Abstract
The retinoic acid syndrome (RAS) is a most important complication in all-trans retinoic acid (ATRA)-based remission induction therapy in acute promelocytic leukemia (APL). Predictive factors associated with the development of RAS have not been fully established. Cyclin A1 is a tissue specific cyclin that is required for G2/M phase transition in spermatogenesis. Abnormal expression of cyclin A1 has association with tumorigenesis. Human cyclin A1 is highly expressed in acute myeloid leukemia, especially in APL. The close association between expression of cyclin A1 and pathogenesis of APL was suggested by previous study, which showed (1) the APL-associated aberrant fusion proteins(PML-RARαor PLZF-RARα) caused overexpression of cyclin A1 mRNA through the direct activation of it’s promoter, (2) the elevation of cyclin A1 expression was reversed by treatment with ATRA in APL cells. To elucidate the clinical relevance between expression of cyclin A1 and APL, we examined expression of cyclin A1 mRNA in 37 APL samples having PML-RARα fusion gene, and correlated the results of clinical features and disease outcomes of the APL patients treated with ATRA-based differentiation therapy. Twenty-nine patients out of 37 were treated with ATRA-based remission induction therapy, including (1) ATRA(45mg/m2) followed by chemotherapy (cytarabine plus idarubicin) if the initial leukocyte count was below 3x109/L or (2) ATRA combined chemotherapy if the initial leukocyte count exceeded 3x109/L. Levels of cyclin A1 transcripts were determined by quantitative real-time RT-PCR. Cyclin A1 mRNA expression index (A1 index)was defined as copy numbers of cyclin A1 in μg RNA which contains 107 copies of GAPDH. Genomic methylation status of cyclin A1 promoter was also analyzed quantitatively using real-time PCR-based method as described previously. All APL samples expressed detectable cyclin A1 mRNA, and A1 index ranged from 4.3 to 202 (A1 index 71.9± 59.5 [mean±SD]), while CD34 +cells isolated from 4 normal bone marrow samples showed below detection levels of cyclin A1 mRNA. In APL cells, initial white blood cell count and A1 index showed significant correlation (r=0.557,p=0.0014). A1 index in 6 APL samples, which developed RAS, was significantly higher compared with those in a group without developing RAS (A1 index 101.5± 65.8 for a group with RAS vs. 72.9±60.1 for a group without RAS, p=0.015). Difference of initial white blood cell count between these 2 groups was not statistically significant (5.0±4.8x 109/L for a group with RAS vs. 9.6±20.6x109/L for a group without RAS, p=0.118). Furthermore, in a group received ATRA-based differentiation therapy, survival probability of the group with high A1 index (A1 index > 130, 9 cases) was significantly shorter compared with the group with low A1 index (A1 index < 130, 20 cases) (p=0.035). Hypermethylation of cyclin A1 promoter was detected in 4 out of 37 APL samples and the degree of CpG methylation varied from 8.6 to 85.1%. In Hela cells, 96% of sequence was methylated. No detectable genomic methylation was found in normal CD34+ cells. However, expression levels of cyclin A1 mRNA did not have significant association with detectable methylation in APL cells. (p=0.3)
In summary, the present study demonstrated that increased expression of cyclin A1 mRNA in APL cells is a novel predictive factor for the development of RAS and an adverse prognostic indicator for APL in differentiation therapy by ATRA.
Disclosure: No relevant conflicts of interest to declare.
Author notes
Corresponding author