A multi-frequency approach of the altered functional connectome for autism spectrum disorder identification.

Autism spectrum disorder stands as a multifaceted and heterogeneous neurodevelopmental condition. The utilization of functional magnetic resonance imaging to construct functional brain networks proves instrumental in comprehending the intricate interplay between brain activity and autism spectrum disorder, thereby elucidating the underlying pathogenesis at the cerebral level. Traditional functional brain networks, however, typically confine their examination to connectivity effects within a specific frequency band, disregarding potential connections among brain areas that span different frequency bands. To harness the full potential of interregional connections across diverse frequency bands within the brain, our study endeavors to develop a novel multi-frequency analysis method for constructing a comprehensive functional brain networks that incorporates multiple frequencies. Specifically, our approach involves the initial decomposition of functional magnetic resonance imaging into distinct frequency bands through wavelet transform. Subsequently, Pearson correlation is employed to generate corresponding functional brain networks and kernel for each frequency band. Finally, the classification was performed by a multi-kernel support vector machine, to preserve the connectivity effects within each band and the connectivity patterns shared among the different bands. Our proposed multi-frequency functional brain networks method yielded notable results, achieving an accuracy of 89.1%, a sensitivity of 86.67%, and an area under the curve of 0.942 in a publicly available autism spectrum disorder dataset.

Pathogenic variants in human DNA damage repair genes mostly arose in recent human history.

BACKGROUND: Genome stability is maintained by the DNA damage repair (DDR) system composed of multiple DNA repair pathways of hundreds of genes. Germline pathogenic variation (PV) in DDR genes damages function of the affected DDR genes, leading to genome instability and high risk of diseases, in particular, cancer. Knowing evolutionary origin of the PVs in human DDR genes is essential to understand the etiology of human diseases. However, answer to the issue remains largely elusive. In this study, we analyzed evolutionary origin for the PVs in human DDR genes. METHODS: We identified 169 DDR genes by referring to various databases and identified PVs in the DDR genes of modern humans from ClinVar database. We performed a phylogenetic analysis to analyze the conservation of human DDR PVs in 100 vertebrates through cross-species genomic data comparison using the phyloFit program of the PHAST package and visualized the results using the GraphPad Prism software and the ggplot module. We identified DDR PVs from over 5000 ancient humans developed a database to host the DDR PVs ( https://genemutation.fhs.um.edu.mo/dbDDR-AncientHumans ). Using the PV data, we performed a molecular archeological analysis to compare the DDR PVs between modern humans and ancient humans. We analyzed evolution selection of DDR genes across 20 vertebrates using the CodeML in PAML for phylogenetic analysis. RESULTS: Our phylogenic analysis ruled out cross-species conservation as the origin of human DDR PVs. Our archeological approach identified rich DDR PVs shared between modern and ancient humans, which were mostly dated within the last 5000 years. We also observed similar pattern of quantitative PV distribution between modern and ancient humans. We further detected a set of ATM, BRCA2 and CHEK2 PVs shared between human and Neanderthals. CONCLUSIONS: Our study reveals that human DDR PVs mostly arose in recent human history. We propose that human high cancer risk caused by DDR PVs can be a by-product of human evolution.

Wavelet transform-based frequency self-adaptive model for functional brain network.

The accurate estimation of functional brain networks is essential for comprehending the intricate relationships between different brain regions. Conventional methods such as Pearson Correlation and Sparse Representation often fail to uncover concealed information within diverse frequency bands. To address this limitation, we introduce a novel frequency-adaptive model based on wavelet transform, enabling selective capture of highly correlated frequency band sequences. Our approach involves decomposing the original time-domain signal from resting-state functional magnetic resonance imaging into distinct frequency domains, thus constructing an adjacency matrix that offers enhanced separation of features across brain regions. Comparative analysis demonstrates the superior performance of our proposed model over conventional techniques, showcasing improved clarity and distinctiveness. Notably, we achieved the highest accuracy rate of 89.01% using Sparse Representation based on Wavelet Transform, outperforming Pearson Correlation based on Wavelet Transform with an accuracy of 81.32%. Importantly, our method optimizes raw data without significantly altering feature topology, rendering it adaptable to various functional brain network estimation approaches. Overall, this innovation holds the potential to advance the understanding of brain function and furnish more accurate samples for future research and clinical applications.

