Progressive Hyperleukocytosis during Initial Therapy Is a Predictive Marker for Differentiation Syndrome and Early Mortality in Adult Patients with Acute Promyelocytic Leukemia

Background: Acute promyelocytic leukemia (APL) is classified into a favorable-risk group and long-term overall survival (OS) is estimated at around 80%. Relapse rate of APL is lower than another acute myeloid leukemia (AML) subtypes, but we confront higher incidence of early deaths caused by fatal complications including bleeding events and differentiation syndromes (DS) during initial therapy. Recently, although arsenic trioxide (ATO) is introduced with a better survival outcome, the results were from data of low to intermediate-risk group. Thus, patients in high-risk group still show poor survival outcome with high probability of early complications and deaths. We calculated the incidence of DS and early deaths, and tried to find out affecting factors for those early events.

Methods: In this single center retrospective study, 259 APL patients (median 42 years old (16-72), follow-up was 65.4 months (11.1 - 170.5) from 2002 to 2014 were analyzed. APL was diagnosed by RT-PCR method for detection of PML-RARa and all patients were available with cytogenetic results. All except 5 patients with normal karyotype was identified with t(15;17)(q22;q21) and 77 showed combination of additional karyotypes. All patients were supported with sufficient transfusion and received ATRA. Our treatment protocol was based on the modified AIDA protocol using ATRA and idarubicin monotherapy (Sanz et al. Blood. 1999; 94: 3015-21) but some patients with comorbidity were treated with ATO, low-dose cytarabine, and ATRA alone for remission induction. For hyperleukocytosis, we conducted leukapheresis when leukocyte counts exceeded 50 (x10⁹/L) and some were treated with hydroxyurea, cytarabine and prophylactic dexamethasone. High-risk group was determined according to the Sanz criteria which presented leukocyte count > 10 (x10⁹/L) at diagnosis. For leukocyte count, we checked diagnostic level (WBCdx) and the maximal level (WBCmax) during initial therapy and identified a group which showed a meaningful increment of WBCmax compared to WBCdx.

Results: ATRA was applied in 258 patients and 217 (84.1%) were treated with idarubicin, 13 (5.0%) were with ATO, 3 (1.2%) were with low-dose cytarabine. Eight-week cumulative incidence of early death and DS was 13.5% and 17.8%, and hematological CR was identified in 222 (86.0%) patients. Five-year OS and EFS was 76.8% and 69.8%, and CIR rate was 15.7%. Six patients showed clonal evolution to therapy-related AML and 3 patients died in CR. FLT3-TKD and FLT3-ITD mutation was identified in 12 (7.3%) and 34 (20.7%) patients, and PML-RARa BCR3 and BCR1 subtype was identified in 70 (36.8%) and 120 (63.2%) patients, respectively. For leukocyte counts, except for WBCdx higher than 43 (x10⁹/L), which showed significantly higher rate of early death and DS, patient groups with WBCdx <10 (x10⁹/L) vs. 10 to 43 (x10⁹/L) showed no differences regarding early death or DS. We identified that the significance of WBCdx has been changed with increment during initial therapy which revealed WBCmax was more influential. Among the patients with WBCdx <43 (x10⁹/L), WBCmax increased higher than 43 (x10⁹/L) was related with higher incidence of early death (35.5%) and DS (30.6%), while more DS (40%) was identified in patients with higher increment ratio from WBDdx <10 (x10⁹/L). Multivariate analysis revealed WBCmax > 43 (x10⁹/L) and low antithrombin III were significant for DS, while old age, WBCmax, and high D-dimer were associated with early death. In our data, dexamethasone prophylaxis did not show a preventive effect for DS or early death, while leukapheresis in patients with WBCmax >43 (x10⁹/L) showed marginally decreased early death rate `resulting superior OS without significant bleeding complications.

Conclusion: Our data revealed WBCmax with higher increment ratio was a significant predictive factor for early death and DS compared to WBCdx even in the low Sanz-risk group. The role of dexamethasone, transfusion support including antithrombin III, leukapheresis or cytoreduction should be evaluated in the specific patient subset for reducing early events in APL.