The detection of circulating tumor DNA (ctDNA) is a valuable method to predict the risk of recurrence and to detect real-time gene changes. The amount of ctDNA is affected by many factors. Moreover, the detection rate of ctDNA varies from report to report.

The present study evaluated differentially expressed genes using a DNA microarray assay for gene expression in tumors with and without detected ctDNA and constructed a prediction model for the detectability of ctDNA in breast tumor tissues. The model, named Cir-Predict, consisted of 126 probe sets (111 genes) and was constructed in a training set of breast cancer patients (n = 35) and validated in a validation set (n = 13).

The accuracy, sensitivity, and specificity in training and validation sets were over 90%, and Cir-Predict was significantly associated with ctDNA detection independently of the other conventional clinicopathological parameters in training and validation sets (P < 0.001, P = 0.014, respectively). Cir-Predict (+) was significantly associated with worse recurrence-free survival (P = 0.006). Pathway analysis revealed that nine pathways including tight junction and cell cycle tended to be related to ctDNA detectability.

Cir-Predict not only provides information useful for breast cancer treatment, but also helps the understanding of the mechanism by which ctDNA is detected.

The diagnostic and therapeutic approach to the axilla in breast cancer patients has changed significantly over the past 30 years, with the replacement of complete axillary lymph node dissection practices by less invasive approaches. Reference is made to clinical findings that have led to practical treatment recommendations and are paving the way to new levels of de-escalation in breast cancer surgery.

CDK4/6 inhibitors in combination with endocrine therapy are widely used to treat HR+/HER2- metastatic breast cancer leading to improved progression-free survival (PFS) compared to single agent endocrine therapy. Over 300 patients receiving standard-of-care CDK4/6 inhibitor combination therapy for metastatic disease were enrolled at a single institution. Clinical, pathological, and gene expression data were employed to define determinants for PFS duration. Visceral disease (HR 1.55, p = 0.0013), prior endocrine therapy (HR 2.34, p < 0.001), and the type of endocrine therapy (HR 2.16, p < 0.001) were highly associated with PFS duration. Multiple pre-defined gene expression signatures were employed to determine association with response to CDK4/6 inhibitor-based therapy. Random survival forest was applied to define key gene expression and clinical features associated with PFS and develop a predictive model. The time to progression predicted by this model was related to the median PFS observed in PALOMA-2/3 and PEARL studies. Interrogating genes identified as highly significant across all studies indicated common enrichment of gene networks associated with cell cycle and estrogen receptor signaling. These findings indicate that there are common features from real-world use of CDK4/6 inhibitors that could be used to infer time to progression and better inform treatment.

Breast cancer (BCa), a leading malignancy among women, is characterized by morphological and molecular heterogeneity. While early-stage, hormone receptor, and HER2-positive BCa are treatable, triple-negative BCa and metastatic BCa remains largely untreatable. Advances in sequencing and proteomic technologies have improved our understanding of the molecular alterations that occur during BCa initiation and progression and enabled identification of subclass-specific biomarkers and therapeutic targets. Despite the availability of abundant omics data in public repositories, user-friendly tools for multi-omics data analysis and integration are scarce. To address this, we developed a comprehensive BCa data analysis platform called MammOnc-DB ( http://resource.path.uab.edu/MammOnc-Home.html ), comprising data from more than 20,000 BCa samples. MammOnc-DB facilitates hypothesis generation and testing, biomarker discovery, and therapeutic targets identification. The platform also includes pre- and post-treatment data, which can help users identify treatment resistance markers and support combination therapy strategies, offering researchers and clinicians a comprehensive tool for BCa data analysis and visualization.

In the context of time-to-event analysis, First hitting time methods consider the event occurrence as the ending point of some evolving process. The characteristics of the process are of great relevance for the analysis, which makes this class of models interesting and particularly suitable for applications where something about the degradation path is known. In cases where the degradation can only worsen, a monotonic process is the most suitable choice. This paper proposes a boosting algorithm for first hitting time models based on an underlying homogeneous gamma process to account for the monotonicity of the degradation trend. The predictive power and versatility of the algorithm are shown with real data examples from both engineering and biomedical applications, as well as with simulated examples.

Recent advancements in multi-omics and big-data technologies have facilitated the discovery of numerous cancer prognostic biomarkers and gene signatures. However, their clinical application remains limited due to poor reproducibility and insufficient independent validation. Despite the availability of high-quality datasets, achieving reliable biomarker identification across multiple cohorts continues to be a significant challenge. To address these issues, we developed a comprehensive platform, SurvivalML, designed to support the discovery and validation of prognostic biomarkers and gene signatures using large-scale and harmonized data from 21 cancer types. Through SurvivalML, we identified DCLRE1B as a novel prognostic biomarker for hepatocellular carcinoma, with experimental confirmation of its role in promoting tumor progression. Additionally, we developed the Chinese glioblastoma prognostic signature (CGPS) and its simplified version, SCGPS, a three-gene model. Both demonstrated superior predictive performance compared to other glioblastoma signatures in our in-house cohort and five independent Chinese datasets. The SCGPS model was further validated in 109 clinical samples using multiplex immunofluorescence, showing strong consistency with the original CGPS model. Overall, SurvivalML provides a robust platform for the identification and validation of prognostic biomarkers and gene signatures, offering a valuable resource for advancing cancer research and clinical application.

Gene network models provide a foundation for graph theory approaches, aiding in the novel discovery of drug targets, disease genes, and genetic mechanisms for various biological functions. Disease genetics must be interpreted within the cellular context of disease-associated cell types, which cannot be achieved with datasets consisting solely of organism-level samples. Single-cell RNA sequencing (scRNA-seq) technology allows computational distinction of cell states which provides a unique opportunity to understand cellular biology that drives disease processes. Importantly, the abundance of cell samples with their transcriptome-wide profile allows the modeling of systemic cell-type-specific gene networks (CGNs), offering insights into gene-cell-disease relationships. In this review, we present reference-based and de novo inference of gene functional interaction networks that we have recently developed using scRNA-seq datasets. We also introduce a compendium of CGNs as a useful resource for cell-type-resolved disease genetics. By leveraging these advances, we envision single-cell network biology as the key approach for mapping the gene-cell-disease axis.

