Machine Learning Applied to Pre-Operative Computed-Tomography-Based Radiomic Features Can Accurately Differentiate Uterine Leiomyoma from Leiomyosarcoma: A Pilot Study.

BACKGROUND: The accurate discrimination of uterine leiomyosarcomas and leiomyomas in a pre-operative setting remains a current challenge. To date, the diagnosis is made by a pathologist on the excised tumor. The aim of this study was to develop a machine learning algorithm using radiomic data extracted from contrast-enhanced computed tomography (CECT) images that could accurately distinguish leiomyosarcomas from leiomyomas. METHODS: Pre-operative CECT images from patients submitted to surgery with a histological diagnosis of leiomyoma or leiomyosarcoma were used for the region of interest identification and radiomic feature extraction. Feature extraction was conducted using the PyRadiomics library, and three feature selection methods combined with the general linear model (GLM), random forest (RF), and support vector machine (SVM) classifiers were built, trained, and tested for the binary classification task (malignant vs. benign). In parallel, radiologists assessed the diagnosis with or without clinical data. RESULTS: A total of 30 patients with leiomyosarcoma (mean age 59 years) and 35 patients with leiomyoma (mean age 48 years) were included in the study, comprising 30 and 51 lesions, respectively. Out of nine machine learning models, the three feature selection methods combined with the GLM and RF classifiers showed good performances, with predicted area under the curve (AUC), sensitivity, and specificity ranging from 0.78 to 0.97, from 0.78 to 1.00, and from 0.67 to 0.93, respectively, when compared to the results obtained from experienced radiologists when blinded to the clinical profile (AUC = 0.73 95%CI = 0.62-0.84), as well as when the clinical data were consulted (AUC = 0.75 95%CI = 0.65-0.85). CONCLUSIONS: CECT images integrated with radiomics have great potential in differentiating uterine leiomyomas from leiomyosarcomas. Such a tool can be used to mitigate the risks of eventual surgical spread in the case of leiomyosarcoma and allow for safer fertility-sparing treatment in patients with benign uterine lesions.

Molecular Characterization of Advanced-Stage Melanomas in Clinical Practice Using a Laboratory-Developed Next-Generation Sequencing Panel.

Cutaneous melanoma is one of the most lethal tumors among skin cancers, characterized by complex genetic and molecular alterations that result in uncontrolled cell proliferation and metastatic spread. Next-generation sequencing (NGS) enables the simultaneous examination of numerous genes, making this molecular technique essential for melanoma diagnosis, prognostic stratification, and therapy planning. Herein, we present the experience with our laboratory-designed NGS panel for the routine assessment of advanced-stage melanoma. A total of 260 specimens of advanced-stage melanomas were evaluated utilizing a laboratory-developed multi-gene NGS panel, which allowed the investigation of 229 amplicons in 25 oncogene/oncosuppressor genes. The NGS panel proved to be a reliable tool, failing to produce results in only 1.2% of the samples tested. BRAF and TERT were the two more commonly altered genes in 44.0% and 59.9% of samples, respectively. In 59.3% of the mutated cases, at least two concomitant variants were detected. In eight cases, both primary lesion and metastatic disease were analyzed by NGS. In all specimens (8/8, 100%), a perfect concordance in variants harbored by the primary and recurrence lesions was observed. Finally, this study described the validity of a laboratory-developed multi-gene NGS panel built specifically for advanced-stage melanomas in ordinary clinical practice.

Unveiling intrinsic material and extrinsic pinning dimensionality in superconductors: Why Fe(Se,Te) is able to mimic YBCO.

Discovery of iron-based superconductors paved the way to a competitor of high-temperature superconductors, easier to produce, better performing in high fields, and promising to be less expensive. Critical parameters are investigated by resistivity measurements as a function of temperature, field, and angle R(T,H,θ). This work presents a deep analysis of H-θ phase diagram of PLD-processed Fe(Se,Te) superconducting films, thus revealing material and pinning anisotropy at once. By selecting different thresholds along the R(T,H,θ) curves, all possible regimes emerge. Surprisingly, anisotropy arises moving from the upper critical field toward the irreversibility line: gradually a non-monotonous transition from 3D to 2D, and backward to 3D occurs. Although Fe(Se,Te) appears as a 3D superconductor, its anisotropic pinning landscape shows up similarities with an intrinsic layered superconductor and Fe(Se,Te) definitively mimics YBCO. We propose a general method to disentangle, in any other superconductor, material dimensionality and pinning anisotropy that are key constraints for applications.