[Research Progress of Fibroblast Growth Factor Receptor Signaling Pathway in Breast Cancer].

Breast cancer,one of the common malignant tumors in women,has shown rising incidence in recent years,posing a serious threat to women's health.The advancement of molecular biology facilitates the revealing of the relationships between signaling pathways and breast cancer.Fibroblast growth factor receptor (FGFR) signaling pathway plays an important role in the proliferation,survival,differentiation,migration,and apoptosis of breast cancer cells.Strategies targeting the FGFR signaling pathway thus exhibit a promising prospect in breast cancer treatment.

SEMPANet: A Modified Path Aggregation Network with Squeeze-Excitation for Scene Text Detection.

Recently, various object detection frameworks have been applied to text detection tasks and have achieved good performance in the final detection. With the further expansion of text detection application scenarios, the research value of text detection topics has gradually increased. Text detection in natural scenes is more challenging for horizontal text based on a quadrilateral detection box and for curved text of any shape. Most networks have a good effect on the balancing of target samples in text detection, but it is challenging to deal with small targets and solve extremely unbalanced data. We continued to use PSENet to deal with such problems in this work. On the other hand, we studied the problem that most of the existing scene text detection methods use ResNet and FPN as the backbone of feature extraction, and improved the ResNet and FPN network parts of PSENet to make it more conducive to the combination of feature extraction in the early stage. A SEMPANet framework without an anchor and in one stage is proposed to implement a lightweight model, which is embodied in the training time of about 24 h. Finally, we selected the two most representative datasets for oriented text and curved text to conduct experiments. On ICDAR2015, the improved network's latest results further verify its effectiveness; it reached 1.01% in F-measure compared with PSENet-1s. On CTW1500, the improved network performed better than the original network on average.

Geometric Algebra-Based ESPRIT Algorithm for DOA Estimation.

Direction-of-arrival (DOA) estimation plays an important role in array signal processing, and the Estimating Signal Parameter via Rotational Invariance Techniques (ESPRIT) algorithm is one of the typical super resolution algorithms for direction finding in an electromagnetic vector-sensor (EMVS) array; however, existing ESPRIT algorithms treat the output of the EMVS array either as a "long vector", which will inevitably lead to loss of the orthogonality of the signal components, or a quaternion matrix, which may result in some missing information. In this paper, we propose a novel ESPRIT algorithm based on Geometric Algebra (GA-ESPRIT) to estimate 2D-DOA with double parallel uniform linear arrays. The algorithm combines GA with the principle of ESPRIT, which models the multi-dimensional signals in a holistic way, and then the direction angles can be calculated by different GA matrix operations to keep the correlations among multiple components of the EMVS. Experimental results demonstrate that the proposed GA-ESPRIT algorithm is robust to model errors and achieves less time complexity and smaller memory requirements.

Decoding the dynamic DNA methylation and hydroxymethylation landscapes in endodermal lineage intermediates during pancreatic differentiation of hESC.

Dynamic changes in DNA methylation and demethylation reprogram transcriptional outputs to instruct lineage specification during development. Here, we applied an integrative epigenomic approach to unveil DNA (hydroxy)methylation dynamics representing major endodermal lineage intermediates during pancreatic differentiation of human embryonic stem cells (hESCs). We found that 5-hydroxymethylcytosine (5hmC) marks genomic regions to be demethylated in the descendent lineage, thus reshaping the DNA methylation landscapes during pancreatic lineage progression. DNA hydroxymethylation is positively correlated with enhancer activities and chromatin accessibility, as well as the selective binding of lineage-specific pioneer transcription factors, during pancreatic differentiation. We further discovered enrichment of hydroxymethylated regions (termed '5hmC-rim') at the boundaries of large hypomethylated functional genomic regions, including super-enhancer, DNA methylation canyon and broad-H3K4me3 peaks. We speculate that '5hmC-rim' might safeguard low levels of cytosine methylation at these regions. Our comprehensive analysis highlights the importance of dynamic changes of epigenetic landscapes in driving pancreatic differentiation of hESC.

