Abstract Study question Does revascularization of xenotransplanted human ovarian grafts occur equally efficient from both the cortical and medullary side of the cortex tissue? Summary answer Murine vessels were established equally efficient from both sides of human ovarian grafts, demonstrating that the fibrous cortical side does not hinder the neovascularization process. What is known already There is a lack of consensus regarding the surgical approach for transplanting frozen-thawed ovarian tissue and various techniques are being used clinically worldwide. Some centers attach ovarian tissue with stitches or Interceed® on top of the decorticated remaining ovary to mimic normal ovarian architecture. Other centers transplant ovarian tissue in subcortical pockets made by longitudinal or transverse incisions in the remaining intact ovary to facilitate vascularization from both sides of the graft. While other centers transplant to sub-peritoneal pelvic sites only. It is unknown which techniques and transplantation sites provide the most efficient revascularization of the ovarian grafts. Study design, size, duration Three pieces of cryopreserved ovarian cortex (approximately 5x5x1 mm in size) were donated and thawed from 12 women and two of the pieces were transplanted to subcutaneous pockets on the dorsal side of ovariectomized immunodeficient female mice. The third piece was processed as an ungrafted control. Ovarian grafts were retrieved after 8 days to analyze the spatial distribution of graft-revascularization using immunohistochemical detection of murine CD31, and after 8 weeks to evaluate follicle density (follicles/mm3). Participants/materials, setting, methods Ovarian cortex was donated from 12 women aged 27-31 years having their tissue frozen for fertility preservation. The CD31 positive vessel area and density were quantified using a custom designed application (APP) from Visiopharm®. Three regions of interest (ROIs) were defined in each tissue section; the cortical side, the center, and the medullary side. Vessels were sub-divided into three categories according to size: micro-vessels (<300 µm2), small vessels (300-1000 µm2), and large vessels (1000-3000 µm2). Main results and the role of chance After 8 days xenotransplantation, a statistical significant lower density of vessels was found in the center of the human ovarian grafts compared to the cortical and medullary sides (cortical side: 323 ± 14 vessels/mm2; center: 240 ± 12 vessels/mm2; medullary side: 301 ± 18 vessels/mm2; p < 0.001). The mean percentage of CD31 positive vessel area in the human ovarian grafts was similar in the three ROIs, but lowest in the center of the grafts (cortical side: 3.9% ± 0.2 (SEM); center: 3.5% ± 0.2; medullary side: 4.0% ± 0.3; p = 0.17). Micro-vessels comprised 89-91% of all vessels in the three ROIs, demonstrating that the vast majority of vessels were newly formed. Heatmaps were generated for each section based on the CD31 positive area to visualize the localization of vessels in the ovarian grafts and the maps showed highly variable vascularization patterns in the grafts from different patients. Overall, vascularization appeared to occur equally efficient in all peripherical areas of the ovarian grafts. The follicle density in ungrafted cortex tissue was 51.8 ± 17.3 follicles/mm3 and 14.7 ± 3.7 follicles/mm3 after 8 weeks xenografting, resulting in an average follicle survival rate of 28%. Limitations, reasons for caution The current study was not designed to directly compare clinically used transplantation techniques, which should be studied in larger animal models, like sheep, in which the structure of ovarian tissue is similar to human. Wider implications of the findings Revascularization was established equally efficient from both sides of xenotransplanted human ovarian cortex, suggesting that transplantation techniques ensuring revascularization from both sides of the ovarian graft, such as sub-cortical or peritoneal pockets, may facilitate a more efficient revascularization of the graft than techniques allowing revascularization from only the medullary side. Trial registration number not applicable
Read full abstract