Background Freely accessible large language models provide health information, but their communication characteristics require evaluation. We compared information quality, transparency, readability and perceived empathy for questions about polyendocrine metabolic ovarian syndrome (PMOS; formerly PCOS).Methods Thirty-three clinician-formulated English questions using PCOS terminology were submitted once to each of three free-access configurations during 20–25 July 2026. Two clinicians rated DISCERN, EQIP, JAMA benchmarks, global quality and perceived empathy. Six readability indices were calculated. Matched comparisons used Friedman tests, paired signed-rank permutation comparisons and Holm correction.Results Across 99 archived responses, the DeepSeek-V4 configuration received higher final ratings than GPT-5.5 and Gemini 3.5 on all five clinician-rated measures (Holm-adjusted pairwise P < 0.05). Median DISCERN scores were 67.0, 62.0 and 61.0, respectively. DeepSeek-V4 had higher median ARI, GFI, FKGL and SMOG values, whereas GPT-5.5 had the highest median Coleman–Liau score. FRES showed no overall difference (adjusted P = 0.100). Initial empathy agreement was low (ICC 0.05–0.43).Conclusion Ratings differed for the specific configurations and single responses examined in this treatment-heavy question set. The findings neither establish stable model rankings nor demonstrate clinical correctness, safety or effectiveness for patients.
Background This study evaluated the independent associations of four prespecified variables readily available at the initial assessment (gestational age at preterm prelabour rupture of membranes [PPROM], admission C-reactive protein [CRP], white blood cell [WBC] count and oligohydramnios) with NICU admission in singleton pregnancies between 34 + 0 and 36 + 6 weeks. A secondary exploratory objective was to assess the discriminative performance of a multivariable logistic regression model incorporating these four variables.Methods This retrospective cohort study included 345 singleton pregnancies complicated by PPROM and managed at a tertiary care centre between January 2020 and December 2025. The four prespecified variables were entered simultaneously into a multivariable logistic regression model to evaluate their independent associations with NICU admission. Antenatal corticosteroid exposure was added in a sensitivity analysis. Model discrimination, calibration and bootstrap internal validation were assessed exploratorily.Results NICU admission occurred in 89 neonates (25.8%). Gestational age at PPROM was independently associated with NICU admission (adjusted odds ratio 0.607, 95% confidence interval 0.445–0.830; p = 0.002), corresponding to a 39.3% reduction in the odds of NICU admission per 1-week increase. Admission CRP, WBC count and oligohydramnios were not independently associated with NICU admission. The association with gestational age persisted after adjustment for antenatal corticosteroid exposure. Model discrimination was modest (apparent AUC 0.611, 95% CI 0.541–0.681; optimism-corrected AUC 0.584); after bootstrap correction, the calibration slope was 0.796 and the Brier score was 0.191.Conclusions Gestational age at PPROM was the only prespecified variable independently associated with subsequent NICU admission, whereas admission CRP, WBC count and oligohydramnios were not independently associated with NICU admission. The exploratory model showed limited overall performance and should not be used for individual-level risk prediction.
Background Although mismatch repair-deficient (dMMR) endometrial cancers are highly immunogenic and commonly responsive to immune checkpoint inhibitors, their heterogeneous outcomes indicate that dMMR status alone is insufficient to predict therapeutic benefit. The RPL22 gene – frequently mutated in microsatellite instability-high tumours – represents a candidate biomarker whose prognostic significance in dMMR endometrial cancer remains largely unexplored.Methods In this retrospective cohort study, we analysed 250 patients with immunohistochemically confirmed dMMR endometrial cancer diagnosed between 2008 and 2022. The RPL22 p.K15Rfs*5 mutation was detected by Sanger sequencing and analysed in relation to MMR protein expression pattern, histologic subtype, treatment modality, recurrence pattern, cancer-specific survival (CSS), and progression-free survival (PFS).Results RPL22 mutations were identified in 78.4% (196/250) of dMMR tumours, at a comparable frequency in MLH1-PMS2-deficient (79.2%) and MSH2-MSH6-deficient (76.1%) cases (p = 0.596), and were unrelated to FIGO stage, treatment, or recurrence pattern. RPL22-mutant tumours had significantly better CSS (p < 0.001) and PFS (p = 0.002) than wild-type tumours. MLH1-PMS2-deficient tumours showed worse CSS than MSH2-MSH6-deficient tumours (p = 0.008); within this subgroup, RPL22-mutant tumours had better CSS and PFS than wild-type tumours (p < 0.001 and p = 0.005, respectively). In aggressive histologic subtypes, RPL22 mutations were associated with better CSS (p = 0.003). In multivariable analysis including FIGO stage, RPL22 mutations remained independently associated with better CSS (HR = 0.14; 95% CI, 0.05–0.43; p = 0.001) and PFS (HR = 0.43; 95% CI, 0.23–0.81; p = 0.009).Conclusions In this hypothesis-generating retrospective cohort, the RPL22 p.K15Rfs*5 mutation identified a favourable prognostic subset within dMMR endometrial cancers, particularly among MLH1-PMS2-deficient tumours and aggressive histologic subtypes. Because treatment was not modelled and cancer-specific deaths were few, prospective validation in treatment-annotated cohorts is required before routine clinical implementation.