SHON expression predicts response and relapse risk of breast cancer patients after anthracycline-based combination chemotherapy or tamoxifen treatment.

BACKGROUND: SHON nuclear expression (SHON-Nuc+) was previously reported to predict clinical outcomes to tamoxifen therapy in ERα+ breast cancer (BC). Herein we determined if SHON expression detected by specific monoclonal antibodies could provide a more accurate prediction and serve as a biomarker for anthracycline-based combination chemotherapy (ACT). METHODS: SHON expression was determined by immunohistochemistry in the Nottingham early-stage-BC cohort (n = 1,650) who, if eligible, received adjuvant tamoxifen; the Nottingham ERα- early-stage-BC (n = 697) patients who received adjuvant ACT; and the Nottingham locally advanced-BC cohort who received pre-operative ACT with/without taxanes (Neo-ACT, n = 120) and if eligible, 5-year adjuvant tamoxifen treatment. Prognostic significance of SHON and its relationship with the clinical outcome of treatments were analysed. RESULTS: As previously reported, SHON-Nuc+ in high risk/ERα+ patients was significantly associated with a 48% death risk reduction after exclusive adjuvant tamoxifen treatment compared with SHON-Nuc- [HR (95% CI) = 0.52 (0.34-0.78), p = 0.002]. Meanwhile, in ERα- patients treated with adjuvant ACT, SHON cytoplasmic expression (SHON-Cyto+) was significantly associated with a 50% death risk reduction compared with SHON-Cyto- [HR (95% CI) = 0.50 (0.34-0.73), p = 0.0003]. Moreover, in patients received Neo-ACT, SHON-Nuc- or SHON-Cyto+ was associated with an increased pathological complete response (pCR) compared with SHON-Nuc+ [21 vs 4%; OR (95% CI) = 5.88 (1.28-27.03), p = 0.012], or SHON-Cyto- [20.5 vs. 4.5%; OR (95% CI) = 5.43 (1.18-25.03), p = 0.017], respectively. After receiving Neo-ACT, patients with SHON-Nuc+ had a significantly lower distant relapse risk compared to those with SHON-Nuc- [HR (95% CI) = 0.41 (0.19-0.87), p = 0.038], whereas SHON-Cyto+ patients had a significantly higher distant relapse risk compared to SHON-Cyto- patients [HR (95% CI) = 4.63 (1.05-20.39), p = 0.043]. Furthermore, multivariate Cox regression analyses revealed that SHON-Cyto+ was independently associated with a higher risk of distant relapse after Neo-ACT and 5-year tamoxifen treatment [HR (95% CI) = 5.08 (1.13-44.52), p = 0.037]. The interaction term between ERα status and SHON-Nuc+ (p = 0.005), and between SHON-Nuc+ and tamoxifen therapy (p = 0.007), were both statistically significant. CONCLUSION: SHON-Nuce+ in tumours predicts response to tamoxifen in ERα+ BC while SHON-Cyto+ predicts response to ACT.

Plasma cell-free DNA chromosomal instability analysis by low-pass whole-genome sequencing to monitor breast cancer relapse.

BACKGROUND: Chromosomal instabilities (CIN) of plasma cell-free DNA (cfDNA) are common in breast cancer. We aimed to investigate the value of cfDNA CIN in monitoring the breast cancer relapse and additionally to compare it with the traditional biomarkers (CA15-3 and CEA). METHODS: Overall 62 recurrent breast cancer patients and 20 healthy controls were recruited. Low-pass whole-genome sequencing (LPWGS) was performed to detect cfDNA CIN. A CIN score was calculated. The performance of CA15-3, CEA, and CIN score in monitoring the recurrence was investigated with receiver operating characteristic (ROC) curve and the area under curve (AUC). Multivariable Cox proportional hazard model was established to analyze the correlations between copy number gain/loss and disease-free survival (DFS). RESULTS: cfDNA CIN achieved the positive rate of 77.6% [(95% confidence interval (CI) 73.4-95.3%)] among recurrent breast cancer patients, with an AUC value of 0.933, superior to CA15-3 (positive rate: 38.7%; AUC: 0.864) and CEA (positive rate: 41.93%; AUC: 0.878) (P < 0.01). The combination of cfDNA CIN with two biomarkers further increased the positive rate to 88.7% (95% confidence interval 77.5-95.0%). cfDNA CIN achieved better performance in patients with shorter DFS (≤ 41 months), with an AUC value of 0.975. CONCLUSIONS: cfDNA CIN yields a higher accuracy in monitoring breast cancer recurrence compared to traditional biomarkers (CA15-3 and CEA), especially for biomarker-negative patients. The combination of cfDNA CIN to traditional biomarkers further improved the detection rate of recurrence, which may provide a new method for monitoring the early relapse of breast cancer, though further investigations are warranted.

