Prediction of Lymph Node Metastasis in Breast Cancer by Gene Expression and Clinicopathological Models: Development and Validation within a Population-Based Cohort

Multimodal AI model for preoperative prediction of axillary lymph node metastasis in breast cancer using whole slide images

In breast cancer management, predicting axillary lymph node (ALN) metastasis using whole-slide images (WSIs) of primary tumor biopsies is a challenging and underexplored task for pathologists. We developed METACANS, an multimodal artificial intelligence (AI) model that integrates WSIs with clinicopathological features to predict ALN metastasis. METACANS was trained on 1991 cases and externally validated across five cohorts with a total of 2166 cases. Across all validation cohorts, METACANS achieved an area under the curve (AUC) of 0.733 (95% CI, 0.711-0.755), with an overall negative predictive value of 0.846, sensitivity of 0.820, specificity of 0.504, and balanced accuracy of 0.662. Without additional annotations, METACANS identified pathological imaging patterns linked to metastatic status, such as micropapillary growth, infiltrative patterns, and necrosis. While its predictive performance may not yet support immediate clinical application, METACANS addresses the task of predicting ALN metastasis using WSIs and clinicopathological features, and demonstrates the feasibility of multimodal AI approaches for preoperative axillary staging in breast cancer.

Applications of gene pair methods in clinical research: advancing precision medicine

The rapid evolution of high-throughput sequencing technologies has revolutionized biomedical research, producing vast amounts of gene expression data that hold immense potential for biological discovery and clinical applications. Effectively mining these large-scale, high-dimensional data is crucial for facilitating disease detection, subtype differentiation, and understanding the molecular mechanisms underlying disease progression. However, the conventional paradigm of single-gene profiling, measuring absolute expression levels of individual genes, faces critical limitations in clinical implementation. These include vulnerability to batch effects and platform-dependent normalization requirements. In contrast, emerging approaches analyzing relative expression relationships between gene pairs demonstrate unique advantages. By focusing on binary comparisons of two genes' expression magnitudes, these methods inherently normalize experimental variations while capturing biologically stable interaction patterns. In this review, we systematically evaluate gene pair-based analytical frameworks. We classify eleven computational approaches into two fundamental categories: expression value-based methods quantifying differential expression patterns, and rank-based methods exploiting transcriptional ordering relationships. To bridge methodological development with practical implementation, we establish a reproducible analytical pipeline incorporating feature selection, classifier construction, and model evaluation modules using real-world benchmark datasets from pulmonary tuberculosis studies. These findings position gene pair analysis as a transformative paradigm for mining high-dimensional omics data, with direct implications for precision biomarker discovery and mechanistic studies of disease progression.

Breast Cancer Specificities of Patients With Anti-Ri Paraneoplastic Neurologic Syndromes

BACKGROUND AND OBJECTIVES

Breast cancers (BCs) of patients with paraneoplastic neurologic syndromes and anti-Yo antibodies (Yo-PNS) overexpress human epidermal growth factor receptor 2 (HER2) and display genetic alterations and overexpression of the Yo-onconeural antigens. They are infiltrated by an unusual proportion of B cells. We investigated whether these features were also observed in patients with PNS and anti-Ri antibodies (Ri-PNS).

METHODS

Using clinicopathologic data, DNA sequencing, and whole-transcriptome analysis, 28 BCs associated with Ri-PNS were characterized regarding oncological characteristics. Genetic alteration of the onconeural antigens and differential gene expression profiles were analyzed in the 12 available tumor samples and compared with those of 5 Yo-PNS tumors.

RESULTS

Ri-PNS BCs were mainly luminal B invasive carcinomas that did not overexpress HER2 and were a subtype of BCs different from the ones observed in Yo-PNS BCs. They had a low expression of wild-type TP53 and deletions of 1p chromosome. Neither overexpression nor genetic alteration of the Ri onconeural antigens was found in Ri-PNS BCs. Conversely, the nature of the antitumor immune reaction in Ri-PNS BCs was similar to the one found in Yo-PNS BCs. Ri-BCs also had a high propension for early nodal regional metastasis.

DISCUSSION

BCs associated with Ri-PNS are uncommon and the tumor subtype and their molecular and oncological characteristics are different from those of Yo-PNS and controls. Overexpression or sequence variation in the gene of the onconeural antigen is not mandatory. Conversely, Ri-PNS-associated and Yo-PNS-associated BCs display the same atypical B-cell-mediated intratumoral immune response, and tumor escape in the lymph nodes is frequent.

