Successful treatment of a PNH patient non‐responsive to eculizumab with the novel complement C5 inhibitor coversin (nomacopan)

Abstract

Paroxysmal Nocturnal Haemoglobinuria (PNH) is a rare acquired life-threatening disease characterised by complement-induced haemolysis and a high incidence of thrombosis. The monoclonal antibody eculizumab binds to complement C5 and prevents its activation and cleavage into C5a and C5b. Treatment of PNH patients with eculizumab decreases haemolysis, transfusion requirements and the risk of thrombosis (Hillmen et al., 2004, 2006, 2007; Brodsky et al., 2008; Kelly et al., 2011). Failure to respond to eculizumab has been reported in a subgroup of Asian PNH patients (Nishimura et al., 2014). These patients had a genetic variant of C5, a missense mutation leading to c.2654G>A (p.Arg885His), which occurs in approximately 3·5% of Japanese and 1% of Chinese Han populations. This genetic variant interferes with the binding of eculizumab to C5 (Nishimura et al., 2014). For PNH patients with this genetic variant there is no effective treatment. A new small (17 kDa) protein complement inhibitor named coversin is now in phase 2 clinical development. Like eculizumab, coversin prevents cleavage and activation of C5, but it binds C5 at a different site (Jore et al., 2016). Administration to healthy volunteers proved safe and demonstrated inhibition of terminal complement activation. We report safety, efficacy, pharmacokinetic and pharmacodynamic data from the first ever PNH patient treated with coversin. The patient, a 30-year-old Caucasian man (BMI 21 kg/m2) with a stable PNH granulocyte clone of 90% (monocyte and erythrocyte clone sizes shown in Table SI), suffered from severe haemolysis. LDH (lactate dehydrogenase) and haemoglobin levels before start of treatment are shown in Fig 1A. There were several episodes of very severe haemolysis with LDH levels >2500 U/l (ULN 250 U/l). Three episodes of increased haemolysis were accompanied by a transient deterioration in kidney function. During one episode (LDH 2813 U/l), eGFR decreased to 24 ml/min. Main symptoms were extreme fatigue and muscle dystonia. He had no history of thrombosis, no prior transfusions and no underlying aplastic anaemia or myelodysplastic syndrome. He started treatment with eculizumab four years after diagnosis. Unexpectedly, he remained severely haemolytic despite adequate trough drug levels and no antidrug antibodies (ADA). Other causes of haemolysis were excluded. The PNH clone size had not changed (Table SI). Treatment was discontinued when, several months into treatment, the patient experienced increased haemolysis (LDH 9 × ULN) and macroscopic haemoglobinuria one day after receiving 900 mg eculizumab. DNA analysis of the coding region of C5 showed a heterozygous genetic variant in C5, c.2653C>A, which predicts p.Arg885Ser (Fig 2A). The amino acid substitution is different from that described previously (Nishimura et al., 2014), but occurs with the same amino acid of the C5 protein. The same genetic variant was demonstrated in the DNA of the patient's healthy father and not in his healthy mother. In vitro studies confirmed a diminished response to eculizumab (measured by inhibition of CH50) compared to a normal control (NC), whereas coversin was equally effective in the patient and in the normal control, reducing CH50 to less than 8 U Eq/ml (Fig 2B)(the lower limit of quantification). The AK578 study (ClinicalTrials.gov, NCT02591862) was an open label phase 2 study designed for this and potentially up to five other patients with genetic variants of C5, which rendered them unresponsive to treatment with Eculizumab (Data S1). The study was approved by the independent ethics committee of the Radboudumc and was conducted in agreement with Good Clinical Practice guidelines and according to the Declaration of Helsinki. When the inclusion period closed, the patient described here was the only participant in this trial. Coversin was administered by subcutaneous injection at an ablating dose of 0·57 mg/kg/day on day 1, followed by a maintenance dose of 0·14 mg/kg/day thereafter. Protocol defined that blood samples were drawn for CH50 activity, free coversin levels and anti-drug antibodies (ADA). The primary efficacy goal was a reduction in serum LDH from day 0 (pre-dose) to day 28. Secondary goals were haemoglobin at days 28, 90 and 180 compared to baseline, LDH at days 90 and 180, haptoglobin