Pharmacokinetic analysis of vilobelimab, anaphylatoxin C5a and antidrug antibodies in PANAMO: a phase 3 study in critically ill, invasively mechanically ventilated COVID-19 patients.

BACKGROUND: Vilobelimab, a complement 5a (C5a)-specific monoclonal antibody, reduced mortality in critically ill COVID-19 patients in a phase 3 multicentre, randomized, double-blind, placebo-controlled study. As part of the study, vilobelimab concentrations and C5a levels as well as antidrug antibodies (ADAs) to vilobelimab were analysed. RESULTS: From Oct 1, 2020 to Oct 4, 2021, 368 invasively mechanically ventilated COVID-19 patients were randomized: 177 patients were randomly assigned to receive vilobelimab while 191 patients received placebo. Pharmacokinetic sampling was only performed at sites in Western Europe. Blood samples for vilobelimab measurements were available for 93 of 177 (53%) patients in the vilobelimab group and 99 of 191 (52%) patients in the placebo group. On day 8, after three infusions, mean vilobelimab (trough) concentrations ranged from 21,799.3 to 302,972.1 ng/mL (geometric mean 137,881.3 ng/mL). Blood samples for C5a measurements were available for 94 of 177 (53%) patients in the vilobelimab group and 99 of 191 (52%) patients in the placebo group. At screening, C5a levels were highly elevated and comparable between groups. In the vilobelimab group, median C5a levels were 118.3 ng/mL [IQR 71.2-168.2 ng/mL] and in the placebo group, median C5a levels were 104.6 ng/mL [IQR 77.5-156.6 ng/mL]. By day 8, median C5a levels were reduced by 87% in the vilobelimab group (median 14.5 ng/mL [IQR 9.5-21.0 ng/mL], p < 0.001) versus an 11% increase in the placebo group (median 119.2 ng/mL [IQR 85.9-152.1 ng/mL]). Beyond day 8, though plasma sampling was sparse, C5a levels did not reach screening levels in the vilobelimab group while C5a levels remained elevated in the placebo group. Treatment-emergent ADAs were observed in one patient in the vilobelimab group at hospital discharge on day 40 and in one patient in the placebo group at hospital discharge on day 25. CONCLUSIONS: This analysis shows that vilobelimab efficiently inhibits C5a in critically ill COVID-19 patients. There was no evidence of immunogenicity associated with vilobelimab treatment. Trial registration ClinicalTrials.gov, NCT04333420. Registered 3 April 2020, https://clinicaltrials.gov/ct2/show/NCT04333420.

Anti-C5a antibody vilobelimab treatment and the effect on biomarkers of inflammation and coagulation in patients with severe COVID-19: a substudy of the phase 2 PANAMO trial.

We recently reported in the phase 3 PANAMO trial that selectively blocking complement 5a (C5a) with vilobelimab led to improved survival in critically ill COVID-19 patients. C5a is an important contributor to the innate immune system and can also activate the coagulation system. High C5a levels have been reported in severely ill COVID-19 patients and correlate with disease severity and mortality. Previously, we assessed the potential benefit and safety of vilobelimab in severe COVID-19 patients. In the current substudy of the phase 2 PANAMO trial, we aim to explore the effects of vilobelimab on various biomarkers of inflammation and coagulation. Between March 31 and April 24, 2020, 17 patients with severe COVID-19 pneumonia were enrolled in an exploratory, open-label, randomised phase 2 trial. Blood markers of complement, endothelial activation, epithelial barrier disruption, inflammation, neutrophil activation, neutrophil extracellular trap (NET) formation and coagulopathy were measured using enzyme-linked immunosorbent assay (ELISA) or utilizing the Luminex platform. During the first 15 days after inclusion, change in biomarker concentrations between the two groups were modelled with linear mixed-effects models with spatial splines and compared. Eight patients were randomized to vilobelimab treatment plus best supportive care (BSC) and nine patients were randomized to BSC only. A significant decrease over time was seen in the vilobelimab plus BSC group for C5a compared to the BSC only group (p < 0.001). ADAMTS13 levels decreased over time in the BSC only group compared to the vilobelimab plus BSC group (p < 0.01) and interleukin-8 (IL-8) levels were statistically more suppressed in the vilobelimab plus BSC group compared to the BSC group (p = 0.03). Our preliminary results show that C5a inhibition decreases the inflammatory response and hypercoagulability, which likely explains the beneficial effect of vilobelimab in severe COVID-19 patients. Validation of these results in a larger sample size is warranted.