Comprehensive mutation detection of BRCA1/2 genes reveals large genomic rearrangements contribute to hereditary breast and ovarian cancer in Chinese women.

BACKGROUND: Mutated BRCA1/2 genes are associated with hereditary breast and ovarian cancer (HBOC). So far most of the identified BRCA1/2 pathogenic variants are single nucleotide variants (SNVs) or insertions/deletions (Indels). However, large genomic rearrangements (LGRs) such as copy number variants (CNVs) are also playing an important role in HBOC predisposition. Their frequency and spectrum have been well studied in western populations but remain largely unknown for Chinese population. METHODS: Peripheral blood samples were collected from 218 unrelated familial breast and/or ovarian cancer (FBOC) patients living in Eastern China. PCR-based Sanger sequencing and panel-based next-generation sequencing (NGS) were performed to detect pathogenic SNVs and Indels in BRCA1/2 genes. For the patients lacking small pathogenic variants, multiplex ligation dependent probe amplification (MLPA) assay was conducted to screen for LGRs. RESULTS: In total, we identified 44 samples (20.1%) carrying small pathogenic variants (26 in BRCA1 and 18 in BRCA2, respectively). Among the rest of 174 samples, five were found carrying novel deleterious LGRs in BRCA1 which are exon5-7dup (1 patient), exon13-14dup (2 patients), and exon1-22del (2 patients). No LGR was found in BRCA2. Overall, LGRs accounted for 16.1% (5/31) of BRCA1 pathogenic variants, and were detected in 2.3% (5/218) of all FBOC patients. , CONCLUSIONS: LGR variants in BRCA1 gene play a significant role in Chinese HBOC patients. MLPA or other similar LGR-detecting methods should be recommended along with nucleotide sequencing as the initial screening approach for Chinese HBOC women.

MicroRNA and LncRNA Expression Profiles in Human Estrogen Receptor Positive Breast Cancer.

BACKGROUND: Recent studies have established non-coding RNAs, which include microRNAs and lncRNAs, are aberrantly expressed in breast cancer. In this study, we explore the expression profile of microRNAs and lncRNAs in ER positive (ER+) breast and paracancerous tissues and define their possible correlations. METHODS: We collected ER+ breast cancer patients' and paracancerous tissues from the specimen bank of Zhejiang Cancer Hospital to extract total RNA for obtaining the expression level of microRNAs and lncRNAs by qRT-PCR. RESULTS: The relative expression results indicated that microRNAs such as MIR-191, MIR-213, MIR-122A had significantly higher expression and MIR-125B-1, MIR-125B-2, MIR-145 had lower expression in ER positive breast cancer compared to normal breast. The interaction of microRNA and lncRNA results exhibited upregulated MIR382-5P and lncRNA 362 in ER+ breast cancer compared to non-cancerous breast. By contrast, MIR222 and NFIA-AS1 are down-regulated. Furthermore, MIR222 and NFIA-AS1 showed different expressions in different TNM stages. CONCLUSIONS: MicroRNAs and lncRNAs are aberrantly expressed in ER+ breast cancer. It is inferred that these microRNAs and lncRNAs may be promising biomarkers for ER positive breast cancer.

A New Application of Multimodality Radiomics Improves Diagnostic Accuracy of Nonpalpable Breast Lesions in Patients with Microcalcifications-Only in Mammography.