Breast cancer (BC) is the most prevalent cancer among women and a leading cause of cancer-related deaths worldwide. Emerging evidence suggests that DNA methylation, a well-studied epigenetic modification, regulates various cellular processes critical for cancer development and progression and holds promise as a biomarker for cancer diagnosis and prognosis, potentially enhancing the efficacy of precision therapies.

We developed a robust prognostic model for BC based on DNA methylation and clinical data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). We analyzed the association of the model with clinicopathological features, survival outcomes, and chemotherapy drug sensitivity.

A set of 216 differentially methylated CpGs was identified by intersecting three datasets (TCGA, GSE22249, and GSE66695). Using univariate Cox proportional hazard and LASSO Cox regression analyses, we constructed a 14-CpG model significantly associated with progression-free interval (PFI), disease-specific survival (DSS), and overall survival (OS) in BC patients. Kaplan-Meier (KM) survival analysis, receiver operating characteristic (ROC) analysis, and nomogram validation confirmed the clinical value of the signature. The Cox analysis showed a significant association between the signature and PFI and DSS in BC patients. KM analysis effectively distinguished high-risk from low-risk patients, while ROC analysis demonstrated high sensitivity and specificity in predicting BC prognosis. A nomogram based on the signature effectively predicted 5- and 10-year PFI and DSS. Additionally, combining our model with clinical risk factors suggested that patients in the I-II & M+ subgroup could benefit from adjuvant chemotherapy regarding PFI, DSS, and OS. Gene Ontology (GO) functional enrichment and KEGG pathway analyses indicated that the top 3,000 differentially expressed genes (DEGs) were enriched in pathways related to DNA replication and repair and cell cycle regulation. Patients in the high-risk group might benefit from drugs targeting DNA replication and repair processes in tumor cells.

The 14-CpG model serves as a useful biomarker for predicting prognosis in BC patients. When combined with TNM staging, it offers a potential strategy for individualized clinical decision-making, guiding personalized therapeutic regimen selection for clinicians.

Early breast cancer subtypes classification improves the survival rate as it facilitates prognosis of the patient. In literature this problem was prominently solved by various Machine Learning and Deep Learning techniques. However, these studies have three major shortcomings: huge Trainable Weight Parameters (TWP), suffer from low performance and class imbalance problem.

This paper proposes a lightweight model named EpiBrCan-Lite for classifying breast cancer subtypes using DNA methylation data. This model encompasses three blocks namely Data Encoding, TransGRU, and Classification blocks. In Data Encoding block, the input features are encoded into equal sized chunks and then passed down to TransGRU block which is a modified version of traditional Transformer Encoder (TE). In TransGRU block, MLP module of traditional TE is replaced by GRU module, consisting of two GRU layers to reduce TWP and capture the long-range dependencies of input feature data. Furthermore, output of TransGRU block is passed to Classification block for classifying breast cancer into their subtypes.

The proposed model is validated using Accuracy, Precision, Recall, F1-score, FPR, and FNR metrics on TCGA breast cancer dataset. This dataset suffers from the class imbalance problem which is mitigated using Synthetic Minority Oversampling Technique (SMOTE). Experimentation results demonstrate that EpiBrCan-Lite model attained 95.85 % accuracy, 95.96 % recall, 95.85 % precision, 95.90 % F1-score, 1.03 % FPR, and 4.12 % FNR despite of utilizing only 1/1500 of TWP than other state-of-the-art models.

EpiBrCan-Lite model is efficiently classifying breast cancer subtypes, and being lightweight, it is suitable to be deployed on low computational powered devices.

The landscape of breast cancer management has been revolutionized by advancements in molecular diagnostics, yet accessibility and cost remain significant barriers for many patients. Limited-resource settings often face significant challenges in accessing expensive molecular tests, which can impact timely diagnosis and treatment decisions. The Magee equations present a practical, cost-effective solution that can bridge this gap, ensuring that patients, regardless of their financial or geographic limitations, receive appropriate and timely treatment. Originally developed at the University of Pittsburgh Medical Center, these equations offer a reliable alternative for estimating recurrence scores without the prohibitive costs associated with genomic assays.

Cutaneous melanoma is a type of malignant tumor that is challenging to predict and is readily stimulated by various factors. Oxidative stress can induce damage and alterations in melanocytes, subsequently triggering immune responses. Given that oxidative stress is a prevalent tumor stimulus, we aimed to enhance melanoma prediction by identifying lncRNA signatures associated with oxidative stress.

We screened for oxidative stress-related lncRNAs that could improve melanoma patient prognosis using the TCGA and GTEx databases. Utilizing differentially expressed oxidative stress-related lncRNAs (DE-OSlncRNAs), we constructed a Lasso regression model. The accuracy of the model was validated using univariate and multivariate regression, Kaplan-Meier (K-M) curves, and ROC curves. Subsequently, we conducted immune infiltration analysis, immune checkpoint differential analysis, IC50 pharmaceutical analysis, and gene set enrichment analysis. Investigating the effects of the target gene on melanoma using fluorescence in situ hybridization (FISH), quantitative real-time PCR (qRT-PCR), Edu assay, wound healing assay, transwell assay, flow cytometry, and reactive oxygen species (ROS) detection.

Thirteen lncRNAs were identified as significant prognostic factors. Four oxidative stress-related lncRNAs (COPDA1, LINC02132, LINC02812, and MIR205HG) were further validated by fluorescence in situ hybridization (FISH), with results consistent with our data analysis. LINC02132 and COPDA1 can influence the proliferation, invasion, migration, and apoptosis of melanoma.

