Characterizing C5 Inhibition with the SMART-Ig Anti-hC5 Antibody Crovalimab in PNH Patients Using Free Available Paratopes

Crovalimab (RO7112689) is a novel anti-human C5 antibody engineered with Sequential Monoclonal Antibody Recycling Technology (SMART Ig)(Fukuzawa et al., Sci Rep. (2017) 7(1):1080), resulting in significant half-life extension and enabling infrequent SC dosing using small volumes (1mL - 4mL) in C5 mediated diseases. To establish pharmacokinetics (PK), pharmacodynamics (PD), safety, efficacy, and optimal dose of crovalimab, we conducted a four-part adaptive clinical trial (Figure 1) in healthy volunteers (Part 1), treatment naïve PNH patients (Part 2, Part 4) and in eculizumab pre-treated patients (Part 3, Part 4). The conventional PK metrics to determine optimal target inhibition is drug concentration, however this has some limitations when considering an abundant fluctuating soluble target such as C5. Therefore, we proposed to characterize C5 inhibition by quantifying the level of crovalimab free paratopes (i.e. the concentration of free crovalimab antigen-binding sites not bound to C5) and to illustrate how it can be used to estimate the available binding capacity reserve of crovalimab.

Total crovalimab concentration, free C5 and total C5 protein levels (measuring free C5, C5 bound to one crovalimab Fab arm, or two C5s bound to the two crovalimab Fab arms) were measured using validated assays. In assuming steady state conditions at every measurement time for the binding of crovalimab with C5 (due to the rapid binding of C5 with crovalimab), a mathematical model based on law mass action principle was used to describe the equilibrium between free C5, free crovalimab and crovalimab bound to one or two C5 molecules. Available crovalimab free paratope was estimated from the model in using total C5 and total crovalimab concentrations collected in the COMPOSER trial. Paratope level was expressed in C5-binding ability concentration equivalent (i.e. how much additional C5 could be bound). Relationships between total concentration time course, free C5 and available free epitopes were evaluated using graphical analysis. To complement these analyses and using the mathematical model, a sensitivity analysis was performed to identify key antibody properties driving the level of available free crovalimab paratopes.

In COMPOSER Part 2, the longitudinal time course of crovalimab free paratope concentration (Figure 2) shows that 170mg SC weekly provided a median reserve of free paratopes that allows binding around 120µg/mL of additional C5 corresponding approximatively to two times the baseline C5 levels observed in Part 2 (i.e. 129µg/mL). Therefore, if the level of C5 in the circulation doubled from baseline, e.g. during infection, there would be a sufficient binding capacity to block the activity of these new C5 molecules at least for 75% of the patients (Figure 2). In COMPOSER Part 3, the observed median reserve of free paratopes was lower during the first 60 days after crovalimab initiation (Figure 2). This is driven by the remaining presence of eculizumab and the binding of both crovalimab and eculizumab to C5. Eculizumab and crovalimab bind different C5 epitopes and switching patients from eculizumab to crovalimab induces formation of drug target drug complexes (DTDC). DTDCs clearance was estimated to be ten times faster than crovalimab-C5 complexes clearance resulting in a drop of free paratope level during the first 60 days. Washout from eculizumab is not feasible in PNH patients. To increase the availability of crovalimab free paratopes after switching, crovalimab dose and regimen was optimized and is currently tested in COMPOSER Part 4 (Figure 1).The sensitivity analysis demonstrated that the main driver of free paratopes availability were the recycling efficiency of free crovalimab after endocytosis and the clearance of crovalimab while the affinity of crovalimab to C5 had little effect.

Free paratope concentration enables characterization of available binding reserve of crovalimab. The capacity to adequately control C5 increase due to the SMART-Ig engineering is expected to result in a better control of breakthrough hemolysis in PNH patients compared with an antibody without this technology. For soluble targets, this approach provides more stringent criteria than antibody concentration as it defines the capacity of the drug to bind free target at any time. Therefore, this metric should prove helpful in guiding dose selection for monoclonal antibodies binding a soluble target.

View largeDownload slide

View largeDownload slide

Close modal