How can deep endometriotic stromal cells proliferate and persist in a fibrotic environment? The serine/threonine kinase AKT and extracellular regulated kinase (ERK) signaling pathways may co-operate to support growth of deep endometriotic lesions by enhancing endometriotic stromal cell proliferation and survival in a fibrotic microenvironment in vitro. Endometriosis, particularly deep infiltrating endometriosis, is characterized histologically by dense fibrous tissue that is primarily composed of type I collagen. This tissue may cause pelvic pain and infertility, which are major clinical issues associated with endometriosis. Proliferation of normal fibroblasts is tightly regulated, and fibrillar, polymerized type I collagen inhibits normal fibroblast proliferation. However, no studies to date have investigated how deep endometriotic stromal cells can proliferate and persist in a fibrotic environment. Endometrial and/or endometriotic tissues from 104 patients (61 with and 43 without endometriosis) of reproductive age with normal menstrual cycles were analyzed. A total of 25 nude mice received a single injection of endometrial fragments from a total of five samples. We evaluated cell proliferation, caspase 3/7 activity, and the AKT and ERK signaling pathways in endometrial and endometriotic stromal cells on three-dimensional (3D) polymerized collagen matrices in vitro. In addition, to determine whether aberrant activation of the AKT and ERK pathways is involved during progression of fibrosis in endometriosis in vivo, we evaluated the expression of phosphorylated AKT and ERK1/2 in endometriotic implants in a nude mouse model of endometriosis. Finally, we evaluated the effects of MK2206 (an AKT inhibitor) and U0126 (a MEK inhibitor) on cell proliferation, caspase 3/7 activity, and phosphorylation of AKT and ERK1/2 of endometriotic stromal cells on 3D polymerized collagen matrices. Proliferation of endometriotic stromal cells was significantly less inhibited than that of endometrial stromal cells (P < 0.05) on 3D polymerized collagen. Levels of phosphorylated AKT, phosphorylated p70S6K and phosphorylated ERK1/2 were significantly higher in endometriotic stromal cells than in endometrial stromal cells at 24 h (P < 0.05) and at 72 h (P < 0.05) on 3D polymerized collagen. Phosphorylated AKT expression was significantly increased on Days 21 and 28 compared with those on Days 3 and 7 (all P < 0.05) in endometriotic implants during progression of fibrosis in a nude mouse model of endometriosis. Inhibition of AKT or ERK1/2 with MK2206 or U0126, respectively, did not significantly increase caspase 3/7 activity in endometriotic stromal cells on either two-dimensional or 3D collagen matrices. Western blot analysis showed that MK2206 alone decreased levels of phosphorylated AKT; however, it increased levels of phosphorylated ERK in endometriotic cells compared with vehicle-treated cells (both P < 0.05). In addition, U0126 treatment decreased levels of phosphorylated ERK; however, it resulted in increased levels of phosphorylated AKT in endometriotic stromal cells compared with vehicle-treated cells (both P < 0.05). Endometriosis involves a number of processes, such as invasion, metastasis, angiogenesis, and apoptosis resistance, and a variety of signaling pathways may be involved in promoting development and progression of the disease. In addition, further animal experiments are required to determine whether the AKT and ERK signaling pathways co-operate to support growth of endometriotic lesions in a fibrotic microenvironment in vivo. Co-targeting the AKT and ERK pathways may be an effective therapeutic strategy for endometriosis treatment. This study was supported in part by Karl Storz Endoscopy & GmbH (Tuttlingen, Germany). No competing interests are declared.