Baseline Assessment of Patients with Newly-Diagnosed Acquired Bone Marrow Failure Enrolled in a Prospective Observational Study, "Search for Unidentified Links between PNH Clone Size and the Related Clinical Manifestations By High Precision Flow Cytometry (SUPREMACY)"

Introduction:Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal stem cell disorder characterized by an acquired mutation(s) of the phosphatidylinositol glycan class A (PIGA) gene, leading to expansion of glycosylphosphatidylinositol-anchored protein deficient blood cells (PNH-type cells). By using high-resolution flow cytometry established by our group, PNH-type cells can also be detected at low frequencies (less than 1%) in some patients with acquired bone marrow failure (BMF) such as aplastic anemia (AA) and lower-risk myelodysplastic syndromes (MDS). Although the presence of a small number of PNH-type cells has been shown to be associated with a favorable response to immunosuppressive therapy and good prognosis in patients with BMF (Blood 2006, 107:1308-1314; Br J Haematol 2014, 164:546-54), the clinical significance of the PNH-type cell population in the clinical course of BMF remains unclear. Japan PNH Study Group is conducting a nation-wide multicenter prospective observational study for newly-diagnosed patients with BMF in Japan, SUPREMACY (Search for unidentified links between PNH clone size and the related clinical manifestations by high precision flow cytometry). We herein report an interim analysis of the baseline data of this study.

Methods:Patients newly diagnosed with PNH, AA, MDS or unclassified BMF are recruited to the study from April 2016 until March 2020 in Japan. The percentage of PNH-type cells in peripheral blood is measured by high-resolution flow cytometry at the time of enrollment and repeated every year for 3 years. Combination of antibodies against the lineage markers and liquid FLAER or cocktail antibodies against CD55 and CD59 were used for the detection of PNH-type cells in granulocytes (Gr) (≥0.003%) or erythrocytes (Ery) (≥0.005%) and monocytes (Mo) (≥0.01%), respectively. Moreover, to explore the relationship between the presence of PNH-type cells and specific clinical manifestations, we collected clinical laboratory test data on iron metabolism, coagulation, renal function and cardiac function and surveyed physical conditions of the patients using the health-related quality of life (HRQOL) questionnaires targeted to the management of chronic illness in those patients at baseline; measurements will also be repeated every year for 3 years. This study was approved by the research ethics board of each institution.

Results: As of March 2017, 314 patients (162 female [52%]) were examined; 129 (41.1%) were positive for PNH-type cells in all of Gr, Ery and Mo, and 38 (12.1%) had PNH-type cells ≥1%. The median age (min, max) was 67 (18, 86) vs 65 (16, 100) years for patients with vs without PNH clones, respectively. There was no significant difference in physical conditions between two groups. Out of 314, 162 (52%) patients were diagnosed to have AA (n=72), MDS (n=78), and PNH (n=12) based on the case report forms. PNH-type cells were detected in 59.7%, 29.5% and 100% of patients with AA, MDS (33.3% for RCUD, 26.5% for RCMD, and 0% for RARS and RAEB) and PNH, respectively. PNH clone size was significantly correlated between each cell lineage (correlation coefficient, 0.861 for Gr vs Ery, 0.886 for Gr vs Mo and 0.855 for Ery vs Mo). The PNH clone size in Ery was smaller than those in Gr or Mo. Half of the patients with PNH-type cells ≥1%, the serum lactate dehydrogenase levels exceeded the ≥1.5×upper limits of normal. Interestingly, the total iron binding capacity (TIBC) significantly correlated with the PNH-type cell population in Gr, Ery or Mo (p<0.0001). In addition, there was a negative correlation between other iron metabolism markers (serum ferritin and iron) and the PNH-type Ery population (p<0.01). These results may be explained by iron loss due to clinical or subclinical intravascular hemolysis in a manner dependent upon the PNH clone size.

Conclusion: High-resolution flow cytometry could identify PNH-type cells not only in Gr/Ery, but also Mo in about 40% of BMF patients enrolled in SUPREMACY. Frequency of patients positive for PNH-type cells in each disease was similar to those reported in previous studies (about 60% in AA, 20~30% in lower-risk MDS, and 0% in RARS or RAEB). The study will continue to enroll up to 1500 patients, and further monitor the PNH clone size and the concomitant clinical information in the following three years.