発作性夜間ヘモグロビン尿症

The RACE study (NCT02009747) compared horse antithymocyte globulin (hATG) plus cyclosporine A (CsA) ± eltrombopag as initial immunosuppressive treatment (IST) for severe aplastic anemia. Here we report the final 2-year analysis of this prospective randomized phase III study. One hundred ninety-seven treatment-naive patients were randomized to standard IST (hATG 40 mg/kg × 4 days and CsA 5 mg/kg/day; arm A; n = 101) or standard IST + eltrombopag at the dose of 150 mg/day (arm B; n = 96) from day +14 until 6 months (or 3 months, in case of complete response). The median follow-up was 23.2 months. The 2-year cumulative incidence of complete response was significantly superior in arm B (62.4% vs. 35.3%; p < 0.001). The 2-year overall survival (OS) was 86% in arm A and 91% in arm B (p = 0.081), with hazard ratio (HR), adjusted for age and disease severity, of 0.54 (p = 0.064). The 2-year disease-free survival (DFS) was 56% vs. 37% (adjusted HR = 0.49; p < 0.001), while event-free survival (EFS) was 48% vs. 32% (p < 0.001), with adjusted HR = 0.54 (p < 0.001), both significantly superior for arm B. The cumulative incidence of relapse was comparable in the two arms, while evolution to clinical paroxysmal nocturnal hemoglobinuria was 8% in arm A and 1% in arm B (p = 0.041). The risk of clonal evolution remained negligible, with one patient in arm A and two in arm B developing karyotypic abnormalities. The initial hematological response benefit of eltrombopag added to IST as front-line treatment of AA is associated with better 2-year OS, DFS, and EFS without increased risk of secondary myeloid malignancies.

Paroxysmal nocturnal hemoglobinuria (PNH) is a rare clonal hematopoietic stem cell disorder characterized by uncontrolled complement-mediated intravascular hemolysis and a disproportionately high thromboembolic risk. Although terminal complement inhibition (CIT) has substantially modified the disease course, reducing thromboembolic events (TE), the residual risk remains elevated and evidence-based anticoagulation guidelines are lacking. To characterize real- world practice, we conducted a nationwide, cross-sectional survey involving 24 Italian Centers caring for 291 patients, evaluating strategies for primary antithrombotic prophylaxis (PAP), secondary prophylaxis (SAP), and management of breakthrough hemolysis (BTH). Marked heterogeneity emerged. PAP prior to CIT was routinely employed in only one-third of Centers, whereas an equal proportion did not use it; the remainder adopted risk-based approaches integrating thrombophilia traits or clone size. SAP was universally administered after TE, with 75% of Centers maintaining indefinite anticoagulation regardless of CIT response. Anticoagulation during BTH was likewise variable, with LMWH as the preferred agent and duration largely dictated by biochemical resolution. These findings demonstrate substantial inter-center variability, underscoring the lack of harmonized national frameworks. The results highlight the need for structured expert consensus and standardized clinical algorithms to optimize thrombosis prevention and management in PNH in the era of complement inhibition.

Paroxysmal nocturnal hemoglobinuria (PNH) is an ultra-orphan disease. In 2026, approved complement inhibitors (CIs) for PNH include three C5 inhibitors (C5i) and three proximal inhibitors (PIs). Clinical trials for the approved PI pegcetacoplan, iptacopan, and C5i + danicopan had clear protocols for changing from terminal to PI, extrapolated into real-world practice. There is no guidance for patients changing from a PI to a different CI. We present the largest international cohort to date. The inclusion criteria were as follows: patients established on a PI, who have changed treatment to a different CI. Sixty-five patients with a median age at PNH diagnosis of 40 years were included. Indications for CI switch were as follows: hemolysis (extravascular/recurrent breakthrough/not specified), side effects, trial termination, patient choice, and other. CI changes (149) were classified as (1) terminal-to-proximal (75/149), 6 hemolytic events reported within 14 days of CI change; (2) proximal-to-proximal (45/149), 2 hemolytic events reported within 14 days of CI change; and (3) proximal-to-terminal (29/149), 13 hemolytic events within 14 days of CI change. This is the largest cohort of PNH patients facing multiple CI changes. With the increasing availability of different CI classes, patients may experience several treatment changes over their disease course. CI should not be interrupted. The cohort would suggest that overlapping treatment for patients changing from PI to another CI is not required. Tapering of treatment does not seem to be effective at preventing hemolysis complications. Patients should be monitored closely for hemolysis, especially when changing from proximal-to-terminal inhibition. Prospective clinical/laboratory analysis is recommended for further clarification management for this patient group.