The trial was initiated in rapid response to the Covid-19 public health emergency, at which time there was very limited information about clinical outcomes in hospitalized patients with Covid-19. The original total sample size was set at 160, since it would provide the trial with 80% power to detect a difference, at a two-sided significance level of α=0.05, of 8 days in the median time to clinical improvement between the two groups, assuming that the median time in the standard-care group was 20 days and that 75% of the patients would reach clinical improvement. The planned enrollment of 160 patients in the trial occurred quickly, and the assessment at that point was that the trial was underpowered; thus, a decision was made to continue enrollment by investigators. Subsequently, when another agent (remdesivir) became available for clinical trials, we decided to suspend enrollment in this trial.
Primary efficacy analysis was on an intention-to-treat basis and included all the patients who had undergone randomization. The time to clinical improvement was assessed after all patients had reached day 28, with failure to reach clinical improvement or death before day 28 considered as right-censored at day 28 (right-censoring occurs when an event may have occurred after the last time a person was under observation, but the specific timing of the event is unknown). The time to clinical improvement was portrayed by Kaplan–Meier plot and compared with a log-rank test. Hazard ratios with 95% confidence intervals were calculated by means of the Cox proportional-hazards model. Five patients who had been assigned to the lopinavir–ritonavir group did not receive any doses (three of them died within 24 hours) but were included in the intention-to-treat analysis, since no reciprocal removals occurred in the standard-care group. A modified intention-to-treat analysis that excluded three early deaths was also performed. Post hoc analyses include subgroup analysis for National Early Warning Score 2 (NEWS2)19 of 5 or below or greater than 5 and those who underwent randomization up to 12 days or more than 12 days after the onset of illness.
Because the statistical analysis plan did not include a provision for correcting for multiplicity in tests for secondary or other outcomes, results are reported as point estimates and 95% confidence intervals. The widths of the confidence intervals have not been adjusted for multiplicity, so the intervals should not be used to infer definitive treatment effects for secondary outcomes. Safety analyses were based on the patients’ actual treatment exposure. Statistical analyses were conducted with SAS software, version 9.4 (SAS Institute).
Primary efficacy analysis was on an intention-to-treat basis and included all the patients who had undergone randomization. The time to clinical improvement was assessed after all patients had reached day 28, with failure to reach clinical improvement or death before day 28 considered as right-censored at day 28 (right-censoring occurs when an event may have occurred after the last time a person was under observation, but the specific timing of the event is unknown). The time to clinical improvement was portrayed by Kaplan–Meier plot and compared with a log-rank test. Hazard ratios with 95% confidence intervals were calculated by means of the Cox proportional-hazards model. Five patients who had been assigned to the lopinavir–ritonavir group did not receive any doses (three of them died within 24 hours) but were included in the intention-to-treat analysis, since no reciprocal removals occurred in the standard-care group. A modified intention-to-treat analysis that excluded three early deaths was also performed. Post hoc analyses include subgroup analysis for National Early Warning Score 2 (NEWS2)19 of 5 or below or greater than 5 and those who underwent randomization up to 12 days or more than 12 days after the onset of illness.
Because the statistical analysis plan did not include a provision for correcting for multiplicity in tests for secondary or other outcomes, results are reported as point estimates and 95% confidence intervals. The widths of the confidence intervals have not been adjusted for multiplicity, so the intervals should not be used to infer definitive treatment effects for secondary outcomes. Safety analyses were based on the patients’ actual treatment exposure. Statistical analyses were conducted with SAS software, version 9.4 (SAS Institute).
