Background: Bone mineral density (BMD) loss is a consequence of Aromatase Inhibitors (AI) therapy. This study aims to identify variants of genes in the vitamin D and estrogen signaling pathways associated with AI-related BMD loss in the B-ABLE cohort. Methods: B-ABLE is a prospective cohort of women recruited at the time they initiate AI therapy in a bone metabolism unit. Bisphosphonate users were excluded for these analyses. Selected single nucleotide polymorphisms (SNPs) were genotyped. Multivariate linear regression was used to test their association with relative lumbar spine (LS) and femoral neck (FN) BMD loss after 1 and 2 years of follow-up. All models were adjusted for age, body mass index, previous tamoxifen and chemotherapy. Further, potential confounding for baseline 25(OH)-VITD concentrationsand AI used was assessed. Results: 391 participants were included. For estrogen signaling, two SNPs in CYP11A1 (rs2959008 and rs7174179) for the association with FN BMD loss at both one (P=0.003 and P=0.012) and two years (P=0.004 and P=0.002) of follow-up were significant. For LS BMD lossthree SNPs in HSD3B2 (rs2854964), CYP2C19 (rs12248560) and CYP2C9 (rs28371674) were significantly associated at one year of follow-up (P=0.026, P=0.019 and P=0.011 respectively). Only rs12248560 remained also significant at 2 years of follow-up (P=0.014). Regarding vitamin D signaling genes, rs11907350 in CYP24A1 was associated with FN BMD loss at one year. As for LS BMD loss one SNP in GC (rs11907350) at one year of follow-up and one SNP in VDR (rs2544037) at two years of follow-upreached significant p-values (P =0.020 and P=0.024, respectively). Only the SNP rs7174179 in CYP11A1 for FN BMD loss association at 2 years of follow-up remained significant after Bonferroni correction. Conclusions: Several genes in estrogen and vitamin D signaling appeared involved in BMD loss in AI-treated women, suggesting a complex regulation of this outcome.