Genetic instabilities as potential biomarkers for pregnancy-associated breast cancer.

Abstract

e12559 Background: Pregnancy associated breast cancer (PABC), defined as breast cancer diagnosed during or within one year after pregnancy, occurs approximately once in every 3,000 pregnancies and accounts for up to 6.9% of all breast cancers in women ≤ 45 years. Compared to age-matched non-PABC patients, PABC is generally characterized by a particularly aggressive histopathologic profile (more ER-, PR- and HER2 negative tumors and tumors of higher grade) with a higher mortality rate. Whether these cancers arise before or during pregnancy, and whether their genetic profile differs from non-PABC tumors, is currently unknown. This study assesses the genetic background of PABC by detection of specific DNA copy number alterations (CNA). Methods: We assembled 29 triple-negative PABC patients from the Dutch nationwide pathology database (PALGA). From their formalin-fixed and paraffin embedded (FFPE) tumor tissue blocks, DNA was extracted. Multiplex Ligation-Dependent probe Amplification (MLPA) analysis was performed for detection of CNAs in 20 oncogenes and 2 tumor suppressor genes using the P078-D2 kit (MRC Holland). Individual gene copy number loss (MLPA ratio < 0.7), gain (ratio 1.3-2.0) and amplification (ratio > 2.0) was determined and unsupervised hierarchical cluster analysis was performed to identify differences in copy number (CN) patterns between PABC patients. Chi square statistics were used to interrogate associations with clinicopathologic characteristics. Log-Rank test was used to compare Kaplan Meier survival curves between CN subgroups. Results: Triple negative PABC tumors showed frequent copy number loss on chromosome 17q ( CPD, MED1, TOP2A, MAPT, PPM1D) and 8p ( ZNF703 and ADAM9), as well as frequent copy number gain/amplification of MYC (8q) and CCND1 (11q). Cluster analysis identified 2 clusters of PABC patients based on their tumor copy number patterns: a CN-neutral cluster and a CN-high cluster, the latter demonstrating significantly lower chromosome 17 copy numbers than the former (for PPM1D, ERBB2, CPD, MED1, TOP2A, CDC6 and MAPT). Patients with CN-neutral tumors more often showed lymph node metastases (p = 0.028), were of significantly higher stage (p = 0.031), and demonstrated a worse overall survival (p = 0.036) compared with PABC patients harboring CN-high tumors. Conclusions: We demonstrate different tumor CN patterns within a small group of triple negative PABC-patients, associated with distinct aggressiveness. This will form the basis for further in-depth genetic profiling within a larger Dutch PABC cohort, rendering insight in the disease progression and contributing to the development of more targeted and personalized treatments.