Using an administrative dataset (Hygie), we apply a difference-in-differences with dynamic matching estimation method to the onset of breast cancer. The employment probability decreases by 10 percentage points (pp) one year after the onset of cancer compared to the not-treated group. The detrimental effect of breast cancer on employment significantly increases over time, up to 12 pp after five years.Our study also aims to identify some socio-demographic and work-related protective factors against adverse effects of breast cancer on labour market outcomes. We stress out four potential protective factors related to the negative breast cancer effect. Firstly, a young age at occurrence reduces this deleterious effect. Secondly, a high first job wage appears as a protective factor. Thirdly, having faced less unemployment in the past is associated with a weaker negative breast cancer effect on employment in the short run. Finally, we find a moderate “generation effect” after a stratification by year of cancer onset.