Abstract 1972

Poster Board I-995

The only successful immunologic approach in acute myeloid leukemia (AML) is allogeneic transplantation. However, this treatment modality is restricted by age and donor availability. It is also associated with significant toxicity. Therefore, autologous approaches are being sought. Among those, harnessing the patients' immune system seems most appealing. We have previously demonstrated successful generation of leukemia derived-dendritic cells (DC) in acute lymphoblastic leukemia and their ability to induce autologous cytotoxic T cell responses against the unmodified blasts. Similar approaches in AML have proven to be less successful. We have previously shown that constitutive signal transducer and activator of transcription (STAT) 3 activity is present in blasts of approximately half of AML patients and correlates with poor prognosis. STAT3 regulates a variety of cellular events and has been shown to play a role in immunosuppression by inhibiting DC differentiation, resulting in induction of T cell tolerance rather than T cell activation. Taken together, we hypothesize that DC derived from AML blasts with constitutive STAT3 activity will have impaired ability to fully differentiate and efficiently stimulate T cells, resulting in an ineffective vaccine. Our study aims to generate AML-derived DC with reduced STAT3 to create a more efficient vaccine and gain a better understanding of how STAT3 regulates DC differentiation and T cell activation. To test this question, primary blasts from four newly diagnosed AML patients were differentiated to DC with granulocyte-macrophage colony stimulating factor, interleukin 4 and tumor necrosis factor á and transduced with commercially available non-silence control or STAT3 shRNAmirs. Following eight days of culture, non-silenced and STAT3 knock-down cells were analyzed by flow cytometry for changes in immunophenotype. Cells were further analyzed by western blot for STAT5, extracellular signal-regulated kinase (ERK) 1/2 and protein kinase C-bII, known to regulate differentiation through interactions with STAT3. Actin was used as a loading control. To verify the leukemic origin of the DC, a sample with trisomy 8 was analyzed by fluorescent in situ hybridization (FISH). ImageStream cytometry was used to monitor endocytosis of dextran molecules by immature DC. Finally, control and STAT3 knock-down cells were compared for the ability to stimulate allogeneic T cells in a mixed lymphocyte reaction (MLR). The STAT3 knock-down cells compared to controls had an increased percentage of cells expressing the DC markers CD40, CD80, CD86, CD11c and HLA-DR, in two of the four samples, indicating a mature phenotype capable of more efficient T cell stimulation. The two remaining samples had an unchanged phenotype, but retained expression of the molecules necessary for T cell stimulation. Western blotting revealed reduced levels of total STAT3 protein (23-65%) in all knock-down samples compared to controls, while total STAT5, phosphorylated ERK1/2 and actin levels remained unchanged. Protein kinase C-βII was elevated in two of the STAT3 knock-down samples, lysates from the remaining samples were no longer available for analysis. FISH analysis of trisomy 8 confirmed the leukemic origin of the DC. The STAT3 knock-down cells demonstrated an increased capacity to take up dextran molecules in 3 out of 4 samples. Finally, MLR revealed increased T cell stimulation by the STAT3 knock-down cells in all 4 samples. The increase was statistically significant in two (p=0.019, p=0.013) and showed a trend in the other two (p=0.057 and p=0.152). These data demonstrate that AML-derived DC have improved function after reducing STAT3 levels. We propose to target STAT3 in order to improve the function of AML-derived DC vaccines.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution