A growing panoply of options for patients with paroxysmal nocturnal hemoglobinuria

In this issue, the COMMODORE 1¹ and COMMODORE 2² trials of crovalimab for previously treated and treatment-naïve paroxysmal nocturnal hemoglobinuria, respectively, are presented, demonstrating efficacy and safety for both patient populations in comparison with the benchmark, eculizumab.

Paroxysmal nocturnal hemoglobinuria (PNH) is a hemolytic anemia characterized by sometimes painful intravascular hemolysis (IVH), pancytopenia with increased infections, and significant risk of potentially fatal venous thrombosis especially of the sagittal sinuses and hepatoportal system. Instead of autoantibodies, PNH is a clonal disorder driven by the acquisition of somatic mutations in PIGA. These mutations, as well as rarer mutations in related genes, lead to loss of the glycosylphosphatidylinositol (GPI) anchor required for presenting an array of cell-surface proteins. Loss of GPI anchorage has wide ranging consequences, key among them is loss of CD55 and CD59, proteins essential for regulating cell surface-aspects of complement (Figure 1). Without CD55 and CD59, complement complexes cannot be removed from the cell surface, leading to C5 activation, membrane attack complex (MAC) formation, and cell lysis. This causes erythrocyte destruction, thrombosis through inappropriate platelet activation, and infections secondary to leukopenia. As the PIGA-deficient clone or clones expand—patients often acquire multiple, separate mutations—the risk and severity of these complications increase. The causes and mechanisms underlying this expansion are complex and beyond the scope of this commentary.^3-5

Historically, non-transplant management of PNH consisted of anticoagulation—typically anti-vitamin K—and supportive care. However, lifelong anticoagulation has obvious hemorrhagic risks and does not address ongoing hemolysis. Recognition that the clinical manifestations of PNH arise from inappropriate complement-mediated cell membrane attack led to the development of the terminal complement inhibitor eculizumab. By targeting C5, the key enzyme downstream of all complement pathways, eculizumab prevents the final step in cell lysis: membrane disruption by MAC (Figure 1). Provided as fortnightly intravenous infusions, eculizumab has proven revolutionary in the treatment of PNH, eliminating anemia and transfusion requirements for 49% of subjects in the TRIUMPH study⁶ (increasing to 88% in long-term follow-up⁷) and reducing the risks of thrombosis and its recurrence by 85%.⁸ It received FDA approval as the first PNH-specific treatment in 2007. As such, it has become the gold standard against which novel treatments are assessed.

Crovalimab is a humanized anti-C5 monoclonal antibody engineered to be recycled instead of cleared following Fcγ receptor binding, leading to a prolonged half-life and the ability to be administered by subcutaneous injection. After the phase 1/2 COMPOSER trial demonstrated safety and potential efficacy in PNH, two phase 3 (COMMODORE) trials were conducted, the results of which are presented in this issue. COMMODORE 1 examined crovalimab as an alternative treatment for PNH in adult patients previously receiving eculizumab. Subjects were randomized to continue maintenance eculizumab or to receive crovalimab as injections every 4 weeks for 24 weeks after the initial loading doses. All subjects were offered extension on or crossover to crovalimab after the primary treatment period. Both arms maintained CH50 reduction. Subjects on crovalimab experienced further reduction of C5 levels from pre-study maintenance levels (0.18 g/L from 0.3 g/L baseline vs. 0.28 g/L throughout for eculizumab). Clinically, 92.9% of subjects maintained hemolysis control on crovalimab versus 93.7% on eculizumab. Breakthrough hemolysis (BTH) was similar across arms (10.3% vs. 13.5%).

COMMODORE 2 similarly examined the safety and efficacy of crovalimab as first-line (complement inhibition naïve) treatment for PNH. Patients underwent the same randomization and treatment protocols as COMMODORE 1, including optional extension with crossover to crovalimab. In treatment-naïve patients, crovalimab was non-inferior to eculizumab with regard to hemolysis control (79.3% vs. 79.0% for eculizumab), LDH normalization, and transfusion avoidance (65.7% vs. 68.1%). Rates of BTH were similar (10.4% vs. 14.5%), as was stabilization of hemoglobin (63.4% vs. 60.9%). The slightly lower response rates for COMMODORE 2 almost certainly reflect differences in enrolling treatment-naïve instead of previously controlled patients.

In both trials, crovalimab had an excellent safety profile. Injection site reactions—of particular note as crovalimab is a subcutaneous medication—were uncommon (COMMODORE 1: 9.1%; COMMODORE 2: 5.2%), mild to moderate, and did not require dose adjustments. Infusion-related reactions during the initial loading dose(s) occurred with frequencies similar to eculizumab (COMMODORE 1: 13.6%; COMMODORE 2: 15.6% vs. 13.0% for eculizumab). Transient immune complex reactions (crovalimab-C5-eculizumab complexes, a known risk when changing immunotherapies that recognize different epitopes of a target) were seen in 15.9% of subjects, were generally mild to moderate, managed symptomatically, and did not require therapeutic adjustments. Only a single treatment-related severe adverse event (thrombocytopenia) was identified across either trial. Infectious rates were similar across the arms of both trials (COMMODORE 1: 40.9% vs. 35.7% for eculizumab; COMMODORE 2: 24% vs. 36%).