Our findings suggest that upregulation of the LINC02132 or COPDA1 genes elevates intracellular reactive oxygen species (ROS) levels in melanoma cells, suppresses tumor cell proliferation, migration, and invasion, and promotes apoptosis. These results suggest a novel therapeutic strategy for melanoma treatment.

As the field of artificial intelligence evolves rapidly, these hallmarks are intended to capture fundamental, complementary concepts necessary for the progress and timely adoption of predictive modeling in precision oncology. Through these hallmarks, we hope to establish standards and guidelines that enable the symbiotic development of artificial intelligence and precision oncology.

NATALEE assessed efficacy and tolerability of 3 years of adjuvant ribociclib plus a nonsteroidal aromatase inhibitor (NSAI) compared with an NSAI alone in a broad population of patients with hormone receptor (HR)-positive/human epidermal growth factor 2 (HER2)-negative early breast cancer, including a select group without nodal involvement. This is the final preplanned analysis of invasive disease-free survival (iDFS).

Premenopausal/postmenopausal women and men were randomized 1 : 1 to ribociclib (n = 2549; 400 mg/day, 3 weeks on/1 week off for 36 months) plus NSAI (letrozole 2.5 mg/day or anastrozole 1 mg/day for 60 months) or NSAI alone (n = 2552). Men and premenopausal women also received goserelin (3.6 mg once every 28 days). Patients had anatomical stage IIA (N0 with additional risk factors or N1), IIB, or III disease. The primary endpoint was iDFS. Secondary efficacy endpoints were recurrence-free survival (RFS), distant DFS, and overall survival. This final iDFS analysis was planned after ∼500 events.

At data cut-off (21 July 2023), ribociclib was stopped for 1996 patients (78.3%); 1091 (42.8%) completed 3 years of ribociclib, and ribociclib treatment was ongoing for 528 (20.7%). Median follow-up for iDFS was 33.3 months. Overall, 226 and 283 iDFS events occurred with ribociclib plus NSAI versus NSAI alone, respectively. Ribociclib plus NSAI demonstrated significant iDFS benefit over NSAI alone [hazard ratio 0.749, 95% confidence interval (CI) 0.628-0.892; P = 0.0012]. The 3-year iDFS rates were 90.7% (95% CI 89.3% to 91.8%) versus 87.6% (95% CI 86.1% to 88.9%). A consistent benefit was observed across prespecified subgroups, including stage (II/III) and nodal status (positive/negative). Distant DFS and RFS favored ribociclib plus NSAI. Overall survival data were immature. No new safety signals were observed.

With longer follow-up and most patients off ribociclib, NATALEE continues to demonstrate iDFS benefit with ribociclib plus NSAI over NSAI alone in the overall population and across key subgroups. Observed adverse events remained stable.

The MammaPrint (MP) prognostic assay categorizes breast cancers into high- and low-risk subgroups, and the high-risk group can be further subdivided into high-1 (MP-H1), and very high-risk high-2 (MP-H2). The aim of this analysis was to assess clinical and molecular differences between the hormone receptor-positive (HR+)/HER2-negative MP-H1, -H2, and triple-negative (TN) MP-H1 and -H2 cancers.

Pretreatment gene expression data from 742 HER2-negative breast cancers enrolled in the I-SPY2 neoadjuvant trial were used. Prognostic risk categories were assigned using the MP assay. Transcriptional similarities across the four receptor and prognostic groups were assessed using principal component analyses and by identifying differentially expressed genes. We also examined pathologic complete response rates and event-free survivals by risk group.

Principal component analysis showed that HR+/MP-H2 tumors clustered with TN/MP-H2 cancers. Only 125 genes showed differential expression between the HR+/MP-H2 and TN/MP-H2 cancers, whereas 1,465 genes were differentially expressed between HR+/MP-H2 and -H1. Gene set analysis revealed similarly high expression of cell cycle, DNA repair, and immune infiltration-related pathways in HR+/MP-H2 and TN/MP-H2 cancers. HR+/MP-H2 cancers also showed low estrogen receptor-related gene expression. Pathologic complete response rates were similarly high in TN/MP-H2 and HR+/MP-H2 cancers (42% vs. 30.5%; P = 0.11), and MP-H2 cancers with residual cancer had similarly poor event-free survival regardless of estrogen receptor status.

In conclusion, HR+/MP-H2 cancers closely resemble TN breast cancers in transcriptional and clinical features and benefit from similar treatment strategies.

Intense research on founding members of the RAS superfamily has defined our understanding of these critical signalling proteins, leading to the premise that small GTPases function as molecular switches dependent on differential nucleotide loading. The closest homologs of H/K/NRAS are the three-member RRAS family, and interest in the MRAS GTPase as a regulator of MAPK activity has recently intensified. We show here that MRAS does not function as a classical switch and is unable to exchange GDP-to-GTP in solution or when tethered to a lipid bilayer. The exchange defect is unaffected by inclusion of the GEF SOS1 and is conserved in a distal ortholog from nematodes. Synthetic activating mutations widely used to study the function of MRAS in a presumed GTP-loaded state do not increase exchange, but instead drive effector binding due to sampling of an activated conformation in the GDP-loaded state. This includes nucleation of the SHOC2-PP1Cα holophosphatase complex. Acquisition of NMR spectra from isotopically labeled MRAS in live cells validated the GTPase remains fully GDP-loaded, even a supposed activated mutant. These data show that RAS GTPases, including those most similar to KRAS, have disparate biochemical activities and challenge current dogma on MRAS, suggesting previous data may need reinterpretation.

