Allogeneic Hematopoietic Cell Transplantation for Myelodysplastic Neoplasms with Accompanying Paroxysmal Nocturnal Hemoglobinuria - a Study on Behalf of Cmwp and Saawp of the EBMT

Introduction The presence of paroxysmal nocturnal hemoglobinuria (PNH) clones in a proportion of myelodysplastic neoplasms (MDS) has been demonstrated by various investigators. In the report by Fattizzo et al., Leukemia 2021, it was shown that the presence of PNH clones, regardless of size, in the setting of MDS is associated with improved outcomes. Here we present the detailed results of allogeneic hematopoietic cell transplantation (allo-HCT) for MDS patients with a PNH clone, being the first such description in the literature. Methods: Patients of any age, diagnosed with MDS and harboring a PNH clone of any size, who underwent their first allo-HCT between 2000 and 2020 were selected from the EBMT database. Additional data on diagnoses of MDS and PNH was requested from participating centers. Overall survival (OS), progression-free survival (PFS), and GvHD-free, progression-free survival (GFRS) were analyzed using Kaplan-Meier (KM) methods. Composite endpoint of GFRS was defined as survival without experiencing grade III-IV aGvHD, extensive cGvHD, or relapse/progression. Cumulative incidence (CI) of relapse (RI) and non-relapse mortality (NRM) were analyzed together in a competing risks framework. Prognostic factors were evaluated in univariate KMs, and subsequently in Cox Proportional Hazards (PH) models. Results: Sixty patients from 34 centers were included in the analysis. There were 35 males (58.3%); the median age at allo-HCT was 45.0 (range, 13.7-67.0) years. The median interval between MDS diagnosis and allo-HCT was 6.8 (range, 0.5-302.7) months, and between PNH and allo-HCT 20 (range, 0.9-515.4) months. A total of 46.7% patients had a PNH diagnosis prior to MDS, while 30.4% had an aplastic anemia diagnosis prior to MDS. KPS was ≥90% in 59.3% patients with data available. The IPSS-R score was known for 22 patients, and 45.5% had either high or very high risk; cytogenetic risk according to IPSS-R was intermediate in 52.0% and poor in 40.0%. The median PNH clone size at allo-HCT was 7.5% (0.0-98.0%) on granulocytes (n=16), and 10% (0.0-99.6%) on monocytes (n=15). Active hemolysis was present in 34.4% of the patients, and 25.0% had a history of thrombosis. Most patients received previous treatment for their disease before receiving allo-HCT; the most common treatment was immunosuppression (48.6% cyclosporine, 34.2% ATG), while 16.7% of patients were treated with eculizumab due to clinical PNH, and 8.5% with hypomethylating agents. Allo-HCT was performed from an identical sibling in 33.3% of patients. The source of stem cells was peripheral blood in 76.7%, bone marrow in 15.0%. Reduced intensity conditioning was used in 49.1% patients. The CI of engraftment at 30 days was 92% (95% CI, 84-99%) for neutrophils; 85% (95% CI, 76-95%) for platelets ≥20×109/L and 74% (95% CI, 62-86%) for platelets ≥50×109/L. After allo-HCT 97.8% of patients achieved complete response. The median PNH clone size assessed by flow cytometry was 0% (0-0.1) on granulocytes (n=12) and 0% (0-0) on monocytes (n=8). The CI of acute GvHD grade II-IV at 100 days was 36% (95% CI, 24-49%), while for grade III-IV was 12% (95% CI, 4-20%). The cumulative incidence of any chronic GvHD at 1 year was 38% (95% CI, 25-50%), while for extensive cGvHD was 23% (95% CI, 12-35%). The 3-year GRFS rate was 48% (95% CI, 35-61%). With median follow-up of 7.7 years (95% CI, 6.1-10.3 years), the 3-year OS was 66% (95% CI, 54-78%), the 3-year PFS 66% (95% CI, 54-78%), the 3-year NRM amounted to 26% (95% CI, 15-38%), and the 3-year RI to 5% (95% CI, 0-11%). Twenty-two patients died during follow-up. The most frequent causes of death were infections (8, 40.0%), including 3 patients with infections and chronic GvHD and 1 patient with infection and acute GvHD. Patients' age was the only prognostic for both OS and PFS. Each additional year of age at allo-HCT was associated with a HR of 1.04 (95% CI, 1.01-1.08) for both OS and PFS, with p-values 0.015 and 0.017, respectively. No evidence was found for an impact of GvHD on OS nor PFS. Conclusions Allo-HCT in patients with PNH clones in the context of MDS is associated with very good outcomes in comparison to patients diagnosed solely with MDS, which may at least partially result from the younger age of the transplanted patients or the different etiology of MDS. Relapse incidence is low, and non-relapse mortality associated mostly with infections is responsible for treatment failures. The incidence of GvHD is within the range reported for MDS.