Introduction: To investigate the clinical difference between CD56+AML and CD56AML, we examined the incidence of CD56+AML and analyzed several clinical and biological characteristics, cytogenetic features, immunophenotypic features, and clinical prognosis of CD56+AML.

Method: 213 patients (177 adults, 36 children) with de novo AML in Seoul National University Hospital were enrolled. FAB subtypes were 4 M0 (2.3%), 26 M1, (12.2%), 78 M2 (36.6%), 27 M3 (12.8%), 43 M4 (20.2%), 15 M5 (7.1%), 14 M6 (6.6%), and 5 M7 (2.4%). Immunophenotyping was performed by immunohistochemical stain or multiparameter flow cytometry. The association between CD 56 expression and variables (age, sex, WBC count, blast, cytogenetic features, extramedullary involvement, disease free survival, and overall survival) was analyzed by the fisher’s exact test, chi-square test and T test, and Kaplan-Meier method. All statistical calculation were performed using SPSS system.

Results: CD56 was detected in 6 of 177 adults (3.4%) and in 9 (25.0%) of children, showing significantly higher in children group (p = 0.000). The CD56 expression was significantly frequent in M7 subtype (p=0.003). The other clinical characteristics, such as sex, WBC count, % blasts in peripheral blood was not significantly associated with CD56 positivity. Frequency of extramedullary involvement did not differ between CD56+AML and CD56AML. However, the frequency of CNS involvement at initial diagnosis or relapse were significantly higher in CD56+AML (p=0.018). (Table 1) Among cytogenetic changes, AML1/ETO rearrangement was significantly associated with CD56 expression (p=0.024). Aberrant expression of one or more lymphoid markers was observed in 42.9% (6/15 CD56+AML) and 45.2% (80/198 CD56+AML), showing no difference between CD56+AML and CD56AML. Median disease free survival was 7 months in CD56+AML and 19 months in CD56AML, showing no significant difference in two groups (p=0.2466). Median overall survival was not reached in CD56+AML and 25 months in CD56AML (p=0.1591).

Conclusion: The incidence of CD56 positive AML was higher in childhood AML and association of CD56+AML with AML1/ETO rearrangement, CNS involvement and M7 subtype was observed, suggesting the different biologic behavior of CD56+AML. Furthermore, we suggest that 3rd lineage of leukemic cells, would be taken into consideration for the biphenotypic leukemia.

Extramedullary involvement of CD56+ vs. CD56−

CD56+CD56-p
* chi-square test or fisher’s exact test (level of significance: p<0.05) 
 n=15 (7.0%) n=198 (93.0%)  
Extramedullary involvement 3 (20.0%) 27 (13.6%) 0.450 
Adult 19  
Child  
CNS involvement 3 (20.0%) 7 (3.5%) 0.025 
Adult  
Child  
CD56+CD56-p
* chi-square test or fisher’s exact test (level of significance: p<0.05) 
 n=15 (7.0%) n=198 (93.0%)  
Extramedullary involvement 3 (20.0%) 27 (13.6%) 0.450 
Adult 19  
Child  
CNS involvement 3 (20.0%) 7 (3.5%) 0.025 
Adult  
Child  

Author notes

Disclosure: No relevant conflicts of interest to declare.

Sign in via your Institution