Endosomes play a pivotal role in cellular biology, orchestrating processes such as endocytosis, molecular trafficking, signal transduction, and recycling of cellular materials. This study aims to construct an endosome-related gene (ERG)-derived risk signature for breast cancer prognosis. Transcriptomic and clinical data were retrieved from The Cancer Genome Atlas and the University of California Santa Cruz databases to build and validate the model. A Lasso Cox regression model was employed for risk signature construction. The immune landscape was assessed using CIBERSORT and ESTIMATE algorithms, while drug sensitivity was evaluated via the pRRophetic algorithm. Gene set enrichment analysis and gene set variation analysis were applied to evaluate gene expression patterns. A nomogram was constructed and validated for predicting breast cancer outcomes. The expression of ERGs in breast cancer cells and tissues was further validated. Sixty-one ERGs associated with breast cancer prognosis were identified, with 23 selected for constructing the risk signature. This signature stratified breast cancer patients into high- and low-risk groups, where the low-risk group exhibited significantly better prognosis. Notably, younger patients tended to have lower risk scores compared to older ones. The low-risk group exhibited enhanced sensitivity to the majority of the drugs tested, accompanied by increased infiltration of T cells and M1 macrophages. Additionally, cell cycle pathways were suppressed in the low-risk group, whereas antigen binding functions were significantly activated. Ultimately, risk score and age were identified as independent prognostic factors for breast cancer, and these factors were incorporated into a nomogram that demonstrated excellent performance in prognosis assessment. Finally, external cohort validated the dysregulation of the risk score-associated ERGs in breast cancer cells and tissues. This study successfully established an ERG-derived breast cancer risk signature and nomogram, elucidating their potential value in prognosis prediction and evaluation of therapeutic response.

What sense does it make to say that a new program implemented in a community with roots as old as evolution caused an observed health benefit? Evaluation of community approaches has often sought to isolate the causal roles of interventions. Central to this is the assumption that there are causes to be proven and isolated. Benedict Spinoza (1632-1677) dismissed the concept of cause, arguing that all things, "substances," are not caused but simply are. Actions of things in nature can influence each other, e.g., erosion of a mountain, but their substance, the mountains simply are. For Spinoza, satisfaction in life comes from realizing and acting in accord with our substance, but this requires communities that support such realization and action. Thus, communities and the vast influences they contain are central to human welfare. Interventions within them do not cause benefits but join with the history, culture, and numerous other features of the community in becoming part of how the community influences its members. Implications include a) expanding the social ecological model fully to embrace multiple influences - including innovative programs - and interactions among them, and c) varied research methods to identify practical lessons about how communities may adopt and incorporate innovations to engender change, rather than a catalogue of interventions that are supposed to change them.

Lung metastasis is a critical and often fatal progression in cancer patients, with monocyte-derived macrophages (Mo-macs) playing multifaceted roles in this process. Despite the recognized importance of Mac-macs, most studies focus on these cells themselves, while the precise mechanisms through which tumor cells manipulate Mo-macs to promote metastasis remain poorly understood.

We developed an in vivo CRISPR screening system to identify genes involved in macrophage-dependent metastasis by depleting Mo-macs. Osteoprotegerin (OPG) was identified as the factor significantly enhances lung metastasis. We validated its function in lung metastasis by modulating the expression of OPG in an array of cell lines and performed spontaneous and experimental lung metastasis assays. Genetically engineered mice were utilized to confirm the role of RANKL-RANK signaling in OPG-mediated metastasis. Additionally, we employed different neutralizing antibodies to elucidate the roles of Mo-macs and NK cells and inhibitor to clarify the role of CXCL10 signaling.

Employing in vivo screening techniques, we elucidate the role of OPG, a protein secreted by cancer cells, in driving lung metastasis, contingent upon regulating Mo-mac activity. OPG blocks the signaling cascade between receptor activator of nuclear factor kappa-B ligand (RANKL) and its receptor RANK on Mo-macs, thereby hindering Mo-macs from secreting CXCL10, a chemokine crucial for recruiting natural killer (NK) cells that help control lung metastasis. Moreover, we observe an enrichment of OPG amplifications in metastatic cancer patients, and elevated levels of OPG expression in lung metastatic sites compared to paired primary breast cancer samples.

Our work revealed that OPG works as a lung metastasis promoting factor by blocking the RANKL-RANK-CXCL10 axis to drive the paucity of NK cells, which could be a therapeutic target for lung metastatic cancer patients.

The full list of funding supporting this study can be found in the Acknowledgements section.

Hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) breast cancer is the most common subtype, representing over two-thirds of new diagnoses. Adjuvant therapy, which encompasses various medications and treatment durations, is the standard approach for managing early stage HR+ HER2- breast cancer. Optimizing treatment is essential to minimize unnecessary side effects while addressing the biological variability inherent in HR+/HER2- breast cancers. Incorporating biological biomarkers into treatment decisions, alongside traditional clinical factors, is vital. Gene expression assays can identify patients unlikely to benefit from adjuvant chemotherapy, thereby refining treatment strategies and improving risk assessment. This paper reviews evidence for several genomic tests, including Oncotype DX, MammaPrint, Breast Cancer Index, RucurIndex, and EndoPredict, which assist in tailoring adjuvant therapy. Additionally, we explore the role of liquid biopsies in personalizing treatment, emphasizing the importance of considering late relapse risks and potential benefits of extended systemic therapy for HR+/HER2- breast cancer patients.