BACKGROUND The aim of this study was to assess a radiomic scheme that combines image features from digital mammography and dynamic contrast-enhanced MRI to improve classification accuracy of nonpalpable breast lesion (NBL) with Breast Imaging-Reporting and Data System (BI-RADS) 3-5 microcalcifications-only in mammography. MATERIAL AND METHODS This retrospective study was approved by the Internal Research Review and Ethical Committee of our hospital. We included 81 patients who underwent a three-dimensional digital breast X-ray wire positioning for local resection between October 2012 and November 2016. All patients underwent breast MRI and mammography before the treatment, and all obtained pathological confirmation. According to the pathological results, 41 patients with benign lesions were assigned to the benign group and 40 patients with malignant lesions were assigned to the malignant group. We used the random forest algorithm to select significant features and to test the single and multimodal classifiers using the Leave-One-Out-Cross-Validation method. An area under the receiver operating characteristic curve was also used to evaluate its discriminating performance. RESULTS The multimodal classifier achieved AUC of 0.903, with a sensitivity of 82.5% and a specificity of 80.48%, which was better than any single modality. CONCLUSIONS Multimodal radiomics classification shows promising power in discriminating malignant lesions from benign lesions in NBL patients with BI-RADS 3-5 microcalcifications-only in mammography.

Novel germline mutations and unclassified variants of BRCA1 and BRCA2 genes in Chinese women with familial breast/ovarian cancer.

BACKGROUND: Germline mutations in the BRCA1 and BRCA2 genes greatly increase a woman's risk of developing breast and/or ovarian cancer. The prevalence and distribution of such mutations differ across races/ethnicities. Several studies have investigated Chinese women with high-risk breast cancer, but the full spectrum of the mutations in these two genes remains unclear. METHODS: In this study, 133 unrelated Chinese women with familial breast/ovarian cancer living in Zhejiang, eastern China, were enrolled between the years 2008 and 2014. The complete coding regions and exon-intron boundaries of BRCA1 and BRCA2 were screened by PCR-sequencing assay. Haplotype analysis was performed to confirm BRCA1 and BRCA2 founder mutations. In silico predictions were performed to identify the non-synonymous amino acid changes that were likely to disrupt the functions of BRCA1 and BRCA2. RESULTS: A total of 23 deleterious mutations were detected in the two genes in 31 familial breast/ovarian cancer patients with a total mutation frequency of 23.3% (31/133). The highest frequency of 50.0% (8/16) was found in breast cancer patients with a history of ovarian cancer. The frequencies of BRCA1 and BRCA2 mutations were 13.5 % (18/133) and 9.8% (13/133), respectively. We identified five novel deleterious mutations (c.3295delC, c.3780_3781delAG, c.4063_4066delAATC, c.5161 > T and c.5173insA) in BRCA1 and seven (c.1-40delGA, c.4487delC, c.469_473delAAGTC, c.5495delC, c.6141T > A, c.6359C > G and c.7588C > T) in BRCA2, which accounted for 52.2% (12/23) of the total mutations. Six recurrent mutations were found, including four (c.3780_3781delAG, c.5154G > A, c.5468-1del8 and c.5470_5477del8) in BRCA1 and two (c.3109C > T and c.5682C > G) in BRCA2. Two recurrent BRCA1 mutations (c.5154G > A and c.5468-1del8) were identified as putative founder mutations. We also found 11 unclassified variants, and nine of these are novel. The possibility was that each of the non-synonymous amino acid changes would disrupt the function of BRCA1 and BRCA2 varied according to the different algorithms used. CONCLUSIONS: BRCA1 and BRCA2 mutations accounted for a considerable proportion of hereditary breast/ovarian cancer patients from eastern China and the spectrum of the mutations of these two genes exhibited some unique features. The two BRCA1 putative founder mutations may provide a cost-effective option to screen Chinese population, while founder effects of the two mutations should be investigated in a lager sample size of patients.

Serum protein S100A9, SOD3, and MMP9 as new diagnostic biomarkers for pulmonary tuberculosis by iTRAQ-coupled two-dimensional LC-MS/MS.

