Time and residual hematopoiesis are crucial for PNH clones escape in hepatitis-associated aplastic anemia

The presence of paroxysmal nocturnal hemoglobinuria (PNH) clones in aplastic anemia (AA) suggests immunopathogenesis, but when and how PNH clones emerge and proliferate are unclear. Hepatitis-associated aplastic anemia (HAAA) is a special variant of AA, contrarily to idiopathic AA, in HAAA the trigger for immune activation is clearer and represented by the hepatitis and thus serves as a good model for studying PNH clones. Ninety HAAA patients were enrolled, including 61 males and 29 females (median age 21 years). Four hundred three of idiopathic AA have been included as controls. The median time from hepatitis to cytopenia was 50 days (range 0–180 days) and from cytopenia to AA diagnosis was 26 days (range 2–370 days). PNH clones were detected in 8 HAAA patients (8.9%) at diagnosis and in 73 patients with idiopathic AA (IAA) (18.1%). PNH cells accounted for 4.2% (1.09–12.33%) of red cells and/or granulocytes and were more likely to be detected in patients with longer disease history and less severe disease. During follow-up, the cumulative incidence of PNH clones in HAAA increased to 18.9% (17/90). Nine HAAA patients newly developed PNH clones, including six immunosuppressive therapy (IST) nonresponders. The clone size was mostly stable during follow-up, and only 2 of 14 patients showed increased clone size without proof of hemolysis. In conclusion, PNH clones were infrequent in newly diagnosed HAAA, but their frequency increased to one that was similar to the IAA frequency during follow-up. These results suggest that the PNH clone selection/expansion process is dynamic and takes time to establish, confirming that retesting for PNH clones during follow-up is crucial.