Studies have reported risk factors for a single-squamous cell carcinoma(Single-SCCs). However, the shared common germline genetic factors and environmental factors have not been well elucidated with respect to augmented risk of pan-squamous cell carcinoma(Pan-SCCs). By integrating a large-scale genotype data of 1,928 Pan-SCCs cases and 7,712 age- and sex-matched controls in the UK Biobank cohort, as well as multiple transcriptome and protein databases, we conducted a multi-omics analysis. Genome-wide association analysis (GWAS) was used to identify genetic susceptibility loci of SCCs. High resolution human leucocyte antigen (HLA) alleles and corresponding amino acid sequences were imputed using SNP2HLA and tested for association with SCCs. Credible risk variants (CRVs) were combined risk SNPs reported in GWAS Catalog and our study, followed by comprehensive bioinformatics analyses. We identified six novel index SNPs in the progression of SCCs, which were also strongly interacted with fresh fruit intake. Moreover, our study systematically characterize the HLA variants and their relationship to SCCs susceptibility. We identified HLA-A*01 and six HLA-A amino acid position were associated independently with SCCs. Credible risk variants were annotated to 469 target genes, further GO and KEGG Pathway Enrichment Analysis showed that SCCs genes were primarily involved in immune-related pathways, espechially regulated by HLA region. The transcriptome analysis showed that there were 270 differentially expressed genes(DEGs), with the upregulated genes were enriched in the regulation of stem cell differentiation, proliferation, development, and maintenance. The PPI Network and Modular Analysis uncovered the Keratin(KRT) genes may serve as a potential marker in SCCs. Our results illustrate the molecular basis of both well-studied and new susceptibility loci of SCCs, providing not only novel insights into the genetic commonality among SCCs but also a set of plausible gene targets for post-GWAS functional experiments.

The positive predictive value (PPV) and negative predictive value (NPV) can be expressed as functions of disease prevalence (ρ $$ \rho $$ ) and the ratios of two binomial proportions (ϕ $$ \phi $$ ), whereϕ ppv = 1 - specificity sensitivity $$ {\phi}_{ppv}=\frac{1- specificity}{sensitivity} $$ andϕ npv = 1 - sensitivity specificity $$ {\phi}_{npv}=\frac{1- sensitivity}{specificity} $$ . In prospective studies, where the proportion of subjects with the disease in the study cohort is an unbiased estimate of the disease prevalence, the confidence intervals (CIs) of PPV and NPV can be estimated using established methods for single proportion. However, in enrichment studies, such as case-control studies, where the proportion of diseased subjects significantly differs from disease prevalence, estimating CIs for PPV and NPV remains a challenge in terms of skewness and overall coverage, especially under extreme conditions (e.g.,NPV = 1 $$ \mathrm{NPV}=1 $$ ). In this article, we extend the method adopted by Li, where CIs for PPV and NPV were derived from those ofϕ $$ \phi $$ . We explored additional CI methods forϕ $$ \phi $$ , including those by Gart & Nam (GN), MoverJ, and Walter and convert their corresponding CIs for PPV and NPV. Through simulations, we compared these methods with established CI methods, Fieller, Pepe, and Delta in terms of skewness and overall coverage. While no method proves universally optimal, GN and MoverJ methods generally emerge as recommended choices.

The prognostic role of preoperative carcinoembryonic antigen (CEA) in breast cancer is recognized, but the impact of postoperative CEA levels on survival in early breast cancer is uncertain.

We conducted a retrospective study of 921 non-metastatic breast cancer patients treated at anonymized. Patients were categorized as normal (CEA ≤3 µg/L) or elevated (CEA >3 µg/L).

Elevated postoperative CEA levels were associated with shorter disease-free survival (DFS) (median, 174.6 vs. 239.8 months; hazard ratio (HR): 1.80; 95% confidence interval (CI): 1.27-2.56; p < 0.001) and overall survival (OS) (median, 174.6 vs. 261.1 months; HR:2.34; 95% CI: 1.59-3.45; p < 0.001). Elevated CEA was associated with shorter DFS (median, 174.6 months vs. not reached (NR); HR:2.30; 95% CI: 1.03-5.19; p = 0.043) and OS (NR vs. NR; HR: 2.81; 95% CI: 1.06-7.48; p = 0.039) in stage 1, shorter DFS (median, 239. 8 vs. 141.1 months; HR: 1.95; 95% CI: 1.28-2.98; p = 0.002) and OS (median, 169 vs. 261.1 months; HR: 2.56; 95% CI: 1.6-4.12; p < 0.001) in stage 2 and shorter OS (median, 65 vs. 183.1 months; HR: 3.25; 95% CI: 1.19-8.83; p = 0.021) in stage 3.

Elevated postoperative CEA indicates worse DFS and OS in patients with HR-positive/HER2-negative early breast cancer.

In breast cancer treatment, accurately predicting how long a patient will survive is crucial for decision-making. This information guides treatment choices and supports patients' psychological recovery. To address this challenge, we introduce a novel predictive model to forecast breast cancer prognosis by leveraging diverse data sources, including clinical records, copy number variation, gene expressions, DNA methylation, microRNA (miRSeq) sequencing, and whole slide image data from the TCGA Database. The methodology incorporates graph contrastive learning with cross-modality attention (CAGCL), considering all possible combinations of the six distinct data modalities. Feature embeddings are enhanced through graph contrastive learning, which identifies subtle differences and similarities among samples. Further, to learn the complementary nature of information across multiple data modalities, a cross-attention framework is proposed and applied to the graph contrastive learning-based extracted features from various data sources for breast cancer survival prediction. It performs a binary classification to anticipate the likelihood of short- and long-term breast cancer survivors, delineated by a five-year threshold. The proposed model (CAGCL) showcases superior performance compared to baseline models and other state-of-the-art models. The model attains an accuracy of 0.932, a sensitivity of 0.954, a precision of 0.958, an F1 score of 0.956, and an AUC of 0.948, underscoring its effectiveness in predicting breast cancer survival.

