Acute T-Cell Leukemias Remain Dependent On Notch Signaling Despite PTEN and INK4A/ARF Loss.

Abstract 8

T-cell acute lymphoblastic leukemia (T-ALL) is a malignancy of immature T cell progenitors in which activating mutations of NOTCH1 occur in over 50% of cases and loss-of-function mutations in FBW7, which degrades activated NOTCH1, occur in 8-16% of cases. We and others have characterized c-MYC as a critical downstream transcriptional target of Notch signaling in this context, and loss of FBW7 also contributes here as intact FBW7 marks c-MYC protein for proteosomal degradation. Inhibition of Notch signaling in T-ALL cells by various means including small molecule gamma-secretase inhibitors (GSI) causes cell cycle/growth arrest and in some cases apoptosis, providing rationale for NOTCH1 as a therapeutic target.

The tumor suppressor PTEN is also mutated or lost in up to 20% of T-ALL cases. PTEN activity converts PIP3 to PIP2, and thus loss of PTEN potentiates PI3K signaling. It was observed among human T-ALL cell lines that PTEN loss correlated with resistance to Notch inhibition with GSI, leading to the idea that constitutive PI3K/AKT signaling relieves T-ALL cells of their “addiction” to oncogenic Notch signaling. Since GSI and other NOTCH1 signaling inhibitors are presently under investigation as targeted therapeutic agents for T-ALL and other cancers including breast, brain, colorectal, pancreatic, and melanoma, this idea that PTEN loss could confer resistance to NOTCH1 inhibition raises concern that patients with PTEN-negative (or presumably PIK3CA-mutated) diseases may fail Notch inhibitor therapy.

As the studies linking GSI-resistance/Notch-independence to PTEN loss were limited to established cell lines, we sought to address this issue using a well established and genetically defined mouse retroviral transduction/bone marrow transplantation model. Briefly, we generated primary murine T-cell acute leukemias using mutated NOTCH1 retroviruses on both wild-type and PTEN-null backgrounds and tested these tumors for response to GSI. Unexpectedly, we observed these primary murine leukemias to undergo growth arrest and cell size reduction with GSI treatment on both wild-type and PTEN-null backgrounds. We even observed wild-type background tumors which had lost PTEN spontaneously after either serial transplantation or extended culture in vitro to remain sensitive to GSI. Given that most cases of human T-ALL have also lost expression of p16^INK4A/p14^ARF tumor suppressors either by deletion or silencing, and since p16^INK4A/p14^ARF are critical regulators at the G1/S cell cycle checkpoint, we also generated primary murine tumors on a Pten, Ink4a/Arf double-null background. Even when both PTEN and Ink4a/Arf were deleted, we still observed the murine tumors to remain sensitive to GSI.

To confirm these findings were not unique to murine tumors, we examined a panel of 13 primary human T-ALL cases including 7 PTEN-positive cases (6 of which were NOTCH1 mutated) and 6 PTEN-negative cases (4 of which were NOTCH1 mutated). We found 11 of these 13 cases to be GSI sensitive (6 PTEN-positive, 5 PTEN-negative) as illustrated by decreased proliferation and reduction in cell size. We found only 2 cases to be GSI resistant, including 1 PTEN-positive and 1 PTEN-negative. Our panel contained only one case with FBW7 mutation; this case was PTEN-positive and GSI-sensitive. Thus, in contrast to previous studies with established cell lines, we find in this study using primary human T-ALL samples and genetically defined primary murine leukemias that PTEN loss indeed does not confer resistance to Notch inhibition. Of note, other groups have postulated a role for FBW7 mutation in conferring resistance to Notch inhibition; however, we encountered only 2 GSI-resistant cases in our panel of primary human samples, neither of which was FBW7 mutated.

In generating the murine Notch leukemias, we also had the opportunity to compare disease penetrance/latency and clonality among the various genetic backgrounds. We noted that PTEN loss was associated with accelerated disease onset and multiclonal tumors as compared to wild-type, whereas INK4A/ARF loss alone had no such effect. These findings suggest NOTCH1 activation and PTEN loss may collaborate in leukemia induction; however, technical caveats related to potentially enhanced retroviral transduction efficiency of PTEN-null bone marrow progenitors limit this conclusion, and thus more definitive experiments to address this issue are needed.