Treatment and prognosis of paroxysmal nocturnal hemoglobinuria

INTRODUCTION — Paroxysmal nocturnal hemoglobinuria (PNH) is a rare disorder in which hematopoietic stem cells and their cellular progeny have lost the ability to anchor certain proteins to the cell surface. Loss of the complement inhibitors, CD55 and CD59, on the surface of red blood cells leads to chronic and/or paroxysmal intravascular hemolysis and a propensity for thrombosis. Patients may also have hypocellular or dysplastic bone marrow, and a subset have clinically significant aplastic anemia or myelodysplastic syndromes/neoplasms [1]. The availability of biologic therapies that target the underlying complement-mediated hemolysis has dramatically improved the prognosis of PNH.

This topic discusses the treatment and prognosis of PNH.

Clinical manifestations, diagnosis, and pathophysiology of PNH are presented separately. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria" and "Pathogenesis of paroxysmal nocturnal hemoglobinuria".)

BASELINE EVALUATION

Clinical/laboratory — The diagnostic evaluation for PNH, including clinical and laboratory studies, flow cytometry to document hematopoietic clonality, and evaluation for associated aplastic anemia (AA) or myelodysplastic syndromes/neoplasms (MDS) is described separately. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Diagnosis and classification'.)

The following should be performed prior to initiating treatment for PNH:

●Clinical – The history and physical examination should elicit the nature and severity of hemolysis-associated findings (eg, fatigue, dyspnea, tachycardia, jaundice, dark or red urine), muscle or bone pain, and recurrent infections or excessive bleeding/bruising from cytopenias associated with bone marrow failure.

The patient should also be evaluated for findings of venous thrombosis, including atypical sites such as mesenteric or hepatic veins (eg, abdominal pain, increasing abdominal girth, ascites, esophageal or gastric varices, hepatosplenomegaly) or cerebral vein thrombosis (eg, intractable headache, abnormal neurologic findings).

Comorbid conditions and performance status should be assessed (table 1).

●Laboratory studies

•Hematology

-Complete blood count (CBC) with differential

-Reticulocyte count

-D-dimer

•Chemistries

-Serum electrolytes, glucose, kidney function tests, and liver function tests, including lactate dehydrogenase (LDH), fractionated bilirubin, haptoglobin

-Serum iron, total iron binding capacity, and ferritin

Previous laboratory results should be reviewed, as they may reflect the duration and degree of hemolysis, size of the PNH clone, and the trajectory of the condition.

●Flow cytometry – The population of PNH-affected cells is measured by flow cytometry to assess the percentage of CD55-negative and CD59-negative cells. Both granulocytes and red blood cells (RBC) should be analyzed, as the population of PNH cells will be underestimated if only RBCs are analyzed due to selective destruction of PNH RBCs by complement coupled with dilution from recent RBC transfusions. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Flow cytometry/FLAER'.)

●Bone marrow examination – Bone marrow aspirate and biopsy are important for defining cellularity, dysplasia, and other abnormalities associated with AA and/or MDS. Bone marrow examination should include microscopy, karyotype, fluorescence in situ hybridization (FISH; for MDS-related genetic abnormalities), quantitative CD34 cell count, and iron staining. (See "Aplastic anemia: Pathogenesis, clinical manifestations, and diagnosis", section on 'Bone marrow examination' and "Clinical manifestations, diagnosis, and classification of myelodysplastic syndromes (MDS)", section on 'Bone marrow examination'.)

●Imaging – Imaging is used to evaluate clinical findings that may suggest thromboembolism.

•Symptomatic for thrombosis – Symptoms or signs that may suggest thromboembolic (TE) abnormalities, such as deep vein thrombosis or pulmonary embolism, should also consider atypical sites, such as mesenteric or hepatic veins (eg, abdominal pain, increasing abdominal girth, ascites, esophageal or gastric varices, hepatosplenomegaly) and cerebral vein thrombosis (eg, intractable headache, abnormal neurologic findings).

Diagnostic evaluation of findings suggestive of TE abnormalities is described separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity" and "Acute portal vein thrombosis in adults: Clinical manifestations, diagnosis, and management" and "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

•Asymptomatic – We do not routinely perform imaging for TE abnormalities in asymptomatic patients, as there is no evidence to support a benefit of routine imaging for deep vein thrombosis in the absence of a clinical suspicion.

Evaluation for transplantation — For individuals who have PNH with co-existing severe bone marrow failure (ie, severe AA or higher-risk MDS), we refer the patient to evaluate eligibility for allogeneic hematopoietic cell transplantation (HCT). (See "Determining eligibility for allogeneic hematopoietic cell transplantation".)

OVERVIEW — Management is guided by the nature and severity of PNH-related symptoms and laboratory abnormalities.

Categories of PNH — Classification of PNH is based on the severity of hemolysis-related symptoms and the degree of bone marrow hypoplasia. PNH is a dynamic condition, and the category can evolve. As examples, patients who were originally classified as having subclinical PNH or hemolytic PNH may later evolve to PNH with bone marrow failure. Details of clinical manifestations, diagnosis, and classification of PNH are presented separately. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Diagnostic criteria'.)

Although surrogate markers such as the serum level of serum lactate dehydrogenase (LDH) or the size of the leukocyte PNH clone differ between categories, they are not used as classification criteria because they do not correlate precisely with the severity of symptoms and bone marrow findings.

PNH is classified as follows:

●Hemolytic (classical) PNH

•Prominent symptoms related to hemolysis (eg, fatigue, dyspnea, transfusion-dependence, episodic hemoglobinuria, thrombosis, pain, and/or organ dysfunction)

•LDH typically >1.5 x upper limit of normal (ULN)

•PNH granulocyte clone size typically >50 percent

•Normal white blood cell (WBC) count and platelet count or modest, asymptomatic leukopenia and/or thrombocytopenia

•Cellular bone marrow with erythroid hyperplasia and no significant dysplasia

Management of hemolytic PNH is discussed below. (See 'Symptomatic hemolytic PNH' below.)

●PNH with thrombosis

•PNH clone, usually >30 percent PNH granulocytes and LDH >1.5 ULN

plus

•Current thrombosis or a history of thromboembolic event(s)

Management of thrombosis in a patient with PNH is discussed below. (See 'PNH with thrombosis' below.)

●Subclinical PNH

•No substantial hemolysis-related symptoms, thrombosis, pain, organ dysfunction, and no need for transfusions

•LDH typically ≤1.5 x ULN

•PNH granulocytes typically ≤20 percent

•Normal WBC count and platelet count or modest, asymptomatic leukopenia/thrombocytopenia

•Normal or near-normal bone marrow cellularity and morphology

Management of subclinical PNH is discussed below. (See 'Subclinical PNH' below.)

