The prognostic effects of somatic mutations in ER-positive breast cancer

Obi L. Griffith, Nicholas C. Spies, Meenakshi Anurag, Malachi Griffith, Jingqin Luo, Dongsheng Tu, Belinda Yeo, Jason Kunisaki, Christopher A. Miller, Kilannin Krysiak, Jasreet Hundal, Benjamin J. Ainscough, Zachary L. Skidmore, Katie Campbell, Runjun Kumar, Catrina Fronick, Lisa Cook, Jacqueline E. Snider, Sherri Davies, Shyam M. KavuriEric C. Chang, Vincent Magrini, David E. Larson, Robert S. Fulton, Shuzhen Liu, Samuel Leung, David Voduc, Ron Bose, Mitch Dowsett, Richard K. Wilson, Torsten O. Nielsen, Elaine R. Mardis, Matthew J. Ellis

Research output: Contribution to journalArticlepeer-review

83 Scopus citations


Here we report targeted sequencing of 83 genes using DNA from primary breast cancer samples from 625 postmenopausal (UBC-TAM series) and 328 premenopausal (MA12 trial) hormone receptor-positive (HR+) patients to determine interactions between somatic mutation and prognosis. Independent validation of prognostic interactions was achieved using data from the METABRIC study. Previously established associations between MAP3K1 and PIK3CA mutations with luminal A status/favorable prognosis and TP53 mutations with Luminal B/non-luminal tumors/poor prognosis were observed, validating the methodological approach. In UBC-TAM, NF1 frame-shift nonsense (FS/NS) mutations were also a poor outcome driver that was validated in METABRIC. For MA12, poor outcome associated with PIK3R1 mutation was also reproducible. DDR1 mutations were strongly associated with poor prognosis in UBC-TAM despite stringent false discovery correction (q = 0.0003). In conclusion, uncommon recurrent somatic mutations should be further explored to create a more complete explanation of the highly variable outcomes that typifies ER+ breast cancer.

Original languageEnglish
Article number3476
JournalNature communications
Issue number1
StatePublished - Dec 1 2018


Dive into the research topics of 'The prognostic effects of somatic mutations in ER-positive breast cancer'. Together they form a unique fingerprint.

Cite this