Background: Acute myeloid leukemia (AML) with t(8;21) has considerable clinical heterogeneity. Only 50% of these patients receiving multiple cycles of high-dose cytarabine as post-remission treatments could achieve long term survival, and relapse occures in up to 40% of them. Therefore, further risk stratification is needed to guide appropriate post-remission treatment for t(8;21) AML patitents in first complete remission(CR1). Measurable residual disease (MRD) monitored by RUNX1-RUNX1T1 transcript levels after treatment is established as a powerful marker to predict relapse and guide treatment. Recent studies revealed cell-surface antigen CD19 negativity (CD19-) was a risk factor for relapse in t(8;21) AML. However, reports about the impact of CD19 expression combined with RUNX1-RUNX1T1 MRD on post-remission treatment remains absent to date.Methods: A total of 155 consecutive patients diagnosed with t(8;21) AML were enrolled in this study, including 69 patients who received chemotherapy (CMT) and 86 who received allogeneic stem cell transplantation (allo-SCT) as post remission treatment in CR1. MRD+ was defined as a < 3 log reduction of RUNX1/RUNX1T1 level after the second consolidation therapy.Results:With a median follow-up of 40 months (range, 4-133 months), the 3-year OS, LFS and CIR rates were 75.57% ,70.46% and 27.50% for the entire population, respectively. The 3-year CIR was dramatically decreased in allo-SCT than in CMT group (13.47% vs 49.19%, p<0.001). Allo-SCT demonstrated a significantly superior 3-year OS (85.78% vs 58.89%, p< 0.001) and LFS (85.64% vs 45.84%, p< 0.001) over CMT, respectively. The 3-year cumulative incidence of TRM was 9.27% and 7.04% in the CMT and allo-HSCT groups, respectively (p=0.890). KIT exon 17 mutation was a risk factor for OS in multivariate analysis, CD19 negativity was a risk factor for RFS in univariate analysis. Allo-SCT was a beneficial factor for LFS and OS in multivariate analysis when taking CMT as a reference. For CD19- patients (n=59), allo-HSCT demonstrated significantly lower CIR (17.20% vs 51.03%, p=0.006) , better OS (95.24% vs 51.55%, p < 0.001) and LFS (82.8% vs 39.71%, p = 0.001) than CMT. On the other hand, CMT had a comparable CIR (26.70% vs 11.40, p=0.064 ), OS (73.19% vs 86.01%, p=0.153) and LFS (71.58% vs 87.76%, p=0.067) with allo-SCT for CD19+ patients (n=80). Notably, when combined with RUNX1-RUNX1T1 transcripts after the second consolidation, allo-SCT preserved a significantly lower CIR (6.67% vs 66.67%, p=0.002), improved OS (100% vs 33.33%, p< 0.001) and LFS (93.33% vs 33.33%, p=0.001) than CMT in CD19-MRD+ group (n=22), respectively. In contranst, CMT revealed similar OS (80.0% vs 87.5%, p=0.677 ), LFS (60% vs 62.5%, p=0.663) and CIR (40.00% vs 37.50%, p=0.795 ) with allo-SCT in CD19-MRD- patients (n=13), respectively. For CD19+ patients, addition of MRD status had no impact on the outcome of allo-SCT and CMT. CMT preserved comparable OS (75% vs 84.66%, p =0.516), LFS (85.71% vs 96.30%, p =0.357 ) and CIR (12.50% vs 3.70%, p =0.367) with allo-SCT for CD19+MRD+ patients (n=36), and comparable OS (100% vs 92.31%, p =0.459), LFS (83.33% vs 85.29%, p =0.940 ) and CIR (16.67% vs 13.54%, p =0.998) with allo-SCT in CD19+MRD- patients (n=33).Conclusion: Combination of CD19 expression and RUNX1-RUNX1T1 MRD improved risk stratification and treatment guidance for t(8;21) AML in CR1. Allo-SCT might be recommended for the CD19-MRD+ patients, while CMT might be a reasnonable choice for CD19-MRD+ and CD19+ patients. Multicenter prospective studies are warranted to confirm the current results.Key words: t(8;21) AML, CD19, measurable residual disease, post-remission treatment, allo-SCT, chematherapy DisclosuresNo relevant conflicts of interest to declare.
Read full abstract