Gene mutations in acute promyelocytic leukemia early death in patients treated with arsenic trioxide alone.

purposeAPL patients have recurrent alterations in FLT3, WT1, NRAS and KRAS. Gene mutations have a strong potential for involvement in pathogenesis and may have potential effects on the clinical manifestations. Gene mutations may even be associated with early death (ED) in APL patients. However, there is little published information on mutations in APL patients and whether they are attributed to early death.

methodsIn this study, we retrospectively analyzed the clinical data and gene mutations of 134 de novo APL patients. We detected the gene mutations by next-generation sequencing (NGS) to investigate the genetic predictors of early death in APL patients. According to the number of gene mutations per patient, the 134 APL patients were divided into three groups. All patients received arsenic trioxide (ATO) alone as induction therapy. The clinical data and gene mutations were compared and analyzed.

resultsA total of 134 APL patients were involved in the study. The clinical data of sex, WBC, PT, and DD, UA, and LDH level were significantly different between the three groups (P = 0.000, P = 0.000, P = 0.009, P = 0.020, P = 0.030, P = 0.001 and P = 0.014, respectively). Meanwhile, among them, the Sanz risk stratification and early death rate were significantly different (P = 0.001). The early death rate was 10.4%, and the median time to early death was 6.6 days (range 2-15 days). For the next-generation sequencing, a mean of 1.28 ± 1.06 mutations per patient was detected (range: 0-5). The univariate and the multivariate regression analysis showed that age > 50[HR = 1.666, CI (1.027-2.702), P = 0.039], high WBC count [HR = 4.702, CI (1.026-21.543), P = 0.046] and low ALB levels [HR = 4.547, CI (1.088-18.995), P = 0.038] were independent risk factors for early death in APL patients. Furthermore, Kaplan-Meier survival analysis, univariate analysis, and the multivariate regression analysis showed that patients with multiple gene mutations [HR = 2.258, CI (1.115-4.571), P = 0.024], KRAS [HR = 5.136, CI (1.356-19.455), P = 0.016] and/or GATA2 [HR = 4.070, CI (1.287-12.877), P = 0.017] have a significantly higher early death rate.

conclusionThe results of this investigation show that both molecular markers and clinical variables should be used as potential predictors for early death in APL patients. Our results suggested that age > 50, high WBC count, low ALB levels, and the presence of multiple gene mutations, KRAS and/or GATA2 at the time of diagnosis were independent risk factors for early death in APL patients. For these patients, clinicians should be more cautious during the course of induction treatment.