Cancer immunotherapy has emerged as an effective, personalized treatment for certain patients, particularly for those with hematological malignancies. However, its efficacy in breast cancer has been marginal-perhaps due to cold, immune-excluded, or immune-desert tumors. Natural killer T (NKT) cells play a critical role in cancer immune surveillance and are reduced in cancer patients. Thus, we hypothesized that NKT cells could serve as a surrogate marker for immune function. In order to assess which breast cancer patients would likely benefit from immune cell-based therapies, we have developed a quantitative method to rapidly assess NKT function using stimulation with artificial antigen presenting cells followed by quantitative real-time PCR for IFN-γ. We observed a significant reduction in the percentage of circulating NKT cells in breast cancer patients, compared to healthy donors; however, the majority of patients had functional NKT cells. When we compared BC patients with highly functional NKT cells, as indicated by high IFN-γ induction, to those with little to no induction, following stimulation of NKT cells, there was no significant difference in NKT cell number between the groups, suggesting functional loss has more impact than physical loss of this subpopulation of T cells. In addition, we assessed the percentage of tumor-infiltrating lymphocytes and PD-L1 expression within the tumor microenvironment in the low and high responders. Further characterization of immune gene signatures in these groups identified a concomitant decrease in the induction of TNFα, LAG3, and LIGHT in the low responders. We next investigated the mechanisms by which breast cancers suppress NKT-mediated anti-tumor immune responses. We found that breast cancers secrete immunosuppressive lipids, and treatment with commonly prescribed medications that modulate lipid metabolism, can reduce tumor growth and restore NKT cell responses.

The aim of this study was to assess the prognostic performance of the 70-gene signature, MammaPrint, in an Italian single-center prospective cohort of early-stage intermediate-risk breast cancer (BC) patients.

A total of 195 eligible early BC cases were tested for genomic risk between 2006 and 2013. In this retrospective analysis, the association of genomic risk with distant metastasis-free survival (DMFS) and overall survival (OS) were assessed using Cox regression models, adjusting for clinical and pathological tumor characteristics.

MammaPrint identified 118 (60.5 %) patients with genomically Low Risk tumors and 77 (39.5 %) patients with genomically High Risk tumors. Age, menopausal status, tumor size, receptor status, and nodal status were comparable between MammaPrint Risk categories. The median follow-up was 8.4 years for DMFS and 9.3 years for OS; 8-year follow-up was reported for both endpoints. The 8-year DMFS was 90.4 % (95 % CI 84.9-95.9) in patients with MammaPrint Low Risk tumors compared to 60.8 % (95 % CI 49.8-71.8) for patients with High Risk tumors. Patients with MammaPrint Low Risk tumors exhibited significantly superior 8-year OS (97.3 %; 95 % CI 94.4-100) compared with MammaPrint High Risk tumors (89.5 %; 95 % CI 82.6-96.4; p = 0.028). Multivariate analyses identified MammaPrint as significantly associated with 8-year DMFS and MammaPrint together with Progesterone Receptor positivity with 8-year OS.

The prognostic performance of MammaPrint was demonstrated in early-stage clinically intermediate to high-risk BC patients. Moreover, patients with MammaPrint Low Risk tumors had good outcome regardless of treatment regimen, thus supporting personalized treatment choices.

Tumors, as chronic malignant diseases that account for about 20% of all deaths worldwide, are the number one threat to human health. Until now there is no reliable treatment for most types of tumors. Tumorigenesis and cellular carcinogenesis remain difficult challenges due to their complex etiology and unknown mechanisms. As stress process regulating molecules and protein folding promoters, heat shock proteins (HSPs) play an important role in cancer development. Most studies have shown that HSPs are one of the major anticancer drug targets. HSPs are not only modulators of the cellular stress response, but are also closely associated with tumor initiation, progression, and drug resistance, so understanding the mechanism of the HSP family involved in cellular carcinogenesis is an important part of understanding tumorigenesis and enabling anticancer drug development. In this review, we discuss the functions and mechanisms of key members of the HSP family (HSP70, HSP90, and HSP110) in participating in the process of tumorigenesis and cell carcinogenesis, and look forward to the prospect of key members of the HSP family in targeted cancer therapy.

Molecular profiling is a crucial aspect of cancer therapy selection, underscoring the necessity for representative sampling of both tumor and normal tissues. While much attention has been given to representative tumor sampling, there has been a notable lack of exploration into the issue of poor RNA quality in normal breast tissue processing. Normal breast tissue from the same patient is often used as a negative control for most "-omics" experiments. RNA extracted from normal breast tissues frequently contains nucleic acids from surrounding adipocytes, endothelial cells, and immune cells, leading to a low representation of ductal elements and skewed results. Therefore, ensuring a complete representation of breast glandular tissue is imperative. The study aimed to investigate the variations in RNA enrichment between a random sampling technique and a targeted sampling approach when visually selecting normal breast tissue sections as negative controls for "-omics" experiments. Fifteen female breast cancer subjects who underwent Modified Radical Mastectomy were selected for the study. Normal Breast tissue was visually examined, and samples were collected from random fat pockets (random sampling) and fibromuscular grey-white streak areas (targeted sampling). RNA was isolated, followed by spectrophotometric analysis, agarose gel electrophoresis and Agilent Tape station analysis. Histopathological assessments and a gene expression study for housekeeping genes were performed on both subsets. Tissues collected through targeted sampling exhibited significantly higher RNA quality than those obtained via random sampling. Histopathological analysis revealed cellular areas abundant in terminal ductular units within the targeted samples, and a final validation qPCR showed that the targeted samples were the most representative of normal breast glandular tissue. The comparative analysis of the two sampling methods clearly indicates that the targeted approach, with its superior accuracy and reliability, is the more practical choice for obtaining representative normal breast glandular tissue for "-omics" experiments.

There is a need for biomarkers that predict late recurrence risk and extended endocrine therapy (EET) benefit among patients with early-stage breast cancer (EBC). MammaPrint, a 70-gene expression risk-of-recurrence assay, has been found to project significant EET benefit in patients with assay-classified low-risk tumors.

To determine the test's utility in identifying which patients with EBC in the IDEAL (Investigation on the Duration of Extended Adjuvant Letrozole) trial could benefit from 5-year vs 2.5-year letrozole treatment.

This secondary analysis of the IDEAL randomized clinical trial evaluated postmenopausal women with hormone receptor-positive EBC who were assigned to either 2.5 or 5 years of EET, with 10 years of follow-up after randomization. A 70-gene assay was used to classify tumors as high, low, or ultralow risk. Adverse event (AE) frequency and treatment compliance were evaluated. Statistical analyses were performed from April 2022 to September 2024.

