Non-invasive prenatal testing (NIPT) based on fetal free DNA is a non-invasive technique to screen for common fetal aneuploidies by analyzing cell-free fetal DNA (cffDNA) in the peripheral blood of pregnant women. This technique has opened a new era of prenatal screening for its high safety and reliability. In recent years, it has been shown that NIPT can not only screen for fetal aneuploidies, but may also reveal maternal genomic abnormalities. The incidental detection of maternal tumors has aroused widespread concern in the clinical settings. The aim of this review is to systematically summarize the research progress of NIPT technique in incidental detection of maternal tumors, and to discuss its clinical significance, technical challenges, and future development direction. It has been found that multiple chromosome aneuploidies (MCAs) in NIPT detection is one of the important biomarkers suggesting occult maternal malignant tumors. In this paper, the relevant progress of NIPT technique in the incidental discovery of maternal tumors were reviewed in order to provide a reference for individualized and standardized application of NIPT technique in maternal health monitoring.
Humans ; Female ; Pregnancy ; Cell-Free Nucleic Acids/blood* ; Prenatal Diagnosis/methods* ; Incidental Findings ; Neoplasms/genetics* ; Noninvasive Prenatal Testing/methods* ; Aneuploidy ; Fetus/metabolism*

OBJECTIVE: To assess the clinical value of non-invasive prenatal testing (NIPT) for identifying maternal malignant tumors. METHODS: A retrospective analysis was carried out on pregnant women undergoing Non-invasive prenatal testing (NIPT) at Taizhou Hospital in Zhejiang Province from January 2018 to December 2022. The criteria included maternal copy number variations for at least two chromosomes. Clinical follow-up data were obtained for the high-risk population of maternal malignant tumors through telephone follow-up and review of electronic medical records. This study was approved by the Medical Ethics Committee of the Hospital (Ethics No.: K20250339). RESULTS: Among 45 141 NIPT samples, 6 (0.013%) were suggested to have maternal malignant tumors. Follow-up information was available for 5 patients (83.3%). Two cases were diagnosed with maternal malignant tumors, including 1 myelodysplastic syndrome and 1 pelvic malignant tumor. Two cases were found to have multiple uterine fibroids and 1 was lost during follow-up. CONCLUSION The abnormal copy number indicated by NIPT may serve as an early signal for maternal malignant tumors. To establish a systematic follow-up protocol and multidisciplinary collaboration are conducive to achieving early diagnosis of tumors and improving the prognosis of patients. Based on the results of this study, it is recommended that for pregnant women with unexplained copy number variations and suspected maternal tumors by NIPT, targeted tumor screening program should be implemented to optimize their clinical management.
Humans ; Female ; Pregnancy ; Adult ; Retrospective Studies ; DNA Copy Number Variations/genetics* ; Follow-Up Studies ; Neoplasms/diagnosis* ; Noninvasive Prenatal Testing/methods* ; Pregnancy Complications, Neoplastic/diagnosis* ; Prenatal Diagnosis/methods*

OBJECTIVES: To evaluate the performance of a multi-constraint representation learning classification model for identifying ovarian cancer with missing laboratory indicators. METHODS: Tabular data with missing laboratory indicators were collected from 393 patients with ovarian cancer and 1951 control patients. The missing ovarian cancer laboratory indicator features were projected to the latent space to obtain a classification model using the representational learning classification model based on discriminative learning and mutual information coupled with feature projection significance score consistency and missing location estimation. The proposed constraint term was ablated experimentally to assess the feasibility and validity of the constraint term by accuracy, area under the ROC curve (AUC), sensitivity, and specificity. Cross-validation methods and accuracy, AUC, sensitivity and specificity were also used to evaluate the discriminative performance of this classification model in comparison with other interpolation methods for processing of the missing data. RESULTS: The results of the ablation experiments showed good compatibility among the constraints, and each constraint had good robustness. The cross-validation experiment showed that for identification of ovarian cancer with missing laboratory indicators, the AUC, accuracy, sensitivity and specificity of the proposed multi-constraints representation-based learning classification model was 0.915, 0.888, 0.774, and 0.910, respectively, and its AUC and sensitivity were superior to those of other interpolation methods. CONCLUSIONS The proposed model has excellent discriminatory ability with better performance than other missing data interpolation methods for identification of ovarian cancer with missing laboratory indicators.
Female ; Humans ; Ovarian Neoplasms/diagnosis* ; Machine Learning ; ROC Curve