●PNH with bone marrow failure

•Variable anemia-related symptoms, thrombosis, pain, or organ dysfunction

•Variable LDH level

•Variable granulocyte PNH clone size

•Prominent severe, symptomatic leukopenia and/or thrombocytopenia that meets criteria for severe aplastic anemia (AA) or higher-risk myelodysplastic syndromes/neoplasms (MDS)

Management of PNH with bone marrow failure is discussed below. (See 'PNH with bone marrow failure' below.)

Further details of the diagnosis and classification of PNH are presented separately. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Diagnostic criteria'.)

Stratification of management — Management of PNH is guided by the severity of symptoms, current or prior thromboses, degree of bone marrow failure, and extent of hemolysis, as described in the sections that follow. All symptomatic patients with PNH can benefit from supportive care, as described below. (See 'Supportive care' below.)

Our approach to management of PNH (algorithm 1) is consistent with recommendations of expert panels from the American Society of Hematology [2], Belgium [3], and Canada [4].

SYMPTOMATIC HEMOLYTIC PNH — For patients with symptomatic hemolytic PNH (eg, pain, thrombosis, organ dysfunction), who do not have severe bone marrow failure (BMF), we recommend treatment with a complement inhibitor, rather than supportive care alone (algorithm 1). Complement inhibitors are effective for symptom relief and prevention of thromboses and are associated with only modest toxicity. Although supportive care alone can alleviate pain and anemia-related symptoms, it may be associated with transfusion-associated iron overload, alloimmunization, and thrombotic events, which are common, difficult to predict, and may recur or progress despite anti-thrombotic therapy [5-9].

Selection of a complement inhibitor for initial treatment of symptomatic hemolytic PNH is discussed below. (See 'Choice of initial agent' below.)

Complement inhibitors have proven efficacy and safety for hemolytic PNH. Based on a randomized trial, the C5 complement inhibitor (C5i), eculizumab, was more effective than placebo for relieving symptoms and reducing red blood cell (RBC) transfusions [10,11]. Other randomized trials reported that another C5i, ravulizumab, was associated with efficacy and safety that were comparable to eculizumab [10-13]. At present, there are limited reports of initial treatment of hemolytic PNH with pegcetacoplan or other complement inhibitors.

●Eculizumab versus placebo – In the TRIUMPH trial, which randomly assigned 87 patients with severe PNH to eculizumab versus placebo for 26 weeks, eculizumab more effectively reduced transfusion-dependence and improved quality of life (QoL); there were no deaths or serious adverse events (AEs) related to eculizumab [10,11]. Eculizumab more effectively achieved RBC transfusion-independence (TI; 49 versus 0 percent, respectively) and led to clinically meaningful improved QoL, using two different assessment instruments [10,11]. There was no difference in overall survival (OS) between treatment arms during the 26-week trial and no deaths or serious AEs were attributed to eculizumab; one thrombotic event was reported in the placebo group.

Other studies of eculizumab are presented below. (See 'Eculizumab' below.)

●Ravulizumab versus eculizumab – In an open-label randomized trial in 246 treatment-naïve patients with PNH, ravulizumab and eculizumab achieved similar rates of TI and normalization of serum lactate dehydrogenase (LDH) [12]. In another phase 3 trial, among 195 clinically stable patients with PNH who were previously treated with eculizumab and then randomly assigned to continue eculizumab or switch to ravulizumab, the two agents achieved similar results [13].

Other studies of ravulizumab are presented below. (See 'Ravulizumab' below.)

Importantly, treatment with a complement inhibitor does not mitigate symptoms and complications of PNH-associated BMF, such as aplastic anemia (AA) or myelodysplastic syndromes/neoplasms (MDS). Management of patients who have severe BMF is discussed below. (See 'PNH with bone marrow failure' below.)

Choice of initial agent — For patients who require a complement inhibitor, we suggest initial treatment with ravulizumab rather than eculizumab, based on comparable efficacy and toxicity, but greater convenience, lower overall expense, and fewer episodes of pharmacokinetic breakthrough hemolysis [14]. Importantly, treatment of PNH with these agents must be continued indefinitely, is expensive, and does not ameliorate PNH-associated BMF; in addition, all complement inhibitors can be associated with increased risk for serious and life-threatening infections, especially those caused by encapsulated bacteria. (See 'Adverse effects' below.)

We favor a C5i, rather than other complement inhibitors for initial treatment of hemolytic PNH, because of their proven efficacy for symptom relief and thrombosis prevention, low toxicity, convenience of periodic infusions, long track record, and limited reports of initial treatment with other agents. (See 'Symptomatic hemolytic PNH' above.)

Administration of individual complement inhibitors is presented below. (See 'Complement inhibitors' below.)

The terminal half-life of ravulizumab is four times longer than that of eculizumab, which enables longer intervals between treatments (eg, eight weeks versus two weeks) and reduces annual costs; the safety profiles are similar. Both agents bind the same epitope on C5 and inhibit complement activity by preventing C5 convertase from cleaving C5 into C5a and C5b. (See "Pathogenesis of paroxysmal nocturnal hemoglobinuria".)

●Comparable efficacy – In an open-label randomized trial in 246 treatment-naïve patients with PNH, ravulizumab and eculizumab achieved similar rates of TI and normalization of serum LDH [12]. The trial randomly assigned 246 patients with LDH ≥1.5 times the upper limit of normal (ULN) to receive ravulizumab or eculizumab for 183 days. The two agents achieved similar rates of TI and normalization of LDH, improvement of fatigue (change from baseline in Functional Assessment of Chronic Illness Therapy [FACIT]-Fatigue score), and rate of breakthrough hemolysis.

In another phase 3 trial, among 195 clinically stable patients with PNH who were previously treated with eculizumab and then randomly assigned to continue eculizumab or switch to ravulizumab, the two agents achieved similar results [13].

●Similar toxicity – The two C5i agents have similar safety profiles [12,13]. Infections are the predominant AEs of both agents, but no meningococcal infections were reported among >200 patients in these trials; headache was reported in one-quarter of patients [12,13].

Treatment with a C5i (and other complement inhibitors) is contraindicated in patients with active meningococcal or other severe infections. Both ravulizumab and eculizumab have boxed warnings about an increased risk for meningococcal infections. Patients who will be treated with a C5i should receive both meningococcal vaccinations plus antibiotic therapy, as discussed below. (See 'Adverse effects' below.)

