A personalized functional assay that predicts clinical outcomes in embryo transfer

Objective: We aimed to develop a personalized assay based on patient-derived endometrial organoids to predict frozen embryo transfer (FET) outcomes after in vitro fertilization (IVF). Design: We collected endometrial biopsies from over 100 patients preparing for FET. The biopsies were used to grow endometrial organoids, three-dimensional models which represent a source of personalized tissue that can be grown in culture indefinitely. We combined these organoids with blastoids (standardized pluripotent stem cell-based models of the blastocyst-staged embryo) to simulate the process of implantation. Imaging and biochemical measurements of this culture system were compared with patient histories and FET outcomes. Materials and Methods: Organoids and blastoids were established from patient samples and human pluripotent stem cells, respectively, using previously described procedures. To standardize, organoid cells were grown on micropatterned substrates which precisely control the geometry of the cultures. Blastoids were then introduced. We performed confocal imaging of the resulting aggregates using staining of actin and markers of blastoid and organoid cell origin to judge the structure of integration. We also performed measurements of hCG levels in culture media. Results: We found that naive endometrial organoids repelled blastoids, while those primed with a cycle of estrogen and progesterone interacted with blastoids, causing integration of the blastoid and organoid and hCG production by the blastoid. We showed that quantitative measurements of hCG concentration from culture media were predictive of clinical outcomes. Patients with 3 or more previous unsuccessful FETs uniformly showed poor integration and hCG production while those with 0-2 previous failures showed varied outcomes that correlate with clinical success (p = 0.03). Conclusions: Our data show that this assay can diagnose recurrent implantation failure (failure to achieve a clinical or biochemical pregnancy after 3 or more embryo transfers) with specificity and sensitivity between 80 and 90 percent. Interacting patient-derived endometrial organoids with high quality blastoids represents a promising approach to personalized testing for endometrial receptivity. Support: This work was supported by Simbryo Technologies.