To study the role of epidermal growth factor (EGF) , epidermal growth factor receptor(EGFR), extracellular signal-regulated kinase 1/2 (p-ERK1/2) in the pathogenesis of endometriosis under estrogen deprivation conditions. The estrogen was quickly-stripped in medium and the female nude mice were castrated by bilateral oophorectomy to build estrogen deprivation in vitro and in vivo experimental models, respectively. (1) In vitro experiments:according to different treatments the estrogen deprived ectopic endometrial cells were classified into 4 groups: a. EGF group:the ectopic endometrial cells were cultured for 72 hours with different concentrations of EGF (0.01, 0.1, 1, 10, 50, 100 ng/ml), the results of EGF group were represented by the result of cells treated by 10 ng/ml EGF cultured for 72 hours; b. EGF+PD98059 group:the ectopic endometrial cells were cultured for 72 hours with 5×10(-2) mol/L PD98059 (inhibitor of ERK), followed by a cultivation for 72 hours treated by 10 ng/ml EGF+5×10(-2) mol/L PD98059; c. EGF+ ICI182780 group: the ectopic endometrial cells were cultured for 72 hours with 10(-6) mol/L ICI182780 [inhibitor of estrogen receptor(ER)], followed by a cultivation for 72 hours treated by 10 ng/ml EGF+10(-6) mol/L ICI182780; d. Blank control group:the ectopic endometrial cells were cultured with no treatment. The proliferation activity of ectopic endometrial cells in all groups after treatment were examined by methyl thiazolyl tetrazolium (MTT) method represented by absorbance value (A). The expression of p-ERK1/2 protein were detected by western blot. (2) In vivo experiments: 64 female nude mice were randomly divided into control and castration groups (both n=32) using random number chart. The mice in castration group were castrated by bilateral oophorectomy on 3 weeks after the endometriosis model was established. The levels of EGF, EGFR, p-ERK1/2 protein in ectopic lesions of both groups were measured on 4, 6, 8 and 10 weeks after the endometriosis model was established by western blot. (1) The proliferation activity of ectopic endometrial cells:the proliferation activity of ectopic endometrial cells treated by different concentrations of EGF (0.01, 0.1, 1, 10, 50, 100 ng/ml) for 72 hours were 0.310±0.010, 0.340±0.020, 0.670±0.010, 0.980±0.030, 1.360±0.020, 1.670±0.020, respectively, the proliferation activity was increased along with of EGF concentrations.The proliferation activity was 0.680±0.030 at EGF+ PD98059 group, the differences exhibited significant difference when compared with that at EGF group with 100 ng/ml for 72 hours (P<0.01) .The proliferation activity of EGF+ ICI182780 and blank control groups were 0.330±0.030 and 0.310±0.030, respectively, which did not reached statistical differences (P>0.05). (2) The expression of EGF, EGFR, pERK1/2 protein: a. In vitro experiments:the levels of p-ERK1/2 protein in EGF and blank control groups were 0.670±0.020 and 0.600±0.010, respectively, which reached statistical differences (P<0.05). The level of p-ERK1/2 protein in EGF+ PD98059 group was 0.610±0.020, which exhibited significant differences with that at blank control group (P>0.05). b. In vivo experiments:at 4, 6 and 8 weeks after the endometriosis models were established, the expression of EGF protein in the ectopic lesions of castration group and control group were (0.530±0.015 versus 0.610±0.015), (0.400±0.029 versus 0.620±0.018), (0.560±0.026 versus 0.630±0.021), respectively, the levels of EGFR protein were (0.500±0.030 versus 0.640±0.030), (0.470±0.020 versus 0.630±0.020), (0.510±0.030 versus 0.610±0.020) respectively, and the level of p-ERK1/2 protein were (0.500±0.020 versus 0.580±0.020), (0.490±0.020 versus 0.580±0.020), (0.570±0.020 versus 0.590±0.020), respectively. The difference of EGF, EGFR, p-ERK1/2 protein expression levels between two groups did not exhibited significant difference (P<0.01, P<0.01, P<0.05). At 10 weeks after the endometriosis models were established, the levels of EGF protein in castration group and control group were both 0.620±0.020, the levels of EGFR protein were both 0.610±0.020, and the level of p-ERK1/2 protein were 0.590±0.010 and 0.600±0.020. No statistical difference (P>0.05) was found between those two groups (P>0.05). EGF could stimulate the proliferation of ectopic endometrial cells by activating the ERK pathway under estrogen deprivation conditions. The inhibition of EGF signaling system in ectopic lesions was alleviated along with the prolongation of the period of estrogen deprivation.