Notably, in both trials, a third to half of patients were successfully self-administering after 2 months of initiation, the majority of which were performed at home without the assistance of a healthcare provider, yet without loss of efficacy. Furthermore, approximately 85% of all subjects in COMMODORE 1 and crossed-over subjects in COMMODORE 2 preferred crovalimab to eculizumab.

As occasionally happens during the conduct of large trials, while the COMMODORE trials were ongoing several additional PNH agents received regulatory approval and reached market. The first of these was ravulizumab, an eculizumab biosimilar that also employs antibody recycling technology similar to crovalimab, and has safety and efficacy profiles similar to eculizumab with improved BTH control while on an other-monthly infusion schedule.^{9, 10} With its approval in 2018, ravulizumab quickly became the standard of care, although eculizumab remains the benchmark given its (currently) greater clinical experience.

In addition to alternative C5 inhibitors, proximal complement inhibition targeting C3 has proven attractive. Inhibition of C5 leads to suppression of IVH, but ongoing opsonization by C3 leads to phagocytic extravascular hemolysis (EVH) within the reticuloendothelial system.¹¹ Pegcetacoplan was examined in the PRINCE¹² and PEGASUS¹³ trials, demonstrating efficacy and superiority to eculizumab and leading to its approval in 2021.

More recently, two oral agents have become available. Danicopan targets Factor D, a cofactor essential for C3 opsonization and alternative pathway activation, was found to be effective as an adjuvant therapy for breakthrough EVH while on C5 inhibitors in the ALPHA trial leading to its approval.¹⁴ More intriguingly, iptacopan, which targets Factor B, another C3 cofactor, was examined as a monotherapy in the APPOINT-PNH and APPLY-PNH trials and found to be at least as effective as eculizumab in controlling IVH and more effective in preventing BTH.¹⁵ This promises patients the freedom of an entirely oral monotherapy, and small early-phase studies have suggested similar applications for danicopan as well, hinting at a new paradigm in PNH.^{16, 17}

Yet the completion of any clinical trial, regardless of competing studies, should always be considered cause for celebration and publication, reflecting the effort and commitment of the researchers, providers, and patient-subjects. Despite the advent of the above agents, the COMMODORE trials represent important advances in the treatment of PNH. Although intravenous ravulizumab has a more favorable administration schedule (subcutaneous ravulizumab is administered weekly), the COMMODORE trials demonstrate that crovalimab can be effectively self-administered subcutaneously, providing an alternative to lengthy infusions and a flexibility rivaled only by oral administration. It is also critical to note that because the current oral agents target cofactors of the complement pathway, the core components of the complement system remain uninhibited, with the risk of sudden, rapid hemolysis in the case of missed doses.¹⁸ This raises adherence issues that are less acute for agents such as crovalimab. Furthermore, crovalimab targets a different C5 epitope, overcoming eculizumab/ravulizumab resistance due to C5 polymorphisms, notably R885H found in Japanese populations. Finally, whereas at its market introduction eculizumab was titled “the most expensive drug in the world”—prior to the advent of cell and gene therapies—the expansion of treatment options should exert economic pressures upon drug pricing in a field where, for better or worse, treatment decisions are greatly influenced by financial considerations.

As more agents see broader use, data will gradually become available comparing their outcomes, whether through direct head-to-head trial comparisons, through meta-analyses of existing trials, or through compilation of post-marketing data. In the meantime, the choice of agent will ultimately require an informed discussion between provider and patient, weighing the relative benefits and risks that each agent entails, including availability, schedule, route of administration, side effect profiles, response to prior lines of treatment, and economics.

Although the focus of these trials was PNH, eculizumab has proven helpful in other conditions, especially atypical hemolytic uremic syndrome and transplant-associated thrombotic microangiopathy (TA-TMA). What role newer agents may play in treating these conditions is an area of active research and trials. For example, in TA-TMA, data indicate that inadequate exposure to eculizumab leads to inferior outcomes,¹⁹ suggesting roles for longer-acting agents such as ravulizumab or crovalimab. Trials are currently examining such. A role for C3 inhibition is less understood, and little data exist regarding non-C3/C5 therapies, especially as monotherapy. Consequently, the availability of an expanding range of options for targeting complement is nothing other than a welcome development, and promises to further advance our treatments for PNH and other life-threatening conditions.

The author has no financial, competing, or other relevant interests to disclose.