Sentinel lymph node-related lncRNA typing affects breast cancer prognosis and treatment response through the immune cell microenvironment

The sentinel lymph node (SLN) plays a crucial role in the early treatment of breast cancer. The present study aims to investigate the impact of SLN-associated long noncoding RNAs (lncRNAs) on breast cancer and the influence of molecular subtyping based on related genes on prognosis. To identify SLN-associated lncRNAs, we conducted differential expression analysis using 2 high-throughput sequencing techniques. In addition, ConsensusClusterPlus was employed to establish lncRNA molecular subtypes. Subsequently, comprehensive analysis using LASSO regression was performed to construct an optimal model for predicting breast cancer prognosis. Finally, various functional annotation databases were utilized to elucidate the potential functions of the predictive model. Through differential expression analysis, we identified 14 SLN-associated lncRNAs. These genes primarily influence TNF signaling pathways. Furthermore, we found that lncRNA H19 is a prominent regulatory factor among these 14 gene expressions. By utilizing ConsensusClusterPlus, we successfully stratified the IR samples into 2 distinct subtypes. Through LASSO regression, we established a prognosis model predominantly impacting various immune cells and drug resistance. After verifying 10 pairs of organizations through PCR, we found differences in 6 lncRNAs between the 2 groups of SNLs. At the same time, in the subsequent analysis of immune infiltration and drug targets, it was found that TRPC2 plays a very critical role in breast cancer. Our study highlights the significance of SLN-associated lncRNAs, unveiling the intricate mechanisms underlying the progression of breast cancer. These findings provide novel insights and potential targets for future therapeutic interventions.

Construction of a risk prediction model for axillary lymph node metastasis in breast cancer based on gray-scale ultrasound and clinical pathological features

Purpose

This study aimed to develop a model to predict the risk of axillary lymph node (ALN) metastasis in breast cancer patients, using gray-scale ultrasound and clinical pathological features.

Methods

A retrospective analysis of 212 breast cancer patients who met the inclusion criteria from January 2011 to December 2021 was carried out. Clinical and pathological characteristics, including age, tumor size, pathological type, molecular subtype, estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and proliferation cell nuclear antigen (Ki-67), were examined. Preoperative ultrasound examinations were performed, and ultrasound radiomics features of breast cancer lesions were extracted using Pyradiomics software. The data was divided into training (70%) and testing (30%) sets. A predictive model for axillary lymph node metastasis (ALNM) was established by combining clinical and ultrasound features. The diagnostic performance of the model was evaluated using receiver operating characteristic (ROC) curves and five-fold cross-validation.

Results

The rate of lymph node metastasis was 41.51%. Using LASSO algorithm, 17 features linked to ALN metastasis were extracted from a comprehensive databank of 8 clinical features and 1314 ultrasound radiomic attributes. Of these, four were clinical-pathological features (tumor size, tumor type, age, and expression levels of the Ki-67 protein), and 13 were radiomic features. And the following features exhibited both high weights and correlation coefficients: tumor size (R=0.29, weight=0.071), tumor type (R=-0.24, weight=-0.048), wavelet-LH_glcm_Imc1 (R=0.28, weight=0.029363), wavelet-LH_glszm_SZNUN (R=-0.20, weight=-0.028507), and squareroot_ firstorder_ Minimum (R= -0.25, weight= -0.059). The ROC area under the curve for the model in the training and testing sets was 0.882 (95% CI: 0.830-0.935) and 0.853 (95% CI: 0.762-0.945), respectively. The predictive model demonstrated a sensitivity of 87.5% on the training set and 79.2% on the test set, with corresponding specificities of 75.0% and 77.5%, accuracy of 80.4% and 78.1%, respectively. When evaluated using 5-fold cross-validation, the model achieved an average test set area under the curve (AUC) of 0.799 and a training set AUC of 0.852.

Conclusion

The clinical-radiomic model has the potential to predict axillary lymph node metastasis in breast cancer.