Circulating miRNAs as Novel Clinical Biomarkers in Temporal Lobe Epilepsy.

Temporal lobe epilepsy (TLE) represents the most common form of refractory focal epilepsy. The identification of innovative clinical biomarkers capable of categorizing patients with TLE, allowing for improved treatment and outcomes, still represents an unmet need. Circulating microRNAs (c-miRNAs) are short non-coding RNAs detectable in body fluids, which play crucial roles in the regulation of gene expression. Their characteristics, including extracellular stability, detectability through non-invasive methods, and responsiveness to pathological changes and/or therapeutic interventions, make them promising candidate biomarkers in various disease settings. Recent research has investigated c-miRNAs in various bodily fluids, including serum, plasma, and cerebrospinal fluid, of TLE patients. Despite some discrepancies in methodologies, cohort composition, and normalization strategies, a common dysregulated signature of c-miRNAs has emerged across different studies, providing the basis for using c-miRNAs as novel biomarkers for TLE patient management.

Accuracy of World Health Organisation-grade parameters (necrosis and mitotic activity) and foci of vascular invasion in predicting prognosis of papillary thyroid carcinoma. A case-control validation study.

AIMS: Tumour necrosis and/or increased mitoses define high-grade papillary thyroid carcinoma (PTC). It is unclear whether angioinvasion is prognostic for PTC. Cut-offs at five or more mitoses/2 mm2 and four or more angioinvasive foci have been empirically defined based upon data from all forms of aggressive non-anaplastic thyroid carcinomas. Performance of tumour necrosis, mitoses and vascular invasion in predicting distant metastases when specifically applied to PTC is undefined. METHODS: We analysed 50 consecutive PTC cases with distant metastases (DM-PTC): 16 synchronous and 34 metachronous. A total of 108 non-metastatic PTC (N-DM-PTC, 15.0-year median follow-up) were used as controls. Invasive encapsulated follicular variant PTC was excluded. Necrosis, mitoses and angioinvasion were quantified. Receiver operating characteristics (ROC) and area under the curve (AUC) analyses determined best sensitivity and specificity cut-offs predictive of distant metastases. RESULTS: Metastases correlated with necrosis (any extent = 43.8% all DM-PTC, 53.1% metachronous DM-PTC versus 5% N-DM-PTC; P < 0.001), mitoses (P < 0.001) and angioinvasion (P < 0.001). Mitoses at five or more per 2 mm2 was the best cut-off correlating with distant metastases: sensitivity/specificity 42.9%/97.2% all DM-PTC (AUC = 0.78), 18.8%/97.2% synchronous DM-PTC (AUC = 0.63), 54.6%/97.2% metachronous DM-PTC (AUC = 0.85). Angioinvasive foci at five or more was the best cut-off correlating with distant metastases: sensitivity/specificity 36.2%/91.7% all DM-PTC (AUC = 0.75), 25%/91.7% synchronous DM-PTC (AUC = 0.79) and 41.9%/91.7% metachronous DM-PTC (AUC = 0.73). Positive/negative predictive values (PPV/NPV) were: necrosis 22.6%/98.2%; five or more mitoses 32.3%/98.2%; five or more angioinvasive foci 11.8%/97.9%. After multivariable analysis, only necrosis and mitotic activity remained associated with DM-PTC. CONCLUSION: Our data strongly support PTC grading, statistically validating World Health Organisation (WHO) criteria to identify poor prognosis PTC. Angioinvasion is not an independent predictor of DM-PTC.

Primary vulvar adenocarcinoma of intestinal type: Report of two cases showing molecular similarity with colorectal adenocarcinoma.