OBJECTIVES: We propose a new melanoma segmentation model, SG-UNet, to enhance the precision of melanoma segmentation in dermascopy images to facilitate early melanoma detection. METHODS: We utilized a U-shaped convolutional neural network, UNet, and made improvements to its backbone, skip connections, and downsampling pooling sections. In the backbone, with reference to the structure of VGG, we increased the number of convolutions from 10 to 13 in the downsampling part of UNet to achieve a deepened network hierarchy that allowed capture of more refined feature representations. To further enhance feature extraction and detail recognition, we replaced the traditional convolution the backbone section with self-calibrated convolution to enhance the model's ability to capture both spatial and channel dimensional features. In the pooling part, the original pooling layer was replaced by Haar wavelet downsampling to achieve more effective multi-scale feature fusion and reduce the spatial resolution of the feature map. The global attention mechanism was then incorporated into the skip connections at each layer to enhance the understanding of contextual information of the image. RESULTS: The experimental results showed that the SG-UNet model achieved significantly improved segmentation accuracy on ISIC 2017 and ISIC 2018 datasets as compared with other current state-of-the-art segmentation models, with Dice reached 92.41% and 86.62% and IoU reaching 92.31% and 86.48% on the two datasets, respectively. CONCLUSIONS The proposed model is capable of effective and accurate segmentation of melanoma from dermoscopy images.
Melanoma/diagnosis* ; Humans ; Neural Networks, Computer ; Dermoscopy/methods* ; Skin Neoplasms ; Image Processing, Computer-Assisted/methods* ; Calibration ; Algorithms

OBJECTIVES: To construct a bone tumor classification model based on feature decoupling and fusion for processing modality loss and fusing multimodal information to improve classification accuracy. METHODS: A decoupling completion module was designed to extract local and global bone tumor image features from available modalities. These features were then decomposed into shared and modality-specific features, which were used to complete the missing modality features, thereby reducing completion bias caused by modality differences. To address the challenge of modality differences that hinder multimodal information fusion, a cross-attention-based fusion module was introduced to enhance the model's ability to learn cross-modal information and fully integrate specific features, thereby improving the accuracy of bone tumor classification. RESULTS: The experiment was conducted using a bone tumor dataset collected from the Third Affiliated Hospital of Southern Medical University for training and testing. Among the 7 available modality combinations, the proposed method achieved an average AUC, accuracy, and specificity of 0.766, 0.621, and 0.793, respectively, which represent improvements of 2.6%, 3.5%, and 1.7% over existing methods for handling missing modalities. The best performance was observed when all the modalities were available, resulting in an AUC of 0.837, which still reached 0.826 even with MRI alone. CONCLUSIONS The proposed method can effectively handle missing modalities and successfully integrate multimodal information, and show robust performance in bone tumor classification under various complex missing modality scenarios.
Humans ; Bone Neoplasms/diagnosis* ; Multimodal Imaging/methods* ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Image Processing, Computer-Assisted/methods* ; Algorithms

OBJECTIVES: To explore the correlation of ELFN1 expression level with prognosis of colorectal cancer and its regulatory role in colorectal cancer cell proliferation and metastasis. METHODS: We analyzed the expression levels of ELFN1 across 33 cancer types using publicly available databases and identified differential genes related to ELFN1 in colorectal cancer. Gene function annotation and enrichment analysis were used to identify the involved signaling pathways. Logistic analysis, Kaplan-Meier analysis and Cox regression analysis were performed to evaluate the correlation between ELFN1 expression and clinicopathological parameters and survival of colorectal cancer patients. qPCR and Western blotting were used to validate the expression levels of ELFN1 in different colorectal cancer cell lines and tissues, and Transwell and EDU experiments were carried out to assess the effect of ELFN1 knockdown on biological behaviors of SW480 cells. RESULTS: ELFN1 was highly expressed in 14 cancers, and its expression was significantly higher in colon cancer tissues than in adjacent tissues. A high expression of ELFN1 mRNA was associated with a poorer overall survival of colorectal cancer patients. Cox regression analysis indicated that ELFN1 expression was an independent prognostic factor for overall survival of the patients. ELFN1 was significantly enriched in tumor metastasis and proliferation and participated in several tumor signaling pathways. The colon cancer cell lines showed significantly higher expression levels of ELFN1 than normal cells, ELFN1 knockdown obviously inhibited proliferation and migration of SW480 cells in vitro. CONCLUSIONS ELFN1 is overexpressed in colorectal cancer and is associated with poor clinical prognosis of the patients. A high ELFN1 expression is associated with malignant phenotypes of colorectal cancer and promotes cancer cell proliferation and metastasis, suggesting its potential as a prognostic biomarker for colorectal cancer.
Humans ; Colorectal Neoplasms/diagnosis* ; Cell Proliferation ; Prognosis ; Cell Line, Tumor ; Biomarkers, Tumor/metabolism* ; Neoplasm Metastasis ; Gene Expression Regulation, Neoplastic ; Nerve Tissue Proteins/metabolism* ; Female ; Male

