5-Azacytidine Potentiates the Anti-Proliferative Activity of Anti-CD33 Monoclonal Antibodies (mAb) in Acute Myeloid Leukemia in Part through Induction of Syk and SHP-1 Expression.

Abstract

We have previously demonstrated that anti-CD33 monoclonal antibodies have anti-proliferative and pro-apoptotic effects on AML cells and that this ability is associated with the expression and functional activity of the protein tyrosine kinase Syk and protein tyrosine phosphatase SHP-1. However, about 30% of primary AML samples are Syk-negative and 15% are SHP-1-negative.Since absence of protein expression and unresponsiveness of tumor cells could be caused by silencing by hypermethylation, we tested whether the demethylating agent 5-azacytidine (5-aza) could restore Syk and SHP-1 and, therefore, enhance CD33 ligation mediated inhibition of leukemia cell growth. 40 primary AML samples were tested for their response to submaximal anti-proliferative concentrations of 5-aza (100nM for 48 h) and, subsequently, were cultured in the presence of anti-CD33 or control anti-CD13 mAb. In the majority of cases, 5-aza alone induced various degree of inhibition of proliferation (ranging from 0 to 90%). Based on the level of inhibition, samples were divided into 3 groups: low (<25%), medium (25%-50%) and high (>50%). 52.5% of samples were Low Responders, 22.5% -medium R and 25% -high R. The average level of inhibition was determined for each group: (LR-6.4.+1.8, MR-34.5+1.9, HR-69.9+6.9 Incremental doses (0.01–1 mg/ml) of anti-CD33 mAb (alone) inhibited proliferation of AML samples variably (15–35%). No differences between groups were determined. However, the addition of anti-CD33 mAb to the 5-aza pre-treated samples revealed differences between groups. In the LR group response more than doubled (from 20 to 58%), in the MR group the increase was about 30%. However, in the HR group, no difference was observed. The level of expression of protein kinase Syk and protein phosphatase SHP-1 was determined for each sample by Western blot. Of 40 samples 25 (62.5%) were Syk-positive and 15 (37.5%) were Syk-negative. 35 samples (87.5%) were SHP-1-positive and 5 samples (12.5%) were SHP1-negative. The majority of Syk-negative samples (11 of 15) were in LR group, while the remaining 4 samples were distributed between MR and HR groups and demonstrated significant levels of inhibition. This difference was statistically significant (p=0.003) and suggested a correlation between the Syk+/Syk− samples and their response to the inhibitory activity of 5-Aza. Similarly, analysis of SHP-1 expression and the response of samples to 5-aza inhibition revealed an even stronger correlation (p= 0.023) since all 5 SHP-1-negative samples were in LR group. Meanwhile, in 36.3% (4 of 11 Syk-negative cases), and 40% (2 of 5 SHP-1-negative samples) 5-aza treatment resulted in restoration of protein expression. Moreover, in Syk-positive cases combined treatment induced a 20% increase in inhibition from 36% (5-aza alone) to 55% (5-aza+ a -CD33). In Syk-negative samples, this increase was significant (about 40%). In SHP-1-positive samples combined treatment also induced about a 20% increase in inhibition. However, in SHP-1-negative samples the increase was dramatic (about 70%) In all cases, control a -CD13 mAb did not result in any increases in 5-aza-induced inhibition of proliferation. These data support the notion of a clinical trial of combination 5-aza and anti-CD33 mAb therapy for patients with AML. In addition, biomarkers for response may help select patients in whom the combination therapy will be most efficacious. A phase I/II clinical trial is under development at this time.