Primary vulvar adenocarcinoma is a particularly rare tumor with poorly understood histogenesis and unclear clinical characteristics and prognosis. Vulvar adenocarcinoma of intestinal type (VAIt) is a very uncommon subtype of primary vulvar adenocarcinoma and only 27 cases have been described in the literature in the past. Of these cases, two have been described as human papillomavirus (HPV)-associated VAIt. The current report presents two additional cases of primary VAIt showing variants in the KRAS, TP53, and DPYD genes and no evidence of HPV DNA by real-time polymerase chain reaction (RT-PCR). Next-generation sequencing (NGS) revealed TP53 pathogenic variants in both cases, but only one case had aberrant p53 protein immunohistochemical characteristics. KRAS and DPYD mutations were identified separately in the two cases. Due to their capacity to imitate the spread of more prevalent gastrointestinal carcinomas, these tumors may present diagnostic issues. Additional cases can contribute to a better understanding of the pathophysiology and prognosis of VAIt.

Novel prognostic nomogram for predicting recurrence-free survival in medullary thyroid carcinoma.

AIMS: Recently, there have been attempts to improve prognostication and therefore better guide treatment for patients with medullary thyroid carcinoma (MTC). In 2022, the International MTC Grading System (IMTCGS) was developed and validated using a multi-institutional cohort of 327 patients. The aim of the current study was to build upon the findings of the IMTCGS to develop and validate a prognostic nomogram to predict recurrence-free survival (RFS) in MTC. METHODS AND RESULTS: Data from 300 patients with MTC from five centres across the USA, Europe, and Australia were used to develop a prognostic nomogram that included the following variables: age, sex, AJCC stage, tumour size, mitotic count, necrosis, Ki67 index, lymphovascular invasion, microscopic extrathyroidal extension, and margin status. A process of 10-fold cross-validation was used to optimize the model's performance. To assess discrimination and calibration, the area-under-the-curve (AUC) of a receiver operating characteristic (ROC) curve, concordance-index (C-index), and dissimilarity index (D-index) were calculated. Finally, the model was externally validated using a separate cohort of 87 MTC patients. The model demonstrated very strong performance, with an AUC of 0.94, a C-index of 0.876, and a D-index of 19.06. When applied to the external validation cohort, the model had an AUC of 0.9. CONCLUSIONS: Using well-established clinicopathological prognostic variables, we developed and externally validated a robust multivariate prediction model for RFS in patients with resected MTC. The model demonstrates excellent predictive capability and may help guide decisions on patient management. The nomogram is freely available online at https://nomograms.shinyapps.io/MTC_ML_DFS/.

Impact of micronutrients and nutraceuticals on cognitive function and performance in Alzheimer's disease.

Alzheimer's disease (AD) is a major global health problem today and is the most common form of dementia. AD is characterized by the formation of ß-amyloid (Aß) plaques and neurofibrillary clusters, leading to decreased brain acetylcholine levels in the brain. Another mechanism underlying the pathogenesis of AD is the abnormal phosphorylation of tau protein that accumulates at the level of neurofibrillary aggregates, and the areas most affected by this pathological process are usually the cholinergic neurons in cortical, subcortical, and hippocampal areas. These effects result in decreased cognitive function, brain atrophy, and neuronal death. Malnutrition and weight loss are the most frequent manifestations of AD, and these are also associated with greater cognitive decline. Several studies have confirmed that a balanced low-calorie diet and proper nutritional intake may be considered important factors in counteracting or slowing the progression of AD, whereas a high-fat or hypercholesterolemic diet predisposes to an increased risk of developing AD. Especially, fruits, vegetables, antioxidants, vitamins, polyunsaturated fatty acids, and micronutrients supplementation exert positive effects on aging-related changes in the brain due to their antioxidant, anti-inflammatory, and radical scavenging properties. The purpose of this review is to summarize some possible nutritional factors that may contribute to the progression or prevention of AD, understand the role that nutrition plays in the formation of Aß plaques typical of this neurodegenerative disease, to identify some potential therapeutic strategies that may involve some natural compounds, in delaying the progression of the disease.

miRNA levels are associated with body mass index in endometrial cancer and may have implications for therapy.

Endometrial cancer (EC) is the most prevalent gynecological cancer in high-income countries. Its incidence is skyrocketing due to the increase in risk factors such as obesity, which represents a true pandemic. This study aimed to evaluate microRNA (miRNA) expression in obesity-related EC to identify potential associations between this specific cancer type and obesity. miRNA levels were analyzed in 84 EC patients stratified based on body mass index (BMI; ≥30 or <30) and nine noncancer women with obesity. The data were further tested in The Cancer Genome Atlas (TCGA) cohort, including 384 EC patients, 235 with BMI ≥30 and 149 with BMI <30. Prediction of miRNA targets and analysis of their expression were also performed to identify the potential epigenetic networks involved in obesity modulation. In the EC cohort, BMI ≥30 was significantly associated with 11 deregulated miRNAs. The topmost deregulated miRNAs were first analyzed in 84 EC samples by single miRNA assay and then tested in the TCGA dataset. This independent validation provided further confirmation about the significant difference of three miRNAs (miR-199a-5p, miR-449a, miR-449b-5p) in normal-weight EC patients versus EC patients with obesity, resulting significantly higher expressed in the latter. Moreover, the three miRNAs were significantly correlated with grade, histological type, and overall survival. Analysis of their target genes revealed that these miRNAs may regulate obesity-related pathways. In conclusion, we identified specific miRNAs associated with BMI that are potentially involved in modulating obesity-related pathways and that may provide novel implications for the clinical management of obese EC patients.

Association of the Genomic Profile of Medullary Thyroid Carcinoma with Tumor Characteristics and Clinical Outcomes in an International Multicenter Study.

Purpose: The prognostic importance of RET and RAS mutations and their relationship to clinicopathologic parameters and outcomes in medullary thyroid carcinoma (MTC) need to be clarified. Experimental Design: A multicenter retrospective cohort study was performed utilizing data from 290 patients with MTC. The molecular profile was determined and associations were examined with clinicopathologic data and outcomes. Results: RET germ line mutations were detected in 40 patients (16.3%). Somatic RET and RAS mutations occurred in 135 (46.9%) and 57 (19.8%) patients, respectively. RETM918T was the most common somatic RET mutation (n = 75). RET somatic mutations were associated with male sex, larger tumor size, advanced American Joint Committee Cancer (AJCC) stage, vascular invasion, and high International Medullary Thyroid Carcinoma Grading System (IMTCGS) grade. When compared with other RET somatic mutations, RETM918T was associated with younger age, AJCC (eighth edition) IV, vascular invasion, extrathyroidal extension, and positive margins. RET somatic or germ line mutations were significantly associated with reduced distant metastasis-free survival on univariate analysis, but there were no significant independent associations on multivariable analysis, after adjusting for tumor grade and stage. There were no significant differences in outcomes between RET somatic and RET germ line mutations, or between RETM918T and other RET mutations. Other recurrent molecular alterations included TP53 (4.2%), ARID2 (2.9%), SETD2 (2.9%), KMT2A (2.9%), and KMT2C (2.9%). Among them, TP53 mutations were associated with decreased overall survival (OS) and disease-specific survival (DSS), independently of tumor grade and AJCC stage. Conclusions: RET somatic mutations were associated with high-grade, aggressive primary tumor characteristics, and decreased distant metastatic-free survival but this relationship was not significant after accounting for tumor grade and disease stage. RETM918T was associated with aggressive primary tumors but was not independently associated with clinical outcomes. TP53 mutation may represent an adverse molecular event associated with decreased OS and DSS in MTC, but its prognostic value needs to be confirmed in future studies.

Cenobamate enhances the anticonvulsant effect of other antiseizure medications in the DBA/2 mouse model of reflex epilepsy.

