Paroxysmal Nocturnal Hemoglobinuria (PNH) is a heterogeneous disease characterized by complement-mediated hemolysis and an extremely high risk of thrombosis. Eculizumab, a monoclonal antibody to complement factor C5, effectively blocks hemolysis and reduces thrombotic risk. The mechanism of thrombosis in PNH, and particularly the possible contribution of PNH neutrophils, is still unknown. Recently, neutrophil activation in form of neutrophil extracellular traps (NETs) characterized by the exposure of nucleosomes and neutrophil proteases on a meshwork of DNA was described. NETs are highly procoagulant. Whether they also play a role in the pathogenesis of thrombosis in PNH is not known. Here, we have assessed markers of neutrophil activation and NET formation in untreated PNH patients, and in PNH patients before and after starting eculizumab treatment.


Plasma markers of neutrophil activation such as elastase-α1-antitrypsin (EA) complexes and circulating nucleosomes as a marker of NET formation were measured by ELISA in a cohort of 51 patients with a median PNH clone size of 63% (range 5-100%) and 17 healthy controls. Median time (range) since diagnosis in patients with and without a history of thrombosis (n=12 and 39, respectively) was 8 (1-33) and 2 (0.2-24) years. The median (range) interval between the most recent thrombotic event and the time of sampling was 7.5 (0-240) months. In 20 of 51 PNH patients (clone size 83 (50-100)%), levels immediately before and at various time points during eculizumab treatment (1 hour, 2 hours, 1 week, 4 weeks and ≥ 12 weeks after the first dose) were compared.


In the total cohort, nucleosomes and EA complexes were not significantly different between PNH patients and healthy controls. However, levels of circulating nucleosomes were significantly higher in PNH patients with a history of thrombosis (n = 12, median 16, range 7-264 U/ml) than in PNH patients without (n=39; median 6, range 6-35 U/ml, p < 0.001) and healthy controls (n=17, median 6, range 6-23 U/ml, p < 0.05). Levels of EA complexes did not differ between PNH patients with or without a history of thrombosis and controls. No correlation was found between PNH granulocyte clone size and nucleosome or EA complex levels.

In 20 PNH patients with an indication for treatment with eculizumab, based upon clinical characteristics and clone size, baseline levels (median and range) of nucleosomes and EA complexes were 10 (6-264) U/ml and 27.5 (9-263) ng/ml respectively and did not differ from those in controls (8.1 (6-22.7) U/ml and 24.8 (15.8-44.2) ng/ml). Interestingly EA and nucleosome levels significantly decreased, starting as early as 1 hour after the first dose of eculizumab (p=0.0038 and p=0.002, respectively). EA complexes were still significantly decreased at ≥12 weeks (p 0.0023), whereas this was not the case for nucleosomes. When comparing patients with and without a history of thrombosis, similar as in the total cohort, baseline nucleosome levels (median, range) were significantly higher in patients with a history of thrombosis (n=9; 25 (6-264) U/ml) than in patients without (n=11; 6 (6-101) U/ml; p=0.035), whereas EA complexes were comparable. Immediately after commencement of eculizumab treatment, EA complexes significantly decreased in both groups (p<0.05), whereas for nucleosomes this was only the case in the thrombosis group (p=0.023).


Similar levels of EA complexes in PNH patients and healthy controls argue against marked neutrophil activation in PNH. However, significantly elevated nucleosomes in PNH patients with a history of thrombosis may suggest low-grade NET formation. The prompt and persistent decrease in EA complex levels upon commencement of eculizumab treatment may point at a role of complement in neutrophil activation. The subtle effects of eculizumab on nucleosome levels in the thrombosis group may suggest an inhibitory role of eculizumab on NET formation in PNH patients. The clinical implications of our findings remain to be studied.


Muus:Alexion Pharmaceuticals: Membership on an entity’s Board of Directors or advisory committees.

Author notes


Asterisk with author names denotes non-ASH members.