This study aimed to discover the novel noninvasive biomarkers for the diagnosis of pulmonary tuberculosis (TB). We applied iTRAQ 2D LC-MS/MS technique to investigate protein profiles in patients with pulmonary TB and other lung diseases. A total of 34 differentially expressed proteins (24 upregulated proteins and ten downregulated proteins) were identified in the serum of pulmonary TB patients. Significant differences in protein S100-A9 (S100A9), extracellular superoxide dismutase [Cu-Zn] (SOD3), and matrix metalloproteinase 9 (MMP9) were found between pulmonary TB and other lung diseases by ELISA. Correlations analysis revealed that the serum concentration of MMP9 in the pulmonary TB was in moderate correlation with SOD3 (r = 0.581) and S100A9 (r = 0.471), while SOD3 was in weak correlation with S100A9 (r = 0.287). The combination of serum S100A9, SOD3, and MMP9 levels could achieve 92.5% sensitivity and 95% specificity to discriminate between pulmonary TB and healthy controls, 90% sensitivity and 87.5% specificity to discriminate between pulmonary TB and pneumonia, and 85% sensitivity and 92.5% specificity to discriminate between pulmonary TB and lung cancer, respectively. The results showed that S100A9, SOD3, and MMP9 may be potential diagnostic biomarkers for pulmonary TB, and provided experimental basis for the diagnosis of pulmonary TB.

HGCMorph: joint discontinuity-preserving and pose-learning via GNN and capsule networks for deformable medical images registration.

Objective.This study aims to enhance medical image registration by addressing the limitations of existing approaches that rely on spatial transformations through U-Net, ConvNets, or Transformers. The objective is to develop a novel architecture that combines ConvNets, graph neural networks (GNNs), and capsule networks to improve the accuracy and efficiency of medical image registration, which can also deal with the problem of rotating registration.Approach.We propose an deep learning-based approach which can be utilized in both unsupervised and semi-supervised manners, named as HGCMorph. It leverages a hybrid framework that integrates ConvNets and GNNs to capture lower-level features, specifically short-range attention, while also utilizing capsule networks (CapsNets) to model abstract higher-level features, including entity properties such as position, size, orientation, deformation, and texture. This hybrid framework aims to provide a comprehensive representation of anatomical structures and their spatial relationships in medical images.Main results.The results demonstrate the superiority of HGCMorph over existing state-of-the-art deep learning-based methods in both qualitative and quantitative evaluations. In unsupervised training process, our model outperforms the recent SOTA method TransMorph by achieving 7%/38% increase on Dice score coefficient (DSC), and 2%/7% improvement on negative jacobian determinant for OASIS and LPBA40 datasets, respectively. Furthermore, HGCMorph achieves improved registration accuracy in semi-supervised training process. In addition, when dealing with complex 3D rotations and secondary randomly deformations, our method still achieves the best performance. We also tested our methods on lung datasets, such as Japanese Society of Radiology, Montgoermy and Shenzhen.Significance.The significance lies in its innovative design to medical image registration. HGCMorph offers a novel framework that overcomes the limitations of existing methods by efficiently capturing both local and abstract features, leading to enhanced registration accuracy, discontinuity-preserving, and pose-learning abilities. The incorporation of capsule networks introduces valuable improvements, making the proposed method a valuable contribution to the field of medical image analysis. HGCMorph not only advances the SOTA methods but also has the potential to improve various medical applications that rely on accurate image registration.

ER-positive and BRCA2-mutated breast cancer: a literature review.

BRCA2-mutated carriers have a high lifetime risk of breast cancer (BC), an early age of onset, and an increased risk of other cancers (including ovarian, pancreatic, and prostate cancer). Almost 70-80% of BRCA2-mutated BC are estrogen receptor (ER)-positive, which is a particular type of ER-positive BC that differs from sporadic ER-positive BC. This article reviews the clinicopathological features, treatment, and prognosis of ER-positive and BRCA2-mutated BC to provide a reference for clinical decision-making.

EviD-GAN: Improving GAN With an Infinite Set of Discriminators at Negligible Cost.