Clinical studies documented that cenobamate (CNB) has a marked efficacy compared to other antiseizure medications (ASMs) in reducing focal seizures. To date, different aspects of CNB need to be clarified, including its efficacy against generalized seizures. Similarly, the pattern of drug-drug interactions between CNB and other ASMs also compels further investigation. This study aimed to detect the role of CNB on generalized seizures using the DBA/2 mouse model. We have also studied the effects of an adjunctive CNB treatment on the antiseizure properties of some ASMs against reflex seizures. The effects of this adjunctive treatment on motor performance, body temperature, and brain levels of ASMs were also evaluated. CNB was able to antagonize seizures in DBA/2 mice. CNB, at 5 mg/kg, enhanced the antiseizure activity of ASMs, such as diazepam, clobazam, levetiracetam, perampanel, phenobarbital, topiramate, and valproate. No synergistic effects were observed when CNB was co-administered with some Na+ channel blockers. The increase in antiseizure activity was associated with a comparable intensification in motor impairment; however, the therapeutic index of combined treatment of ASMs with CNB was more favorable than the combination with vehicle except for carbamazepine, phenytoin, and oxcarbazepine. Since CNB did not significantly influence the brain levels of the ASMs studied, we suggest that pharmacokinetic interactions seem not probable. Overall, this study shows the ability of CNB to counteract generalized reflex seizures in mice. Moreover, our data documented an evident synergistic antiseizure effect for the combination of CNB with ASMs including phenobarbital, benzodiazepines, valproate, perampanel, topiramate, and levetiracetam.

Molecular pathology of endocrine gland tumors: genetic alterations and clinicopathologic relevance.

Tumors of the endocrine glands are common. Knowledge of their molecular pathology has greatly advanced in the recent past. This review covers the main molecular alterations of tumors of the anterior pituitary, thyroid and parathyroid glands, adrenal cortex, and adrenal medulla and paraganglia. All endocrine gland tumors enjoy a robust correlation between genotype and phenotype. High-throughput molecular analysis demonstrates that endocrine gland tumors can be grouped into molecular groups that are relevant from both pathologic and clinical point of views. In this review, genetic alterations have been discussed and tabulated with respect to their molecular pathogenetic role and clinicopathologic implications, addressing the use of molecular biomarkers for the purpose of diagnosis and prognosis and predicting response to molecular therapy. Hereditary conditions that play a key role in determining predisposition to many types of endocrine tumors are also discussed.

Microbiome preterm birth DREAM challenge: Crowdsourcing machine learning approaches to advance preterm birth research.

Every year, 11% of infants are born preterm with significant health consequences, with the vaginal microbiome a risk factor for preterm birth. We crowdsource models to predict (1) preterm birth (PTB; <37 weeks) or (2) early preterm birth (ePTB; <32 weeks) from 9 vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from public raw data via phylogenetic harmonization. The predictive models are validated on two independent unpublished datasets representing 331 samples from 148 pregnant individuals. The top-performing models (among 148 and 121 submissions from 318 teams) achieve area under the receiver operator characteristic (AUROC) curve scores of 0.69 and 0.87 predicting PTB and ePTB, respectively. Alpha diversity, VALENCIA community state types, and composition are important features in the top-performing models, most of which are tree-based methods. This work is a model for translation of microbiome data into clinically relevant predictive models and to better understand preterm birth.

Point-Contact Spectroscopy in Bulk Samples of Electron-Doped Cuprate Superconductors.

Point-contact spectroscopy was performed on bulk samples of electron-doped high temperature superconductor Nd2-xCexCuO4-δ. The samples were characterized using X-ray diffraction and scanning electron microscopy equipped with a wavelength-dispersive spectrometer and an electron backscatter diffraction detector. Samples with Ce content x = 0.15 showed the absence of spurious phases and randomly oriented grains, most of which had dimensions of approximately 220 µm2. The low-bias spectra in the tunneling regime, i.e., high-transparency interface, exhibited a gap feature at about ±5 meV and no zero-bias conductance, despite the random oriented grains investigated within our bulk samples, consistent with most of the literature data on oriented samples. High-bias conductance was also measured in order to obtain information on the properties of the barrier. A V-shape was observed in some cases, instead of the parabolic behavior expected for tunnel junctions.

Prognostic Role of p53 Immunohistochemical Status in invasive Breast Cancer. A Retrospective Review of 1387 Cases With Luminal-Like/Her2 Negative Breast Tumors.