level at days 28, 90 and 180 compared to baseline, transfusion independency and quality of life, assessed by the FACIT and EORTC QL-C30 instruments on days 0, 28, 90 and 180. The patient started with the ablating dose of 0·57 mg/kg/day. There was a good initial response with CH50 decreasing from 100 U Eq/ml to <8 U Eq/ml (Fig 1B). He continued with the maintenance dose of 0·14 mg/kg every 24 h. Clinical symptoms and laboratory markers of haemolysis improved during the first five days of treatment. However, on day 6, the patient experienced haemolysis-associated symptoms with dark urine, hours before the next injection, and no further decrease of his LDH occurred (Fig 1B). This was accompanied by an increase in CH50 and decrease in free coversin (Fig 1C,D). According to the protocol, the dose was doubled to 0·29 mg/kg/day (a total once-daily dose of 19 mg). However, this dose was not sufficient to cover the 24 h interval between injections. A dose of 0·14 mg/kg every 12 h was therefore used, providing long-term stable complement inhibition with CH50 levels <8 U Eq/ml and trough coversin levels of 25–78 ng/ml (Fig 1). LDH at day 28 was 474 U/l (ULN 250 U/l) compared to 1400 U/l immediately before the first dose of coversin. LDH levels at days 90 and 180 were 404 and 373 U/l respectively. Haemoglobin remained stable (12·6 g/dl on day 0 and 12·1, 12·6 and 12·9 g/dl after 28, 90 and 180 days). The relatively high haemoglobin level in this patient, even before start of coversin, together with a high reticulocyte count, proves that the bone marrow is compensating for the haemolysis. Reticulocyte counts did not significantly decrease after start of coversin (Figure S1). This may be caused by C3-mediated extravascular haemolysis, which occurs frequently during treatment with complement C5 inhibitors (Notaro & Sica, 2018). Clinical symptoms improved substantially with only mild complaints of fatigue remaining. His quality of life improved (data not shown). Treatment was well tolerated. At one point the patient had a transient skin reaction grade 2 at the injection site. Adverse events were CTCAE grade 1 otherwise (Table SI). ADA were detected from day 15 onwards but did not result in decreased free coversin levels or decreased complement inhibition (Fig 1). Recently, treatment with coversin was shown to suppress complement activity in the setting of post-HSCT (haematopoietic stem cell transplantation) thrombotic microangiopathy in a one-year-old patient suffering from HSCT-associated TMA (thrombotic microangiopathy). This patient carried the variant described in the Asian patients (c.2654 G>A (p.Arg885His) in C5, rendering him resistant to eculizumab treatment (Goodship et al., 2017). Our PNH patient continued treatment with coversin in the AK580 extension study (ClinicalTrial.gov, NCT03588026) and has been stable for more than 1·5 years (Fig 1). After two years, he experienced one transient episode of break-through haemolysis during an influenza B infection for which no additional coversin administration was necessary. Based on our results, we conclude that coversin may be a useful s.c. therapy for PNH patients who are resistant to eculizumab therapy due to genetic variants of C5. The AK585 study (ClinicalTrials.gov, NCT03427060) will investigate this further. S.S. performed research, analysed data and wrote the manuscript. M. N. performed research, analysed data and contributed vital analytical tools. I.M. performed research, analysed data and contributed vital analytical tools. W.W. designed research. J.N. designed and performed research and analysed data. Y.K. designed and performed research and analysed data. N.B. designed research. P.M. designed and performed research, analysed data and wrote the manuscript. S.L. designed and performed research, analysed data and wrote the manuscript. S.S, N.B. and S.L. report no conflicts of interests. M.N. and W.W. are employed by Akari Therapeutics. I.M. has received research-funding from Akari Therapeutics. P.M. has received travel support and fees for consultancy from Akari Therapeutics. Y.K. and J.N. have participated in advisory committee meetings for Alexion Pharma. Fig S1. Haemoglobin and reticulocyte count before and after start of coversin treatment. Table SI. PNH clone size of the patient. Table SII. Adverse events (possibly) related to coversin. Data S1. Supplementary methods. 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.