Ebola virus, though this did not translate to clinical benefit for Ebola. Recently published studies show antiviral activity against COVID-19 and MERS in monkeys and improved clinical results when used early in infection. [4, 5]

Without commenting further on other issues about study design and leaking of early data remdesivir levels of coronavirus have already dropped once COVID-19 has become a severe inflammatory disease. So there is plausibility that earlier use might show a benefit. This is the second time that early data has been leaked before being peer-reviewed and published by STAT news. [6]

A simplified description of COVID-19 has three different stages (which might also overlap), each lasting around a week, and getting progressively worse.[7]

1. An initial viral infection in throat and upper respiratory tract. This will be positive to PCR throat swabs and is when someone will also be very infectious.
2. Move to upper and then lower lungs where this becomes an inflammatory (not viral) disease. The immune response steadily blocks the lungs with pus and dead cells as a result of a cytokine storm, blocking oxygen from getting into the blood.
3. A severe stage of immune activation, with oxygen depletion, risk of major organ failure, use of mechanical ventilators (after medically induced coma) etc.

comment

Mainstream press have jumped on the negative results in articles that fail to analyse the mechanism of action for remdesivir or the stage of COVID-19. The report in the Guardian went further by questioning early access to investigational compounds. [8, 9, 10]

Expanded access programmes were an achievement of AIDS activism that kept thousands of HIV positive people alive until the drugs were approved. If remdesivir – or any other antiviral – shows benefits in early infection, then demands for expanded access before approval will be just as important now.

This highlights the urgency of wider access to testing – and also to run studies in earlier COVID-19 infection.

Two phase 3 remdesivir studies are already ongoing in the UK in moderate and severe stage COVID-19. [11] The Data and Safety Monitoring Boards (DSMB) for these studies should be alert for any signals of earlier benefit.

Note: the paper has now been published in the Lancet. [12]

Although the full results include a numerically faster time to clinical improvement in the remdesivir vs placebo arms for people who started treatment earlier (less than ten days since symptoms), this was not statistically significant (hazard ratio 1·52; 95%CI: 0·95 to 2·43).

Of greater concern is the lack of difference in viral load reductions between the active and placebo group, given that remdesivir would be expected to have a direct antiviral effect.

The EMA has already started a rolling review process for evaluating any decision on remdesivir approval in the European Union. This is a process to help faster evaluation of a new drug submission where additional data can be submitted during the review process. [13]

Compassionate access to remdesivir is already available in many European countries. [14]

NOTE: this article has been updated following publication of the full paper online – see comment and ref 12 below and to update information about approval and access in the EU.. Originally published on 29 April 2020.

References