The European Medicines Agency (EMA) and the US Food and Drug Administration (FDA) have approved ravulizumab, eculizumab, and pegcetacoplan for treatment of PNH; the FDA also approved ravulizumab for treatment of PNH in children ≥1 month and adolescents. In the United States, these agents are available only through a restricted program under a risk evaluation and mitigation strategy (REMS).

For patients with PNH-associated severe BMF, management is discussed below. (See 'PNH with bone marrow failure' below.)

Complement inhibitors — Following are details of administration of individual complement inhibitors.

Reasons for choosing a particular agent for initial treatment of hemolytic PNH are presented above. (See 'Choice of initial agent' above.)

AEs of all complement inhibitors are discussed below. (See 'Adverse effects' below.)

Ravulizumab — Ravulizumab is the preferred complement inhibitor based on its more convenient treatment schedule and similar efficacy and toxicity profile, compared with eculizumab. (See 'Choice of initial agent' above.)

●Administration – Ravulizumab dosing is weight-based, as follows:

•Loading dose administered on days 1 and 15:

-≥40 to <60 kg – 2400 mg intravenously (IV)

-≥60 to <100 kg – 2700 mg IV

-≥100 kg – 3000 mg IV

•Maintenance dosing is administered every eight weeks and continued indefinitely:

-≥40 to <60 kg – 3000 mg IV

-≥60 to <100 kg – 3300 mg IV

-≥100 kg – 3600 mg IV

Ravulizumab is continued indefinitely in responding patients; no discontinuation studies of ravulizumab have been reported.

Treatment with ravulizumab for patients <40 kg is described in the FDA label [15]. Ravulizumab was approved for children by the FDA based on interim results from a phase 3 study that showed it was effective in achieving complete complement inhibition through 26 weeks in children and adolescents ≤18 years. Ravulizumab was not associated with any treatment-related severe AEs and no patients discontinued therapy during the primary evaluation period nor experienced breakthrough hemolysis.

●Toxicity – Ravulizumab is associated with an increased risk for Neisseria meningitidis infection and other AEs. Prevention and management of AEs are described below. (See 'Adverse effects' below.)

●Monitoring – Monitoring the response to a C5i is described below. (See 'Response to complement inhibition' below.)

●Treatment breakthrough – For patients who experience a recurrence of hemolysis and related symptoms while receiving ravulizumab, management is discussed below. (See 'Treatment breakthrough' below.)

●Outcomes – Phase 3 trials that directly compared ravulizumab versus eculizumab are discussed above. (See 'Choice of initial agent' above.)

Eculizumab — Eculizumab is administered as a fixed dose (ie, treatment is not based on weight or body surface area).

●Administration – The standard schedule for eculizumab in adults is 600 mg IV once per week for the first four weeks, followed by 900 mg IV one week later, followed by 900 mg IV once every two weeks thereafter. Eculizumab is generally continued indefinitely in responding patients.

●Toxicity – Eculizumab is associated with an increased risk for N. meningitidis infection and other adverse effects; prevention and management of AEs are described below. (See 'Adverse effects' below.)

●Monitoring – Monitoring the response to a C5i is described below. (See 'Response to complement inhibition' below.)

●Treatment breakthrough – For patients who experience a recurrence of hemolysis and related symptoms while receiving eculizumab, management is discussed below. (See 'Treatment breakthrough' below.)

●Outcomes – A phase 3 trial that directly compared eculizumab versus placebo (TRIUMPH) [10,11] and studies of eculizumab versus ravulizumab [12,13] are discussed above. (See 'Choice of initial agent' above.)

Other informative studies of eculizumab in PNH include:

•In the SHEPHERD study, treatment of 97 patients with eculizumab for one year was associated with reduced RBC transfusion requirements (from 8 units/person/year to 0 units/person/year), improved hemoglobin (from 9.3 to 10.2 g/dL), less hemolysis, and improved QoL [16]. Two thromboses occurred (but only in patients with prior thrombotic events) and grade ≥3 headache, abdominal distention, and viral infection that were potentially related to therapy were reported in seven patients.

•An open-label extension study of 187 patients enrolled in TRIUMPH, SHEPHERD, and an earlier pilot study reported 98 percent three-year survival and reduced transfusion requirements; the four deaths were unrelated to therapy [17]. Fewer patients experienced thromboembolic (TE) events while receiving eculizumab (4 percent), compared with their prior management (32 percent), and renal function improved or stabilized in 93 percent; there was no evidence of cumulative toxicity. Therapy was discontinued in 10 percent (half due to an adverse event). OS in patients treated with eculizumab was comparable to age- and sex-matched controls from the general population [17,18]. Other studies also reported that eculizumab is effective for preventing TE events in patients with PNH [17,19,20].

•A study of 345 Japanese patients with PNH reported that 3.2 percent did not improve with eculizumab; these patients were heterozygous for a single missense mutation in the gene that encodes C5, which prevents eculizumab binding and is present in 3.5 percent of the general Japanese population [21]. A polymorphism in the complement receptor 1 (CR1) gene, which regulates the binding of C3 to RBCs, has also been associated with reduced efficacy of eculizumab [22].

Pegcetacoplan — Pegcetacoplan is a pegylated peptide that targets the proximal complement protein, C3, and can inhibit both intravascular and extravascular hemolysis. By contrast, ravulizumab and eculizumab target C5, which affects only intravascular hemolysis, and treatment with a C5i is associated with breakthrough extravascular hemolysis in a subset of patients. (See 'Treatment breakthrough' below.)

●Administration – Pegcetacoplan 1080 mg requires refrigerated storage; it is administered subcutaneously twice weekly. There are two methods of administration [23]:

•Subcutaneous infusion using a commercially available pump.

•Single-use, disposable on-body injector – The on-body injector is for abdominal subcutaneous injection only. Tender, bruised, red, or hard skin should be avoided, as should tattoos, scars, and stretch marks. The site of each administration should be rotated. The patient and/or caregiver should be trained to prepare and administer the on-body injector prior to use.

●Outcomes – There are no randomized trials of pegcetacoplan versus placebo or trials that compared pegcetacoplan versus a C5i (ie, eculizumab or ravulizumab) for initial treatment of hemolytic PNH. Reasons for selecting a C5i rather than pegcetacoplan or another complement inhibitor are discussed above. (See 'Choice of initial agent' above.)

Use of pegcetacoplan for breakthrough hemolysis with a C5i is discussed below. (See 'Management of breakthrough hemolysis' below.)

Pegcetacoplan is approved for treatment of PNH in adults by the FDA and for adults who remain anemic after at least three months of stable C5i therapy by the EMA. Pegcetacoplan should not be administered to pregnant patients.

