Abstract
Abstract 3575
ALL cells are highly dependent on bone marrow stromal support for in vitro proliferation and survival. The major regulators of patient-derived ALL cell growth and survival convey their proliferative and survival signals through the phosphoinositide 3-kinase (PI-3K) pathway. It has been recently demonstrated that signalling through PI-3K and AKT is the most important pathway for the proliferative responses of ALL cells to CXCL12, the chemokine predominantly responsible for stromal dependent growth of ALL cells. In addition, inhibition of the mTOR signalling molecule downstream of PI-3K with RAD001 has been shown to inhibit proliferation and induce cell death resulting in extended survival in a NOD/SCID xenograft model of human ALL. This work has supported the initiation of clinical trials of RAD001 in adults with relapsed ALL. Although PI-3K and mTOR have similar and overlapping functions, mTOR can be activated independently of PI-3K, and proliferation and survival can be stimulated by PI-3K in an mTOR independent manner. Therefore combining PI-3K and mTOR inhibition is likely to be advantageous over inhibition of either kinase alone, suggesting disruption of PI-3K/AKT/mTOR signalling will provide a new approach for the treatment of ALL. We investigated the dual kinase inhibitors BEZ235 and BGT226.
We have previously demonstrated that PI-3K and mTOR inhibition with the dual kinase inhibitor BEZ235 and BGT226 significantly inhibits the proliferation of ALL cell lines and patient derived stromal dependent lines in vitro, a 3 log greater potency in comparison to the mTOR inhibitor RAD001. The ability to induce cell death differed between the dual mTOR and PI-3K inhibitors. BGT226 potently induced cell death at 1.6μM, while more than 30μM of BEZ235 was required to kill ALL cells within 24 hours, compared to RAD001 which has an IC50 of 16μM. Interestingly, the IC50 for BEZ235 and BGT226 both declined when cell death was measured at 48 and 72 hours, a feature not observed with RAD001. BEZ235 induced extensive caspase 3 cleavage while BGT226 had only a minor effect on the activation of caspase 3. Preliminary data using 3H-thymidine assays, suggests that the addition of dual PI-3K/mTOR inhibitors may not be beneficial with antagonistic interactions being observed with some agents including etoposide and doxorubicin. Further studies will be required to define interactions between kinase inhibitors and chemotherapeutic agents.
We also demonstrate the activity of these dual kinase inhibitors in a NOD/SCID xenograft model of human ALL, with treatment commencing once 1% ALL was detected in the blood. The dual kinase inhibitors significantly extended survival in all 4 xenografts compared to control treated animals. However they only produced superior results to RAD001 in one of these xenografts (2032) and were clearly inferior in another (1345). In xenograft 2023 exposure to RAD001 resulted in increased AKT phosphorylation on Ser473, suggesting RAD001 induced activation of mTORC2 through the mTORC1 feedback loop, potentially diminishing responses to mTOR inhibitors. This particularly highlights the benefit of the dual PI-3K/mTOR inhibitors' ability to block mTORC2 signalling through PI-3K inhibition and could possibly reflect the results seen with xenograft 2032 in vivo. The reason for reduced efficacy in xenograft 1345 is also not apparent at this stage but may possibly reflect greater bioavailability of RAD001 as compared to the dual kinase inhibitors.
Interestingly, Akt signalling differed between xenografts, suggesting that dependence on PI-3K/Akt signalling upstream and potentially independent of mTOR may impact on the efficacy of the dual PI-3K/mTOR inhibitors however, this does not appear to provide a complete explanation of the different responses observed. Dual kinase inhibitors may offer improved therapeutic outcomes for a subset of ALL patients. More importantly some patients actually respond better to single mTOR inhibitors than dual inhibitors. We need a greater understanding of how these inhibitors work, so that patients that will benefit from dual kinase inhibitors and those where dual inhibitors will be less effective than single mTOR inhibitors can be identified prior to treatment.
Disclosures:
No relevant conflicts of interest to declare.
Dual mTORC1/mTORC2 inhibition as a Host-Directed Therapeutic Target in Pathologically Distinct Mouse Models of Tuberculosis
