A combination of immune cell types identified through ensemble machine learning strategy detects altered profile in recurrent pregnancy loss: a pilot study

Risk prediction model based on machine learning for predicting miscarriage among pregnant patients with immune abnormalities

Introduction: It is known that patients with immune-abnormal co-pregnancies are at a higher risk of adverse pregnancy outcomes. Traditional pregnancy risk management systems have poor prediction abilities for adverse pregnancy outcomes in such patients, with many limitations in clinical application. In this study, we will use machine learning to screen high-risk factors for miscarriage and develop a miscarriage risk prediction model for patients with immune-abnormal pregnancies. This model aims to provide an adjunctive tool for the clinical identification of patients at high risk of miscarriage and to allow for active intervention to reduce adverse pregnancy outcomes. Methods: Patients with immune-abnormal pregnancies attending Sichuan Provincial People's Hospital were collected through electronic medical records (EMR). The data were divided into a training set and a test set in an 8:2 ratio. Comparisons were made to evaluate the performance of traditional pregnancy risk assessment tools for clinical applications. This analysis involved assessing the cost-benefit of clinical treatment, evaluating the model's performance, and determining its economic value. Data sampling methods, feature screening, and machine learning algorithms were utilized to develop predictive models. These models were internally validated using 10-fold cross-validation for the training set and externally validated using bootstrapping for the test set. Model performance was assessed by the area under the characteristic curve (AUC). Based on the best parameters, a predictive model for miscarriage risk was developed, and the SHapley additive expansion (SHAP) method was used to assess the best model feature contribution. Results: A total of 565 patients were included in this study on machine learning-based models for predicting the risk of miscarriage in patients with immune-abnormal pregnancies. Twenty-eight risk warning models were developed, and the predictive model constructed using XGBoost demonstrated the best performance with an AUC of 0.9209. The SHAP analysis of the best model highlighted the total number of medications, as well as the use of aspirin and low molecular weight heparin, as significant influencing factors. The implementation of the pregnancy risk scoring rules resulted in accuracy, precision, and F1 scores of 0.3009, 0.1663, and 0.2852, respectively. The economic evaluation showed a saving of ¥7,485,865.7 due to the model. Conclusion: The predictive model developed in this study performed well in estimating the risk of miscarriage in patients with immune-abnormal pregnancies. The findings of the model interpretation identified the total number of medications and the use of other medications during pregnancy as key factors in the early warning model for miscarriage risk. This provides an important basis for early risk assessment and intervention in immune-abnormal pregnancies. The predictive model developed in this study demonstrated better risk prediction performance than the Pregnancy Risk Management System (PRMS) and also demonstrated economic value. Therefore, miscarriage risk prediction in patients with immune-abnormal pregnancies may be the most cost-effective management method.

Biopsy-free profiling of the uterine immune system in patients with recurrent pregnancy loss and unexplained infertility

RESEARCH QUESTION

What are the differences in menstrual blood lymphocytes between controls, patients with recurrent pregnancy loss (RPL) and patients with unexplained infertility (uINF)?

DESIGN

Prospective study including 46 healthy controls, 28 RPL and 11 uINF patients. A feasibility study compared lymphocyte compositions of endometrial biopsies and menstrual blood collected during the first 48 h of menstruation in seven controls. In all patients, peripheral and menstrual blood from the first and subsequent 24 h were analysed separately by flow cytometry, focusing on the main lymphocyte populations and natural killer (NK) cell subsets.