PURPOSE: The aim of this study was to evaluate the prognostic role of p53 immunohistochemical (IHC) expression in a large cohort of patients with hormone receptors (HR)-positive/Her2-negative primary invasive breast cancer. METHODS: Retrospective review of consecutive cases treated at our Breast Unit between 2003 and 2013. Patients were divided into 3 subgroups based on p53 IHC expression: null (0%), low (0.1%-49%), and high (≥50%) p53 expression. RESULTS: A total of 1387 patients were included in the study with a median follow-up of 86 months. After adjusting for age, size, node status, lymphovascular invasion, progesterone, and Ki67 expression, only null p53 immunophenotype was associated with worse disease-free survival (DFS) (OR 1.74, 95% IC, 1.11-2.71, P = .015) and distant recurrence-free survival (DRFS) (OR 1.73, 95% IC, 1.04-2.90, P = .036). Null p53 impacted significantly DFS and DRFS also in patients with early breast cancer. CONCLUSIONS: p53 IHC expression affects survival and, thus can be a valuable tool in the management of patients with HR-positive/Her2-negative breast cancer.

Ki67 proliferation index in medullary thyroid carcinoma: a comparative study of multiple counting methods and validation of image analysis and deep learning platforms.

AIMS: The International Medullary Thyroid Carcinoma Grading System, introduced in 2022, mandates evaluation of the Ki67 proliferation index to assign a histological grade for medullary thyroid carcinoma. However, manual counting remains a tedious and time-consuming task. METHODS AND RESULTS: We aimed to evaluate the performance of three other counting techniques for the Ki67 index, eyeballing by a trained experienced investigator, a machine learning-based deep learning algorithm (DeepLIIF) and an image analysis software with internal thresholding compared to the gold standard manual counting in a large cohort of 260 primarily resected medullary thyroid carcinoma. The Ki67 proliferation index generated by all three methods correlate near-perfectly with the manual Ki67 index, with kappa values ranging from 0.884 to 0.979 and interclass correlation coefficients ranging from 0.969 to 0.983. Discrepant Ki67 results were only observed in cases with borderline manual Ki67 readings, ranging from 3 to 7%. Medullary thyroid carcinomas with a high Ki67 index (≥ 5%) determined using any of the four methods were associated with significantly decreased disease-specific survival and distant metastasis-free survival. CONCLUSIONS: We herein validate a machine learning-based deep-learning platform and an image analysis software with internal thresholding to generate accurate automatic Ki67 proliferation indices in medullary thyroid carcinoma. Manual Ki67 count remains useful when facing a tumour with a borderline Ki67 proliferation index of 3-7%. In daily practice, validation of alternative evaluation methods for the Ki67 index in MTC is required prior to implementation.

Audiogenic epileptic DBA/2 mice strain as a model of genetic reflex seizures and SUDEP.

Epilepsy is a chronic neurological disease characterized by abnormal brain activity, which results in repeated spontaneous seizures. Sudden unexpected death in epilepsy (SUDEP) is the leading cause of seizure-related premature death, particularly in drug-resistant epilepsy patients. The etiology of SUDEP is a structural injury to the brain that is not fully understood, but it is frequently associated with poorly controlled and repeated generalized tonic-clonic seizures (GTCSs) that cause cardiorespiratory and autonomic dysfunctions, indicating the involvement of the brainstem. Both respiratory and cardiac abnormalities have been observed in SUDEP, but not much progress has been made in their prevention. Owing to the complexity of SUDEP, experimental animal models have been used to investigate cardiac and/or respiratory dysregulation due to or associated with epileptic seizures that may contribute to death in humans. Numerous rodent models, especially mouse models, have been developed to better understand epilepsy and SUDEP physiopathology. This review synthesizes the current knowledge about dilute brown agouti coat color (DBA/2) mice as a possible SUDEP model because respiratory arrest (RA) and sudden death induced by audiogenic generalized seizures (AGSs) have been observed in these animals. Respiratory/cardiac dysfunction, brainstem arousal system dysfunction, and alteration of the neurotransmitter systems, which are observed in human SUDEP, have also been observed in these mice. In particular, serotonin (5-HT) alteration and adenosine neurotransmission appear to contribute to not only the pathophysiological mechanisms of medication but also seizure-related respiratory dysfunctions in this animal model. These neurotransmitter systems could be the relevant targets for medication development for chronic epilepsy and SUDEP prevention. We reviewed data on AGSs in DBA/2 mice and the relevance of this model of generalized tonic-clonic epilepsy to human SUDEP. Furthermore, the advantages of using this strain prone to AGSs for the identification of possible new therapeutic targets and treatment options have also been assessed.

