Hyperactivated mTOR Signaling Is a Characteristic Feature of High Risk Acute Lymphoblastic Leukemia and Can Be Effectively Targeted in a Preclinical NOD/SCID/huALL Xenograft Mouse Model

Abstract 2572

Treatment of acute lymphoblastic leukemia (ALL), the most frequent malignant disease in children and adolescents, is increasingly successful. Stratification based on prognostic factors assigning the patients to treatment protocols of different intensity and intensification of chemotherapy regimens have improved patient outcome with present cure rates above 80%. However, about 20% of the patients suffer from relapse which is associated with inferior outcome, especially if occurring early. Importantly, despite the efforts achieved by stratification strategies including detection of minimal residual disease, the majority of relapse patients are initially stratified into non-high risk groups and not identified upfront by current markers. This clearly emphasizes the need for additional prognostic markers which should ideally reflect features of leukemia biology pointing to novel therapeutic targets.

In a recent study we described a strong association of the engraftment phenotype observed in a series of transplanted primary patient B cell precursor (BCP) ALL samples in a NOD/SCID/huALL mouse model and early patient relapse. This rapid engraftment phenotype (time to leukemia, TTL^short) is characterized by a gene expression profile pointing to pathways involved in cell growth and proliferation.

Since mammalian target of rapamycin (mTOR) is a key survival pathway and was identified in the gene profile we now investigated this pathway on a functional level. For this purpose, ALL xenograft samples previously characterized with respect to their engraftment phenotype (TTL^short/early relapse, n=3 or TTL^long/no or late relapse, n=4) were re-transplanted onto NOD/SCID mice. At disease onset ALL cells were harvested from leukemia bearing recipients and mTOR pathway activation was analyzed by flow cytometry assessing phosphorylation of ribosomal protein S6 (P-S6), a molecule downstream of mTOR. Increased P-S6 levels were detected in TTL^short compared to TTL^long xenografts (T-test, P=.009) indicating constitutive mTOR hyperactivation in TTL^short/early relapse leukemia. Furthermore, the effects of the mTOR inhibitor rapamycine and the PI3K/mTOR dual inhibitor BEZ235 were investigated. Xenograft leukemia cells were incubated with rapamycine, BEZ235 or solvent and P-S6 was analyzed. Interestingly, both inhibitors led to a significant decrease of pathway activation in TTL^short but not in TTL^long xenografts indicating that the hyperactivated mTOR pathway of this high risk ALL subtype can be successfully targeted.

Moreover, the effect of mTOR inhibition on individual patient derived ALL was also analyzed in vivo. Since in a clinical setting high risk disease is unlikely to be treated employing a novel strategy on its own, we investigated the effect of mTOR inhibition in addition to multiagent drug treatment resembling ALL induction therapy. NOD/SCID mice were transplanted with TTL^short (n=2) or TTL^long (n=1) leukemia. Upon appearance of leukemia in the peripheral blood treatment with either rapamycine, multiagent chemotherapy, combination of chemotherapy and rapamycin or solvent was initiated (8 mice per group) and the time until reoccurrence of leukemia defined by presence of 25% or more human ALL cells in the recipients blood was assessed. In both TTL^short leukemias a significant delay of ALL onset was observed upon combination treatment compared to chemotherapy alone, however the TTL^long leukemia showed onset at similar time points irrespective of treatment modalities. Importantly, also a significant reduction in leukemia load (as measured by decreased spleen weights despite similar high percentage leukemia infiltration) was observed in TTL^short but not TTL^long leukemias upon in vivo combination therapy.

Taken together, we demonstrated that TTL^short/early relapse leukemia is associated with an hyperactivated mTOR pathway and can effectively be targeted by mTOR inhibitors ex vivo. Most importantly, a significant delay of leukemia onset upon in vivo mTOR inhibition combined with conventional chemotherapy was achieved in a preclinical NOD/SCID/huALL xenograft model. Thus, mTOR inhibition is a successful novel therapeutic strategy for the treatment of high risk ALL.