An ongoing dialogue on HIV/AIDS, infectious diseases,
June 21st, 2020
Dexamethasone Improves Survival in COVID-19 — Why This Should Be Practice Changing Even Before the Paper is Published
When the news broke last week that the dexamethasone component of the RECOVERY randomized clinical trial was halted because those receiving the drug were significantly more likely to survive, I posted the following:
– Very welcome news, dex is cheap, widely available!
– Demonstrates the power of RCTs vs obs studies, which were conflicting
– How will the numerous ongoing studies of immunomodulators be modified?
– Rx guidelines — act now or wait for more info?https://t.co/qBsaZ1csH2
— Paul Sax (@PaulSaxMD) June 16, 2020
Note my last point, about “guidelines”. These committees have a responsibility to get what they recommend right, and might be slower than clinicians to recommend an intervention with limited information — even if it is potentially life-saving.
But my assumption was that clinical practice would change quickly, awaiting the updating of guidelines. After all, this is what we’ve been waiting for — data from a randomized trial demonstrating a clear benefit. Even better, it’s a readily available, inexpensive strategy — a course of corticosteroids — familiar to us all.
I confess the responses to my post, and comments elsewhere, surprised me. Lots of skepticism. Wow.
The comments fell into several interrelated categories:
Let’s wait for the study to be peer-reviewed and published in an established medical journal before changing clinical practice.
Really? Even when the sickest patients — those requiring oxygen or ventilatory support — were more likely to survive?
(Yes, I keep italicizing that endpoint. Emphasis, you know.)
For the record, here are the results:
Dexamethasone reduced deaths by one-third in ventilated patients (rate ratio 0.65 [95% confidence interval 0.48 to 0.88]; p=0.0003) and by one fifth in other patients receiving oxygen only (0.80 [0.67 to 0.96]; p=0.0021).
When a study stops because of a survival benefit for a life-threatening disease, take note. It’s because continuing the study as originally designed is unethical — those randomized to receive “usual care” would be deprived of a potentially life-saving treatment.
The steering committee has a responsibility of ensuring the safety of trial participants. And remember, they have access to all the study data, even if we don’t.
It’s critical that this information be made available as soon as possible. Patients are being treated today who might benefit, and writing papers and subsequent peer review take time — typically weeks, even with the “warp speed” of COVID-19.
To quote one of the investigators: “Dexamethasone is inexpensive, on the shelf, and can be used immediately to save lives worldwide.”
Why are we getting critical information via press release? I’m inherently distrustful. A press release doesn’t represent actual data.
It’s reasonable to be skeptical of clinical trial press releases, especially when issued by private pharmaceutical companies with multi-million dollar marketing divisions.
These notoriously exaggerate the importance of study results, especially when focused on surrogate markers of disease that may or may not predict clinical outcome.
But consider — this isn’t a press release by a giant company, citing a minor change in an inflammatory cytokine or quality-of-life metric in an open-label study. It’s a respected clinical trials group, funded by the government of Great Britain, and they are reporting a survival benefit from their clinical trial.
To their credit, they early on started doing randomized trials of various COVID-19 interventions while the rest of the globe practiced the therapeutic equivalent of throwing drugs against the wall hoping some of them would stick.
Lopinavir-ritonavir! Interferons! Oseltamivir! Hydroxychloroquine! Azithromycin! Ivermectin!
And it’s not just antimicrobials — virtually every immunomodulator under the sun, some extremely expensive, has found its way to off-label use for critically ill patients with COVID-19. Tocilizumab! Sarilumab! Anakinra! Ruxolitinib! Eculizumab! Any-other-mab! And more …
Yes, it’s hard to keep up — see Table 1 in this recent review for all the various anti-inflammatory approaches tried off-label, with many of these now under study.
If we’re using some of these unproven therapies — and many of us have — why not dexamethasone, which in the RECOVERY trial improved survival?
Here we go again! Haven’t we been burned already multiple times with research on COVID-19, only later to have this information questioned, or retracted?
Quite reasonable to be cautious in this very fast-moving area.
But the infamous research that has “burned” us involved much weaker levels of evidence — little more than anecdotal observations at one extreme and observational studies with likely falsified data at the other.
None has been a randomized clinical trial with a survival benefit.
(Have I noted that result enough times already? Nah.)
I need more details about the study. What were the primary endpoints? The specifics of the intervention? What were the patient characteristics of those enrolled? Did some subgroups benefit more than others? What were the toxicities?
All very reasonable questions! But good news — we have the full protocol available for review. This can answer some of these queries, including the endpoints and description of the exact interventions studied.
It’s a highly valuable document that may allay some concerns that the investigators somehow didn’t conduct the study or analyze the data properly.
And I share the interest in seeing the fully published paper to examine the study results in more detail.
But until it is published, we have now highly favorable results in the most important clinical endpoint in any interventional study — improved survival.
So aside from those patients for whom corticosteroids would be contraindicated, it’s hard to imagine not offering dexamethasone — today — to a person with COVID-19 that requires supplemental oxygen or ventilatory support. They might live longer!
And that’s good news.
Just like getting a fresh video from the sporting archives of Olive and Mabel.
[Originally this post cited the action of a Data Safety Monitoring Board; rather the study Steering Committee assessed that enough patients had been enrolled to answer the question of whether dexamethasone provides benefit. Have modified to reflect this difference.]