Gilead's Remdesivir Clinical Trials - What Do Drastic Design Changes Mean?

In the midst of the Covid-19 pandemic, people are rushing to find effective treatments for Covid-19 infected patients. Among the drugs being investigated is remdesivir, an experimental antiviral made by the US drug company Gilead Sciences. It has been characterized as one of the most promising by health authorities, including WHO officials—though that optimism is inspired only by anecdotal information.

There are four randomized, controlled trials with remdesivir that may have the results available in very near term. Two studies conducted in China should have the results available very imminent (before the end of April). Two studies conducted by Gilead itself should have results sometime in May, 2020. 

We noticed that for two pivotal studies run by Gilead, the study design was significantly modified. According to the clinicaltrials.gov, the major changes are listed below: sample sizes are increased; the primary efficacy endpoint is revised, the treatment arms are tweaked.

See the article "Gilead supersizes remdesivir trials, changes primary endpoint".

Protocol Title	Previous Version (before April 6, 2020)	Revised Version (after April 6, 2020)
A Phase 3 Randomized Study to Evaluate the Safety and Antiviral Activity of Remdesivir (GS-5734™) in Participants With Severe COVID-19	400 Subjects Two Strata and Three arms: Remdesivir for 5 days versus Standard of care; Remdesivir for 10 days versus Standard of care	2400 Subjects Four Strata and Five arms: Remdesivir for 5 days (Not Mechanically Ventilated) versus Standard of care; Remdesivir for 10 days (Not Mechanically Ventilated) versus Standard of care Remdesivir, 5 or 10 Days (Extension) versus Standard care Remdesivir 10 days (Mechanically Ventilated) versus Standard of care
	Proportion of Participants With Normalization of Fever and Oxygen Saturation Through Day 14 This is a composite outcome measure. Criteria for fever normalization: Temperature < 36.6 °C armpit, < 37.2 °C oral, or < 37.8 °C rectal sustained for at least 24 hours and criteria for oxygen normalization: peripheral capillary oxygen saturation (Sp02) > 94% sustained for at least 24 hours.	The Odds of Ratio for Improvement on a 7-point Ordinal Scale on Day 14. The odds ratio represents the odds of improvement in the ordinal scale between the treatment groups. The ordinal scale is an assessment of the clinical status at a given day. Each day, the worst score from the previous day will be recorded. The scale is as follows: 1. Death 2. Hospitalized, on invasive mechanical ventilation or Extracorporeal Membrane Oxygenation (ECMO) 3. Hospitalized, on non-invasive ventilation or high flow oxygen devices 4. Hospitalized, requiring low flow supplemental oxygen 5. Hospitalized, not requiring supplemental oxygen - requiring ongoing medical care (coronavirus (COVID-19) related or otherwise) 6. Hospitalized, not requiring supplemental oxygen - no longer required ongoing medical care (other than per protocol Remdesivir administration 7. Not hospitalized.
A Phase 3 Randomized Study to Evaluate the Safety and Antiviral Activity of Remdesivir (GS-5734™) in Participants With Moderate COVID-19 Compared to Standard of Care Treatment	600 Subjects Three arms: Remdesivir for 5 days Remdesivir for 10 days Standard of care	1600 Subjects Added a fourth arm: Experimental: Part B: Extension Treatment, Remdesivir 5 or 10 days Participants will receive continued standard of care therapy together with RDV 200 mg on Day 1 followed by RDV 100 mg on Days 2, 3, 4, 5, 6, 7, 8, 9, and 10.
	Proportion of Participants Discharged by Day 14	The Odds of Ratio for Improvement on a 7-point Ordinal Scale on Day 11. The odds ratio represents the odds of improvement in the ordinal scale between the treatment groups. The ordinal scale is an assessment of the clinical status at a given day. Each day, the worst score from the previous day will be recorded. The scale is as follows: 1. Death 2. Hospitalized, on invasive mechanical ventilation or Extracorporeal Membrane Oxygenation (ECMO) 3. Hospitalized, on non-invasive ventilation or high flow oxygen devices 4. Hospitalized, requiring low flow supplemental oxygen 5. Hospitalized, not requiring supplemental oxygen - requiring ongoing medical care (coronavirus (COVID-19) related or otherwise) 6. Hospitalized, not requiring supplemental oxygen - no longer required ongoing medical care (other than per protocol Remdesivir administration 7. Not hospitalized.

It is not clear what these protocol changes are based on: the external data from other studies or the cumulative data to date from these two ongoing studies? The fact is that these two Gilead-run studies are randomized, open-label – unless they treated them as the blinded studies, the Sponsor would be able to see the cumulative data and monitor the treatment effects along the way. 

A dramatic increase in sample size seems to suggest that the treatment effect is much smaller and the previous study design may be underpowered.

It may also indicate that two randomized, double-blinded Remdesivir studies in China will most likely be underpowered as well (308 subjects for study in mild-moderate patients and 453 subjects for study in severe patients). Unless there is a magic or luck, these two studies from China may not be able to reach the statistical significance because of the smaller sample size. We hope that there will be at least some trends - at a time of no effective treatment for Covid-19 infection, a study with a trend will be encouraging news. 

For the primary efficacy endpoint, odds ratio should be more sensitive to detect the treatment difference for the ordinal data - ordinal logistic regression will need to be used to calculate the odds ratios.