Abstract 1814: NF1 as an estrogen receptor-α co-repressor in breast cancer

Eric C. Chang,zeyi Zheng,Meenakshi Anurag,Jin Gao,Burcu Cakar,Xinhui Du,Jing Li,Philip Lavere,Jonathan T. Lei,Purba Singh,Sinem Seker,Wei Song,Jianheng Peng,Tiffany Nguyen,Doug Chan,Xi Chen,Kimberly C. Banks,Richarad B. Lanman,Maryam Shafaee,Susan Hilsenbeck,Charles Foulds,Matthew J. Ellis

DOI: https://doi.org/10.1158/1538-7445.am2018-1814

IF: 5.051

2018-07-01

Endocrinology

Abstract:Abstract NF1 has been best known as a GAP (GTPase Activating Protein) that inactivates Ras. However, we are now finding evidence that it also functions as an ER co-repressor, whose loss leads to endocrine therapy resistance. Sequencing tumor DNA from >600 ER+ breast cancers treated by tamoxifen adjuvant monotherapy, we found that frameshift (FS) and nonsense (NS) NF1 mutations, which can create an NF1-null state, strongly correlate with relapse risk (HR=2.6, submitted). Surprisingly, no recurrent missense NF1 mutations inactivating GAP activity were found in our cohort, and such mutations are rare in primary cancers in general. We thus posulated that complete loss of NF1 protein (e.g., caused by NS/FS mutations), but not GAP inactivation alone, is required to drive endocrine therapy resistance. Here we demonstrate that NF1 loss (by gene silencing) in ER+ breast cancer cells greatly enhances ligand-dependent ER transcriptional activity in vitro and in vivo, causing estradiol (E2) hypersensitivity and tamoxifen agonism. Mechanistically we show that NF1 can bind directly to ER, an interaction enhanced by tamoxifen but not by E2. Binding is mediated by leucine/isoleucine-rich motifs in NF1, analogous to other ER co-repressors. Mutations in these motifs (some of which are targeted by somatic mutation in cancer) inhibit ER binding and transcriptional activity without impacting GAP activity; conversely, inactivating GAP activity does not impact ER binding and repression. To validate NF1 as an ER co-repressor, we examined proteomic data from >100 breast cancer patients in the CPTAC data base and found that proteins whose levels are positively correlated with NF1 are highly enriched with factors known to bind nuclear receptors; by contrast, levels of another GAP, p120, which lacks ER binding sites, are negatively correlated with these molecules. Importantly, preclinical treatment studies indicate that while NF1-deficient ER+ breast cancer should not be treated by tamoxifen or aromatase inhibitors, fulvestrant, which degrades ER, remains effective. However, fulvestrant monotherapy can activate the Ras-MAP pathway, which may promote cell survival and acquired fulvestrant resistance unless combined with dabrafinib and trametinib to inhibit Raf and MEK —a clinical trial for this combination is in development. Our data suggest that NF1 is a dual negative regulator at the intersection of two potent oncogenic signaling pathways, Ras and ER. Combination therapy targeting both the ER and the Ras-Raf pathways should be investigated for NF1-deficient cancers driven by ER. Citation Format: Eric C. Chang, zeyi Zheng, Meenakshi Anurag, Jin Gao, Burcu Cakar, Xinhui Du, Jing Li, Philip Lavere, Jonathan T. Lei, Purba Singh, Sinem Seker, Wei Song, Jianheng Peng, Tiffany Nguyen, Doug Chan, Xi Chen, Kimberly C. Banks, Richarad B. Lanman, Maryam Shafaee, Susan Hilsenbeck, Charles Foulds, Matthew J. Ellis. NF1 as an estrogen receptor-α co-repressor in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1814.

endocrinology & metabolism