Abstract. The effect of retinoids on breast cancer has been predominantly studied in vitro, on established cell lines, which in biology differ significantly from primary tumor cells. Little is known on whether early in vitro passages of breast cancer cells (EPBCCs) are differentially sensitive to retinoids and differentially express retinoid acid receptors (RARs) and retinoid X receptors (RXRs). We have previously identified a novel RARβ isoform (RARβ5) and hypothesized that it may serve as a potential target of retinoids in EPBCCs. Breast cancer cells isolated from primary tumors were cultured in vitro for 6-12 passages (EPBCCs) and their epithelial origin was confirmed by a cocktail of antibodies against cytokeratins. EPBCCs were treated for 4 days with 1.0 µM of all-trans retinoic acid (atRA), 9-cis retinoic acid (9cRA) or 4-hydroxy-phenylretinamide (4-HPR) and their viability determined by MTT assay. Among nine EPBCCs consistently grown in vitro, three were resistant to the above retinoids, five were susceptible to atRA, four to 4-HPR and two to 9cRA, suggesting that patients with breast carcinomas may differentially respond to various retinoids. All EPBBCs differentially expressed RARα, RARγ, RXRα, RXRβ proteins and RARβ5 and RARβ2 mRNAs. However, only one EPBCC (BCA-2) expressed RARβ5 at mRNA and protein level and it was resistant to retinoids, both in vitro and in a xenograft tumor assay. RARβ5 suppression by siRNA in BCA-2 cells increased their susceptibility to atRA. No correlation was found between sensitivity of EPBCCs to the above retinoids and RARβ5 and RARβ2 mRNA expression. atRA reduced RARβ expression in most EPBCCs suggesting that this retinoid receptor is most probably the prime target of retinoids in breast cancer. These data may have clinical implication in selecting patients with breast cancer that would benefit the most from clinical trials with retinoids.