Other agents — Complement inhibitors that are not yet approved for management of hemolytic PNH follow. Use of these alternative complement inhibitors for patients with breakthrough hemolysis is discussed below. (See 'Management of breakthrough hemolysis' below.)

●Iptacopan – Iptacopan is a first-in-class oral factor B (FB) inhibitor that can reduce extravascular hemolysis while maintaining control of intravascular hemolysis in patients treated with a C5i.

Due to its short half-life (approximately 20 hours [24]), iptacopan is not a good choice for patients who have a history of medication non-adherence. Missed doses, especially in the setting of complement-amplifying conditions (eg, infections) or excessive consumption of alcohol, could lead to massive hemolysis and risk for thrombosis.

Iptacopan received FDA breakthrough therapy designation in PNH and orphan drug designations from the FDA and EMA.

●Danicopan – Danicopan is a first-in-class oral inhibitor of complement factor D (FD; a component of the alternative complement pathway) that can control signs and symptoms of extravascular hemolysis as an add-on treatment in patients receiving a C5i [25,26].

Preliminary results of a phase 3 trial (published in abstract form) indicate that, for patients with clinically significant extravascular hemolysis while receiving ravulizumab or eculizumab, addition of danicopan was superior to addition of placebo for reducing transfusions and elevating hemoglobin, with little toxicity and continued control of intravascular hemolysis [27].

Danicopan was granted breakthrough therapy designation by the FDA and PRIME (priority medicine) status by the EMA.

●Agents in development – In addition to the drugs listed above, other approaches to blocking complement activation include monoclonal antibodies to other complement proteins, peptide inhibitors, small molecule inhibitors, and decoy receptors that are in various stages of preclinical development [28-31].

Adverse effects — Treatment with any complement inhibitor can cause life-threatening infections with N. meningitidis and other encapsulated organisms. Other toxicity of these agents is relatively modest.

●Meningitis – Complement inhibitors increase the risk of infections or sepsis with N. meningitidis, which can rapidly become life-threatening or fatal. These agents are contraindicated in patients with unresolved neisserial infections, unless the risks of delaying treatment outweigh the risks of developing a meningococcal infection.

We administer meningococcal vaccine plus antibiotic prophylaxis because invasive meningococcal disease (including non-typeable strains that are not included in the vaccine) has occurred despite vaccination [14,32]. Even after vaccination, the risk of Neisseria infection is as high as 0.5 percent per year; this risk is 1000-fold greater than the general population [33,34]. (See "Treatment and prevention of meningococcal infection", section on 'Patients receiving C5 inhibitors'.)

•Neisseria vaccination – We administer meningococcal vaccine (both ACYW135 and serogroup B) immediately after the first dose of the complement inhibitor, because of reports of thrombosis precipitated by vaccine-induced complement amplification [14,35]. Some experts instead favor vaccination two weeks prior to the first dose of a C5 inhibitor.

•Antibiotic prophylaxis – Antimicrobial prophylaxis for prevention of meningococcal infection in patients receiving complement inhibitors is discussed separately. (See "Treatment and prevention of meningococcal infection", section on 'Patients receiving C5 inhibitors'.)

•Treatment of meningococcal infection – Patients should seek immediate medical attention if they develop fever, nausea, vomiting, headache, myalgias, or other symptoms suggestive of meningitis, regardless of vaccination status [14]. (See "Clinical manifestations of meningococcal infection" and "Meningococcal vaccination in children and adults", section on 'Immunization of persons at increased risk'.)

●Other adverse effects – Treatment with ravulizumab and eculizumab can cause generally mild non-infectious AEs, are expensive, and may be associated with extravascular hemolysis and breakthrough symptoms.

•Other toxicity – AEs reported in ≥10 percent of patients in clinical trials included headache, nasopharyngitis, back pain, and nausea [10,16]. Headache initially occurs in up to half of patients, possibly due to increased nitric oxide levels, but it rarely persists after the first several doses [36].

•Extravascular hemolysis with C5 inhibitors – Despite effective control of intravascular hemolysis by C5i agents, up to one-fifth of patients require periodic RBC transfusions due to an increase in extravascular hemolysis [28,37,38]. C5i drugs do not prevent deposition on PNH red cells of C3 fragments, which foster extravascular hemolysis by serving as opsonins. (See "Pathogenesis of paroxysmal nocturnal hemoglobinuria", section on 'Extravascular hemolysis from reduced CD55'.)

Management of extravascular hemolysis is discussed below. (See 'Management of breakthrough hemolysis' below.)

•Cost – Cost is a major barrier to the use of complement inhibitors, which has been estimated to be approximately $400,000 US dollars (€370,000) per year. Continuous treatment is generally required to maintain suppression of intravascular hemolysis and prevent thrombosis.

PNH WITH THROMBOSIS — Thromboembolic (TE) events occur in up to 40 percent of patients with PNH. Prior to the development of complement inhibitors, thromboembolism was the leading cause of death in patients with PNH.

Current thrombosis — For patients with PNH and an acute TE event, we suggest anticoagulation plus treatment with a complement inhibitor, rather than anticoagulation alone. Anticoagulation alone is ineffective in preventing recurrent TE events in patients with PNH.

●Complement inhibition – Treatment with a complement inhibitor is discussed above. (See 'Symptomatic hemolytic PNH' above.)

●Anticoagulation – Acute thromboses should be treated with anticoagulation and/or thrombolysis for life-threatening events, according to the site of thrombosis and severity of clinical findings [14,39]. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)".)

•Duration of anticoagulation – For patients whose PNH is well-controlled with a complement inhibitor (ie, lactate dehydrogenase [LDH] <1.5 x upper limit of normal [ULN]) who have no other thrombosis-predisposing factors, we generally anticoagulate for three to six months, rather than continuing anticoagulation indefinitely. Discontinuing anticoagulation after three to six months reduces the risk of bleeding and complications of thrombocytopenia.

Adequacy of a limited duration of anticoagulation for patients with PNH and a history of a TE event was evaluated in a retrospective single-institution study [19]. Treatment of four patients with a complement inhibitor plus indefinite anticoagulation was associated with an 86 percent relative reduction (RR) of TE events after initiation of the C5i (38.7 TE events/100 patient-years prior to beginning the C5i versus 5.4 events/100 patient-years while receiving C5i therapy plus indefinite anticoagulation). By comparison, among 18 other patients, treatment with a C5i while off anticoagulation was associated with a 94 percent RR of TE events (25.9 TE events/100 patient-years prior to initiating the C5i versus 1.5 TE events/100 patient-years post-C5i therapy while off anticoagulants). Among the entire cohort, TE events prior to C5i therapy included intra-abdominal and cerebrovascular locations; by contrast, TE events while receiving a C5i involved more common locations (ie, deep vein thrombosis, pulmonary emboli) and most were provoked by major surgery or uterine compression. There were two significant gastrointestinal bleeding episodes among patients treated with anticoagulation alone (ie, prior to C5i therapy).