OBJECTIVES: To identify immunosuppressive neutrophil subsets in patients with prostate cancer (PCa) and construct a risk prediction model for prognosis and immunotherapy response of the patients based on these neutrophil subsets. METHODS: Single-cell and transcriptome data from PCa patients were collected from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). Neutrophil subsets in PCa were identified through unsupervised clustering, and their biological functions and effects on immune regulation were analyzed by functional enrichment, cell interaction, and pseudo-time series analyses. Lasso-Cox regression was utilized to construct a prognostic risk model based on the immunosuppressive neutrophil subsets, and survival analysis and ROC curve analysis were used to compare the prognosis of PCa patients with high and low risks stratified using this model. The relationship of the prognostic risk model with PCa immune infiltration and immune response was evaluated using CIBERSORT and TIDE scores. RESULTS: PCa tissues showed a significantly greater proportion of infiltrating neutrophils than the adjacent normal tissues (P<0.05). PCa-associated neutrophils could be clustered into two independent cell subsets: Neu_1 and Neu_2. Neu_2 cells exhibited highly enriched immunoregulatory functions and were highly differentiated and mature, with upregulated immunosuppressive cytokines such as TGFB1, ITGB2, and LGALS3. Based on the genetic characteristics of Neu_2 cell subsets, the prognostic risk model was constructed. The patients in the high-risk group identified by the model had a shorter biochemical recurrence time (P<0.05) and a higher proportion of Tregs and M2-TAMs cell infiltration (P<0.05) with a higher risk of immune rejection and poorer immune response scores. CONCLUSIONS PCa-associated neutrophils are highly heterogeneous. The prognostic risk model constructed based on the immunosuppressive neutrophil Neu_2 subset can effectively predict both the survival outcomes and immune response of PCa patients.
Humans ; Male ; Prostatic Neoplasms/diagnosis* ; Prognosis ; Neutrophils/immunology* ; Immunotherapy

OBJECTIVES: To establish a prognostic model for primary liver cancer (PLC) using bioinformatics methods. METHODS: Based on the data from 404 patients in the Cancer Genome Atlas (TCGA) database, we constructed a prognostic model integrating the differentially expressed genes, anoikis, and immune-related genes (DAIs) using univariate Cox regression and the LASSO-Cox approach. The predictive ability of the model was evaluated using Kaplan-Meier method and receiver-operating characteristic curves, and a nomogram was developed to facilitate its clinical applications. Gene set enrichment analysis (GSEA) was performed to explore the associated pathways and relationship between the DAIs and the tumor immune microenvironment, and the half-maximal inhibitory concentration (IC₅₀) of liver cancer drugs was calculated using the "pRRophetic" R package. We also detected the expression of SEMA7A in paired tumor and adjacent tissues from liver cancer patients. RESULTS: We constructed and validated a prognostic model based on 7 DAIs (NR4A3, SEMA7A, IL11, AR, BIRC5, EGF, and SPP1), and obtained consistent results in both the TCGA training cohort and GEO validation cohort (GSE14520), where the patients in the low-risk group were characterized by more favorable clinical outcomes and immune status. By integrating this prognostic signature with clinical information, a composite nomogram was generated. Somatic mutation analysis showed that TTN, TP53, and CTNNB1 mutations accounted for the largest proportion of total mutations, and the patients in the low-risk-low-TMB group had higher survival rate. Drug sensitivity analysis revealed differences in sensitivity to chemotherapeutic agents between high- and low-risk groups and between TP53 mutations and non-mutations. In clinical tissue specimens, SEMA7A expression was significantly higher in liver cancer tissues than in the adjacent tissues. CONCLUSIONS We established a new prognostic model based on DAIs for predicting clinical outcomes and therapeutic response of patients with primary liver cancer.
Humans ; Liver Neoplasms/diagnosis* ; Prognosis ; Anoikis ; Nomograms ; Computational Biology ; Tumor Microenvironment ; Semaphorins/metabolism*

