Clinical Significance of Mitochondrial DNA Content in Acute Promyelocytic Leukemia

Used in the clinical practice for more than three decades, the all-trans retinoic acid (ATRA) rendered acute promyelocytic leukemia (APL) the most curable subtype of acute myeloid leukemia, and currently, its combination with arsenic trioxide (ATO) exceeded all expectations for a chemotherapy-free protocol. In terms of metabolic importance, ATRA can also modulate the mitochondria-mediated cellular metabolism and promote a shift from a glycolytic-driven metabolism to an oxidative phosphorylation profile, although this effect has never been demonstrated in APL. As part of the cellular metabolic machinery, mitochondrial DNA (mtDNA) content has been reported to be altered in different types of solid tumors with clinical implication on patient treatment outcomes, although its clinical significance in acute leukemias has not been investigated to the same extent. Particularly in acute promyelocytic leukemia (APL), the role of mtDNA content on prognostication is completely unknown. Considering that mostly APL samples display a glycolytic-driven metabolism, it is conceivable that APL patients harboring high mtDNA content may present a better response to ATRA-based therapies. To test this hypothesis, we determined the mtDNA content in samples from patients with APL enrolled in the International Consortium on Acute Promyelocytic Leukemia study (Rego et al. Blood. 2013 Mar 14;121(11):1935-43) and analyzed its relationship to treatment outcomes. Diagnostic bone marrow (BM) mononuclear cells from 156 consecutive patients with APL (median age: 35 years, range: 18-82 years; 45% male) were obtained at diagnosis. For comparison purposes, we also included peripheral blood (PB) from 293 age- and sex-adjusted healthy volunteers. First, we determined whether mtDNA content could be compared between PB mononuclear cells and BM. To do so, we measured the mtDNA content of 22 APL patients, for whom paired samples were available at the time of diagnosis and detected a strong correlation between PB and BM samples (Pearson correlation coefficient, r=0.78, 95% confidence interval, CI: 0.54 to 0.9). Next, we used the values of mtDNA higher than the 95 ^th percentile of healthy subjects (≥1.63. Note: this value represents a fold change relative to healthy control) to define APL patients with high mtDNA content. Patients that presented values within the range of normal control samples (<1.63) were classified as normal mtDNA content. The median follow-up among survivals was 40 months (95%CI: 34-47 months). Of the 131/156 patients who achieved complete remission, 18 patients (14%) relapsed. mtDNA content had no impact on complete remission achievement (84% for normal mtDNA versus 83% for high mtDNA; P=0.924) or overall survival (78% for normal mtDNA versus 80% for high mtDNA; P=0.69). In contrast, patients with high mtDNA content had a significantly high 5-year disease-free survival rate (86%, 95%CI: 78-95%) than patients with normal mtDNA content (61%, 95%CI: 46-82%). Considering non-relapse death as a competing cause of failure, the 5-year cumulative incidence of relapse (CIR) for patients with high and normal mtDNA content were 35% (95%CI: 16-49%) and 10% (95%CI: 2-17%), respectively. The multivariate Cox proportional hazards model showed that mtDNA content was independently associated with CIR (hazard ratio, HR: 0.31, 95%CI: 0.12-0.8) considering PETHEMA/GIMEMA risk of relapse subgroups and age as confounders. To functionally evaluate the metabolic alterations in APL cells upon ATRA treatment, NB4 cell line was treated with ATRA (1 µM) for 48 and 72 hours. In vitro analyses demonstrated (as expected) that the treatment with resulted in increased levels of myeloid maturation markers (CD11b/CD11c/CD15), with morphological changes being only observed at 72 hours. Metabolically, we observed an increase in mitochondrial mass and potential upon ATRA-treatment after 48 hours, which was also reflected by increase in the mtDNA content (2-fold increase in comparison with the vehicle). Together, these findings demonstrate an important, but not completely understood role for mtDNA content in APL.