Crovalimab for treatment of patients with paroxysmal nocturnal haemoglobinuria and complement <scp>C5</scp> polymorphism: Subanalysis of the phase 1/2 <scp>COMPOSER</scp> study

Abstract

Paroxysmal nocturnal haemoglobinuria (PNH) is characterised by the loss of complement regulators CD55 and CD59 on peripheral blood elements, which can result in intravascular haemolysis and thrombosis.1 C3 (pegcetacoplan) and C5 (eculizumab and ravulizumab) inhibitors are approved therapies for PNH and can control intravascular haemolysis in patients with PNH.2-5 However, eculizumab and ravulizumab lack efficacy in patients with inherited nonsynonymous polymorphisms in the C5α subunit (i.e. c.2654G → A and c.2653C → T) affecting Arg885, which corresponds to their target epitope.5, 6 Such polymorphisms are present in ≤3.5% of individuals of Asian descent and have been reported in a patient with no known Asian ancestry.5-8 Hence, patients with PNH and C5 polymorphism have a high unmet clinical need.5, 6 The newly approved C3 inhibitor pegcetacoplan, which blocks the complement cascade upstream of C5, may be effective in treating these patients, but specific data are unavailable.2 Therefore, these patients should be enrolled in an appropriate study. Crovalimab is a novel anti-C5 monoclonal antibody that binds to an epitope on the C5 β chain distinct from that of the eculizumab and ravulizumab binding site.9 It equally binds and inhibits the activity of wild-type C5 and several C5 variants with nonsynonymous single nucleotide polymorphisms (SNPs), including Arg885.9 Here, we report a post hoc subgroup analysis of the efficacy and safety of crovalimab in patients with the C5 codon 885 polymorphism in the phase 1/2 COMPOSER study (NCT03157635) of crovalimab in patients with PNH. Details of COMPOSER have been previously described.10-12 Whole blood samples were collected, and the C5 SNP rs56040400 was genotyped using the TaqMan genotyping assay C__89025427_10 (Thermo Fisher Scientific). Patients with a single missense C5 heterozygous mutation, including c.2654G → A, were eligible for enrolment and included in this exploratory analysis. C5 inhibitor–naive or –experienced patients received crovalimab loading followed by an average subcutaneous maintenance dose of 680 mg every 4 weeks (q4w). Patients were followed up for lactate dehydrogenase (LDH) concentrations, pharmacokinetics, pharmacodynamics, breakthrough haemolysis (BTH) events, occurrence of blood transfusions, and safety. Complete complement inhibition was defined as a liposome immunoassay (LIA) level <10 U/ml. Efficacy and safety data were summarised using descriptive statistics. Four of the 44 patients enrolled in parts 2 to 4 of COMPOSER were found to carry the C5 SNP (Table 1). At clinical cut-off (29 January 2020), follow-up for these patients ranged from 33.4 to 119.3 weeks. All four patients were male, of Asian descent, and diagnosed with PNH between 44 and 734 weeks before enrolment. At enrolment, one patient who displayed evidence of intravascular haemolysis with an LDH level of 4.81 × upper limit of normal (ULN) and an LIA level of 34 U/ml switched from ongoing eculizumab therapy. The other three patients had previously discontinued eculizumab due to poor haemolysis control and were not considered as having switched from eculizumab to crovalimab per study protocol. All four patients achieved sustained terminal complement inhibition. LIA levels were maintained below the limit of quantitation (<10 U/ml) after day 2 (Figure 1A). Free C5 levels were sustained at <0.5 μg/ml after day 43 (Figure 1B). LDH levels declined rapidly from baseline and were subsequently maintained at <1.5 × ULN (Table 1; Figure 1C). Crovalimab concentrations on a maintenance subcutaneous dose were sustained above the target concentration associated with complement activity inhibition (Figure S1).12 The crovalimab pharmacokinetics profiles were similar among these patients and the overall study population (data not shown). Two patients developed antidrug antibodies, with no impact on pharmacokinetics or pharmacodynamics. No BTH events were reported (Table 1). However, one patient had an apparent drop in haemoglobin level from baseline at week 108 along with elevated LDH (1.74 × ULN), reflecting subclinical BTH in the context of a concurrent nasopharyngitis adverse event. This patient's haemoglobin level recovered to 12.0 g/L at the next visit (week 116). Three of the four patients were transfusion avoidant. The other patient required 13 transfusions in the 33 weeks between enrolment and clinical cut-off. This patient had an underlying diagnosis of aplastic anaemia and required 24 transfusions in the year before enrolment. Overall, the response to previous therapy in these patients with C5 SNP was consistent with the typical poor haemolysis control experienced with eculizumab or ravulizumab in such patients.5, 