Abstract PR04: mTOR mutations in cancer

Abstract

Abstract Genes in the PI3K-Akt-mTOR signaling axis are frequently mutated in cancer but relatively few mutations have been observed in the MTOR gene itself. By mining publicly available cancer sequencing data, we recently reported that MTOR is commonly mutated in cancer, with more than 35 previously uncharacterized pathway activating mutations. Most were recurrent, localized within distinct clusters in the C-terminal half of the protein, and induced activation of mTOR Complex 1 (mTORC1), mTORC2, or both. The mutations did not prevent mTOR inhibition by rapamycin or ATP-competitive inhibitors, although several conferred resistance to nutrient withdrawal. Importantly, cancer cell lines with activating mTOR mutations are especially sensitive to rapamycin both in vitro and in vivo, indicating pathway dependency. In a companion report, we also demonstrated that the tumor of a bladder cancer patient with exquisite sensitivity to a rapamycin analog (rapalog) harbored two activating mTOR mutations, likely driving mTOR pathway dependency. However, after 14 months of rapalog response, the patient displayed progressive disease, indicating that the tumor had acquired resistance to the targeted therapy. Here, we demonstrate two mechanisms by which cancers may display resistance to rapamycin through mTOR mutation. In the first example, we acquired a tumor biopsy from an anaplastic thyroid carcinoma patient before they were enrolled on a rapalog clinical trial. The patient displayed a durable 18 month response to the rapalog before progressing, at which time we acquired a second biopsy. Whole exome sequencing of the biopsies revealed the presence of a loss-of-function mutation in the mTORC1 negative regulator Tuberin (TSC2), likely driving pathway addiction and thus sensitivity to the rapalog. Sequencing of the post-resistance biopsy revealed that the tumor had acquired a mutation in the rapamycin-binding domain of mTOR. Cells made to exogenously express this mutation displayed potent resistance to rapamycin-mediated inhibition of cell proliferation and phosphorylation of the mTORC1 substrate S6K1. Importantly, the cells remained sensitive to an ATP-competitive mTOR inhibitor because the two classes of mTOR inhibitors rely upon completely different mechanisms of action. Thus, we show that as patients acquire resistance to rapalogs, they may still be effectively treated with ATP-competitive mTOR inhibitors. The rapalog-resistant mutation described above is an example of an acquired mutation, although due to the genetic instability of cancer cells, a tumor may harbor pre-existing rapamycin-resistant mutations. To determine the prevalence of such mutations, we again turned to publicly available cancer sequencing data, identifying approximately 30 mutations in the rapamycin-binding domain of mTOR. The majority these mutations did not alter mTOR pathway activation and remained rapamycin sensitive. Importantly, one breast cancer sample harbored a rapamycin-resistant mTOR mutation, although it remained sensitive to an ATP-competitive mTOR inhibitor. This rare event is highly relevant given that rapalogs are currently approved for the treatment of breast cancer. Had this patient been treated with a rapalog, they would have only experienced the side-effects and none of the benefits of treatment. Collectively, we demonstrate that mTOR mutations are common in cancer, many of which confer pathway activation and addiction. Additionally, both acquired and pre-existing mTOR mutations can confer resistance towards rapamycin treatment, although this resistance can be overcome with ATP-competitive inhibitors. This work has important ramifications for the treatment of cancer patients with mTOR inhibitors, by identifying both those patients likely to respond or display resistance to such treatments. This abstract is also being presented as Poster A44. Citation Format: Brian Grabiner, Nikhil Wagle, Eliezer Van Allen, Levi Garraway, Jochen Lorch, David Sabatini. mTOR mutations in cancer. [abstract]. In: Proceedings of the AACR Special Conference: Targeting the PI3K-mTOR Network in Cancer; Sep 14-17, 2014; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(7 Suppl):Abstract nr PR04.