ABSTRACT
Introduction: Poor outcomes in COVID-19 patients (pt) are associated with C5a-C5aR axis activation. A C5a-specific monoclonal antibody, vilobelimab (VILO), improves outcomes in critically ill COVID-19 pt in a Phase 3 randomized, double-blind, placebo (PLC)- controlled study [1]. Method(s): COVID-19 pt within 48 h of intubation were randomly assigned to receive 6, 800 mg infusions of VILO or PLC at a 1:1 ratio on top of standard of care. Predefined subgroup analyses by region and country were performed. Result(s): Forty-six (46) hospitals on 4 continents randomized 369 pt: VILO (n = 178), PLC (n = 191). VILO significantly reduced 28- (HR 0.67;95% CI 0.48-0.96;p = 0.027) and 60-Day mortality (HR 0.67;95% CI 0.48-0.93, p = 0.0163) using a predefined, unstratified per protocol analysis. Mortality rates at 28- and 60-days and VILO treatment effects, however, differed substantially between regions: Western Europe HR for 60-day mortality 0.59 [0.37-0.95], South Africa plus Russian Federation HR 0.62 [0.28-1.38] and South America HR 0.80 [0.46-1.39] (Fig. 1). The weak signal in South America is predominately driven by Brazil (n = 74), which showed a significant age imbalance with a median 9-years younger PLC group (44.5-years-old vs 53.5-years-old) with low 60-day mortality of ~ 32.5% in the PLC group versus ~ 43.3% in Western Europe. Adjusting for age group categories (<= 30, 31-40, 41-50, 51-60, > 60;Cox regression) for 60-day mortality changed the HR in Brazil (0.96 [0.44-2.10] for continuous age-adjustment) to values near the estimate for the entire study population (HR 0.77 [0.35-1.69] for age in categories), suggesting a by chance imbalance and not a statistically evident weaker effect in Brazil. Conclusion(s): Regional efficacy differences between the rest of the world and South America were driven by age imbalances between treatment groups, which do not diminish the robust efficacy signal for VILO in severe COVID-19.
ABSTRACT
Introduction: C5a-C5aR axis activation is associated with increased mortality in severe COVID-19. Vilobelimab (VILO), a C5a-specific monoclonal antibody, improved mortality in severe COVID-19 patients (pts) in a Phase 3 multicenter, randomized, double-blind, placebo (PLC)- controlled study [1]. A pharmacokinetic/pharmacodynamic (PK/PD) analysis was undertaken to assess VILO and C5a as well as antidrug antibodies (ADA) levels in the study. Method(s): Forty-six (46) hospitals on four continents randomized 369 COVID-19 pts (VILO [n = 178], PLC [n = 191]) within 48 h of being mechanically ventilated to receive 6, 800 mg infusions of VILO or PLC at a 1:1 ratio on top of standard of care. Blood samples were taken at screening, Day 8 and at hospital discharge for VILO and C5a and at screening and hospital discharge for ADA. Enzyme-linked immunosorbent assays were used to analyze levels. Result(s): Screening blood samples for VILO and C5a were available for VILO (n = 93) and PLC (n = 99) from sites in Western Europe. On Day 8 after 3 infusions, mean VILO trough concentrations were 21799.3- 302972.1 ng/mL (geometric mean 137881.3 ng/mL) (Fig. 1). At screening, C5a was highly elevated and comparable between groups: VILO median 118.3 ng/mL, mean 130.3 ng/mL, PLC median 104.6 ng/mL, mean 123.2 ng/mL. By Day 8, C5a levels were reduced by 84.6% in the VILO group (median 14.5 ng/mL [mean 16.8 ng/mL], p < 0.001) versus a 19.6% increase in the PLC group (median, 119.2 ng/mL, mean 129.8 ng/ mL). Beyond Day 8, though PD sampling was sparse, C5a levels remained elevated for PLC whereas C5a slowly rose but did not reach screening levels for VILO. Treatment-induced ADA were observed in 1 pt in the VILO group (Day 40 discharge) and 1 pt in the PLC group (Day 25 discharge), both appeared independent of treatment. Conclusion(s): The PK/PD analysis shows that VILO efficiently inhibits C5a in pts with severe COVID-19 resulting in a robust clinical effect on mortality reduction without inducing ADA.