Identification of sentinel lymph node macrometastasis in breast cancer by deep learning based on clinicopathological characteristics

The axillary lymph node status remains an important prognostic factor in breast cancer, and nodal staging using sentinel lymph node biopsy (SLNB) is routine. Randomized clinical trials provide evidence supporting de-escalation of axillary surgery and omission of SLNB in patients at low risk. However, identifying sentinel lymph node macrometastases (macro-SLNMs) is crucial for planning treatment tailored to the individual patient. This study is the first to explore the capacity of deep learning (DL) models to identify macro-SLNMs based on preoperative clinicopathological characteristics. We trained and validated five multivariable models using a population-based cohort of 18,185 patients. DL models outperform logistic regression, with Transformer showing the strongest results, under the constraint that the sensitivity is no less than 90%, reflecting the sensitivity of SLNB. This highlights the feasibility of noninvasive macro-SLNM prediction using DL. Feature importance analysis revealed that patients with similar characteristics exhibited different nodal status predictions, indicating the need for additional predictors for further improvement.

EMBER creates a unified space for independent breast cancer transcriptomic datasets enabling precision oncology

Transcriptomics has revolutionized biomedical research and refined breast cancer subtyping and diagnostics. However, wider use in clinical practice is hampered for a number of reasons including the application of transcriptomic signatures as single sample predictors. Here, we present an embedding approach called EMBER that creates a unified space of 11,000 breast cancer transcriptomes and predicts phenotypes of transcriptomic profiles on a single sample basis. EMBER accurately captures the five molecular subtypes. Key biological pathways, such as estrogen receptor signaling, cell proliferation, DNA repair, and epithelial-mesenchymal transition determine sample position in the space. We validate EMBER in four independent patient cohorts and show with samples from the window trial, POETIC, that it captures clinical responses to endocrine therapy and identifies increased androgen receptor signaling and decreased TGFβ signaling as potential mechanisms underlying intrinsic therapy resistance. Of direct clinical importance, we show that the EMBER-based estrogen receptor (ER) signaling score is superior to the immunohistochemistry (IHC) based ER index used in current clinical practice to select patients for endocrine therapy. As such, EMBER provides a calibration and reference tool that paves the way for using RNA-seq as a standard diagnostic and predictive tool for ER+ breast cancer.

Identifying Key Genes Involved in Axillary Lymph Node Metastasis in Breast Cancer Using Advanced RNA-Seq Analysis: A Methodological Approach with GLMQL and MAS

Our study aims to address the methodological challenges frequently encountered in RNA-Seq data analysis within cancer studies. Specifically, it enhances the identification of key genes involved in axillary lymph node metastasis (ALNM) in breast cancer. We employ Generalized Linear Models with Quasi-Likelihood (GLMQLs) to manage the inherently discrete and overdispersed nature of RNA-Seq data, marking a significant improvement over conventional methods such as the t-test, which assumes a normal distribution and equal variances across samples. We utilize the Trimmed Mean of M-values (TMMs) method for normalization to address library-specific compositional differences effectively. Our study focuses on a distinct cohort of 104 untreated patients from the TCGA Breast Invasive Carcinoma (BRCA) dataset to maintain an untainted genetic profile, thereby providing more accurate insights into the genetic underpinnings of lymph node metastasis. This strategic selection paves the way for developing early intervention strategies and targeted therapies. Our analysis is exclusively dedicated to protein-coding genes, enriched by the Magnitude Altitude Scoring (MAS) system, which rigorously identifies key genes that could serve as predictors in developing an ALNM predictive model. Our novel approach has pinpointed several genes significantly linked to ALNM in breast cancer, offering vital insights into the molecular dynamics of cancer development and metastasis. These genes, including ERBB2, CCNA1, FOXC2, LEFTY2, VTN, ACKR3, and PTGS2, are involved in key processes like apoptosis, epithelial-mesenchymal transition, angiogenesis, response to hypoxia, and KRAS signaling pathways, which are crucial for tumor virulence and the spread of metastases. Moreover, the approach has also emphasized the importance of the small proline-rich protein family (SPRR), including SPRR2B, SPRR2E, and SPRR2D, recognized for their significant involvement in cancer-related pathways and their potential as therapeutic targets. Important transcripts such as H3C10, H1-2, PADI4, and others have been highlighted as critical in modulating the chromatin structure and gene expression, fundamental for the progression and spread of cancer.

A radiogenomic multimodal and whole-transcriptome sequencing for preoperative prediction of axillary lymph node metastasis and drug therapeutic response in breast cancer: a retrospective, machine learning and international multicohort study

BACKGROUND

Axillary lymph nodes (ALN) status serves as a crucial prognostic indicator in breast cancer (BC). The aim of this study was to construct a radiogenomic multimodal model, based on machine learning and whole-transcriptome sequencing (WTS), to accurately evaluate the risk of ALN metastasis (ALNM), drug therapeutic response and avoid unnecessary axillary surgery in BC patients.