Background Polyendocrine metabolic ovarian syndrome (PMOS), formerly known as polycystic ovary syndrome, is characterised by hyperandrogenaemia, ovulatory dysfunction, and metabolic disturbances. Steroid 5α-reductase 2 (SRD5A2) converts testosterone to dihydrotestosterone (DHT), a potent androgen receptor agonist. However, the expression profile of SRD5A2 in the ovaries of women with PMOS and its potential contribution to hyperandrogenaemia remain unclear. The aim of this study was to investigate the expression of SRD5A2 and steroidogenic acute regulatory protein (StAR) in the ovaries of a rat PMOS model and to explore their associations with serum hormone levels.Methods Rats were randomly assigned to three groups: PMOS (n = 20), model control (dimethyl sulfoxide [DMSO]; n = 10), and blank control (BC; n = 10). The PMOS model was established via subcutaneous injection of dehydroepiandrosterone. Oestrous cycles were evaluated using vaginal cytology. Serum hormones were quantified using enzyme-linked immunosorbent assay. Protein and mRNA expression levels of SRD5A2 and StAR in ovarian tissues were evaluated using western blotting and reverse transcription polymerase chain reaction, respectively.Results Rats in the PMOS group exhibited significantly increased body weight (P < 0.05); disrupted oestrous cycles; and significantly elevated serum levels of anti-Müllerian hormone, testosterone, luteinising hormone (LH), and LH/follicle-stimulating hormone (FSH) ratio (P < 0.05). The protein and mRNA levels of SRD5A2 and StAR were significantly upregulated in the ovaries of the PMOS group compared with those of the BC and DMSO groups (P < 0.05). Ovarian volume and weight were similar among the groups, with no significant differences (P > 0.05).Conclusions SRD5A2 and StAR are highly expressed in the ovaries of rats with PMOS and are associated with hyperandrogenaemia and elevated LH secretion. These findings indicate that SRD5A2 and StAR may contribute to PMOS pathogenesis, although further studies are required to establish causality.
Objective Recurrent miscarriage (RM) is a significant reproductive health issue affecting many women of childbearing age. The objective of this study was to evaluate the value of microRNA-614 (miR-614) as a potential biomarker for RM and to elucidate its functional mechanism.Methods Differentially expressed miRNAs (DEmiRs) were first screened from the Gene Expression Omnibus (GEO) database using a bioinformatics approach. A total of 135 patients with RM and 135 healthy controls were included. The expression levels and diagnostic performance of miR-614 were evaluated by RT-qPCR and ROC curve analysis. CCK-8 and Transwell were performed to assess the impact of the miRNA on trophoblast cell proliferation, invasion, and migration. For target gene verification, TargetScan 8.0 was utilised for prediction, followed by a luciferase activity assay to confirm direct regulation by miRNA.Results Bioinformatic interrogation of the GSE73025 dataset revealed a significant down-regulation of miR-614. Plasma miR-614 was significantly downregulated in RM patients. The area under the ROC curve (AUC) for miR-614 in distinguishing RM patients from controls was 0.865 (95% CI: 0.823–0.907, sensitivity = 77.78%, specificity = 77.78%). The AUC for predicting adverse pregnancy outcomes was 0.790 (95% CI: 0.708–0.870, sensitivity = 71.08%, specificity = 82.69%). Transfection with miR-614 mimics decreased the proliferation, invasion, and migration of HTR-8 and JEG3 cells. miR-614 Directly targets FOXD1 and negatively modulates its mRNA expression to drive RM progression.Conclusion In RM patients, miR-614 is downregulated, leading to increased FOXD1 expression. miR-614 mimics suppresses trophoblast proliferation, migration, and invasion by directly targeting FOXD1. Thus, miR-614 downregulation contributes to RM pathogenesis by directly targeting FOXD1, thereby upregulating FOXD1 and inhibiting the proliferation, invasion, and migration of trophoblast cells.
