Relevance of Minor PIG-A Mutations In Acquired Aplastic Anemia and Myelodysplastic Syndromes

Aplastic anemia (AA), paroxysmal nocturnal hemoglobinuria (PNH), and myelodysplastic syndromes (MDS) are acquired bone marrow failure diseases. About 15% AA patients eventually transform to classical PNH; however, MDS patients almost never evolve to classical PNH. PNH originates from a multipotent HSC that acquires a PIG-A mutation. PIG-A gene mutations lead to absence of glycosylphosphatidylinositol-anchor proteins (GPI-AP). In PNH, matching mutations can be found in myeloid, erythroid and lymphoid lineages. Roughly 1 in 50,000 (0.002%) colony-forming-cells (CFC) from healthy controls harbor PIG-A mutations. However, PIG-A mutations in healthy controls arise from progenitors rather than multipotent HSC; and are often transient. More than 50% AA patients and 10~30% MDS patients were reported to have small populations (0.01 to 10%) of PNH granulocytes. The objective of this study is to determine the origin of PIG-A mutations in patients with AA and MDS.

A prospective study of 3 AA patients with small PNH granulocyte populations and 17 MDS patients was performed. AA and MDS were diagnosed as per clinical criteria and WHO classification (2008) respectively. The median age for AA patients was 49 (range: 28~66). The median age for MDS patients was 66 (range: 48~79). Two AA patients were managed with supportive care, and 1 received HiCy. Five MDS patients were under observation; 3 each received Aranesp, Revlimid, 5-azacytidine, and MS-275/GM-CSF respectively. PNH granulocyte size was measured by flow cytometry. Bone marrow CD34⁺ cells were isolated via Ficoll-Paque separation and magnetic selection. Proaerolysin is a toxin that selectively kills GPI-AP intact cells but not PNH cells. PNH CFC assay was performed using proaerolysin-containing methylcellulose culture. T cell enrichment culture was performed using peripheral mononuclear cells, RPMI1640 medium with 10% AB serum, micro-beads coated with antibodies against CD2, CD3, and CD28, rIL-2, in the presence of 1nM proaerolysin. DNA was extracted from individual proaerolysin resistant CFC and T cell colonies and was PCR amplified and sequenced for PIG-A mutation analysis. PIG-A mutations were identified using forward/reverses primers.

Of 3 AA patients bearing a small PNH granulocyte population (range: 3.2~3.8%), median granulocyte PIG-A mutation frequency was 1.07% (range: 0.88~1.33%), median bone marrow cellularity was 10% (range: 5~20%), and median CD34⁺ cell was 0.05% (range: 0.019~0.09%). All AA patients had normal karyotypes. PIG-A mutation analysis on 2 of them showed that PIG-A mutations occurred at multipotent HSC level: 1 had a single base substitution mutation at bp549, switching a C to T in both CFU-Gs and T cells; and the other one had a 3 bps (GTC from bp776 to 778) deletion mutation in cells from both lineages. The third patient had a substitution mutation at bp1005 in CFCs, switching an A to T. Her T cell PIG-A mutation analysis is pending. All mutations were confirmed in 3 independent colonies from each lineage.

Of 17 MDS patients, 14 had hypercellular bone marrow, 2 had normal cellularity, and 1 had hypocellularity. Seven had normal karyotypes, 9 had abnormal karyotypes. Only 3 patients (17.6%) had detectable PNH granulocyte populations (range: 0.1~3.2%). Studies on the patient bearing a 3.2% PNH granulocytes indicated that PIG-A mutation occurred at progenitor level and was transient. Proaerolysin-resistant CFC assay only grew 1 proaerolysin-resistant CFU-G in 10 plates with a granulocyte PIG-A mutation frequency of 0.03%. DNA sequence analysis of CFC from that colony demonstrated a PIG-A gene single base T deletion mutation at bp 210. However, no PNH T cells were detected via either flow cytometry or T cell enrichment culture. After 4 months of supportive care, he no longer had detectable PNH cells via either flow cytometry or proaerolysin-resistant CFC colony assay. Another patient, who initially bore a 1.5% PNH granulocyte population, also had his PNH granulocyte population disappeared after 2 months of therapy with 5-azacytidine. The 3rd patient is still on surveillance. No PNH T cells were detected in last 2 patients either.

Similar to classical PNH, PIG-A mutations in AA are clonal and arise from multipotent HSC. However, PIG-A mutations in MDS arise from progenitor cells, which likely explains why PNH populations in MDS patients are transient and seldom evolve into classical PNH.

No relevant conflicts of interest to declare.