An Optimized Crovalimab Dose and Regimen Reduced the Formation of Drug-Target-Drug Complexes in Patients with Paroxysmal Nocturnal Hemoglobinuria from the Phase I/II COMPOSER Trial

Introduction

Crovalimab is a novel anti–complement component 5 (C5) monoclonal antibody engineered with the Sequential Monoclonal Antibody Recycling Technology (SMART-Ig; Fukuzawa et al, Sci Rep. 2017) to extend half-life and enable infrequent, subcutaneous (SC) self-administration in C5-mediated diseases. Crovalimab is being investigated as a therapy for paroxysmal nocturnal hemoglobinuria (PNH), a disease for which C5 inhibition is the standard of care. The Phase I/II COMPOSER trial (NCT03157635; Röth, et al. Blood. 2020) is a global, open-label, multicenter study of crovalimab consisting of 4 sequential parts. Parts 1, 2, and 3 assessed the pharmacokinetics (PK) and safety of crovalimab in healthy volunteers, C5 inhibitor–naive patients, and patients switched from eculizumab, respectively. Part 4 assessed an optimized crovalimab dose and regimen in naive and switched patients with PNH. Because eculizumab and crovalimab bind to different C5 epitopes, drug-target-drug complexes (DTDCs) consisting of eculizumab, C5, and crovalimab motifs can temporarily form in the circulation of patients who switch treatments. DTDCs can form in a range of sizes, from single crovalimab-C5-eculizumab motif to larger complexes with multiple motifs. Larger DTDCs are a concern because they take longer to clear and may be more likely to induce type III hypersensitivity reactions.

Objectives

Describe the impact of DTDC formation on the safety, PK, and pharmacodynamics of crovalimab in patients with PNH who switched from eculizumab to crovalimab and to describe the effect of crovalimab dose on DTDC size distribution and kinetics.

Study Design and Methods

Using data from COMPOSER Parts 1-3, a biochemical mathematical model was developed to investigate the kinetics of the formation and dissociation of DTDCs under the assumption that larger complexes are formed by the reversible binding of smaller complexes. The model was calibrated using concentration-time profiles of total C5, total crovalimab, and the concentration of eculizumab at the time of crovalimab initiation. DTDC size distributions were measured using size-exclusion chromatography coupled to enzyme-linked immunosorbent assay. Using model-based simulations, an optimized crovalimab dosing strategy was identified to reduce the formation of large DTDCs while maintaining serum concentration of crovalimab above the target level of ≈ 100 μg/mL. The optimized dose and regimen were a loading series of 1000 mg intravenously on day 1 and 340 mg SC on days 2, 8, 15, and 22, followed by maintenance dosing of 680 mg SC every 4 weeks starting on day 29. The loading dose series increased the total crovalimab dose received during the first month of treatment to reduce the formation of larger DTDCs, in line with the lattice theory of complex formation. This optimized dosing strategy was investigated in Part 4 patients who switched from eculizumab.

Results

In COMPOSER, 19 patients with PNH were enrolled in Part 3 and switched from eculizumab to crovalimab. DTDCs were observed in all patients from Part 3 (Figure; larger DTDCs are found in fractions 1-4 and smaller crovalimab-containing complexes, such as single motifs and single crovalimab molecules, are found in fractions 5 and 6). Two Part 3 patients experienced clinical manifestations compatible with type III hypersensitivity reactions that were ascribed to DTDCs. The DTDC size distribution in Part 4 patients, who received the optimized dosing strategy, evolved differently than in Part 3 patients, consistent with model predictions. In the switched patients from Part 4, large DTDC levels started to decrease on day 8 and continued to decrease, in contrast to Part 3, in which they started to decrease on day 15. On day 22, the mean percentage of the largest DTDCs was reduced by 56% in patients in Part 4 relative to patients in Part 3. Part 4 patients achieved and maintained serum crovalimab concentrations above ≈ 100 µg/mL throughout follow-up. Despite DTDCs being observed in all Part 4 patients who switched from eculizumab, no adverse events suggestive of a type III hypersensitivity reaction occurred.

Conclusions

The optimized crovalimab regimen resulted in lower concentrations of large DTDCs than in patients who received the Part 3 regimen and reduced the persistence of DTDCs in patients who switched treatment. This regimen is now being evaluated in the Phase III COMMODORE 1 (NCT04432584) and COMMODORE 2 (NCT04434092) studies.

Jun-ichi Nishimura and Antoine Soubret contributed equally to this work.
Ido Paz-Priel and Alexandre Sostelly are co-senior authors.