Deciphering a shared transcriptomic regulation and the relative contribution of each regulator type through endometrial gene expression signatures.

BACKGORUND: While various endometrial biomarkers have been characterized at the transcriptomic and functional level, there is generally a poor overlap among studies, making it unclear to what extent their upstream regulators (e.g., ovarian hormones, transcription factors (TFs) and microRNAs (miRNAs)) realistically contribute to menstrual cycle progression and function. Unmasking the intricacies of the molecular interactions in the endometrium from a novel systemic point of view will help gain a more accurate perspective of endometrial regulation and a better explanation the molecular etiology of endometrial-factor infertility. METHODS: An in-silico analysis was carried out to identify which regulators consistently target the gene biomarkers proposed in studies related to endometrial progression and implantation failure (19 gene lists/signatures were included). The roles of these regulators, and of genes related to progesterone and estrogens, were then analysed in transcriptomic datasets compiled from samples collected throughout the menstrual cycle (n = 129), and the expression of selected TFs were prospectively validated in an independent cohort of healthy participants (n = 19). RESULTS: A total of 3,608 distinct genes from the 19 gene lists were associated with endometrial progression and implantation failure. The lists' regulation was significantly favoured by TFs (89% (17/19) of gene lists) and progesterone (47% (8 /19) of gene lists), rather than miRNAs (5% (1/19) of gene lists) or estrogen (0% (0/19) of gene lists), respectively (FDR < 0.05). Exceptionally, two gene lists that were previously associated with implantation failure and unexplained infertility were less hormone-dependent, but primarily regulated by estrogen. Although endometrial progression genes were mainly targeted by hormones rather than non-hormonal contributors (odds ratio = 91.94, FDR < 0.05), we identified 311 TFs and 595 miRNAs not previously associated with ovarian hormones. We highlight CTCF, GATA6, hsa-miR-15a-5p, hsa-miR-218-5p, hsa-miR-107, hsa-miR-103a-3p, and hsa-miR-128-3p, as overlapping novel master regulators of endometrial function. The gene expression changes of selected regulators throughout the menstrual cycle (FDR < 0.05), dually validated in-silico and through endometrial biopsies, corroborated their potential regulatory roles in the endometrium. CONCLUSIONS: This study revealed novel hormonal and non-hormonal regulators and their relative contributions to endometrial progression and pathology, providing new leads for the potential causes of endometrial-factor infertility.

A Radiomic-Based Machine Learning Model Predicts Endometrial Cancer Recurrence Using Preoperative CT Radiomic Features: A Pilot Study.

BACKGROUND: Current prognostic models lack the use of pre-operative CT images to predict recurrence in endometrial cancer (EC) patients. Our study aimed to investigate the potential of radiomic features extracted from pre-surgical CT scans to accurately predict disease-free survival (DFS) among EC patients. METHODS: Contrast-Enhanced CT (CE-CT) scans from 81 EC cases were used to extract the radiomic features from semi-automatically contoured volumes of interest. We employed a 10-fold cross-validation approach with a 6:4 training to test set and utilized data augmentation and balancing techniques. Univariate analysis was applied for feature reduction leading to the development of three distinct machine learning (ML) models for the prediction of DFS: LASSO-Cox, CoxBoost and Random Forest (RFsrc). RESULTS: In the training set, the ML models demonstrated AUCs ranging from 0.92 to 0.93, sensitivities from 0.96 to 1.00 and specificities from 0.77 to 0.89. In the test set, AUCs ranged from 0.86 to 0.90, sensitivities from 0.89 to 1.00 and specificities from 0.73 to 0.90. Patients classified as having a high recurrence risk prediction by ML models exhibited significantly worse DSF (p-value < 0.001) across all models. CONCLUSIONS: Our findings demonstrate the potential of radiomics in predicting EC recurrence. While further validation studies are needed, our results underscore the promising role of radiomics in forecasting EC outcomes.