The aim of this thesis was to evaluate whether the use of the 70-gene signature (MammaPrint) can be improved by taking into account other factors to better define subgroups of breast cancer patients who are at risk of overtreatment, and for whom adjuvant systemic treatment (chemotherapy and/or endocrine therapy) could be safely omitted. In women older than 50 years with unfavorable clinical pathological characteristics, chemotherapy can be safely omitted in case of a low-risk 70-gene signature. In women younger than 50 years there may be some benefit from chemotherapy, but further research into the optimal treatment of these younger women is needed. Patients with stage I ER+/HER2- breast cancers who received no adjuvant systemic treatment had excellent survival outcomes, but had higher cumulative incidences of locoregional recurrences and contralateral breast cancers than patients who received endocrine therapy. Patients with 70-gene signature ultralow risk tumors had the best prognosis, and could be candidates for further de-escalation of treatment. Both the 70-gene signature and method of detection were significantly associated with breast cancer outcome, and incorporating both into (online) clinical risk prediction and decision-making tools will further optimize the prognostication for individual breast cancer patients. We evaluated the influence of the 70-gene signature on risk assessment and chemotherapy recommendation, and found that knowledge of the 70-gene signature result led to fewer patients being assessed as high risk, and fewer recommendations for chemotherapy, among European breast cancer specialists. Furthermore, we evaluated the association between a polygenic risk score (PRS) for breast cancer and outcome. The PRS313 was associated with favorable tumor characteristics and a low risk 70-gene signature. The PRS313 was not independently associated with prognosis.