PP2A inhibition is a common event in AML. Restoration of PP2A activity induces both arrest of cell growth and caspase-dependent apoptosis, suggesting that PP2A inactivation has an important role in AML and can be a promising therapeutic target in hematological malignancies [17].
CIP2A is an endogenous PP2A inhibitor that interacts directly with the oncogenic transcription factor c-Myc, inhibiting PP2A-mediated dephosphorylation of c-Myc and preventing its proteolytic degradation, resulting in c-Myc stabilization, promotion of cell growth, and tumor formation [18]. CIP2A is expressed in few normal tissues, but it is overexpressed in many human cancers, and is associated with clinically aggressive cancer behavior. CIP2A is known to be overexpressed in head and neck squamous-cell carcinoma, breast cancer, colon cancer, and gastric cancer [14].
To date, CIP2A expression levels in patients with AML have been investigated in two studies. In a study involving Asian patients with AML, conventional RT-PCR identified higher expression levels (77.14%) of CIP2A in newly diagnosed AML patients as well as in patients with relapsed AML (70.86%) compared to 6.25% in AML patients in CR and 2.86% in healthy controls. This finding points to a relation between CIP2A overexpression and disease activity in AML [18].
In a cohort of 203 European patients diagnosed with a normal karyotype AML, qRT-PCR was done to find out the prevalence of CIP2A in patients with NK-AML. In that study, the median expression of CIP2A was 0.005 (range 0.0027–0.007) in the control group and 0.004 (range 0.00012–0.59) in the AML group. The authors chose a cut-off value of 0.0072 to define high-CIP2A expression which corresponds to the mean value in the control [9].
In this study, qRT-PCR was used to detect the expression of CIP2A in de novo AML patients. The median expression in the control group which we used as a cut-off was much higher than the value reported by Barragan et al. [9].
Based on this cut-off value, we found that 35% of the patients had a high-CIP2A level; a value higher than that reported by Barragan et al. [9], and much less than that reported by Wang et al. [18] who used conventional RT-PCR. This may be explained by the fact that the relative quantitation method and the threshold for determining high and low expression in our study were different from those used by the other two studies.
In the current study, no significant relations was found between CIP2A expression and the clinical and laboratory data; a finding consistent with previous findings [9]. Similarly, we found that the level of expression of CIP2A did not have a significant effect on the response to induction chemotherapy or relapse rate (P = 0.57 and P = 0.27, respectively), which is in line with previous findings [9].
Noteworthy, we found a trend towards improved survival in patients with low-CIP2A expression with a 3-year survival probability of 36% and 25% for patients with low- and high-CIP2A expression, respectively. Barragan et al. [9] reported a close percentage (27%) in patients with high CIP2A, but a much higher percentage (50%) in patients with low expression.
At 3 years, the current study recognized relapse in 54.5% and 57.9% of patients with low- and high-CIP2A expression, respectively. These values are higher than those reported by Barragan et al. [9] (30% and 47% for low- and high-CIP2A expression)
The lower PFS and OS reported in this study compared to those reported by Barragan et al. [9] may be attributed to the fact that the European trial was larger and only patients with NK-AML were allowed to participate.