Patient-derived epithelial cell organoids mimic the phenotypic complexity of endometriosis subtypes

Can patient-derived organoid models be reliably established from diverse surgical phenotypes of endometriosis, and how do clinical factors such as hormonal treatment affect their growth success and morphology?

Endometriosis organoids can be established across all major surgical phenotypes with variable efficiency, and hormonal treatment at the time of biospecimen collection significantly reduces organoid establishment success.

Organoid cultures have been developed from eutopic endometrium and select endometriosis tissue biospecimens previously, but their feasibility as pre-clinical models of endometriosis across diverse tissue types and clinical presentations remains unclear.

Twenty-eight endometriosis tissue biospecimens were obtained from 23 patients undergoing surgery, with organoid cultures assessed through successive stages of establishment, passage, and cryopreservation.

Endometriosis biospecimens, including deep infiltrating endometriosis (DIE), ovarian endometrioma (OMA), and superficial peritoneal (SUP) biospecimens, were processed into organoid cultures using a validated low-Wnt culture system. Organoid viability, morphology, hormone receptor expression, and cellular composition were evaluated by microscopy, immunohistochemistry, and quantitative morphometric analysis.

Overall, 22/28 (78.6%) biospecimens established 3-dimensional structures, with 15/28 (53.6%) remaining viable after cryopreservation. Establishment success differed by phenotype (OMA 71.4%, DIE 63.6%, SUP 30%). Progesterone receptor expression was retained in SUP and DIE-derived organoids (7/7, 100%), while OMA-derived organoids showed substantial reductions (4/5 cases). Biospecimens from patients receiving hormonal treatment were smaller (P = 0.038) and had reduced organoid establishment success (3/13, 23.1% vs 12/15, 80.0%, P = 0.003). Organoids exhibited distinct morphological patterns correlating with disease phenotype.

Uniform culture conditions may limit growth of certain subtypes, and the in vitro organoid models may not fully represent in vivo tissue complexity. Sample sizes were modest, and pooling tissues from the same patient could mask intra-patient heterogeneity.

These organoid models offer a promising platform for studying subtype-specific endometriosis biology, including hormone resistance mechanisms, and could inform personalized therapeutic development. The impact of hormonal treatment on organoid viability underscores the need to consider clinical context in pre-clinical models of endometriosis.

This work was supported by the National Endometriosis Clinical and Scientific Trials (NECST) Network, funded by the Australian Government Department of Health and Aged Care (Grant 4-I66SNMA), and by a research grant from Endometriosis Australia to C.E.F., D.L., and J.A.A. K.G. is supported by an Australian Government Research Training Program Scholarship and a NECST Network Top-Up Scholarship, which did not influence the conduct or outcomes of this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. J.A.A. has received consulting fees from Hologic, Gedeon Richter, and BD, personal payments from Hologic, Bayer, Organon, and Gedeon Richter, travel support from Gedeon Richter, and participated on data safety monitoring advisory boards for Hologic and Gideon Richter. He was the former chair of the Australian Endometriosis Guideline Committee and is the Co-Editor-in-Chief of the Journal of Minimally Invasive Gynaecology. All other authors declare no competing interests.

N/A.