RESULTS

The first 24 h of menstrual blood resembles the uterine immune milieu as tested by endometrial biopsy. RPL patients showed significantly higher menstrual blood CD56+ NK cell numbers than controls (mean ± SD: 31.13 ± 7.52% versus 36.73 ± 5.4%, P = 0.002). Menstrual blood CD56dimCD16bright NK cells within the CD56+ NK cell population were decreased in RPL (16.34 ± 14.65%, P = 0.011) and uINF (15.7 ± 5.91%, P = 0.02) patients versus control (20.42 ± 11.53%). uINF patients had the lowest menstrual blood CD3+ T cell counts (38.81 ± 5.04%, control versus uINF: P = 0.01) and cytotoxicity receptors NKp46 and NKG2D on CD56brightCD16dim cells were higher in uINF (68.12 ± 11.84%, P = 0.006; 45.99 ± 13.83%, P = 0.01, respectively) and RPL (NKp46: 66.21 ± 15.36%, P = 0.009) patients versus controls. RPL and uINF patients had higher peripheral CD56+ NK cell counts versus controls (11.42 ± 4.05%, P = 0.021; 12.86 ± 4.29%, P = 0.009 versus 8.4 ± 3.5%).

CONCLUSIONS

Compared with controls, RPL and uINF patients had a different menstrual blood-NK-subtype profile, indicating an altered cytotoxicity. In future studies, this non-invasive analysis might enable identification and monitoring of patients receiving immunomodulatory medications.

Correlation of peripheral blood and endometrial immunophenotyping in ART: is peripheral blood sampling useful?

PURPOSE

Using a comprehensive flow cytometric panel, simultaneously obtained mid-luteal immunophenotypes from peripheral blood and endometrium were compared and values correlated. Is a peripheral blood evaluation of reproductive immunophenotype status meritorious relative to local endometrial evaluation to directly assess the peri-implantation environment?

METHODS

Fifty-five patients had a mid-luteal biopsy to assess the local endometrial immunophenotype, while simultaneously providing a peripheral blood sample for analysis. Both samples were immediately assessed using a comprehensive multi-parameter panel, and lymphocyte subpopulations were described and compared.

RESULTS

Distinct lymphocyte proportions and percentage differences were noted across the two compartments, confirming the hypothesis that they are distinct environments. The ratio of CD4 + to CD8 + T cells were reversed between the two compartments, as were Th1 and Th2-type CD4 + T cell ratios. Despite these differences, some direct relationships were noted. Positive Pearson correlations were found between the levels of CD57 + expressing natural killer cells, CD3 + NK-T cells and CD4 + Th1 cells in both compartments.

CONCLUSIONS

Flow cytometric evaluation provides a rapid and objective analysis of lymphocyte subpopulations. Endometrial biopsies have become the gold standard technique to assess the uterine immunophenotype in adverse reproductive outcome, but there may still a place for peripheral blood evaluation in this context. The findings demonstrate significant variations in cellular proportions across the two regions, but some positive correlations are present. Immunological assessment of these specific peripheral blood lymphocyte subtypes may provide insight into patients with potential alterations of the uterine immune environment, without the risks and inconveniences associated with an invasive procedure.

ID3 mediates BMP2-induced downregulation of ICAM1 expression in human endometiral stromal cells and decidual cells

Recurrent pregnancy loss (RPL) remains an unsolved problem in obstetrics and gynecology, and up to 50% of RPL cases are unexplained. Unexplained RPL (uRPL) is widely considered to be related to an aberrant endometrial microenvironment. BMP2 is an important factor involved in endometrial decidualization and embryo implantation, and intercellular adhesion molecule 1 (ICAM1) is a critical inflammatory regulator in the endometrium. In this study, we found that endometrial samples obtained from Unexplained RPL patients have significantly lower BMP2 and higher ICAM1 levels than fertile controls. For further research on the relationship between BMP2 and ICAM1 and the potential molecular mechanisms in Unexplained RPL, immortalized human endometrial stromal cells (HESCs) and primary human decidual stromal cells (HDSCs) were used as study models. Our results showed that BMP2 significantly decreased ICAM1 expression by upregulating DNA-binding protein inhibitor 3 (ID3) in both HESCs and HDSCs. Using kinase receptor inhibitors (dorsomorphin homolog 1 (DMH-1) and dorsomorphin) and siRNA transfection, it has been found that the upregulation of ID3 and the following downregulation of ICAM1 induced by BMP2 is regulated through the ALK3-SMAD4 signaling pathway. This research gives a hint of a novel mechanism by which BMP2 regulates ICAM1 in the human endometrium, which provides insights into potential therapeutics for unexplained RPL.