BACKGROUND: Despite evidence of estrogen's mood-enhancing effects, the association between estrogen receptor (ER) gene variants and lifetime major depression has been insufficiently studied. METHODS: 3987 community-dwelling women aged 65years and over were recruited in France as part of the Three City Study. Current and past major depressive disorders (MDD) were diagnosed using the Mini-International Neuropsychiatry Interview, according to DSM-IV criteria. The association between two common estrogen receptor alpha (ESR1) polymorphisms with lifetime MDD was examined using adjusted logistic regression models, taking into account the age at first depressive episode and the recurrence of depression. RESULTS: Women homozygous for the variant G allele of ESR1 rs9340799 had a 1.6-fold increased risk of MDD across their lifetime compared with women who were homozygous for the A allele (p=0.009). There was a similar non-significant trend for the C allele of rs2234693 being associated with an increased risk (p=0.09). Polytomous regression analysis further indicated that the GG genotype of rs9340799 was specifically associated with an increased risk of recurrent depressive episodes, regardless of the age at first onset of depression relative to the menopause. LIMITATIONS: The duration and severity of depressive episodes was not considered in the analysis. CONCLUSIONS: This is the first study to examine the association between ESR1 gene variants and lifetime MDD. Our findings indicate a significant association between common variants and the risk of recurrent depressive episodes. This suggests that certain depressed women could be most responsive to hormone-based treatment.