High-Level IGF1R Expression Is Required for Leukemia Stem Cell Activity In T-ALL and Is Supported by Notch Signaling.

Abstract 3649

T-cell acute lymphoblastic leukemia (T-ALL) is a lethal cancer of immature T cells that often shows aberrant activation of NOTCH1 and PI3K/Akt pathways. Although PI3K/Akt signaling is potentiated by PTEN mutations which occur in up to 20% of cases, upstream signals which initiate PI3K/Akt activation remain unclear. IGF1R is a receptor tyrosine kinase upstream of PI3K/Akt and Ras/Raf/MAPK signaling pathways with known roles in cancer cell growth and is expressed broadly in primary human T-ALL cells and established cell lines. We thus considered that IGF1R may represent a biologically relevant upstream activator of the PI3K/Akt pathway in T-ALL. Consistent with this idea, we demonstrate that treatment of human and mouse T-ALL cells with small molecule IGF1R kinase inhibitors and IGF1R blocking antibodies resulted in growth arrest of bulk cells. As well, genetic deletion of IGF1R by inducible Cre expression in mouse T-cell leukemias also leads to growth arrest and apoptosis of bulk cells.

IGF1R and PI3K/Akt/mTOR signaling have also been described to promote self-renewal of normal and cancer stem cells, and thus we addressed whether IGF1R may play a role in leukemia stem cell (LSC) activity in T-ALL. Since outgrowth of bulk leukemia cells is required to detect LSC activity in vivo, we were not able to address IGF1R function in LSCs by genetic deletion; however, we elected to take advantage of a well-characterized hypomorphic allele of IGF1R carrying an integrated neo cassette within the second intron. IGF1R^neo/neo mice express full-length IGF1R protein, but at reduced levels (30-60% of wild-type). Despite this reduced expression level, we were able to generate lethal T-cell leukemias with 100% penetrance in primary recipients of IGF1R^neo/neo bone marrow transduced with activated NOTCH1 retrovirus, similar to what is observed with IGF1R^+/+ bone marrow. Upon IV or intra-medullary injection into secondary congenic recipients, however, IGF1R^neo/neo leukemias exhibited a striking transplantation defect. Whereas IGF1R^+/+ leukemias are consistently serially transplantable, 5 of 16 primary IGF1R^neo/neo leukemias failed to reconstitute disease in any secondary recipients, 5 produced disease in only a subset of secondary recipients, and 6 produced disease in all secondary recipients. Furthermore, treatment of IGF1R^+/+ leukemia cells in vitro with IGF1R inhibitor prior to injection into secondary recipients prevented their transplantability, whereas treatment with PI3K and ERK inhibitors did not.

Finally, given that inhibition of Notch signaling in human T-ALL cells compromises their ability to recapitulate disease in NOD/Scid recipients, we wondered whether this effect could be mediated by downregulation of IGF1R. In fact, we observed inhibition of Notch signaling, either by small molecule gamma-secretase inhibitor or transduction with dominant negative Mastermind, to result in a 2–3 fold decrease in IGF1R protein expression, corresponding to a 5 to 20-fold decrease in response to stimulation by IGF-1 ligand. Thus, these studies demonstrate high-level IGF1R signaling is required for LSC activity in T-ALL, and that Notch signaling may promote LSC function by supporting IGF1R expression. IGF1R inhibitors may therefore prove clinically useful in preventing disease relapse by compromising LSC self-renewal.