A COVID trial that recruited 29 of its target, and what it still reports

Design

randomised controlled trial, open to standard care comparator, terminated early

Population

29 men with mild COVID-19 in Qatar (target sample size not reached)

Primary outcome

feasibility, safety and preliminary efficacy of transdermal 17β-estradiol

Effect

well tolerated, no thromboembolic events; underpowered for any efficacy estimate

Early observational work suggested women on oestrogen therapy had less severe COVID-19, with a plausible mechanism in oestrogen-mediated immunomodulation and downregulation of ACE2, the receptor SARS-CoV-2 uses to enter cells. A randomised trial in Qatar set out to test transdermal 17β-estradiol added to standard care.

It did not get there. Approvals came in March 2021, protocol amendments delayed recruitment to February 2022, and by then vaccination coverage and revised national quarantine policy had removed most of the eligible hospitalised population. The trial was terminated in June 2022 with 29 men with mild COVID-19 enrolled, 16 to estradiol and 13 to standard care. The treatment duration had already been cut from 10 days to 7 to fit changed quarantine rules.

What can be said is that it was tolerated, with no adverse safety signals and no thromboembolic events - a relevant negative in men given transdermal oestrogen. What cannot be said is anything about whether it works. The value of publishing this is the operational record: a pandemic trial that was overtaken by the pandemic, which is a recurring and under-reported reason that infectious disease questions go unanswered.

• Draw no efficacy conclusion: 29 participants against a much larger target cannot support one.
• The tolerability observation - no thromboembolic events in men on transdermal oestrogen - is small but worth noting.
• Treat the ACE2 mechanism as hypothesis, unchanged by this trial.
• When designing time-limited infection trials, build in the possibility that the eligible population disappears.
• Publishing terminated trials prevents the same question being funded again without knowing why the last attempt failed.

Why it matters

It documents a failure mode that keeps recurring in outbreak research: the trial is ready just after the patients stop existing.

Don't overread it

This was underpowered by termination, so 'no efficacy shown' here means 'not tested', not 'does not work'.

The statistics, in plain English

With 16 and 13 participants per arm, the trial has almost no ability to detect a difference in any clinical outcome - an effect would have to be enormous to register, and its absence tells you nothing. The safety observation is on firmer, though still limited, ground: rare events would not appear in 16 people, but common ones would.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.