A few researchers have previously doubted remdesivir efficacy against Covid-19; they are more right now.
“World Health Organization Headquarters and Flag” by US Mission Geneva is licensed under CC BY-ND 2.0Gilead Sciences, a biotech company, developed remdesivir in 2017 to treat Ebola. But it did not work and later repurposed for Covid-19. Remdesivir is very expensive and costs 2–3k USD for a five-day course.
But remdesivir was never approved by the Food and Drug Administration (FDA). The FDA has only granted remdesivir with an Emergency Use Authorization (EUA), which is not the same as official approval that requires more stringent requirements.
The EUA permission was based on only one randomized controlled trial (RCT) funded by Gilead Sciences. But, as more RCTs on remdesivir were published, scientists started doubting the drug’s efficacy, which I detailed here. In brief, remdesivir can work, but not as good as Gilead’s RCT shows, and all published RCTs produce results that differ from one another.
Now that the results of the World Health Organization (WHO)-organized Solidarity trial is out on 15 October 2020, what does it say about remdesivir?
“It’s [Solidarity Trial] a reliable result, don’t let anybody tell you otherwise, because they’ll try to…This is real world evidence.”
It should be noted that the study is still a pre-print. This means that the study is yet to be peer-reviewed and published in a formal journal — a process that takes weeks to months. Regardless, it is highly unlikely that a WHO-organized study would not pass peer-review.
Dissecting the Solidarity Trial
Methods
The phase III-IV trial investigated the effects of four drugs — remdesivir, hydroxychloroquine (HCQ), lopinavir, and interferonβ1a — on the clinical outcomes of hospitalized Covid-19 patients.
Researchers randomized 11,266 patients across 405 hospitals from different countries into either remdesivir (2,750 patients), HCQ (954), lopinavir (1,411), interferon (1,412), interferon plus lopinavir (651), or control (4,088) groups.
The control group received standard care and none of the studied drugs. The control group was further randomized into four groups to match each of the study drug groups in terms of age, disease severity at admission (i.e., oxygen support requirement and degree of lung lesions), geographical location, sex, smoking status, and comorbidities (i.e., diabetes, asthma, and chronic heart, liver, and lung diseases). This ensures that treatment and comparison groups for a given drug are as similar as possible so that results are explained by one variable only — the study drug.
The drugs’ dosages were per standard guidelines. Notably, HCQ has two guidelines: A higher dose for amoebic liver abscess and a lower dose for malaria. The trial used a higher dosage to maximize efficacy. For remdesivir, 200mg was administered on the day of hospitalization and then 100mg for the following nine days.
Taking a step further, this study did a meta-analysis of their own data plus three existing RCTs on remdesivir. The pooled results found that remdesivir does not lower Covid-19 mortality.
Results
HCQ and lopinavir groups were terminated early due to a lack of efficacy. Interferon was also terminated later on, and only remdesivir remains. Patients’ compliance was high at 94–96%, which means most took the drug at least until midway through the trial.
Results found that, the authors wrote, “no study drug definitely reduced mortality (in unventilated patients or any other subgroup of entry characteristics), initiation of ventilation, or hospitalization duration.”
Specifically, taking the remdesivir group as a reference:
- Death within 28 days: 10.97% (301 out of 2,743 patients) vs. 11.19% in the control group — no significant differences (p>0.05).
- Need for ventilation: 10.75% (295 out of 2,743 patients) vs. 10.49% in the control group — no significant differences (p>0.05).
- Still hospitalized at day 7: 69% (out of 2,743 patients) vs. 59% in the control group — no significant differences (p>0.05).
A p-value of >0.05 means that results are not statistically significant and no effects were observed. The three other drugs showed similar patterns where they neither increased nor decreased Covid-19 mortality, ventilation needs, or hospitalization stay. Results remain unchanged when adjusted for confounding variables, including the use of corticosteroids, or when separated into sub-group analyses: <50 years, 50–69 years, >70 years, moderate disease, or severe disease.
“No study drug definitely reduced mortality (in unventilated patients or any other subgroup of entry characteristics), initiation of ventilation, or hospitalization duration.”
Taking a step further, this study did a meta-analysis of their own data plus three existing RCTs on remdesivir. The pooled results found that remdesivir does not lower Covid-19 mortality. “Combining data appropriately from all 4 trials, the remdesivir vs. control death rate ratio (RR) is 0.91 (95% CI 0.79–1.05),” the study stated. “This absolutely excludes the suggestion that remdesivir can prevent a substantial fraction of all deaths.”
[Note: RR of 0.91 means a 9% reduction in events, as calculated by one minus 0.91 equals 0.09, which is 9%. But this is not statistically significant as the 95% confidence interval (CI) is 0.79 to 1.05, which crosses one.]
The WHO vs. Gilead Sciences
“It’s a reliable result, don’t let anybody tell you otherwise because they’ll try to,” said Richard Peto, professor of medical statistics and epidemiology at the University of Oxford and an independent statistician hired by the WHO to inspect the Solidarity trial results. “This is real-world evidence.”
And Gilead Sciences did say otherwise. “The emerging (WHO) data appears inconsistent, with more robust evidence from multiple randomized, controlled studies published in peer-reviewed journals validating the clinical benefit of remdesivir,” Gilead told Reuters. Gilead also criticized the trial for its recruitment across hundreds of hospitals: “The trial design prioritized broad access, resulting in significant heterogeneity in trial adoption, implementation, controls, and patient populations and consequently, it is unclear if any conclusive findings can be drawn from the study results.”
While this point of differences in clinical settings makes sense, it could also be a strength as “you’re including many different types of subgroups and populations in different parts of the world,” said Nahid Bhadelia, MD, associate professor at the Boston University School of Medicine. Thus, the trial results are widely generalizable. And, as mentioned, existing RCTs on remdesivir are not very encouraging either.
Further, if remdesivir provided no additional clinical benefits over standard care in over 400 hospitals, the chances are that a person receiving remdesivir in one hospital would not benefit. Clemens Wendtner, professor and chief physician of infectiology and tropical medicine at Munich’s Schwabing Clinic, agreed, “The real disappointment is that remdesivir has also failed in a larger number of cases and in the ‘real-world setting.’” Is remdesivir still useful for Covid-19? At this point, the answer is no.
Further, if remdesivir provided no additional clinical benefits over standard care in over 400 hospitals, the chances are that a person receiving remdesivir in one hospital would not benefit.
The Solidarity trial continues to recruit about 2000 patients every month across multiple countries. But study drugs will change to other antivirals, immunomodulators, and monoclonal antibodies. For example, a cancer and immunomodulatory drug called acalabrutinib is considered for the participants as soon as next week. Remdesivir will also continue “to get more precise evidence,” said one of the Solidary trial executive committee.
Late October 2020 update: Despite conflicting findings, the FDA has just approved remdesivir as a drug for Covid-19 with the reason being that it may be better than nothing. “A safer and more reasonable conclusion, given the data at hand, is that remdesivir may be useful for some, but not all,” Forbes reported. But many other experts disagree, saying that the FDA approval may have been a mistake.
This article was originally published here with modifications.
Comments / 0