For patients with PNH who have additional risk factor(s) for TE events, the duration of anticoagulation should be individualized. Factors that predispose to thromboses are described separately. (See "Overview of the causes of venous thrombosis".)

•New TE event while receiving adequate complement inhibition – For patients who experience a TE event while receiving adequate C5i therapy (ie, LDH <1.5 x ULN), we generally anticoagulate indefinitely.

•Thrombosis with concurrent thrombocytopenia – For patients with severe thrombocytopenia (eg, platelet count <50,000/microL) who experience a TE event, it may be necessary to transfuse platelets to safely administer anticoagulants and/or thrombolytic therapy. (See "Anticoagulation in individuals with thrombocytopenia".)

Thrombotic prophylaxis — For patients with PNH without a current or prior TE event and without an exacerbating thrombophilic condition, we suggest not treating with prophylactic anticoagulation, based on a lack of high-quality evidence that it effectively prevents thromboses.

Exceptions include:

●Hospitalization – Short-term prophylactic anticoagulation is appropriate for patients with PNH who are hospitalized with an acute medical illness or for a surgical procedure, because PNH may exacerbate the known risks of venous thromboembolism. When possible, elective surgery should be scheduled soon after treatment with a C5i to reduce the risk of a TE event.

Selection of a prophylactic anticoagulation regimen is presented separately. (See "Prevention of venous thromboembolic disease in acutely ill hospitalized medical adults" and "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

●Pregnancy – Management of PNH in pregnant patients is discussed below. (See 'Pregnancy' below.)

●Other thrombophilic conditions – For patients with PNH who have additional thrombophilic conditions, the duration of anticoagulation should be individualized.

Other factors that predispose to thromboses are described separately. (See "Overview of the causes of venous thrombosis".)

SUBCLINICAL PNH — For patients who have no substantial hemolysis-associated symptoms, thrombosis, pain, organ dysfunction, or bone marrow failure, we manage with watchful waiting to avoid the inconvenience and expense of a complement inhibitor and the substantial toxicity and possible transplant-related mortality of allogeneic hematopoietic cell transplantation (HCT) (algorithm 1).

Patients with subclinical PNH may receive supportive care, if needed, and should be monitored regularly for worsening of hemolysis-associated findings and for progression of cytopenias that may herald worsening bone marrow failure, as discussed below. (See 'All patients with PNH' below.)

PNH WITH BONE MARROW FAILURE — Many patients with PNH also have a degree of bone marrow failure (BMF). PNH is generally associated with aplastic anemia (AA), but some cases are associated with myelodysplastic syndromes/neoplasms (MDS).

Severity of BMF — Patients with PNH accompanied by severe AA (sAA) or higher-risk MDS are classified as having PNH with BMF.

Briefly, criteria for sAA and higher-risk MDS are:

●Severe AA (sAA) – Bone marrow cellularity <25 percent plus ≥2 of the following:

•Absolute neutrophil count (ANC) <500/microL (<0.5 x 10⁹/L)

•Platelet count <20,000/microL

•Reticulocyte count <60,000/microL

Further details of diagnosis and classification of AA are discussed separately. (See "Aplastic anemia: Pathogenesis, clinical manifestations, and diagnosis", section on 'Diagnostic criteria'.)

●Higher-risk MDS – Risk stratification is based on prognostic score:

•Revised international prognostic scoring system (IPSS-R) (table 2) (calculator 1): ≥4.0 points

or

•Original IPSS (table 3) (calculator 2): ≥1.5 points

Further details of diagnosis and prognostic classification of MDS are discussed separately. (See "Overview of the treatment of myelodysplastic syndromes", section on 'Prognostic category'.)

Indications for transplantation — For patients with PNH accompanied by sAA or higher-risk MDS who are medically fit for transplantation, we suggest allogeneic hematopoietic cell transplantation (HCT). Complement inhibitors are not effective for relieving PNH-associated BMF. However, while awaiting identification of a suitable graft donor, complement inhibitors are safe and effective for managing hemolysis-associated findings [40].

For patients with PNH with BMF who are not suitable for allogeneic HCT and for those patients with lesser degrees of BMF, we individualize management with supportive care and/or a complement inhibitor, according to the presence and severity of hemolysis-associated symptoms. (See 'Stratification of management' above.)

Allogeneic HCT for PNH with BMF — Candidates for allogeneic HCT must have no severe lung, heart, liver, or kidney disease; a suitable graft source; and adequate social supports. Although age limits vary, many institutions restrict allogeneic HCT to patients ≤70 years of age.

Details of eligibility for allogeneic HCT are presented separately. (See "Determining eligibility for allogeneic hematopoietic cell transplantation".)

Preferred conditioning regimens and graft sources for PNH-associated transplantation vary between institutions and have not been directly compared in randomized trials.

●Conditioning regimen – Myeloablative conditioning (MAC) is not required to eradicate the PNH clone [36]. Reduced intensity conditioning (RIC)/nonmyeloablative (NMA) conditioning may yield comparable rates of engraftment as MAC, while reducing transplant-related morbidity and mortality.

●Graft source – Human leukocyte antigen (HLA)-matched siblings are preferred when available, but matched unrelated donors are used successfully for PNH. Results with haploidentical donors are also encouraging [41].

●Bone marrow versus peripheral blood stem cells – We favor bone marrow over peripheral blood stem cells, when possible, because of a lower incidence of graft-versus-host disease (GVHD). A cohort study of 211 patients reported that the incidence of GVHD was lower with bone marrow-derived stem cells compared with peripheral blood stem cells [42].

Outcomes with transplantation for patients with PNH come from observational studies and case reports. No randomized trials have directly compared allogeneic HCT versus complement inhibitors or observation for PNH, and no trials have directly compared various conditioning regimens or graft sources. Long-term overall survival after allogeneic HCT for PNH has been reported from 50 to 70 percent with MAC, while RIC/NMA conditioning may be associated with rates that approach 90 percent [43-47].

SUPPORTIVE CARE — Supportive care refers to management of PNH-associated symptoms without complement inhibitor therapy or transplantation.

●Pain control – Chronic pain is a common symptom with PNH and may persist despite treatment with a complement inhibitor. (See "Approach to the management of chronic non-cancer pain in adults".)