OBJECTIVES: To investigate the role of SF3B3 in gastric cancer (GC) progression and prognosis and its possible mechanisms. METHODS: SF3B3 expression levels in pan-cancer and GC were analyzed using TIMER2.0, GEPIA, and UALCAN databases and validated using immunohistochemistry in GC tissues. Survival curves of GC patients were established using Kaplan-Meier Plotter and the data of a patient cohort our hospital. The independent risk factors for 5-year postoperative survival were identified using Cox regression, and their predictive values were evaluated using ROC analysis. SF3B3-associated biological processes were predicted by bioinformatics enrichment analyses. In GC HGC-27 cells, the effects of lentivirus-mediated SF3B3 knockdown and overexpression on cell proliferation and migration were investigated, and the changes in the key glycolytic proteins and extracellular acidification rate (ECAR) were detected. The influence of SF3B3 expression level on tumorigenesis and glycolytic protein expression in vivo were evaluated in a nude mouse xenograft model. RESULTS: High expression of SF3B3 in GC was associated with poor patient prognosis (P<0.05). The factors affecting 5-year survival outcomes following gastric oncological resection included high SF3B3 expression, a CEA level ≥5μg/L, a CA19-9 level ≥37 kU/L, tumor stage T3-4, and lymph node metastasis stage N2-3 (P<0.05). Bioinformatics analysis showed significant enrichment of SF3B3 in glycolysis. In HGC-27 cells, SF3B3 knockdown significantly inhibited while SF3B3 overexpression enhanced cell proliferation, migration, and invasion. SF3B3 knockdown obviously decreased the expressions of HK2, PKM2 and LDHA proteins and ECAR in HGC-27 cells, whereas SF3B3 overexpression produced the opposite effect. In nude mouse xenograft models, SF3B3 knockdown significantly reduced tumor mass and downregulated expression of HK2, PKM2 and LDHA proteins, and SF3B3 overexpression induced the opposite changes. CONCLUSIONS SF3B3 overexpression is associated with poor prognosis of GC patients and promotes GC cell proliferation, migration and invasion possibly by enhancing glycolysis.
Stomach Neoplasms/diagnosis* ; Humans ; Cell Proliferation ; Prognosis ; Animals ; Mice, Nude ; Cell Line, Tumor ; Mice ; Cell Movement ; Male ; Female

OBJECTIVES: To evaluate the effect of artificial intelligence-assisted compressed sensing (ACS) acceleration on MRI radiomic feature extraction and performance of diagnostic staging models for nasopharyngeal carcinoma (NPC) in comparison with conventional parallel imaging (PI). METHODS: A total of 64 patients with newly diagnosed NPC underwent 3.0T MRI using axial T1-weighted (T1W), T2-weighted (T2W), and contrast-enhanced T1-weighted (CE-T1W) sequences. Both PI and ACS protocols were performed using identical imaging parameters. The total scan time for the 3 sequences in ACS group was 227 s, representing a 30% reduction from 312 s in the PI group. Eighteen first-order and 75 texture features were extracted using Pyradiomics. Intraclass correlation coefficients (ICCs) were calculated to assess the agreement between the two acceleration methods. After feature selection using the least absolute shrinkage and selection operator (LASSO), random forest regression models were constructed to distinguish early-stage (T1 and T2) from advanced-stage (T3 and T4) NPC. The diagnostic performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test. RESULTS: ACS-accelerated images demonstrated good radiomic reproducibility, with 86.0% (240/279) of features showing good agreement (ICC>0.75), with mean ICCs for T1W, T2W and CE-T1W sequences of 0.91±0.09, 0.89±0.13 and 0.88±0.11, respectively. The staging prediction models achieved similar AUCs for ACS and PI (0.89 vs 0.90, P=0.991). CONCLUSIONS The MRI radiomic features extracted using ACS and PI techniques are highly consistent, and the ACS-based model shows comparable diagnostic performance to the PI-based model, but ACS significantly reduces the scan time and provides an efficient and reliable acceleration strategy for radiomics in NPC.
Humans ; Nasopharyngeal Neoplasms/diagnosis* ; Magnetic Resonance Imaging/methods* ; Nasopharyngeal Carcinoma ; Neoplasm Staging ; Artificial Intelligence ; Carcinoma ; Female ; Male ; Middle Aged ; Adult ; Radiomics