6 With crovalimab treatment, all four patients achieved sustained terminal complement inhibition and haemolysis control consistent with all patients in COMPOSER.10 Three serious adverse events (SAEs) were reported in two patients. One patient had two SAEs (bile duct stone and cholelithiasis) that were not treatment related, and the other had a treatment-related SAE (upper respiratory tract infection) that resolved while on treatment. The only adverse event that occurred in >1 patient was nasopharyngitis (n = 3), of which no cases were serious or treatment related. One patient reported mild pruritus on day 23 followed by mild fatigue and pruritus on day 58 that were treatment related and resolved spontaneously without recurrence or dose modification. Generally, crovalimab was well tolerated, and its safety profile in these patients was consistent with that in the overall COMPOSER population.10 In conclusion, crovalimab achieved haemolysis control in patients with the C5 c.2654G → A SNP by targeting a different C5 epitope from that of eculizumab and ravulizumab,9 enabling efficacy in a wider range of patients with PNH than traditional anti-C5 treatment.5, 6 Additionally, crovalimab is injected subcutaneously q4w, thus reducing the treatment burden versus intravenously administered anti-C5 treatments.13 Although only four patients (9%) enrolled in COMPOSER carried the C5 polymorphism, the SNP prevalence was higher than in other studies (≤3.2%).5, 6 Two larger phase 3 trials are underway to evaluate the efficacy and safety of crovalimab versus eculizumab in patients with PNH who were previously treated with complement inhibitors (COMMODORE 1 [NCT04432584]) and those who were not (COMMODORE 2 [NCT04434092]).14, 15 Patients with a known C5 polymorphism will be eligible for enrolment in COMMODORE 1, enabling further assessment of crovalimab in more patients with the SNP. The authors thank the patients and their families as well as the investigators and clinical study sites. Jules Hernández-Sánchez, of F. Hoffmann-La Roche, Ltd, Basel, Switzerland, provided statistical support for this exploratory analysis. Medical writing assistance for this report was provided by Bena Lim, PhD, of MediTech Media, Ltd, and funded by F. Hoffmann-La Roche, Ltd. This study is sponsored by F. Hoffmann-La Roche, Ltd, and Chugai Pharmaceutical Co., Ltd. The study sponsors developed the protocol in cooperation with the investigators. Conceptualisation: JN, RPdL, AR, IPP, and AS; data acquisition: JR, AK, SI, and KU; data analysis: JR, MB, AK, AR, IPP, and AS; interpretation: JN, KU, AR, JR, IPP, and AS; writing (original draft): JR, AS, IPP, and AK; writing (review and editing): all authors. Patient recruitment for the COMPOSER study: JN, KU, SI, RPdL, and AR. Jun-ichi Nishimura received research funding and honoraria from and served as an advisor to Alexion, received funding from F. Hoffman-La Roche to his institution to conduct this trial, is an advisor to Chugai, F. Hoffmann-La Roche, Sanofi KK, BioCryst, Apellis, and Novartis, and holds a patent for WO2020/027279 (anti-C5 antibody dosage regimen). Kensuke Usuki has received fees for speaker bureau/expert testimony from Alexion and Novartis and research grant/funding to his institution from Alexion, Aperi, Chugai, F. Hoffmann-La Roche, and Novartis. Julia Ramos is an employee of Genentech, a Member of the Roche Group, and received fellowship support from Genentech, a Member of the Roche Group. Satoshi Ichikawa has received research funding from F. Hoffmann-La Roche to his institution. Muriel Buri is an employee of F. Hoffmann-La Roche. Anna Kiialainen is an employee of and holds stock in F. Hoffmann-La Roche. Alexandre Sostelly is an employee of F. Hoffmann-La Roche. Régis Peffault de Latour is a consultant to Alexion, Amgen, Apellis, Novartis, Pfizer, and Swedish Orphan Biovitrum AB, received research funding from Alexion, Amgen, Novartis, and Pfizer, and received honoraria from Alexion, Amgen, Apellis, Novartis, Pfizer, and Swedish Orphan Biovitrum AB. Ido Paz-Priel was an employee of Genentech, a Member of the Roche Group, and is currently employed at Graphite Bio. Alexander Röth is a consultant to Alexion, Apellis, BioCryst, Bioverativ, F. Hoffmann-La Roche, and Novartis, received research funding from F. Hoffmann-La Roche and honoraria from Alexion, BioCryst, Bioverativ, F. Hoffmann-La Roche, and Novartis, and sits on advisory committees for Alexion, Apellis, BioCryst, F. Hoffmann-La Roche, and Novartis. For up-to-date details on Roche's Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents, see: https://go.roche.com/data_sharing. Anonymized records for individual patients across more than one data source external to Roche cannot, and should not, be linked due to a potential increase in risk of patient reidentification. Figure S1 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts