Monitoring of patients with paroxysmal nocturnal hemoglobinuria on a complement inhibitor.

Abstract

The management of the rare, life threatening and chronic disease, paroxysmal nocturnal hemoglobinuria (PNH) was revolutionized by the introduction of the terminal complement inhibitor therapy, eculizumab (Soliris), which was licensed in 2007.1 Since that time, numerous studies have highlighted the improvement in organ damage on a C5 inhibitor, such as reduction in thrombotic events, improvement in renal function and pulmonary hypertension and, importantly, improvement in quality of life and survival for patients with PNH.2 These improvements are perhaps not surprising with the understanding that the morbidities and mortality in PNH relate to uncontrolled terminal complement activation. As PNH is a rare disease with most treating physicians having only 1–2 patients, as well as the need for long-term treatment, the monitoring for patients on treatment can be complex. Paroxysmal nocturnal hemoglobinuria arises on the background of bone marrow failure (BMF) which may require concomitant management due to potential evolution, especially due to the longer-life expectancy when on a C5 complement inhibitor. This longer life-expectancy can also result in clonal remission of PNH in some patients. Patients with PNH are often iron deficient at diagnosis but may acquire iron overload after years of terminal complement inhibition. The occurrence of extravascular hemolysis, which manifests only after initiation of C5 inhibition, can be evaluated as described below.3 It is important to be aware of all these nuances for the holistic management of a patient with PNH. We therefore report here, a monitoring strategy to support management of patients optimally worldwide. The algorithm is based on 18 years of experience and applies predominantly to C5 inhibition although could be extrapolated to complement inhibition in general for PNH (Figure 1). When a patient is commenced on a complement inhibitor, the first measure is to ensure that the complement therapeutic achieves terminal complement blockade to prevent the life-threatening complications seen in PNH (Figure 1). In PNH, this can often be done by measuring trough LDH (just prior to dosing, on multiple occasions) in patients with high baseline LDH before commencing treatment. In rare cases, if LDH is not elevated before commencement of therapy (e.g., thrombosis without hemolysis,4 proceed straight to terminal complement activity assays. After the commencement of therapy, if the LDH is unchanged within 6 weeks of commencing then assess for the C5 polymorphism.5 Eculizumab is designed as a terminal complement inhibitor and achieves this in patients with PNH (with up to 20% requiring a dose adjustment from the labeled dose), apart from those with a rare genetic polymorphism to C5. The prevalence of this polymorphism worldwide is <0.01%. The UK PNH Service and Johns Hopkins Hematology had seen no cases of this in over 600 patients treated with complement inhibition, at the time of writing. The patients with the C5 polymorphism are the only true “poor responders” or “non-responders” to eculizumab. “Poor/suboptimal responder” has been widely used in the context of patients having residual anemia on eculizumab. This terminology can be misinterpreted as the risk of thrombosis, the leading cause of death from PNH before the availability of terminal complement inhibitors, is still markedly reduced in these “suboptimal” responders, despite presence of residual anemia. After 6 weeks and regularly thereafter, if the LDH is reduced ideally to a target of <1.5 times the upper limit of normal (xULN), treatment should be continued. If the LDH is >1.5xULN and the CH50 (50% hemolytic complement assay), a measure of terminal complement activity, demonstrates complete blockade, therapy should be continued. Alternate causes of elevated LDH should be evaluated, especially as LDH is released from tissues other than red blood cells such as liver, bone and cardiac muscle. Testing for LDH isoenzymes to identify the source is not routinely available. If the LDH is elevated and CH50 is not blocked, inadequate dosing should be considered, and adjustments made, or if the patient is on eculizumab, a switch to ravulizumab has been shown to control these pharmacokinetic breakthrough events.6, 7 In patients with good control of intravascular hemolysis (i.e., LDH <1.5xULN) with complement inhibition, intermittent episodes of breakthrough intravascular hemolysis (worsening anemia, increased LDH and reappearance of PNH symptoms) may be triggered on occasions, by complement amplifying conditions, for example, infection. Although hemolysis improves on resolution of the inciting trigger, it is important to continue or even administer an early additional dose of C5 inhibition during these episodes and closely monitor patients. The complement inhibitor should not be discontinued at these times as the serious consequences of PNH could appear without terminal complement blockade. Once terminal complement inhibition has been achieved as assessed by LDH and/ or CH50, patient safety from the end organ damage (e.g., renal function, pulmonary pressures) and other specific complications of PNH, for example, thrombosis is more assured. Monitoring should still continue, as should education on complement amplifying conditions for breakthrough of complement blockade events and risk of Neisserial infections. Anticoagulants, if used as primary prophylaxis, can be safely discontinued once stable complement blockade is achieved with C5 inhibitors. Measures to enhance treatment further can now be safely considered. The next measure to take into consideration is the hemoglobin level. If it is satisfactory for the patient for their daily activities and quality of life, we would simply continue the therapy, even if the hemoglobin level is not within the normal range. Reduction in intravascular hemolysis is more predictive of an improvement in fatigue than an improvement in anemia and therefore remains the primary goal. The serious consequences of PNH are also significantly reduced with the assurance earlier of terminal complement control. However, if the hemoglobin is low and the patient is symptomatic or needing transfusions, we would then evaluate comprehensively. There are multiple reasons for anemia in a patient with PNH on a complement inhibitor (Figure 1) and it is important all are systematically explored for optimal care of the patient. The next measure would be the reticulocyte count. If the reticulocyte count is low or normal then concomitant bone marrow failure should be investigated and treated appropriately (according to available guidelines) in conjunction with complement inhibition. The hematinic deficiencies, such as folate deficiency, are a simple, common, reversible cause of continuing low hemoglobin and easily rectified with supplementation. Patients might need exogenous erythropoietin particularly if they have renal dysfunction, an inappropriately low erythropoietin level for the degree of anemia and in the context of evolution/coexistence of” low risk” MDS with erythropoietin level <500 mU/mL. Iron overload, due to chronic transfusions prior to therapy and/ or ongoing transfusion need, can be seen associated with either low or elevated reticulocyte count. When the reticulocyte count is elevated, extravascular hemolysis would be considered as the likely cause for anemia.3 For completeness, others have also been listed (Figure 1), such as hypersplenism (especially following Budd-Chiari syndrome), bleeding as the patients may be thrombocytopenic or have dysfunctional platelets if they have an underlying MDS, presence of alloantibodies from previous transfusions or again ineffective erythropoiesis if they do have underlying MDS rather than AA. Proximal complement inhibitors are in development to aim to address the clinically significant extravascular hemolysis that affects a proportion of patients. Any advancement in therapy for patients with PNH should be without additional risk for the patient, regardless of the target in the complement cascade. Although reticulocytosis may be evident in a significant proportion of patients, this does not translate to an increased mortality or thrombosis risk and such patients were managed supportively over the last 18 years. These same patients still have a significant improvement in survival, particularly as thrombosis risk is reduced.2 Addition of a proximal complement inhibitor to a C5 inhibitor has been shown to significantly improve or even normalize hemoglobin levels while also keeping the safety net of terminal complement inhibition in reducing thrombosis risk and end organ damage in those patients affected significantly enough.8 Although initial promising data are available regarding monotherapy with a proximal complement inhibitor, long-term data for efficacy in achieving and maintaining terminal complement inhibition (the goal in therapy to avoid significant morbidity and mortality), without breakthrough intravascular hemolysis, as well as safety and tolerability are needed.9 Although the direct antiglobulin test can become positive to C3d in patients who are receiving treatment with a C5 inhibitor, this does not necessarily correlate with neither the clinical picture nor with the degree of C3 loading on PNH red cells. Finally, the proportion of PNH cells should be monitored by flow cytometry at regular intervals in patients safely maintained on a complement inhibitor to detect spontaneous remission, which is seen in a variable proportion (5–15%) of patients.10 Rarely, patients are able to stop complement inhibition if the clonal remission is sustained, and no hemolytic activity is evident (the aim is usually at least two consecutive flow cytometric evaluations confirming proportions of PNH granulocytes and monocytes of <10%). Flow cytometry testing should still be continued after discontinuation of therapy to ensure the proportion of PNH cells do not increase which may require re-institution of therapy. In rare instances, PNH clonal regression might herald the onset of dysplastic or leukemic transformation and this should be evaluated with comprehensive bone marrow studies. Therefore, in order to manage a patient with PNH optimally all of these possible causes of anemia must be considered and investigated appropriately (Figure 1). Many physicians today will not have managed PNH before the eculizumab era, which can allow for complacency as the significant complications of terminal complement activity is not seen as frequently in this anti-complement therapy era with prompt institution of C5 inhibition therapy. Hemoglobin level alone is not used as a marker of disease severity in PNH. It is known that some patients with PNH may be able to compensate for significant hemolysis by increasing erythropoiesis and have normal or near normal hemoglobin levels but still have significant risk from the disease due to ongoing terminal complement activity. PNH is a terminal complement mediated disease and includes complement activation of PNH platelets and white cells (likely a major role in inciting thrombotic manifestations) and is therefore more than a red cell disorder - this needs to be highlighted and remembered. Overall, we present a practical and useful clinical tool/ algorithm for monitoring patients on anti-complement therapy based on our clinical experience, which will be useful for physicians treating this rare disease with long-term complement inhibition. A.G.K. has served as on advisory boards for Alexion, Amgen, Apellis, Biocryst, Celgene, Novartis, Ra Pharma, and Regeneron and has received travel grants from Achilleon, Celgene, and Ra Pharma. R.A.B. has received grant support from Alexion. A.H. is an employee of Alexion. The data that support the findings of this study are available from the corresponding author upon reasonable request.