Controlling the anaphylatoxin C5a in diseases requires a specifically targeted inhibition.

The terminal complement split product C5a has been described as an important mediator in inflammatory diseases. C5a is generated upon cleavage of C5 and earlier research suggests that, besides the known C5 convertases formed upon activation of the complement pathways, various enzymes could activate C5 directly. We demonstrate that eculizumab effectively blocks C5 activation when mediated by C5-convertase formation, but fails to block C5a generation resulting from direct enzymatic cleavage by trypsin and thrombin. C5a generated by these enzymes is shown to be fully biologically functional and can be blocked by IFX-1, a specific monoclonal anti-human C5a antibody. We further report clinical cases of atypical hemolytic uremic syndrome (aHUS) and C3 Glomerulonephritis (C3GN) patients under treatment with eculizumab presenting substantially elevated C5a levels. Thus, blocking the C5 convertase mediated activation of C5 may not be efficient to control C5a-mediated effects in human disease and that a targeted approach is warranted.

Treatment with anti-C5a antibody improves the outcome of H7N9 virus infection in African green monkeys.

BACKGROUND: Patients infected with influenza A(H7N9) virus present with acute lung injury (ALI) that is due to severe pneumonia and systemic inflammation. It is often fatal because there are few effective treatment options. Complement activation has been implicated in the pathogenesis of virus-induced lung injury; therefore, we investigated the effect of targeted complement inhibition on ALI induced by H7N9 virus infection. METHODS: A novel neutralizing specific antihuman C5a antibody (IFX-1) was used. This antibody blocked the ability of C5a to induce granulocytes to express CD11b while not affecting the ability of C5b to form the membrane attack complex. African green monkeys were inoculated with H7N9 virus and treated intravenously with IFX-1. RESULTS: The virus infection led to intense ALI and systemic inflammatory response syndrome (SIRS) in association with excessive complement activation. Anti-C5a treatment in H7N9-infected monkeys substantially attenuated ALI: It markedly reduced the lung histopathological injury and decreased the lung infiltration of macrophages and neutrophils. Moreover, the treatment decreased the intensity of SIRS; the body temperature changes were minimal and the plasma levels of inflammatory mediators were markedly reduced. The treatments also significantly decreased the virus titers in the infected lungs. CONCLUSIONS: Antihuman C5a antibody treatment remarkably reduced the ALI and systemic inflammation induced by H7N9 virus infection. Complement inhibition may be a promising adjunctive therapy for severe viral pneumonia.

Early intra-articular complement activation in ankle fractures.

Cytokine regulation possibly influences long term outcome following ankle fractures, but little is known about synovial fracture biochemistry. Eight patients with an ankle dislocation fracture were included in a prospective case series and matched with patients suffering from grade 2 osteochondritis dissecans (OCD) of the ankle. All fractures needed external fixation during which joint effusions were collected. Fluid analysis was done by ELISA measuring aggrecan, bFGF, IL-1 ß, IGF-1, and the complement components C3a, C5a, and C5b-9. The time periods between occurrence of fracture and collection of effusion were only significantly associated with synovial aggrecan and C5b-9 levels (P < 0.001). Furthermore, synovial expressions of both proteins correlated with each other (P < 0.001). Although IL-1 ß expression was relatively low, intra-articular levels correlated with C5a (P < 0.01) and serological C-reactive protein concentrations 2 days after surgery (P < 0.05). Joint effusions were initially dominated by neutrophils, but the portion of monocytes constantly increased reaching 50% at day 6 after fracture (P < 0.02). Whereas aggrecan and IL-1ß concentrations were not different in fracture and OCD patients, bFGF, IGF-1, and all complement components were significantly higher concentrated in ankle joints with fractures (P < 0.01). Complement activation and inflammatory cell infiltration characterize the joint biology following acute ankle fractures.