Abstract Study question Does gender-affirming testosterone therapy alter the composition of the extra-cellular matrix (ECM) within the ovarian stroma and subsequently affect follicle activation in vivo Summary answer Ovarian stroma of trans men is more collagenous and less elastic, indicating fibrotic change. This may affect in vivo follicle growth activation What is known already Changes in the ovarian stroma have been demonstrated in the ovaries of transgender men taking testosterone, including thickening of the tunica albuginea, stromal cell hyperplasia and stromal cell luteinisation. Ovaries of trans men also have increased cortical stiffness. These changes are similar to those seen in female patients with PCOS and in physiological ovarian aging, which has been attributed to accumulation of collagen in the ECM. Increasing stiffness of the supportive follicular microenvironment has been shown to reduce follicle growth activation in vitro Study design, size, duration Whole ovaries were obtained from transgender men (mean age 27.6 ± 1.7 years, n = 8) with informed consent at oophorectomy. All patients had received 1000mg testosterone undecanoate intramuscularly at 12-16 week intervals for a minimum of 18 months pre-operatively (range 18 months-10 years). Cortical tissue was dissected into small fragments (≈1x1x0.5mm) and fixed for histological and immunohistochemical analysis. Testosterone-treated ovaries were compared to cortical biopsies from age-matched healthy women obtained at caesarean section (mean age 31.8±1.5, n = 8). Participants/materials, setting, methods Follicle number, classification of developmental stage, non-growing follicle density (NGFD) and stromal cell density were evaluated by histological analysis of ovarian cortical tissue. Sections were stained with Picrosirius red (PSR) to analyse total collagen content using brightfield microscopy. Polarised light was also used to analyse the collagen birefringence, which allows quantification of collagen fibre thickness into thick, medium or thin. Total elastin content was evaluated using immunofluorescence. Main results and the role of chance 4526 follicles were analysed. Transgender ovary showed a higher proportion of non-growing follicles found compared to control (93.9±1.2% vs 84.6±1.5% p < 0.05): the proportions of primary (4.7±0.9% vs 10.6±1.5%, p = 0.2) and secondary (1.4±0.4% vs 4.6±0.7%, p = 0.1) follicles tended to be lower. Stromal cell density was significantly higher in transgender ovarian cortex than control (2.5±0.1 x106cells/mm3 vs 1.7±0.1 x106cells/mm3), indicating stromal cell hyperplasia. Combined data from control and transgender groups showed a positive correlation between NGFD and stromal density (r = 0.64, p = 0.01). Transgender ovary had a higher total collagen content (77.2±1.2%) compared to control (31.3±3.3%, p < 0.005). Analysis of collagen birefringence showed that transgender ovaries had similar quantities of thick collagen fibres (0.014±0.005 vs 0.010±0.009, p = 0.1), more medium thickness collagen fibres (45.1±6.6%vs 14.4±4.9%, p < 0.05) and fewer thinner fibres (41.5±9.6% vs 27.7±2.8%, p = 0.08) than control. The total elastin content in transgender ovaries was lower than control (1.3±0.1% vs 3.6±0.6%, p < 0.005) and subsequently, the collagen/elastin ratio was significantly higher (63.1±7.9 vs 10±1.3, p < 0.005). Limitations, reasons for caution The impact of these findings on in vivo follicle growth are unclear. The effect of duration of testosterone treatment has not investigated. Wider implications of the findings More collagenous, less elastic ovarian stroma in trans men indicates fibrotic change; these findings are similar to women with PCOS and with reproductive ageing. These stromal changes may alter follicle growth activation and may contribute value to our understanding of the regulation of follicle function in a range of conditions. Trial registration number nil