Real-world data back remdesivir benefit for hospitalized COVID-19 patients

Findings from real-world studies presented at ECCMID 2021 showed improved survival outcomes with remdesivir in hospitalized COVID-19 patients.

In a randomized, double-blind, placebo-controlled trial, remdesivir treatment shortened time to recovery, improved outcomes, and reduced disease progression in hospitalized COVID-19 patients. [N Engl J Med 2020;383:1813-1826]

Early treatment

Using data from the US Premier Healthcare Database, 76,046 met the inclusion criteria* for retrospective analysis. Of these, 28,855 received remdesivir within 2 days of admission. After propensity-score (PS) matching, they were matched with patients who did not receive remdesivir during their hospitalization. More than half (56 percent) of the population was male. [ECCMID 2021, abstract 4704]

After adjusting for covariates**, remdesivir recipients had a significantly lower risk of mortality than non-remdesivir recipients, both at day 14 (hazard ratio [HR], 0.76; p<0.001) and day 28 (HR, 0.89; p=0.003). These effects were sustained even after removing the hospital-level effects*** (HRs, 0.76 and 0.88, respectively; p<0.001 for both).

Although unmeasured confounding cannot be excluded, the benefits of remdesivir withstood removal of hospital-level effects on sensitivity analysis, highlighted Essy Mozaffari from Gilead Sciences, Foster City, California, US, during his presentation. “[The] robust matching methods led to balanced treatment groups… Adjusting for time [also] helped balance the evolution of clinical management of COVID-19 over the course of the pandemic.”

Comparative analysis

Another study evaluated hospitalized patients with newly diagnosed COVID-19. Remdesivir-exposed patients were matched 1:1 with referent patients using risk set sampling (RSS; n=30,352 each arm) and PS matching (n=24,856 each arm). Mean age for both RSS and PS cohorts was 67 years; 52 percent were male. [ECCMID 2021, abstract 4884]

Remdesivir was associated with a significant reduction in all-cause mortality by day 28 in the overall cohort, both in the as-treated primary analysis and intention-to-treat sensitivity analysis (HRs, 0.77 and 0.83, respectively). “This benefit was also observed in all baseline O₂ requirement subgroups (ie, room air, low-flow O₂, high-flow/NIV^#, and IMV/ECMO^#),” said Dr Anand Chokkalingam, also from Gilead Sciences.

The probability of hospital discharge was also increased with remdesivir in the overall cohort (HR, 1.19). The effect appeared strongest among patients requiring no O₂ (HR, 1.24) or low-flow O₂ at baseline (HR, 1.10); results for high-flow/NIV or IMV/ECMO patients were nonsignificant, noted Chokkalingam.

Despite the possible residual confounding, lack of certain data (eg, lab measurements, granular changes in O₂ supplementation), and potential misclassification of O₂ support, the robust methods used ensured balance between exposed and referent patients, noted Chokkalingam. The large sample size also allowed for a well-powered analysis to observe treatment effects, he added.

“With a wide range of COVID-19 patients and minimal inclusion/exclusion criteria, these results represent the real-world treatment effect in hospitalized US COVID-19 patients,” Chokkalingam continued.

“Real-world data can complement results from well-controlled trials during a pandemic, wherein evolution of clinical management can outpace the development of new trials, and when markedly larger sample sizes are needed to assess secondary outcomes or outcomes in important patient subgroups,” said Chokkalingam.

Taken together, the findings augment the growing body of evidence reflecting the benefit of remdesivir in hospitalized COVID-19 patients.