After 5 years of endocrine therapy, patients were randomized to 2.5 or 5 years of EET with letrozole.

Primary end point was distant recurrence (DR). Cox proportional hazard regression models and likelihood ratios tested the interaction between treatment and gene expression assay.

Among 515 women included (mean [SD] age at randomization, 59.9 [9.5] years), 265 were in the 2.5-year treatment arm and 250 in the 5-year treatment arm. Of these patients, 223 (43.3%) patients with 70-gene assay-classified low-risk tumors had a significant absolute benefit of 10.1% for DR (hazard ratio, 0.32; 95% CI, 0.12-0.87; P = .03). Treatment interaction was not significant for DR. Of patients with either 70-gene assay-classified high-risk tumors (259 [50.3%]) or ultralow risk tumors (33 [6.4%]), 5 years vs 2.5 years of EET was not associated with improved benefit for DR. As expected, rates of AEs and treatment discontinuation were comparable among the different 70-gene assay risk groups in each treatment arm.

This secondary analysis of the IDEAL trial found that the 70-gene assay identified patients with low-risk tumors who could benefit from 5-year vs 2.5-year EET. These findings suggest that this gene expression assay could go beyond guiding neoadjuvant and adjuvant chemotherapy decisions to informing the optimal duration of adjuvant endocrine therapy.

EU Clinical Trials Register Eudra CT: 2006-003958-16.

Long-term survival varies among hormone receptor-positive (HR+) and human epidermal growth factor receptor 2-negative (HER2-) breast cancer patients and is seriously impaired by metastasis. Chromosomal instability (CIN) was one of the key drivers of breast cancer metastasis. Here we evaluate CIN and 10-year invasive disease-free survival (iDFS) and overall survival (OS) in HR+/HER2-- breast cancer. In this large-scale, multiple-site, retrospective study, 354 HR+/HER2- breast cancer patients were recruited. Of these, 204 patients were used for internal training, 70 for external validation, and 80 for cross-validation. All medical records were carefully reviewed to obtain the disease recurrence information. Formalin-fixed paraffin-embedded tissue samples were collected, followed by low-pass whole-genome sequencing with a median genome coverage of 1.86X using minimal 1 ng DNA input. CIN was then assessed using a customized bioinformatics workflow. Three or more instances of CIN per sample was defined as high CIN and the frequency was 42.2% (86/204) in the internal cohort. High CIN correlated significantly with increased lymph node metastasis, vascular invasion, progesterone receptor negative status, HER2 low, worse pathological type, and performed as an independent prognostic factor for HR+/- breast cancer. Patients with high CIN had shorter iDFS and OS than those with low CIN [10-year iDFS 11.1% versus 82.2%, hazard ratio (HR) = 11.12, p < 0.01; 10-year OS 45.7% versus 94.3%, HR = 14.17, p < 0.01]. These findings were validated in two external cohorts with 70 breast cancer patients. Moreover, high CIN could predict the prognosis more accurately than Adjuvant! Online score (10-year iDFS 11.1% versus 48.6%, HR = 2.71, p < 0.01). Cross-validation analysis found that high consistency (83.8%) was observed between CIN and MammaPrint score, while only 45% between CIN and Adjuvant! Online score. In conclusion, high CIN is an independent prognostic indicator for HR+/HER2- breast cancer with shorter iDFS and OS and holds promise for predicting recurrence and metastasis.

We investigated the mRNA expression of 124 cuproptosis-associated genes in 7489 biopsies from 20 different tumor types of The Cancer Genome Atlas (TCGA). The KM plotter algorithm has been used to calculate Kaplan-Meier statistics and false discovery rate (FDR) corrections. Interaction networks have been generated using Ingenuity Pathway Analysis (IPA). High mRNA expression of 63 out of 124 genes significantly correlated with shorter survival times of cancer patients across all 20 tumor types. IPA analyses revealed that their gene products were interconnected in canonical pathways (e.g., cancer, cell death, cell cycle, cell signaling). Four tumor entities showed a higher accumulation of genes than the other cancer types, i.e., renal clear cell carcinoma (n = 21), renal papillary carcinoma (n = 13), kidney hepatocellular carcinoma (n = 13), and lung adenocarcinoma (n = 9). These gene clusters may serve as prognostic signatures for patient survival. These signatures were also of prognostic value for tumors with high mutational rates and neoantigen loads. Cuproptosis is of prognostic significance for the survival of cancer patients. The identification of specific gene signatures deserves further exploration for their clinical utility in routine diagnostics.

Benign breast tumors are a nonthreatening condition defined as abnormal cell growth within the breast without the ability to invade nearby tissue. However, benign lesions hold valuable biological information that can lead us toward better understanding of tumor biology. In this study, we have used two pathway analysis algorithms, Pathifier and gene set variation analysis (GSVA), to identify biological differences between normal breast tissue, benign tumors and malignant tumors in our clinical dataset. Our results revealed that one-third of all pathways that were significantly different between benign and malignant tumors were immune-related pathways, and 227 of them were validated by both methods and in the METABRIC dataset. Furthermore, five of these pathways (all including genes involved in cytokine and interferon signaling) were related to overall survival in cancer patients in both datasets. The cellular moieties that contribute to immune differences in malignant and benign tumors were analyzed using the deconvolution tool, CIBERSORT. The results showed that levels of some immune cells were specifically higher in benign than in malignant tumors, and this was especially the case for resting dendritic cells and follicular T-helper cells. Understanding the distinct immune profiles of benign and malignant breast tumors may aid in developing noninvasive diagnostic methods to differentiate between them in the future.

We present a survey of the current state-of-the-art in breast cancer detection and prognosis. We analyze the evolution of Artificial Intelligence-based approaches from using just uni-modal information to multi-modality for detection and how such paradigm shift facilitates the efficacy of detection, consistent with clinical observations. We conclude that interpretable AI-based predictions and ability to handle class imbalance should be considered priority.

