Rare germline complement gene variants in severe aplastic anemia Free

Background: Severe aplastic anemia (SAA) and paroxysmal nocturnal hemoglobinuria (PNH) are strongly associated; the AA/PNH syndrome has been recognized in the clinic for many decades, and approximately 40% of SAA patients have a PNH clone >1% (AA/PNH), as measured by deficiency of glycosylphosphatidylinositol-anchored proteins in granulocytes on flow cytometry. Approximately 5% of patients have clinical PNH at disease presentation, defined as clone size >50%, or intravascular hemolysis or thrombotic events at clone size >20%, while 5% of patients progress to clinical PNH years after immunosuppression (IST). Rare germline variants in complement genes have been found enriched in patients with clinical PNH and to modulate thrombotic risk (Prata et al, 2023; Bravo-Perez et al, 2024). Whether these variants are also present in immune AA and are linked to the presence of sub-clinical PNH clones or clinical PNH evolution is not reported.

Methods: A total of 143 patients with SAA enrolled in different IST-based protocols at the National Institutes of Health (1993-2024) and with DNA sequencing data covering 10 complement genes (C3, C5, CFB, CFD, CFH, CFI, CFP, CFHR1, CFHR5, MASP) were included. Rare germline variants (missense or loss-of-function [LoF]) were scored if they were absent or at maximum frequency <0.1% in gnomAD v.4.1 (Tier 1) or had a maximum frequency >0.1% and <0.5% but were LoF or previously reported as pathogenic (Tier 2). Tier 1/2 variants were correlated with laboratory and clinical characteristics frequently associated with complement disorders, including high LDH, renal impairment (baseline creatinine, acute kidney injury [AKI]), neurologic dysfunction, concurrent autoimmune or rheumatologic disorder, coagulation and thrombosis (including coagulation laboratory parameters), and development of thrombotic microangiopathy (TMA), as well as AA disease-related factors such as hematologic response to IST, and overall survival (OS). We also specifically assessed cyclosporin (CSA)-related complications, including AKI, hypertension, and hypomagnesemia, due to the known relationship between CSA administration and TMA development.

Results: Patients' median age was 31 (4-82). At disease presentation, 55/143 (38%) had a PNH clone >1% (mean size =12% [1-93%]), including six (4%) with clinical PNH (mean size =54% [37-93%]). PNH evolution occurred in 14 patients at a median 3.7 years post-IST. In this cohort, 29/143 (20%) patients had rare complement gene variants classified as Tier 1 (n=19; 13%) or Tier 2 (n=10; 7%). Tier 1/2 variants were mainly in CFH (n=5), C3 (n=5), and C5 (n=4); few cases had variants in either CFI, CFB CFD, or MASP. Patients with or without clinical PNH had similar frequencies of Tier 1/2 variants (25% vs. 19%, respectively; p=0.79); the frequency of these variants increased slightly but not significantly according to PNH clone size at baseline (PNH size <1%=14/87 patients [16%], 1-10%=9/35 patients [25%], ≥10% =6/20 patients [30%]; p=0.25). Rare Tier1 CFH variants frequency was 2.5% vs. 5% in SAA vs. clinical PNH (p=0.45); C3, C5, and CFI variants were absent in the latter. Tier 1/2 variants did not associate with PNH clonal expansion or PNH evolution, hematologic response to IST and OS, or with clinical or laboratory markers of impaired renal function, hemolysis, or thrombosis. There was a history of co-occurring autoimmune or rheumatologic disease in 21% and of neurologic dysfunction in 9%. CSA renal adverse events occurred in 84% patients (AKI 11%, hypertension 54% and hypomagnesemia 67%); none were correlated with Tier 1/2 variants. Only one patient, who did not have a Tier 1/2 variant, developed TMA. Five patients experienced thrombotic events; none had clinical PNH and only one had a Tier 2 variant. Two SAA patients with LoF CFI (p.Q462Hfs) and C3 (c.1480-4C>A) variants had no signs of a complement disorder; however, the latter developed clinical PNH after IST.

Conclusion: In this SAA cohort, rare complement gene variants were seen in approximately 20% of patients, unrelated to the presence of a PNH clone. Given the lack of clinical associations, the relevance of rare complement variants in AA patients should be interpreted with caution. Rare complement variants are present in SAA at a similar frequency to clinical PNH, and they may represent a general feature of the germline background of immune bone marrow failure. Larger cohorts are required to increase our statistical power for analysis.