●Anemia – Anemia in PNH is predominantly caused by hemolysis of red blood cells (RBC), but bone marrow failure (BMF), folate or vitamin B12 deficiency, chronic iron loss in urine, and other causes may also contribute. Patients who have an inappropriately low reticulocyte index should be tested for deficiencies of iron, folate, and vitamin B12 and receive supplementation, as appropriate. If there is an inadequate response to supplementation, the patient should undergo a bone marrow examination to evaluate for aplastic anemia (AA) or myelodysplastic syndromes/neoplasms (MDS) [48].

•Transfusions – Patients with severe anemia (eg, hemoglobin level <7 g/dL or symptoms such as dyspnea, extreme fatigue without another etiology) should receive transfusions with RBCs, as needed. No special modifications (eg, washed RBCs) are needed for patients with PNH. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion".)

•Folate supplementation – All patients with PNH and ongoing hemolysis should be treated with supplemental folic acid (1 to 2 mg daily) because of increased demands created by accelerated erythropoiesis.

•Routine iron supplementation is not needed – Supplemental iron should be administered only to patients who are documented to be iron deficient. Supplementation in patients who are iron replete can contribute to iron overload, because they may also have received substantial amounts of iron from transfusions. (See "Treatment of iron deficiency anemia in adults".)

•Growth factors – Evidence is lacking to support a role for erythropoietin, glucocorticoids, or androgenic hormones. Glucocorticoids may improve hemoglobin levels and reduce hemolysis in some patients, but the substantial long-term adverse effects outweigh their efficacy.

MONITORING — All patients with PNH should be assessed at least annually. Patients who are receiving treatment require clinical evaluation and laboratory studies to assess the response to a complement inhibitor.

Response to complement inhibition — Our approach to assessing the response to complement inhibition follows.

●First four weeks – For the first four weeks of treatment with complement inhibition, patients should be monitored weekly with [14]:

•Complete blood count (CBC)

•Reticulocyte count

•Serum chemistries, including lactate dehydrogenase (LDH)

●Subsequent monitoring – The subsequent schedule for monitoring is guided by the clinical and laboratory response. After hemolysis has stabilized, monitoring the CBC, reticulocyte count, and LDH can generally then be extended to every two months, to coincide with the schedule of ravulizumab infusions.

If treatment with a complement inhibitor is discontinued, the patient should be monitored periodically for ≥16 weeks to detect evidence of increased hemolysis, which may be manifest as worsening fatigue, dyspnea, hemoglobinuria, thrombosis, abdominal pain, decreasing hemoglobin, increasing LDH, erectile dysfunction, or other vascular events.

LDH typically normalizes within days to weeks, the hemoglobin response is variable, and the reticulocyte count can remain elevated indefinitely, due to extravascular hemolysis. There is insufficient evidence to support routine monitoring of CH50 or other measures of complement activity.

Management of patients with breakthrough symptoms is described below. (See 'Treatment breakthrough' below.)

All patients with PNH — All patients, whether or not they are treated with a complement inhibitor, should be monitored at least annually using flow cytometry to assess the PNH clone size, as described separately. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Flow cytometry/FLAER'.)

TREATMENT BREAKTHROUGH — Treatment breakthrough is defined as a return of PNH symptoms (eg, fatigue, hemoglobinuria, abdominal pain, thrombosis) in association with a rise in lactate dehydrogenase (LDH) to ≥2 x upper limit of normal (ULN) after prior reduction to <1.5 x ULN while receiving treatment with a C5 complement inhibitor (C5i).

Categories of breakthrough — Return of PNH-associated symptoms may be due to pharmacokinetic effects of the C5i; complement-amplifying conditions, such as infection, surgery, pregnancy (pharmacodynamic breakthrough); exacerbation of extravascular hemolysis; or progressive bone marrow failure (BMF).

●Pharmacokinetic breakthrough – Pharmacokinetic breakthrough (eg, recurrence prior to the next dose) is less common with ravulizumab than eculizumab because of its longer half-life and weight-based dosing [49]. If hemolysis-related symptoms worsen prior to the next dose of eculizumab, the dosing interval can be shortened (eg, to 12 or 13 days) or the dose can be increased.

●Pharmacodynamic breakthrough – Pharmacodynamic breakthrough refers to events that can increase complement activation, such as infection, surgery, and pregnancy. As a result, PNH red blood cells (RBCs) that are densely coated with C3b can cause C5 to adopt a C5b-like conformation that is not inhibited by eculizumab or ravulizumab [50]. Treating the underlying complement-amplifying condition is necessary to mitigate pharmacodynamic breakthrough. Notably, patients with PNH often experience pharmacodynamic breakthrough within days of receiving coronavirus disease 2019 (COVID-19) vaccination [51].

●Exacerbation of extravascular hemolysis – Virtually all patients with PNH who are treated with a C5i experience mild to moderate extravascular hemolysis that may be accompanied by continued anemia, elevated reticulocyte count, and a positive direct antiglobulin test (DAT; ie, Coombs test) for C3d. Up to one-fifth of patients may require periodic RBC transfusions [28,37,38]. (See "Pathogenesis of paroxysmal nocturnal hemoglobinuria", section on 'Extravascular hemolysis from reduced CD55'.)

●Worsening leukopenia and/or thrombocytopenia – If pancytopenia develops or worsens, the patient should have a repeat bone marrow examination to re-evaluate for BMF.

Management of breakthrough hemolysis — For patients with residual anemia attributable to breakthrough extravascular hemolysis while receiving a C5i, we suggest treatment with pegcetacoplan, or another proximal complement inhibitor, when available (eg, iptacopan as monotherapy or danicopan as an add on to a C5i). This suggestion is based on reduced transfusion needs, increased hemoglobin (Hb), and improved anemia-related symptoms in a randomized trial and other studies [26,52,53].

We manage breakthrough hemolysis by overlapping pegcetacoplan with the C5i by at least three weeks (to enable the new agent to reach a steady state) and then discontinue the C5i. Thus, for patients treated with eculizumab, we treat with two additional doses (every two weeks), while for ravulizumab, there is no need for additional doses because of its longer half-life. There is no need to overlap treatment with a C5i when using iptacopan or danicopan because of their more rapid onset of action.