Ensemble learning improves the capability of convolutional neural network (CNN)-based discriminators, whose performance is crucial to the quality of generated samples in generative adversarial network (GAN). However, this learning strategy results in a significant increase in the number of parameters along with computational overhead. Meanwhile, the suitable number of discriminators required to enhance GAN performance is still being investigated. To mitigate these issues, we propose an evidential discriminator for GAN (EviD-GAN)-code is available at https://github.com/Tohokantche/EviD-GAN-to learn both the model (epistemic) and data (aleatoric) uncertainties. Specifically, by analyzing three GAN models, the relation between the distribution of discriminator's output and the generator performance has been discovered yielding a general formulation of GAN framework. With the above analysis, the evidential discriminator learns the degree of aleatoric and epistemic uncertainties via imposing a higher order distribution constraint over the likelihood as expressed in the discriminator's output. This constraint can learn an ensemble of likelihood functions corresponding to an infinite set of discriminators. Thus, EviD-GAN aggregates knowledge through the ensemble learning of discriminator that allows the generator to benefit from an informative gradient flow at a negligible computational cost. Furthermore, inspired by the gradient direction in maximum mean discrepancy (MMD)-repulsive GAN, we design an asymmetric regularization scheme for EviD-GAN. Unlike MMD-repulsive GAN that performs at the distribution level, our regularization scheme is based on a pairwise loss function, performs at the sample level, and is characterized by an asymmetric behavior during the training of generator and discriminator. Experimental results show that the proposed evidential discriminator is cost-effective, consistently improves GAN in terms of Frechet inception distance (FID) and inception score (IS), and performs better than other competing models that use multiple discriminators.

HR-positive/HER2-negative breast cancer arising in patients with or without BRCA2 mutation: different biological phenotype and similar prognosis.

Background: BRCA2 plays a key role in homologous recombination. However, information regarding its mutations in Chinese patients with breast cancer remains limited. Objectives: This study aimed to assess the clinicopathological characteristics of BRCA2 mutation breast cancer and explore the mutation's effect on hormone receptor (HR)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer survival in China. Design: This hospital-based cohort study prospectively included 629 women with breast cancer diagnosed from 2008 to 2023 at Zhejiang Cancer Hospital in China. Methods: We compared the clinicopathological characteristics and metastatic patterns and analysed the invasive disease-free survival (iDFS), distant relapse-free survival (DRFS) and first-line progression-free survival (PFS1) of patients with HR-positive/HER2-negative breast cancer according to BRCA2 mutations. Results: Among the 629 patients, 78 had BRCA2 mutations (12.4%) and 551 did not (87.6%). The mean age at diagnosis was lower in the BRCA2 mutation breast cancer group than in the non-mutation breast cancer group (38.91 versus 41.94 years, p = 0.016). BRCA2 mutation breast cancers were more likely to be lymph node-positive than non-mutation breast cancers (73.0% versus 56.6%, p = 0.037). The pathological grade was higher in 47.1% of BRCA2 mutation breast cancers than in 29.6% of non-mutation breast cancers (p = 0.014). The proportions of patients with BRCA2 mutations who developed contralateral breast cancer (19.2% versus 8.8%, p = 0.004), breast cancer in the family (53.8% versus 38.3%, p = 0.009) and ovarian cancer in the family (7.6% versus 2.4%, p = 0.022) were higher than those of patients without the mutation. The median follow-up time was 92.78 months. Multivariate analysis showed that BRCA2 mutation was not associated with poorer iDFS [hazard ratio = 0.9, 95% confidence interval (CI) = 0.64-1.27, p = 0.56] and poorer distant relapse-free survival (DRFS) (hazard ratio = 1.09, 95% CI = 0.61-1.93, p = 0.76). There was no significant difference between the two groups with regard to metastatic patterns in the advanced disease setting. In the first-line metastatic breast cancer setting, PFS1 expression was broadly similar between the two groups irrespective of chemotherapy or endocrine therapy. Conclusion: HR-positive/HER2-negative breast cancer with BRCA2 mutations differs from those without mutations in clinical behaviour and reflects more aggressive tumour behaviour. Our results indicate that BRCA2 mutations have no significant effect on the survival of Chinese women with HR-positive/HER2-negative breast cancer.

