Abstract 4615: The hyperinsulinemia caused by PI3K inhibitors attenuates their antineoplastic efficacy, but can be minimized by co-administration of metformin.

Abstract

Abstract As phosphatidylinositol 3-kinase (PI3K) signaling is often inappropriately activated in cancer, targeting this pathway is the goal of many drug development programs. PI3K activation provides an important proliferative and antiapoptotic signal for many normal and transformed cell types, but in tissues such as liver, muscle and fat that are involved in regulation of blood glucose, PI3K activation is a consequence of insulin receptor activation, and leads to cellular glucose uptake and reduction in circulating glucose concentration. Therefore, effective pharmacologic PI3K blockade perturbs glucose homeostasis, leading to insulin resistance associated with hyperglycemia and/or compensatory hyperinsulinemia. In some patients, PI3K blockade leads to hyperinsulinemia with normoglycemia, suggesting that elevation of insulin level overcomes pharmacologic PI3K inhibition in insulin-responsive tissues to an extent sufficient to normalize blood glucose. To investigate the possibility that increased insulin stimulation can also attenuate the antiproliferative effect of PI3K inhibitors in cancer cells, we first developed an in vivo model which demonstrated that conventional antineoplastic doses of GDC-0941 (a potent and selective inhibitor of class 1 PI3K) leads to insulin resistance and compensatory hyperinsulinemia (20 min post 2 g/kg glucose IP, serum glucose rises from 5.1 ± 0.4 to 15 ± 1.8 mM in control conditions, but from 12.4 ± 1.2 to 29.2 ± 1.8 mM with GDC-0941 treatment). Next, we showed that the activity of GDC-0941 on a variety neoplastic cells in vitro, in terms of suppression of both proliferation and signaling downstream of PI3K, is attenuated when insulin concentration is raised (75 nM GDC-0941 reduces proliferation by 50%, 36%, 29% when insulin is 2, 6, or 12 nM, respectively). Metformin, which is often administered clinically to patients on PI3K inhibitors to reduce hyperglycemia, abolished GDC-0941-induced hyperinsulinemia. Thus, PI3K inhibitor-induced hyperinsulinemia (a) represents pharmacodynamic evidence of target inhibition (at least in tissues concerned with glycemia regulation), (b) attenuates antineoplastic efficacy, particularly when inhibitor concentrations are suboptimal and (c) may be minimized by co-administration of metformin. These findings are relevant to ongoing clinical trials of PI3K inhibitors, and indicate that the magnitude of the input signal to PI3K influences the degree of blockade achieved by PI3K inhibitors. Citation Format: Marie-José Blouin, Elena Birman, Yunhua Zhao, Mahvash Zakikhani, Michael N. Pollak. The hyperinsulinemia caused by PI3K inhibitors attenuates their antineoplastic efficacy, but can be minimized by co-administration of metformin. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4615. doi:10.1158/1538-7445.AM2013-4615