Breast cancer (BCa) is one of the most common malignancies among women worldwide. It is a complex disease that is characterized by morphological and molecular heterogeneity. In the early stages of the disease, most BCa cases are treatable, particularly hormone receptor-positive and HER2-positive tumors. Unfortunately, triple-negative BCa and metastases to distant organs are largely untreatable with current medical interventions. Recent advances in sequencing and proteomic technologies have improved our understanding of the molecular changes that occur during breast cancer initiation and progression. In this era of precision medicine, researchers and clinicians aim to identify subclass-specific BCa biomarkers and develop new targets and drugs to guide treatment. Although vast amounts of omics data including single cell sequencing data, can be accessed through public repositories, there is a lack of user-friendly platforms that integrate information from multiple studies. Thus, to meet the need for a simple yet effective and integrative BCa tool for multi-omics data analysis and visualization, we developed a comprehensive BCa data analysis platform called MammOnc-DB (http://resource.path.uab.edu/MammOnc-Home.html), comprising data from more than 20,000 BCa samples. MammOnc-DB was developed to provide a unique resource for hypothesis generation and testing, as well as for the discovery of biomarkers and therapeutic targets. The platform also provides pre- and post-treatment data, which can help users identify treatment resistance markers and patient groups that may benefit from combination therapy.

Recent advancements in image classification have demonstrated that contrastive learning (CL) can aid in further learning tasks by acquiring good feature representation from a limited number of data samples. In this paper, we applied CL to tumor transcriptomes and clinical data to learn feature representations in a low-dimensional space. We then utilized these learned features to train a classifier to categorize tumors into a high- or low-risk group of recurrence. Using data from The Cancer Genome Atlas (TCGA), we demonstrated that CL can significantly improve classification accuracy. Specifically, our CL-based classifiers achieved an area under the receiver operating characteristic curve (AUC) greater than 0.8 for 14 types of cancer, and an AUC greater than 0.9 for 3 types of cancer. We also developed CL-based Cox (CLCox) models for predicting cancer prognosis. Our CLCox models trained with the TCGA data outperformed existing methods significantly in predicting the prognosis of 19 types of cancer under consideration. The performance of CLCox models and CL-based classifiers trained with TCGA lung and prostate cancer data were validated using the data from two independent cohorts. We also show that the CLCox model trained with the whole transcriptome significantly outperforms the Cox model trained with the 16 genes of Oncotype DX that is in clinical use for breast cancer patients. The trained models and the Python codes are publicly accessible and provide a valuable resource that will potentially find clinical applications for many types of cancer.

Breast cancer is a common disease that causes great health concerns to women worldwide. During the diagnosis and treatment of breast cancer, medical imaging plays an essential role, but its interpretation relies on radiologists or clinical doctors. Radiomics can extract high-throughput quantitative imaging features from images of various modalities via traditional machine learning or deep learning methods following a series of standard processes. Hopefully, radiomic models may aid various processes in clinical practice. In this review, we summarize the current utilization of radiomics for predicting clinicopathological indices and clinical outcomes. We also focus on radio-multi-omics studies that bridge the gap between phenotypic and microscopic scale information. Acknowledging the deficiencies that currently hinder the clinical adoption of radiomic models, we discuss the underlying causes of this situation and propose future directions for advancing radiomics in breast cancer research.

Maternal Embryonic Leucine Zipper Kinase (MELK) has been studied intensively in recent years due to its overexpression in multiple cancers. However, the cell biology of MELK remains less characterized despite its well-documented association with mitosis. Here we report a distinctive pattern of human MELK that translocates from the cytoplasm to cell cortex within 3 min of anaphase onset. The cortex association lasts about 30 min till telophase. The spatiotemporal specific localization of MELK depends on the interaction between its Threonine-Proline (TP) rich domain and kinase associated 1 (KA1) domain, which is regulated by CDK1 kinase and PP4 protein phosphatase. KA1 domains are known to regulate kinase activities through various intramolecular interactions. Our results revealed a new role for KA1 domain to control subcellular localization of a protein kinase.

Despite evidence indicating the dominance of cell-of-origin signatures in molecular tumor patterns, translating these genome-wide patterns into actionable insights has been challenging. This study introduces breast cancer cell-of-origin signatures that offer significant prognostic value across all breast cancer subtypes and various clinical cohorts, compared to previously developed genomic signatures.

We previously reported that triple hormone receptor (THR) co-expression patterns of androgen (AR), estrogen (ER), and vitamin D (VDR) receptors are maintained at the protein level in human breast cancers. Here, we developed corresponding mRNA signatures (THR-50 and THR-70) based on these patterns to categorize breast tumors by their THR expression levels. The THR mRNA signatures were evaluated across 56 breast cancer datasets (5040 patients) using Kaplan-Meier survival analysis, Cox proportional hazard regression, and unsupervised clustering.

The THR signatures effectively predict both overall and progression-free survival across all evaluated datasets, independent of subtype, grade, or treatment status, suggesting improvement over existing prognostic signatures. Furthermore, they delineate three distinct ER-positive breast cancer subtypes with significant survival in differences-expanding on the conventional two subtypes. Additionally, coupling THR-70 with an immune signature identifies a predominantly ER-negative breast cancer subgroup with a highly favorable prognosis, comparable to ER-positive cases, as well as an ER-negative subgroup with notably poor outcome, characterized by a 15-fold shorter survival.

The THR cell-of-origin signature introduces a novel dimension to breast cancer biology, potentially serving as a robust foundation for integrating additional prognostic biomarkers. These signatures offer utility as a prognostic index for stratifying existing breast cancer subtypes and for de novo classification of breast cancer cases. Moreover, THR signatures may also hold promise in predicting hormone treatment responses targeting AR and/or VDR.