OBJECTIVES: Severe COVID-19 is associated with exaggerated complement activation. We assessed the efficacy and safety of avdoralimab (an anti-C5aR1 mAb) in severe COVID-19. DESIGN: FOR COVID Elimination (FORCE) was a double-blind, placebo-controlled study. SETTING: Twelve clinical sites in France (ICU and general hospitals). PATIENTS: Patients receiving greater than or equal to 5 L oxygen/min to maintain Sp o2 greater than 93% (World Health Organization scale ≥ 5). Patients received conventional oxygen therapy or high-flow oxygen (HFO)/noninvasive ventilation (NIV) in cohort 1; HFO, NIV, or invasive mechanical ventilation (IMV) in cohort 2; and IMV in cohort 3. INTERVENTIONS: Patients were randomly assigned, in a 1:1 ratio, to receive avdoralimab or placebo. The primary outcome was clinical status on the World Health Organization ordinal scale at days 14 and 28 for cohorts 1 and 3, and the number of ventilator-free days at day 28 (VFD28) for cohort 2. MEASUREMENTS AND MAIN RESULTS: We randomized 207 patients: 99 in cohort 1, 49 in cohort 2, and 59 in cohort 3. During hospitalization, 95% of patients received glucocorticoids. Avdoralimab did not improve World Health Organization clinical scale score on days 14 and 28 (between-group difference on day 28 of -0.26 (95% CI, -1.2 to 0.7; p = 0.7) in cohort 1 and -0.28 (95% CI, -1.8 to 1.2; p = 0.6) in cohort 3). Avdoralimab did not improve VFD28 in cohort 2 (between-group difference of -6.3 (95% CI, -13.2 to 0.7; p = 0.96) or secondary outcomes in any cohort. No subgroup of interest was identified. CONCLUSIONS: In this randomized trial in hospitalized patients with severe COVID-19 pneumonia, avdoralimab did not significantly improve clinical status at days 14 and 28 (funded by Innate Pharma, ClinicalTrials.gov number, NCT04371367).