Glucose-6-phosphatase dehydrogenase [G6PD] serum level predicts acute monocytic leukemia differentiation.

Abstract Introduction: Acute Myeloid Leukemia [AML] leukemogenesis follows differentiation fate that resembles normal hematopoiesis, albeit incompletely. Monocytic AML usually presents with hyperleukocytosis and occasional Leukostasis. Disease may, but not always, be associated with refractory disease. For this particular entity, there is urgent need in uncovering: (a) Leukemogenic vulnerabilities, (b) Biomarkers that could predict either outcome or facilitate post-induction disease monitoring. Interestingly, G6PD is rate limiting enzyme in the pentose-phosphate pathway (PPP) critical for nucleotide synthesis and serve as antioxidant by upregulating glutathione allowing cancer proliferation. In this study, we investigate the association between differential G6PD serum levels and immunophenotypic AML features. Additionally, we explore AML genomic defects associated with deregulated G6PD levels. Methods: After IRB approval, 39/102 [38.2%] of AML cases were included. SAS software was use for descriptive statistics, t-test, Pearson correlations and multilinear regression models. Results: Median age was 68 years (25-93). G6PD level directly correlated with increasing WBC, [r=0.74, p= <.0001, 95% CI 0.55,0.85] suggesting that G6PD is overexpressed in AML blasts with high metabolic and oxidative stress demands. The mean G6PD serum level was 20 U/g Hb (8.6-42). Interestingly, the magnitude of CD34 (r=-0.44, p=0.005) and HLA-DR antigen expression (r=-0.34, p= 0.03) were negatively correlated with G6PD level, albeit CD14 expression was positively correlated, r=0.31, p=0.05, respectively. High CD14 expression [p=0.04] and low CD34 expression [p=0.001] were associated with high G6PD levels when accounting for multiple immunophenotypic antigen expression by multilinear regression [ANOVA, p<.0001]. Since progressive decline in CD34/HLA-DR expression and CD14 antigen acquisition suggests a more differentiated/maturing leukemia subtype [i.e., Monocytic AML], we investigated G6PD expression in AML cases with and without Monocytic default [CD34-, HLA-DR – CD38+ CD14+CD34+]. G6PD level was 24 U/g Hb vs 16 U/g Hb, p=0.007 in Monocytic vs non-Monocytic like AML. In AML cases presenting with G6PD levels >20U/g Hb, a higher proportion of intermediate [0/3 (0%) vs 3/3 (100%)] and lower proportion of adverse [9/28 (32.1%) vs 19/28 (68%)] European Leukemia Network [ELN22] subgroups was observed, p=0.06. Notably, in AML cases with and without G6PD levels >20 U/g Hb, a higher proportion of FLT3 variants [6/9 (67%) vs 3/9 (33%), p=0.07] and lower incidence of P53 mutations [1/11 (9%) vs 10/11 (91%), p=0.01]. Conclusions: Elevated G6PD serum level is associated with propensity for higher WBC and Monocytic AML differentiation. In monocytic FLT3 mutated AML presenting with elevated WBC, it is possible that high G6PD levels support redox adaptation and proliferation. Prior evidence indicates that high G6PD expressing AML stem cells resists cytarabine effect. Translational studies are needed to investigate genetic and pharmacologic inhibition of acute myeloid disorders with high G6PD oncogenic dependency.