L26/O-103 Complex in vitro 3D modelling of endometriosis and endometrial disorders

Abstract Endometriosis is a common and chronic condition causing pelvic pain and subfertility. Despite its prevalence, the pathophysiology of endometriosis remains poorly understood, due in part to disease complexity (superficial, ovarian and deep subtypes) and a lack of laboratory models to study disease mechanisms. Common laboratory animal models, such as rodents, have been successfully utilised to replicate endometriosis-like lesions. However, translation of findings from animal models is hindered by asymmetries in menstrual physiology and spontaneous disease formation between humans and most other mammalian species. Therefore, much research focus of the past decade has prioritised the development of in vitro endometriosis models to elucidate disease pathophysiology and develop therapies. These models have utilised a variety of cell types including cell lines, primary cells, organoids and explant cultures. Primary cells have been isolated from both the eutopic endometrium and endometriotic lesions, as well as the peritoneum and ovaries, to generate subtype-specific models of lesion initiation and progression. Researchers have explored a variety of matrices to support three-dimensional model generation and better recapitulate the in vivo endometriosis niche. Importantly, immune cells have also been incorporated to replicate the proinflammatory endometriosis signature and investigate the role of the immune system in disease progression. More recently, advancements in bioengineering technologies have facilitated the development of complex models including microfluidics and organ-on-a-chip platforms, which allow for multiplex readouts and maximise outputs from precious patient samples. Further, multiorgan systems have been developed to explore the systemic effects of endometriosis. Despite these advancements, challenges still impede translation of model findings to tangible impacts for patients. Most notably, a lack of harmonisation in biosample collection, processing and culture methodologies means that bespoke in vitro models cannot be easily adopted or cross-validated. Overcoming bottlenecks will pave the way for adaptable, reproducible and translational models to advance understanding and expedite novel treatment development. Peritoneal endometriosis is the most common subtype of the disease, characterised by superficial lesions within the abdominal cavity. Endometriosis diagnosis is typically delayed; thus, patient samples are unsuitable to study early endometriosis formation in situ. Furthermore, endometriosis causes an array of symptoms that are not limited to the site of disease establishment. Therefore, a holistic approach is required to delineate endometriosis initiation, progression, and mechanisms of symptom manifestation. Here, we present a co-culture model of early peritoneal endometriosis and complementary organ-on-a-chip platform to explore disease related symptoms. A simple peritoneal model was generated by embedding human peritoneal fibroblasts within a Matrigel-collagen I matrix and seeding a monolayer of donor-matched mesothelial cells on top. Spontaneous basement membrane formation and secretion of tissue plasminogen activator demonstrated functional mesothelial physiology. Endometrial epithelia organoids were cocultured with matched stromal cells to form assembloids mimicking shed menstrual fragments. Assembloids adhered to the peritoneal model, simulating early endometriotic lesion formation. CD10+ stomal cells were observed to invade the peritoneal model following 10 days of culture, suggesting lesion establishment and potential to explore deep endometriosis progression. To generate a high-throughput platform to investigate endometriosis associated symptoms, an organ-on-a-chip model of the uterine wall was created. Endometrial and myometrial biopsies were dissociated to create epithelial, stromal and myometrial fractions. Cell suspensions were seeded into non-adhesive polydimethylsiloxane microfluidic devices containing 5x5 microwell arrays. Tri-cultures were established by sequential seeding of individual or combined cell fractions at various ratios. Incorporation of 5% [v/v] Matrigel improved the reproducibility of spheroids, which exhibited robust self-assembly of a stromal/smooth muscle core encased in epithelium. Functionality was confirmed by endothelin-1 induced contractility and decidual secretions in the presence of steroid hormones. This model will be invaluable in decoding the mechanisms of endometriosis (or adenomyosis/fibroid) induced symptoms including subfertility and aberrant uterine peristalsis.