Human chorionic gonadotropin (hCG), a hormone produced during pregnancy, can elicit life-long refractoriness to carcinogenesis by differentiation of the breast epithelium. Human breast epithelial cells MCF-10F form tubules in collagen, mimicking the normal ductules. We have shown that 17 beta-estradiol (E2) alter the ductulogenic pattern of these cells. The effect of the recombinant hCG (rhCG) in vitro was evaluated on the transformation of MCF-10F induced by E2. MCF-10F cells were treated with 70 nM E2 alone or in combination with 50 IU/ml rhCG during 2 weeks, while the controls were treated with DMSO (the solvent in which E2 was dissolved) or rhCG alone. At the end of treatment, the cells were plated in type I collagen matrix (3D-cultures) for detecting 2 main phenotypes of cell transformation, namely the loss of ductulogenic capacity and the formation of solid masses. Although E2 significantly increased solid mass formation, this effect was prevented when MCF-10F cells were treated with E2 in combination with rhCG. Furthermore, E2 increased the main duct width (p < 0.001), and caused a disruption of the luminal architecture, whereas rhCG increased the length of the tubules (p < 0.001) and produced tertiary branching. In conclusion, rhCG was able to abrogate the transforming abilities of estradiol, and had the differentiating property by increasing the branching of the tubules formed by breast epithelial cells in collagen. These results further support our hypothesis, known as the terminal differentiation hypothesis of breast cancer prevention, that predicts that hCG treatment results in protection from tumorigenic changes by the loss of susceptible stem cells 1 through a differentiation to refractory stem cells 2 and increase differentiation of the mammary gland.