In this work, we found that proliferation-related prognostic gene signatures could aid treatment decision-making independent of age. This is clinically relevant for the younger breast cancer population given the potential long-term side effects of adjuvant systemic chemotherapy and hence the need to identify patients who are less likely to benefit adjuvant chemotherapy. In addition, it was clear that young age at diagnosis adds extra biological complexity, which is independent of differences in breast cancer subtype distribution. This includes enrichment with known breast cancer targets like RANKL. Whilst these results require further validation, either experimentally or in other clinical data sets, it suggests that separate therapeutic approaches may need to be specifically designed in order to improve outcomes for breast cancer arising in young women. In this regard, and based on our results and supportive evidence from other studies, we initiated a proof-of-concept prospective phase II neoadjuvant study investigating the role of denosumab, a RANKL inhibitor on modulating tumor biology in young premenopausal breast cancer patients. <p>We found that diagnosis during pregnancy does not significantly influence the classic pathological features or the prevalence of breast cancer subtypes. We also did not find obvious differences in the distribution of PIK3CA mutations. However, we found that tumors diagnosed during pregnancy have activated serotonin receptor signaling and high expression of potential breast cancer targets; of particular interest IGF1, and PDL1. Such differences appeared to be reflected in the normal pregnant breast underscoring the potential role of the pregnant breast microenvironment on the tumor transcriptome. We were not able to associate these genes with prognosis, which could be partly due to lack of statistical power. Of note, we cannot confirm whether any of these aberrations are key drivers of the biology of tumors diagnosed during pregnancy. Nevertheless, this remains the first study to look into the biology of this relatively rare disease and hence we believe it would serve as a very valuable resource for future research in this field. We are planning to perform targeted gene sequencing to further refine our understanding of the potential effect of pregnancy on the biology of these tumors. <p>In the last part of this work addressing the safety of pregnancy following breast cancer diagnosis, we identified that available studies suffered major limitations related to study design including selection bias and lack of information on patients with history of an ER-positive disease. This has resulted in advising against pregnancy in women with prior history of breast cancer. Our subsequent study has robustly addressed most of the limitations in older studies and clearly showed that pregnancy following breast cancer is safe even in women with a history of ER-positive disease. Hence, this study would provide a very important resource for the oncology community, which would aid adequate fertility counseling for young breast cancer survivors. This work is currently serving as the basis for a new prospective study by the IBCSG to test the safety of early interruption of tamoxifen in young women with early breast cancer seeking subsequent pregnancy. <p> / Doctorat en Sciences biomédicales et pharmaceutiques / info:eu-repo/semantics/nonPublished