Characterizing the Efficacy and Safety of Chemotherapy Plus Everolimus in HER2-Negative Metastatic Breast Cancer Harboring Altered PI3K/AKT/mTOR.

BACKGROUND: The clinical outcomes of chemotherapy (CT) for the treatment of metastatic triple-negative (TN) and hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer (mBC) have proven to be disappointing. The phosphatidylinositol 3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR) pathway, a tumor-promoting signaling cascade frequently mutated in breast cancer (BC), has been implicated in chemoresistance. In this study, our objective is to investigate the efficacy and safety of combining everolimus with chemotherapy in mBC patients exhibiting mutations in the PI3K/AKT/mTOR pathway. METHODS: We conducted a retrospective analysis to characterize the efficacy, safety, and their association with clinical and molecular characteristics of metastatic lesions in 14 patients with HER2- mBC. These patients harbored at least one altered member of the PI3K/AKT/mTOR signaling pathway and were treated with a combination of a chemotherapy agent and the mTOR inhibitor everolimus (CT+EVE). RESULTS: The majority of patients belonged to the triple-negative (TN) subtype (9/14, 64.3%), having already undergone 2 lines of chemotherapy (CT) in the metastatic setting (11, 78.6%). These patients carried altered phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) and were administered a vinorelbine-containing regimen (10, 71.4%). The objective response rate (ORR) was 42.9%, with a disease control rate of 92.9%. The median progression-free survival (PFS) and overall survival (OS) were 5.9 (95% confidence interval (CI): 4.9-13.6) months and 14.3 (95% CI: 8.5-not reached (NR)) months, respectively. Patients with fewer prior treatment lines tended to exhibit longer PFS. OS, PFS, and ORR were comparable between hormone receptor-positive (HR+) and triple-negative breast cancer (TNBC) patients, but numerical improvements were noted in patients with a single PI3K pathway alteration compared to those with more than one alteration. Genomic alterations that surfaced upon progression on CT+EVE included cyclin dependent kinase 4 (CDK4) and epidermal growth factor receptor (EGFR) amplification, as well as neurofibromin 1 (NF1) mutation, suggesting potential mechanisms of acquired resistance. An analysis of adverse events indicated manageable toxicities. CONCLUSIONS: The findings of this study suggest both activity and safety for the combination of chemotherapy and the mTOR inhibitor everolimus (CT+EVE) in patients with HER2- mBC who have alterations in the PI3K pathway, particularly those who have received fewer prior chemotherapy. However, it is crucial to note that large-scale, randomized control studies are warranted to more comprehensively characterize the efficacy and safety of this combination therapy.

FOXK2 amplification promotes breast cancer development and chemoresistance.

Oncogene activation through DNA amplification or overexpression is a crucial driver of cancer initiation and progression. The FOXK2 gene, located on chromosome 17q25, encodes a transcription factor with a forkhead DNA-binding domain. Analysis of genomic datasets reveals that FOXK2 is frequently amplified and overexpressed in breast cancer, correlating with poor patient survival. Knockdown of FOXK2 significantly inhibited breast cancer cell proliferation, migration, anchorage-independent growth, and delayed tumor growth in a xenograft mouse model. Additionally, inhibiting FOXK2 sensitized breast cancer cells to chemotherapy. Co-overexpression of FOXK2 and mutant PI3KCA transformed non-tumorigenic MCF-10A cells, suggesting a role for FOXK2 in PI3KCA-driven tumorigenesis. CCNE2, PDK1, and ESR1 were identified as transcriptional targets of FOXK2 in MCF-7 cells. Small-molecule inhibitors of CCNE2/CDK2 (dinaciclib) and PDK1 (dichloroacetate) exhibited synergistic anti-tumor effects with PI3KCA inhibitor (alpelisib) in vitro. Inhibition of FOXK2 by dinaciclib synergistically enhanced the anti-tumor effects of alpelisib in a xenograft mouse model. Collectively, these findings highlight the oncogenic function of FOXK2 and suggest that FOXK2 and its downstream genes represent potential therapeutic targets in breast cancer.