The effect of remdesivir and nirmatrelvir/ritonavir on mortality in patients hospitalized with COVID-19 during the Omicron era: an emulated target trial

Abstract

Objectives: Despite widespread population immunity, SARS-CoV-2 infection remains a common cause of hospitalization. Although antiviral treatments have shown efficacy in clinical trials, often in outpatient settings, emerging real-world evidence in hospitalized patients is conflicting. Additionally, the impact of immunity status or viral load remains insufficiently studied. Our aim was to assess the effect of remdesivir and/or nirmatrelvir/ritonavir in acute SARS-CoV-2 infection requiring hospital admission, including clinically relevant subgroups.

Methods: We performed a retrospective population-based cohort study designed as an emulated target trial. Patients were included from six acute care hospitals in Stockholm, Sweden, during the Omicron era. All adult patients admitted via the emergency department with a diagnosis indicating infection and a positive polymerase chain reaction SARS-CoV-2 test were assessed for eligibility. Viral burden was estimated using cycle threshold values. Data were analysed using cloning, censoring, and inverse probability weighting. The primary and secondary outcomes were 30- and 90-day all-cause mortality, respectively. Safety outcomes included new-onset kidney failure, liver failure, and cardiac arrhythmia.

Results: Among 4016 included patients, 771 (19%) received antiviral treatment. Overall mortality was 9.1% (n = 365) at 30 days and 13.8% (n = 554) at 90 days. The adjusted risk ratio (aRR) of antiviral treatment was 0.80 (95% CI, 0.62-1.01) for 30-day mortality and 0.78 (95% CI, 0.63-0.97) for 90-day mortality. The treatment effect was greater in unvaccinated individuals without previous confirmed infection (aRR, 0.38 [95% CI, 0.18-0.67] vs. aRR, 0.95 [0.64-1.39] in recently vaccinated individuals). No clear differences were observed in subgroups based on age or cycle threshold value, but precision was limited. Safety analyses revealed no substantial risk with antiviral treatment.

Conclusions: Treatment with remdesivir and/or nirmatrelvir/ritonavir was associated with reduced mortality in hospitalized SARS-CoV-2-infected patients during the Omicron era, primarily in unvaccinated individuals without previous infection. Our findings support prioritizing nonimmune individuals for antiviral treatment.

Keywords: COVID-19/drug therapy; Emulated target trial; Nirmatrelvir; Omicron variant; Real-world evidence; Remdesivir; Ritonavir drug combination; SARS-CoV-2; Vaccination; Viral load.