●Pegcetacoplan – Pegcetacoplan was superior to eculizumab for reducing transfusion-dependence and lessening fatigue in patients with PNH who had Hb <10.5 g/dL despite prior eculizumab therapy [53]. After a four-week run-in phase in which all patients received pegcetacoplan plus eculizumab, patients were randomly assigned to pegcetacoplan monotherapy (41 patients) or eculizumab (39 patients). At week 16, patients who received pegcetacoplan had a nearly 4 g/dL increase in Hb, more transfusion-independence (TI; 85 versus 15 percent, respectively), and greater improvement in fatigue score (FACIT-Fatigue score) compared with eculizumab. The most common adverse events (AEs) during treatment with pegcetacoplan and eculizumab were injection site reactions (37 versus 3 percent, respectively), diarrhea (22 versus 3 percent), breakthrough hemolysis (10 versus 23 percent), headache (7 versus 23 percent), and fatigue (5 versus 15 percent).

●Iptacopan – Among 97 patients with residual anemia while receiving either eculizumab or ravulizumab (for a mean of four years), iptacopan achieved superior RBC TI, improved Hb, and better quality-of-life (QoL), compared with continued C5i therapy, according to preliminary results published in abstract form [52]. Patients with Hb <10 g/dL were randomly assigned (8:5) to 24 weeks of iptacopan monotherapy (200 mg twice daily) versus continued treatment with their previous C5i. Compared with continued C5i therapy, those receiving iptacopan achieved superior TI (85 percent versus 0 percent, respectively), improved Hb at 24 weeks (12.6 g/dL versus 9.2 g/dL), and better QoL (by FACIT-Fatigue score). Iptacopan was well-tolerated and there were no deaths or serious infections with encapsulated bacteria. Iptacopan was associated with more headache (16 versus 3 percent) and diarrhea (15 versus 6 percent) but fewer infections/infestations (39 versus 49 percent) and less breakthrough hemolysis (3 versus 17 percent).

●Danicopan – A phase 2 study added danicopan 100 to 200 mg three times daily for 24 weeks for 11 patients who remained transfusion-dependent while receiving eculizumab [26]. Danicopan was associated with increased Hb (10.3 g/dL at week 24, compared with 7.9 g/dL at baseline), nearly eliminated transfusions (one unit transfused in one patient, compared with mean 4.4 units per patient over a 24-week run-in period), and improved QoL (11-point mean rise in FACIT score). The most common adverse effects were generally mild and included headache, cough, and nasopharyngitis.

SPECIAL POPULATIONS

Children — PNH is rare in children and there is little published evidence to guide therapy. We manage children similarly to adults, using complement inhibitor therapy for hemolysis-related symptoms or thrombosis, watchful waiting for asymptomatic patients, and allogeneic hematopoietic cell transplant (HCT) for PNH with co-existent severe aplastic anemia. (See 'Overview' above.)

Treatment with ravulizumab for patients <40 kg is described in the US Food and Drug Administration (FDA) label [15]. We administer the same dose of eculizumab to children as to adults, based on complete resolution of hemolysis in all seven children (11 to 17 years) in a 12-week pilot study [54]. (See 'Choice of initial agent' above.)

Pregnancy — Women with PNH can have successful pregnancies, but it may be associated with increased maternal and fetal morbidity and mortality [55]. A pregnant patient with PNH should be managed by a hematologist and an obstetrician with expertise in high-risk pregnancy.

The following considerations must be addressed:

●Iron and folate supplementation – Pregnancy increases iron and folate requirements, which may already be high in patients with PNH. Prenatal folate and iron should be administered to all patients. (See "Nutrition in pregnancy: Dietary requirements and supplements".)

●Transfusions – Transfusions are administered as needed for symptomatic anemia and/or thrombocytopenia. Some pregnant patients with PNH had increased transfusion requirements in a case series [56].

●Complement inhibition – Both ravulizumab and eculizumab are acceptable choices for treatment of symptomatic hemolytic PNH in pregnant patients. Although most studies report treatment of pregnant patients with eculizumab, we favor ravulizumab because there is less risk for pharmacokinetic breakthrough. Note that other agents (eg, pegcetacoplan, iptacopan, danicopan) have not been shown to be safe for pregnant patients.

For patients who become pregnant while taking a C5 complement inhibitor (C5i), we continue therapy throughout the pregnancy and while breastfeeding. It may be necessary to increase the dose and/or frequency of treatment during the third trimester because of increases in the volume of distribution. We initiate therapy with a C5i for patients who are newly diagnosed with symptomatic hemolytic PNH while pregnant. (See 'Symptomatic hemolytic PNH' above.)

An industry-sponsored observational study reported outcomes in 61 women (75 pregnancies) who were taking eculizumab while pregnant; 88 percent were also anticoagulated [56]. There were no maternal deaths, two women had a postpartum thromboembolic event while taking eculizumab, and an additional 2 of 10 women who stopped eculizumab postpartum had a thromboembolic event. In 36 of 67 pregnancies that progressed to delivery, breakthrough hemolysis necessitated an increased dose of eculizumab or shortened treatment interval. Live births were reported in 92 percent of the 75 pregnancies, compared with 39 percent in the same cohort before eculizumab therapy; first trimester miscarriages were reported in 8 percent and there were 3 stillbirths. Delivery before 37 weeks of gestation was reported in 29 percent; common reasons included preeclampsia, growth retardation, or planned cesarian delivery. There was no evidence of teratogenicity in this small cohort. Eculizumab was detected in 7 of 20 cord blood samples and in 0 of 10 of breast milk samples tested.

Treatment with a C5i during pregnancy is associated with improved maternal outcomes and no evidence of increased fetal risks [55-58], compared with outcomes prior to availability of complement inhibitor therapy [56,59-61]. Prior to availability of complement inhibitors, studies reported 8 to 20 percent maternal mortality, up to 45 percent spontaneous miscarriage, and premature delivery in more than half of women [56,59-61].

●Thromboembolic risk reduction – Pregnancy increases the already-high thromboembolic risk in PNH. Treatment must be individualized, as there are few data to guide management:

•All pregnant patients – For all pregnant women with PNH, we treat with low molecular weight heparin (eg, 40 mg subcutaneously once or twice daily) during the last trimester and continue treatment for 8 to 12 weeks postpartum.

•Pregnancy while taking a C5i – For women who become pregnant while taking a complement inhibitor, we continue that treatment. For women who are newly diagnosed with symptomatic hemolytic PNH during pregnancy, we initiate treatment with a complement inhibitor.

Further details of management of thrombosis in pregnancy are provided separately. (See "Use of anticoagulants during pregnancy and postpartum".)

Oral contraceptives — Oral contraceptives increase thromboembolic risk in women with PNH. Non-hormonal forms of contraception are preferred for sexually active women with PNH, although there is little information to guide the choice.

