Causes of Induction Failure in Newly Diagnosed Acute Promyelocytic Leukemia Patients Treated with Simultaneous ATRA and Idarubicin (AIDA).

The simultaneous combination of ATRA and anthracycline-based chemotherapy has resulted in a major improvement in remission induction rate and long term survival in patients with APL. This benefit is mainly due to an increased antileukemic efficacy, leading to a virtual absence of leukemic resistance and to a favorable impact on the coagulopathy and on the retinoic acid (RA) syndrome. One of the most successful regimens for induction is the combination of ATRA plus idarubicin (AIDA). In this study we analyze the causes of induction failure observed in two consecutive studies using the AIDA regimen (PETHEMA LPA96 and LPA99). Treatment consisted of oral ATRA (45 mg/m²/d ) until complete hematological remission and idarubicin (12 mg/m²/d) iv on days 2, 4, 6 and 8. In the LPA99 study, the dose of idarubicin on day 8 was omitted for patients older than 70 years of age and all patients received prednisone (0.5 mg/kg/day po for 14 days) for RA syndrome prophylaxis, as well as tranexamic acid (100 mg/kg/day iv) if the platelet count was below 50 × 10⁹/L, in an attempt to improve the control of coagulopathy. The whole series included 642 genetically diagnosed APL patients, 174 in LPA96 and 468 in LPA99 studies, respectively. There were no differences in patient characteristics at presentation between the two groups. Induction results were virtually identical for LPA96 and LPA99 studies with an overall CR rate of 91%. Fifty-two patients (8.4%) failed to achieve CR due to early death. Four additional patients were considered as “resistant” because of persistence of a variable proportion of atypical promyelocytes on days 30 to 41. The distribution of causes of early death was as follows: hemorrhage 30/52 (58%); infection 16/52 (31%); and RA syndrome 5/52 (10%). Mortality due to hemorrhage and infection was significantly different in patients >60 years than in younger patients (37.5% and 56% vs. 72% and 16%, respectively; p=0.01). Neither the addition of prednisone nor tranexamic acid in the LPA99 study had an impact in early death rate. With respect to the chronology of deaths due to hemorrhage, nearly a half (14/30; 47%) occurred during the first week of induction therapy. Central nervous system and pulmonary hemorrhage (17 and 11 patients, respectively) were the only documented sites of lethal bleeding. Univariate analysis showed that age older than 70 years, high WBC counts, clinical and/or laboratory signs of coagulopathy, and abnormally increased level of serum creatinine were associated with an increased risk of lethal bleeding. Multivariate analysis showed that only WBC > 10 × 10⁹/l and serum creatinine > 1.4 mg/dl at presentation retained an independent predictive value of lethal bleeding. In conclusion, despite outcome improvement provided by modern ATRA plus chemotherapy combination, hemorrhagic death remains the primary cause of induction failure in APL. The prophylactic use of prednisone and tranexamic acid failed to reduce the rate of early deaths in newly diagnosed APL. In particular, our study shows that patients presenting with hyperleukocytosis or renal dysfunction are of increased risk of lethal bleeding and should be early managed with more aggressive supportive measures to prevent hemorrhagic death.