PS1115 THE RELATIONSHIP BETWEEN THE PRETREATMENT PNH CLONE SIZE AND CLINICAL COURSE IN PATIENTS WITH BONE MARROW FAILURE SYNDROMES: INTERIM ANALYSIS OF JAPANESE MULTICENTER PROSPECTIVE STUDY

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is caused by a clonal expansion of stem cell(s) with PIGA mutation. PIGA mutation itself doesn’t confer growth advantage on PNH clones in mice models, and what drives PNH clonal expansion has been a long‐standing issue. We have previously reported the high prevalence of increased PNH‐phenotype cells in bone marrow failure syndromes (BMF) including aplastic anemia (AA) and myelodysplastic syndromes (MDS), but the kinetics of PNH clone in BMF remains to be elucidated. Aims: To determine what affects PNH clonal dynamics by conducting a prospective multi‐central clinical trial, SUPREMACY. Methods: Patients (pts) diagnosed with PNH, AA, MDS or other BMF were prospectively recruited to this study. All the pts were eculizumab naive. Blood samples to assess PNH clone size were sent to the single central laboratory, and assessed using high‐precision flow cytometry (Ann Hematol. 2018 97:2289–2297) for the PNH‐type cell positivity (granulocytes, ≧0.003%; red blood cells, ≧0.005%; monocytes, ≧0.01%) every 12 months. All other lab data and clinical information were collected at the time of the PNH‐clone size assessment from each participating institution or hospital, and were analyzed at Japan PNH Study Group. Results: Between April 2016 and September 2018, 1,171 pts were enrolled into this study. The median age was 67 years with 50.0% males. PNH‐type cells were positive in 55.8% (159/285) of AA, 20.3% (49/242) of MDS, 100% (29/29) of PNH, 22.7% (39/172) of suspected PNH, and 36.1% (160/443) of undiagnosed BMF. 235 pts were eligible for this interim analysis of the change in the percentage of PNH‐type cells. At 12 months, 16 out of 102 PNH (+) pts became PNH (‐), whereas none of 133 PNH (‐) pts became PNH (+). Of the 102 PNH (+) pts whose blood samples were serially examined, the percentage of PNH‐type granulocytes increased in 40 pts (PNHg inc pts) , while it decreased in 61 pts (PNHg dec pts) and remained the same in one patient. At the time of registration (0 month), PNHg inc pts had a higher percentage of PNH‐type granulocytes than PNHg dec pts (median: 0.470 % vs 0.059 %, P  < 0.01; mean: 12.534 % vs 4.012 %, P = 0.0362). There were more pts with large (≧1%) PNH‐type cells in PNHg inc pts [40.0% (16/40)] than in PNHg dec pts [21.3% (13/61)] (P = 0.0423). Clinical data regarding response to immunosuppressive therapy (IST) were available with 94 pts. PNH (+) pts showed a better response rate [CR+PR, 57/67 (85.1%)] to IST compared to PNH (‐) pts [16/27 (59.3%)] ( P  < 0.01). The changes of the percentage of PNH‐type granulocytes were not statistically different between those with IST (n = 67, median −0.004%) and those without IST (n = 35, median −0.005 %) (P = 0.9046), or between those who respond to IST (CR+PR: n = 57, median −0.004%) and those who did not (NR: n = 10, median −0.002%)(P = 0.6036). Summary/Conclusion: Increased PNH type‐cells were highly prevalent among pts with BMF, and the presence of increased PNH‐type cells predicted better response to IST, both of which were consistent with previous reports. There was a tendency toward higher chance of PNH clones to expand in pts whose PNH clone size was large than in those whose PNH clone size was small at the time of registration. The effect of IST to the kinetics of PNH‐type cells were not observed. These observations suggest that clonal dynamics of PNH‐type cells might be different depending on their clone size before treatment. This is the first interim analysis of the study, and further accumulation of cases is needed to clarify those issues.