Natural History of Paroxysmal Nocturnal Hemoglobinuria Clones In Patients Presenting as Aplastic Anemia

Acquired aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH) are closely related bone marrow failure disorders. Most AA results from an autoimmune attack directed against hematopoietic stem/progenitor cells. PNH originates from a multipotent hematopoietic stem cell (HSC) that acquires a PIG-A mutation. The PIG-A gene mutation leads to glycosylphosphatidylinositol-anchor protein (GPI-AP) biosynthesis deficiency and subsequent hemolysis secondary to the absence of complement regulatory proteins (CD55 and CD59). Both PNH and AA can be cured by allogeneic bone marrow transplantation (alloBMT), but only a minority of patients is offered this approach due to the potential morbidity and mortality. AA can be treated with immunosuppressive therapy (IST) and PNH can be controlled by eculizumab. It has been estimated that more than 50% of AA patients harbor small, but expandable PNH populations at diagnosis. The natural history of PNH clones in AA patients following non-transplant therapy is not well studied. The purpose of this study is to determine the fate and clinical relevance of these PNH clones in patients with AA who did not receive an alloBMT.

Twenty-seven patients with AA and a detectable PNH clone were monitored for a median of 5.3 years (range,1.5 to 11.5 years). The PNH granulocyte clone sizes were measured using flow cytometry and analyzed via CellQuest software. PE-conjugated anti-CD15 and fluoresceinated aerolysin variant (FLAER) staining were used to define granulocytes and GPI-AP deficient cells respectively. Serum lactate dehydrogenase (LDH) value was used as a surrogate for monitoring hemolysis and 1.5× the upper limit of normal LDH value (330mg/dL) as a cut-off point to define clinically apparent hemolysis. A PNH size change <2.5% was considered as stable. Patients were treated with IST, HiCy, or both.

We found a linear relationship between PNH granulocyte clone size and LDH values (Pearson correlation coefficient=0.73; P<0.0001). A PNH clone size above 23% was the threshold to identify hemolysis as measured by LDH (ROC analysis with AUC=0.88). Higher LDH values over the period of follow-up were associated with larger PNH granulocyte clone size at diagnosis (P=0.03). Patients with small (≤15%) initial PNH granulocytes had lower LDH levels at 5 years after diagnosis (mean±SD: 236.9±109.9 vs 423.1±248.8; P=0.02), and were less likely to develop hemolysis (13.3% vs 55.6%, P=0.06) comparing to those with larger (>15%) initial PNH granulocytes.

Of 9 patients who initially were treated with traditional IST (ATG, CsA, and prednisone), 7 did not respond to treatment and eventually received high-dose cyclophosphamide (HiCy) salvage therapy, 2 achieved a remission and did not require further treatment though one demonstrated PNH clone size expansion to 50% after 37 months. After HiCy salvage, all 7 patients became transfusion independent and 4 of them had no further PNH clone expansion. PNH clone expansion was observed in 7 of 9 patients at a median time of 3 (range: 2 to 87) months after treatment.

Of 15 patients who received HiCy as initial therapy, 14 achieved remissions. Later expansion of PNH size was observed in 7 patients, of which 5 eventually required intermittent blood transfusion but only 1 developed symptomatic hemolysis necessitating eculizumab therapy. The median time to PNH granulocyte clone expansion after HiCy was 52 (range: 18 to 106) months. In 5 patients who received HiCy and then relapsed, their PNH clone size only increased (1±0.7)% in (71±31) months observation during post treatment remission; however, their PNH clone size increase accelerated to (38±14)% in (34±21) months after AA relapse (P=0.04).

Two nSAA patients with an initial PNH clone size ≤15% spontaneously recovered hematopoiesis at 84 and 56 months respectively, neither had PNH clone size expansion.

In this study, 25.9% patients kept a stable PNH size, 48.1% patients increased the size, and 26% patients decreased the size. The group with small initial PNH clone sizes (≤15%) was the most stable over time.

The risk of developing clinically significant PNH over 10 years appears to be low in AA patients with PNH clones, especially for those with small initial PNH granulocyte clones (≤15%) and for those who maintain remission following therapy.

No relevant conflicts of interest to declare.