-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

1553 PI3 Kinase p110 Delta Is Required for Leukemic Cell Survival and Self-Renewal in t(8;21) Acute Myeloid Leukemia

Oncogenes and Tumor Suppressors
Program: Oral and Poster Abstracts
Session: 603. Oncogenes and Tumor Suppressors: Poster I
Saturday, December 3, 2016, 5:30 PM-7:30 PM
Hall GH (San Diego Convention Center)

Leanne Ostrodka, MD1,2, Michael Samuel, MD3*, Taneisha Sinclair1*, Redwan Bhuiyan1* and Kira Gritsman, MD, PhD1

1Department of Medicine-Oncology, Department of Cell Biology, Albert Einstein College of Medicine, Bronx, NY
2Children's Hospital at Montefiore Hospital, Bronx, NY
3Division of Hematology/Oncology, Weill Cornell Medicine, New York Presbyterian Hospital, New York, NY

Acute myeloid leukemia (AML) with t(8;21) comprises about 10% of adult AML cases, and is the most frequent genetic alteration in pediatric AML. In t(8;21) AML, the AML1-ETO transcription factor fusion dysregulates transcription in hematopoietic cells, interfering with apoptosis and differentiation, and leading to enhanced self-renewal. The phosphoinositide-3-kinase (PI3K) signaling pathway is frequently overactivated in AML, representing a promising target in AML. However, PI3K also plays important roles in normal hematopoiesis. Hematopoietic cells express four catalytic subunits of Class I PI3K (p110α, p110β, p110γ, and p110δ). Previous studies have demonstrated that these isoforms have redundant roles in normal hematopoietic stem cells (HSCs), but play specific roles in some genetic subtypes of AML. Therefore, inhibiting individual PI3K isoforms in AML may allow for therapeutic targeting of this pathway with lower toxicity. In fact, isoform-selective PI3K inhibitors are already in clinical development, and the p110δ-selective inhibitor idelalisib is approved by the FDA for lymphoid malignancies.

To determine whether t(8;21) AML cells are dependent upon specific PI3K isoforms, we performed proliferation and apoptosis assays in Kasumi-1 cells, a t(8;21) human AML cell line, with PI3K inhibitors: pictilisib (GDC0941, a pan-PI3K inhibitor), buparlisib (BKM120, a pan-PI3K inhibitor), alpelisib (BYL719, a p110α-selective inhibitor), idelalisib (GS1101,a p110δ-selective isoform inhibitor) and KIN193 or TGX221 (p110β-selective inhibitors). We observed a decrease in proliferation and in Akt/mTOR signaling with the pan-PI3K inhibitors and alpelisib. The pan-PI3K inhibitors, alpelisib, and idelalisib also induced apoptosis in Kasumi-1 cells. In addition, we observed an increase in myeloid differentiation of Kasumi-1 cells with buparlisib or idelalisib. The splice isoform AML1-ETO9a is found in patients with t(8;21) AML, induces AML in mice when introduced into hematopoietic cells by retroviral transduction, and confers serial replating in methylcellulose (Yan et al, Nature Medicine 2006). To investigate the role of p110δ in leukemogenesis genetically, we harvested bone marrow cells from mice with a germline deletion of pik3cd, which encodes p110δ, and transduced them with a retrovirus expressing AML1-ETO9a. To study the ability of AML1-ETO9a leukemic cells to self-renew after p110δ deletion, we performed serial replating assays in methylcellulose.

We found that, while AML1-ETO9a confers self-renewal activity to wild-type (WT) cells past round 8 of serial replating, p110δ-/- cells gradually lost self-renewal activity, and were unable to replate past round 4. Consistent with our genetic data, we found that the p110δ-selective inhibitor idelalisib or the pan-PI3K/mTOR inhibitor dactolisib (BEZ-235) can significantly reduce serial replating by AML1-ETO9a cells. To determine whether p110δ is also important for the initiation and progression of AML1-ETO AML in vivo, we transduced ckit-enriched bone marrow cells from WT or p110δ-/- mice with AML1-ETO9a-GFP retrovirus, and injected 200,000 GFP-sorted cells into lethally irradiated C57 Bl/6 recipient mice. During an interim analysis of this ongoing experiment, we have observed lower white cell counts in recipients of p110δ-/- cells compared with controls, suggesting a decrease in disease burden of AML1-ETO9a AML in the absence of p110δ.

It has been reported that the thrombopoietin (TPO)/MPL/JAK2 signaling pathway is activated in t(8;21) AML, and is important in promoting proliferation and self-renewal of AML1-ETO leukemic cells (Pulikkan et al, Blood, 2012; Lo et al, Blood 2012). We hypothesized that p110δ may be required downstream of the TPO/MPL/JAK2 signaling pathway for leukemic self-renewal in t(8;21) AML. To determine whether TPO signaling depends upon p110δ in AML1-ETO9a cells, we treated WT and p110δ-/- cells transduced with AML1-ETO9a with TPO, and performed replating assays. We found that, although TPO treatment augmented the in vitro self-renewal of WT AML1-ETO9a cells to round 9 with increased colony numbers, p110δ-/- AML1-ETO9a cells treated with TPO failed to replate past round 5. These data suggest that p110δ is important for leukemic survival and self-renewal in AML1-ETO AML, in part by mediating signaling downstream of the TPO/MPL/JAK2 signaling axis.

Disclosures: No relevant conflicts of interest to declare.

*signifies non-member of ASH