Abstract
ΔNp73 is an alternative TP73 gene transcript lacking the transactivation (TA) domain that is generated via alternative splicing and/or P2 promoter. The encoded protein acts as a potent transdominant inhibitor of wild type TP53 and full-length TAp73. In several human malignancies the unbalanced expression of transcriptionally active (TAp73) and inactive (ΔNp73) variants correlates with treatment outcome. We have previously reported that higher ΔN/TA isoform expression ratio was associated with poorer prognosis and resistance to cytarabine induced apoptosis in patients with acute myeloid leukemia (AML) (Lucena-Araujo et al., 2008). In acute promyelocytic leukemia (APL), both isoforms are expressed, but the clinical significance remains unknown. The aim of this study was to determine whether the ΔN/TA expression ratio was associated with treatment outcome of APL patients and to investigate the mechanisms by which ΔNp73 may contribute to PML-RARa+ cell survival. Using isoform-specific probes for ΔNp73 and TAp73, their expression was analyzed in 166 APL patients by Real-time quantitative polymerase chain reaction (RQ-PCR). Patients were divided into tertiles for ΔN/TA expression ratio (median value=23.62; 33rd/66th percentiles=12.8/42.3) and their clinical and laboratory characteristics were compared. Patients in the highest tertile presented higher white blood cells (WBC) counts than those in intermediate/lower tertiles (p <0.001), but no significant differences were observed for age, gender, PML breakpoint, or platelets count. Higher ΔN/TA expression ratio values were significantly associated with the presence of FLT3-ITD (p =0.001). Treatment outcome was obtained for 131 APL patients enrolled in the APL99 (n=41) and IC-APL (n=90) trials. The mean follow-up was 29.1 months, ranging from 1 to 85.5 months. The mean overall survival (OS) of all patients was of 66.8 months [95%CI; 60.8 to 72.8], whereas it was of 67.1 months [56.5 to 77.7] for patients in the lower and 41.7 months [32 to 51.4] for those in the higher tertile for ΔN/TA expression ratio (p=0.014, Figure 1A). Univariate analysis identified WBC counts above 10,000/μl (p =0.003), FLT3-ITD mutation (p =0.011) and ΔN/TA expression ratio (p =0.014) as predictive factors for OS. However, in multivariate Cox analysis, these three prognostic factors were not independent. Until April 2011, a total of eight relapses (6.1%) were recorded. The disease free survival (DFS) rate at five-years for all patients was of 88.3% ± 4.2% and the mean DFS was of 76.1 months [71.2 to 80.9]. DFS was significantly shorter in patients at the higher tertile ΔN/TA expression ratio compared with patients at the lower tertile (72.1 ± 11.2% vs 97.1 ± 2.8%, respectively; p <0.001; Figure 1B) and was the only variable found to be significant in the univariate analysis. To test the functional significance of the association of PML-RARa with high ΔNp73 gene expression, primary murine bone marrow cells from hCG-PML-RARa transgenic mice were transfected with MSCV-based retroviral vector carrying the ΔNp73 cDNA upstream of IRES-GFP cassette (PML-pMIG-ΔN). Expression of ΔNp73 in PML-RARa+ cells increased cell proliferation rate by 2.5-fold compared to PML-RARa+ transfected with the empty vector (p =0.03). This increase resulted from accumulation of cells at the G2/M phase (5.79 ± 0.08% for PML-pMIG vs 9.8 ± 0.35% for PML-pMIG-ΔN, p <0.001), as well as at S phase of the cell cycle (27.74 ± 0.89% for PML-pMIG vs. 36.78 ± 0.81% for PML-pMIG-ΔN, p =0.001). In addition, transfection of ΔNp73 resulted in resistance to cytarabine-induced apoptosis. After 24h of culture with 50μg/ml of cytarabine (ED-50%), the fractional effect for the drug (% Anexin V-positive in (treated – untreated) cells /100 - % Anexin V-positive in non-treated cells) was 32.1% for PML-pMIG-ΔN and 54.8% for PML-pMIG (p <0.001). In conclusion, ΔN/TA expression ratio was associated with shorter OS and DFS in APL, which may reflect increased cell proliferation and apoptosis resistance due to ΔNp73 activity.
Figure 1.
Disclosures:
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
Author notes
*
Asterisk with author names denotes non-ASH members.
© 2011 by The American Society of Hematology
2011