Reply to ‘Heterogeneity within AML with CEBPA mutations; only CEBPA double mutations, but not single CEBPA mutations are associated with favorable prognosis’

Sir,

We read with great interest the recent study by Pabst et al (2009) disclosing that there is relevant prognostic heterogeneity within AML patients with CEBPA mutations and only CEBPA double mutations (CEBPA^double-mut), but not single mutations (CEBPA^single-mut), are associated with favourable prognosis in the AML patients. However, the reason why CEBPA^single-mut patients have a poorer outcome than CEBPA^double-mut patients remains unclear and a comprehensive study to evaluate the biological difference between these two groups is still lacking.

In this study, we investigated the prevalence and clinical relevance of CEBPA^double-mut and CEBPA^single-mut and their association with other genetic changes in a large cohort of 543 consecutive de novo AML patients at the National Taiwan University Hospital (NTUH). This study was approved by the Institutional Review Board of NTUH; written informed consents were obtained from all participants in accordance with the Declaration of Helsinki. CEBPA mutations were detected by genomic-DNA PCR and direct sequencing as described earlier (Lin et al, 2005). Mutational analyses of FLT3/ITD, FLT3/TKD N-RAS, K-RAS, NPM1, CEBPA, KIT, AML1 and MLL/PTD were carried out as previously described (Hou et al, 2008).

Among the 543 AML patients recruited, we identified 71 (13.1%) patients with CEBPA mutations, including 47 CEBPA^double-mut and 24 CEBPA^single-mut. Compared with patients who have CEBPA^double-mut, those with CEBPA^single-mut had lower incidences to express HLA-DR (65 vs 96%, P=0.0014), CD7 (44 vs 79%, P=0.006) and CD15 (35 vs 85%, P<0.0001), but a higher incidence to express CD56 (35 vs 11%, P=0.038) on leukemia cells. Apart from this, there were no differences in other clinical parameters including age, sex, haemogram, LDH level, FAB subtype and karyotype between these two groups. No matter CEBPA^single-mut or CEBPA^double-mut, the mutation disappeared at complete remission in all patients who had paired bone marrow samples for analysis and reappeared at relapse.

Patients with CEBPA^single-mut had a higher incidence of NPM1 mutation than those with CEBPA^double-mut (4/24, 16.7 vs 0%, P=0.0109). There was also a higher incidence of concurrent mutation of FLT3/ITD, FLT3/TKD, AML1/RUNX1 or MLL/PTD in CEBPA^single-mut patients than in CEBPA^double-mut patients (20.8 vs 10.6%, 12.5 vs 4.3%, 8.3 vs 2.1% and 4.2 vs 0%, respectively), but the difference did not reach statistical significance. However, when combined together, simultaneous alteration of any one of these four mutations occurred more frequently in the former group than in the latter (37.5 vs 14.9%, P=0.039). More intriguingly, all four CEBPA^single-mut patients with NPM1 mutation also simultaneously had FLT3/ITD (2 patients), FLT3/TKD (1 patient), or both (1 patient).

In terms of outcome, CEBPA^double-mut patients had a higher complete remission rate than CEBPA^single-mut patients (91 vs 56.3%, P=0.0051). The patients with CEBPA^double-mut had a significant longer overall survival (OS) than those with CEBPA^wild or CEBPA^single-mut (median: not reached vs 29.8 months and 7.5 months; P=0.013 and P=0.001, respectively; among 3 groups, P=0.007, Figure 1A). The same was also true for disease-free survival (DFS) (median: 59 months vs 8 months and 4 months; P=0.016 and P=0.027, respectively; among 3 groups, P=0.037). Among the subgroup of patients with normal karyotype, the differences in OS and DFS between CEBPA^double-mut and CEBPA^single-mut patients were still obvious (P=0.002 and P=0.019, respectively, Figure 1B). The multivariate analysis clearly identified CEBPA^double-mut, but not CEBPA^single-mut as an independent prognostic factor for OS and DFS (hazard ratio 0.362, 95% CI 0.182–0.721, P=0.004 and hazard ratio 0.426, 95% CI 0.263–0.691, P=0.001, respectively, Table 1).

Figure 1

Kaplan–Meier survival curves of overall survival (OS) stratified by different status of CEBPA mutation at diagnosis among total patients (A) and in the subgroup of patients with normal karyotype (B). Only patients receiving standard chemotherapy were enrolled into survival analysis. Among total patients, P-value for OS of CEBPA^double-mut vs CEBPA^wild patients was 0.013, for CEBPA^double-mut vs CEBPA^single-mut patients, 0.001, and among three groups, 0.007 (A). In the subgroup of patients with normal karyotype, P-value for OS of CEBPA^double-mut vs CEBPA^wild patients was 0.01, for CEBPA^double-mut vs CEBPA^single-mut patients was 0.002 and among three groups it was 0.005 (B).

Table 1 Multivariate analysis for overall and disease-free survival^a

From the above findings, we hypothesise that the close association of CEBPA^single-mut with CD56 expression (Raspadori et al, 2001) and other poor-risk genetic alterations, such as FLT3/ITD, FLT3/TKD, MLL/PTD and AML1/RUNX1, (Schnittger et al, 2000; Harada et al, 2004; Whitman et al, 2008) may partially explain why CEBPA^single-mut predisposes to inferior outcome than CEBPA^double-mut. We also observed a trend of shorter OS in CEBPA^single-mut patients who had concurrent FLT3/ITD, FLT3/TKD,MLL/PTD or AML1/RUNX1 mutation than those who did not (P=0.064, Figure 2).

Figure 2

Kaplan–Meier survival curves of overall survival (OS) in the CEBPA^single−mut patients with and without concurrent FLT3/ITD, FLT3/TKD, MLL/PTD or AML1/RUNX1 mutation.

In summary, about one-third of patients with CEBPA mutations had CEBPA^single-mut, which were closely associated with CD56 expression but inversely correlated with HLA-DR, CD7 and CD15 expression. Compared with patients who have CEBPA^double-mut, those with CEBPA^single-mut had a higher incidence of concurrent FLT3/ITD, FLT3/TKD, MLL/PTD or AML1/RUNX1 mutation and had a poorer prognosis. This study provides evidences independently from previous ones, stressing the differences in biological characteristics between CEBPA^single-mut and CEBPA^double-mut AML and their possible prognostic implication. Further studies are necessary to clarify whether the close association of CEBPA^single-mut with CD56 expression and other poor-risk gene alterations contributes to the poorer outcome of this group of patients.

Conflict of interest

The authors declare no conflict of interest.