Abstract P3-06-04: Clonal evolution of non-malignant proliferative lesions into breast cancers

[Introduction] Non-malignant proliferative lesions in the breast have been implicated in the development of invasive breast cancer. Previous studies showed that adjacent atypical proliferative lesions and breast cancers shared common genetic alterations, suggesting that these evolved from the same ancestral cell. However, the clonal structure of atypical proliferative lesions and their clonal dynamics during progression to cancer are poorly understood. In this study, we compared genetic profiles of normal ducts, non-malignant proliferative lesions and cancers in the same patients to illustrate the clonal evolution of cancer from a non-malignant epithelial cell. [Methods] Multiple samples were collected from different proliferative lesions within the cancer-borne breast, including invasive cancers, using micro-dissection from formalin-fixed, paraffin-embedded surgical specimens. Somatic mutations and copy number alterations (CNAs) were then evaluated by whole exome sequencing. [Results] A total of 39 samples from 6 premenopausal females carrying estrogen receptor-positive cancers were analyzed, where the samples were obtained from normal ducts (N = 5), non-malignant proliferative lesions (N = 9), and non-invasive (N = 21) and invasive (N = 4) cancers. The number of somatic mutations per sample was ranging from 1 to 311 and increased with pathological disease progression. Two cases with bilateral cancers had a pathogenic germline mutation of either BRCA2 or TP53, where no somatic mutations or CNAs were shared by individual proliferative lesions, suggesting multifocal independent cancerous evolutions. By contrast, in the remaining four unilateral cases, no pathogenic germline mutations were detected, but all proliferative lesions, which were separated by a distance of 7-33 mm, shared one or more driver alterations, such as an AKT1 mutation (UID: KU01), a GATA3 mutation (UID: KU03 and KU06), a CBFB mutation (UID: KU06) and concurrent 1q gain and 16q loss (UID: KU02, KU03 and KU06), while harboring private mutations/CNAs of their own. The analysis of phylogenic trees based on the number of shared mutations predicted an early origin of these founder mutations, which frequently predated decades before the onset of cancer. [Conclusions] Early breast cancer development is thought to be shaped by a simultaneous evolution of multiple precancerous clones. It may be multi-focally initiated by a germline mutation, frequently terminated in bilateral cancers. By contrast, in unilateral cases, cancer clones might be derived from a common ancestral clone, which has acquired a driver founder mutation long before the onset of cancer, and undergo independent evolution, giving rise to multiple proliferative lesions, from which invasive cancer finally evolves. Our findings provide unique insight into the early development of breast cancer. Citation Format: Tomomi Nishimura, Kenichi Yoshida, Yukiko Kawata, Yasuhide Takeuchi, Nobuyuki Kakiuchi, Yusuke Shiozawa, Kosuke Aoki, Masahiro Hirata, Tatsuki R. Kataoka, Takaki Sakurai, Satoko Baba, Yuichi Shiraishi, Kenichi Chiba, Kengo Takeuchi, Hironori Haga, Satoru Miyano, Masakazu Toi, Seishi Ogawa. Clonal evolution of non-malignant proliferative lesions into breast cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 741.