As with all women, the potential benefits of oral contraceptives in preventing pregnancy must be weighed against the risks, including the risk of unplanned pregnancy, and compared with other contraceptive options. Women with well-controlled PNH (eg, lactate dehydrogenase [LDH] <1.5 x upper limit of normal) are probably at no higher risk of thrombosis than age-matched controls.

Options for non-hormonal contraception and lower-risk hormonal methods are discussed separately. (See "Combined estrogen-progestin oral contraceptives: Patient selection, counseling, and use".)

Surgery — Surgery can precipitate hemolysis in patients with PNH due to increased inflammatory stimuli and/or metabolic abnormalities [36].

Patients who are currently treated with C5i should be scheduled for surgery close to their most recent infusion, if possible, to minimize pharmacokinetic breakthrough hemolysis. For asymptomatic patients with large PNH clones who are not on therapy, some experts initiate Complement inhibition to lessen the thrombotic risk of planned surgery.

The use of routine thromboprophylaxis in surgical patients is discussed in detail separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

PROGNOSIS — PNH is a chronic disease with significant morbidity and mortality. Since the availability of complement inhibitors, mortality appears to be similar to age-matched controls, but the impact will be better-defined with longer follow-up.

●A series of 79 consecutive patients treated with eculizumab over an eight-year period reported that overall survival (OS) was similar to age-matched controls [18].

●A series of 220 patients from the pre-complement inhibitor era (1996) reported 14.6 year median OS, with estimates of 78, 65, and 48 percent at 5, 10, and 15 years after diagnosis, respectively [62]. Eight-year rates of pancytopenia, thrombosis, and myelodysplastic syndrome were 15, 28, and 5 percent, respectively. Adverse prognostic factors included thrombosis, evolution to pancytopenia, myelodysplastic syndrome, or acute leukemia, and age >55 years at disease onset.

●Another series from the pre-complement inhibitor era (2004) reported that outcomes differed between American and Japanese patients [63]. The mean OS for the Japanese patients was 32 years, compared with 19 years for the American population, although the Kaplan-Meier survival curves were not different for the two groups. Japanese patients were more likely to have aplastic anemia, while American patients were more likely to have thrombosis (38 versus 6 percent). Adverse prognostic factors included age >50 years, severe cytopenias at diagnosis, severe infection, thrombosis, and renal failure.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anticoagulation in pregnancy" and "Society guideline links: Bone marrow failure syndromes".)

SUMMARY AND RECOMMENDATIONS

●Description – Paroxysmal nocturnal hemoglobinuria (PNH) is a rare complement-mediated hemolytic anemia associated with thrombosis, pain, and organ dysfunction due to loss of complement inhibitor proteins on hematopoietic cells. PNH may be associated with aplastic anemia (AA) or myelodysplastic syndromes/neoplasms (MDS).

●Baseline evaluation – Should include flow cytometry to estimate PNH clone size and bone marrow examination to evaluate for AA or MDS. (See 'Baseline evaluation' above.)

●Management – Guided by PNH category and severity of symptoms (algorithm 1). (See 'Categories of PNH' above.)

•Symptomatic hemolytic PNH without bone marrow failure (BMF) – For patients with symptomatic hemolytic PNH (eg, thrombosis, organ dysfunction, pain) without severe BMF, we recommend treatment with a complement inhibitor, rather than supportive care alone (Grade 1B). (See 'Symptomatic hemolytic PNH' above.)

-Drug selection – We suggest initial treatment with ravulizumab rather than other complement inhibitors, based on comparable efficacy and toxicity with eculizumab, but greater convenience, lower overall expense, and fewer episodes of pharmacokinetic breakthrough hemolysis (Grade 2C).

We favor a C5 complement inhibitor (C5i) for initial therapy, rather than other complement inhibitors, because of their proven efficacy for symptom relief and thrombosis prevention, low toxicity, convenience, and long track record. However, some clinicians use pegcetacoplan, which has not been investigated in treatment-naïve disease. (See 'Choice of initial agent' above and 'Pegcetacoplan' above.)

-Infectious risk – All patients treated with a complement inhibitor should be vaccinated and receive oral antibiotic prophylaxis, as described above. (See 'Adverse effects' above.)

-Breakthrough hemolysis – For patients with breakthrough hemolysis while receiving a C5i, we suggest treatment with pegcetacoplan (or iptacopan or danicopan, when available), rather than continued treatment with the C5i. (See 'Management of breakthrough hemolysis' above.)

•Thrombosis – Thrombosis or history of thrombosis is an indication for C5i therapy. (See 'PNH with thrombosis' above.)

-Acute thrombosis – Acute thrombosis should be treated with anticoagulation and/or thrombolysis plus complement inhibition.

-Anticoagulant discontinuation – For PNH that is well-controlled with a C5i, we typically continue anticoagulation for three to six months, but duration should be individualized based on other risk factors.

•Subclinical PNH – For patients with no substantial PNH-associated symptoms or BMF, we suggest watchful waiting rather than treatment with a C5i or allogeneic hematopoietic cell transplantation (HCT) (Grade 2C). (See 'Subclinical PNH' above.)

•PNH with severe BMF – For patients with PNH with BMF (ie, severe AA or higher-risk MDS), who are medically fit for transplantation, we suggest allogeneic HCT, rather than supportive care and/or a complement inhibitor (Grade 2C). (See 'PNH with bone marrow failure' above.)

For PNH in association with less severe AA or lower-risk MDS and for patients who are ineligible for transplantation, management is guided by PNH-associated symptoms. (See 'Stratification of management' above.)

●Pregnancy – Pregnant patients with PNH should be managed by a hematologist and an obstetrician specializing in high-risk pregnancy.

•PNH in pregnancy – For pregnant patients with PNH who otherwise meet criteria for complement inhibitor therapy, we suggest initiation or continued treatment with a C5i to decrease maternal morbidity/mortality and improve fetal outcomes (Grade 2C).

•Thrombosis prophylaxis – For pregnant patients with PNH, we suggest anticoagulation using low molecular weight heparin during the last trimester and continue treatment for 8 to 12 weeks postpartum (Grade 2C). (See 'Pregnancy' above.)

Complement inhibitors other than C5is are not proven safe in pregnant patients.

ACKNOWLEDGMENTS

The UpToDate editorial staff acknowledges the contributions of Stanley L Schrier, MD as Section Editor on this topic, his tenure as the founding Editor-in-Chief for UpToDate in Hematology, and his dedicated and longstanding involvement with the UpToDate program.

The UpToDate editorial staff also acknowledges Wendell F Rosse, MD, who contributed to earlier versions of this topic review.