METHODS

In this study, conducted a retrospective analysis of 1078 BC patients from The Cancer Genome Atlas (TCGA), The Cancer Imaging Archive (TCIA), and Foshan cohort. These patients were divided into the TCIA cohort ( N =103), TCIA validation cohort ( N =51), Duke cohort ( N =138), Foshan cohort ( N =106), and TCGA cohort ( N =680). Radiological features were extracted from BC radiological images and differentially expressed gene expression was calibrated using technology. A support vector machine model was employed to screen radiological and genetic features, and a multimodal model was established based on radiogenomic and clinical pathological features to predict ALNM. The accuracy of the model predictions was assessed using the area under the curve (AUC) and the clinical benefit was measured using decision curve analysis. Risk stratification analysis of BC patients was performed by gene set enrichment analysis, differential comparison of immune checkpoint gene expression, and drug sensitivity testing.

RESULTS

For the prediction of ALNM, rad-score was able to significantly differentiate between ALN- and ALN+ patients in both the Duke and Foshan cohorts ( P <0.05). Similarly, the gene-score was able to significantly differentiate between ALN- and ALN+ patients in the TCGA cohort ( P <0.05). The radiogenomic multimodal nomogram demonstrated satisfactory performance in the TCIA cohort (AUC 0.82, 95% CI: 0.74-0.91) and the TCIA validation cohort (AUC 0.77, 95% CI: 0.63-0.91). In the risk sub-stratification analysis, there were significant differences in gene pathway enrichment between high and low-risk groups ( P <0.05). Additionally, different risk groups may exhibit varying treatment responses ( P <0.05).

CONCLUSION

Overall, the radiogenomic multimodal model employs multimodal data, including radiological images, genetic, and clinicopathological typing. The radiogenomic multimodal nomogram can precisely predict ALNM and drug therapeutic response in BC patients.

Identifying Lymph Node Metastasis-Related Factors in Breast Cancer Using Differential Modular and Mutational Structural Analysis

Complex diseases are generally caused by disorders of biological networks and/or mutations in multiple genes. Comparisons of network topologies between different disease states can highlight key factors in their dynamic processes. Here, we propose a differential modular analysis approach that integrates protein-protein interactions with gene expression profiles for modular analysis, and introduces inter-modular edges and date hubs to identify the "core network module" that quantifies the significant phenotypic variation. Then, based on this core network module, key factors, including functional protein-protein interactions, pathways, and driver mutations, are predicted by the topological-functional connection score and structural modeling. We applied this approach to analyze the lymph node metastasis (LNM) process in breast cancer. The functional enrichment analysis showed that both inter-modular edges and date hubs play important roles in cancer metastasis and invasion, and in metastasis hallmarks. The structural mutation analysis suggested that the LNM of breast cancer may be the outcome of the dysfunction of rearranged during transfection (RET) proto-oncogene-related interactions and the non-canonical calcium signaling pathway via an allosteric mutation of RET. We believe that the proposed method can provide new insights into disease progression such as cancer metastasis.

Noninvasive Staging of Lymph Node Status in Breast Cancer Using Machine Learning: External Validation and Further Model Development

BACKGROUND

Most patients diagnosed with breast cancer present with a node-negative disease. Sentinel lymph node biopsy (SLNB) is routinely used for axillary staging, leaving patients with healthy axillary lymph nodes without therapeutic effects but at risk of morbidities from the intervention. Numerous studies have developed nodal status prediction models for noninvasive axillary staging using postoperative data or imaging features that are not part of the diagnostic workup. Lymphovascular invasion (LVI) is a top-ranked predictor of nodal metastasis; however, its preoperative assessment is challenging.

OBJECTIVE

This paper aimed to externally validate a multilayer perceptron (MLP) model for noninvasive lymph node staging (NILS) in a large population-based cohort (n=18,633) and develop a new MLP in the same cohort. Data were extracted from the Swedish National Quality Register for Breast Cancer (NKBC, 2014-2017), comprising only routinely and preoperatively available documented clinicopathological variables. A secondary aim was to develop and validate an LVI MLP for imputation of missing LVI status to increase the preoperative feasibility of the original NILS model.

METHODS

Three nonoverlapping cohorts were used for model development and validation. A total of 4 MLPs for nodal status and 1 LVI MLP were developed using 11 to 12 routinely available predictors. Three nodal status models were used to account for the different availabilities of LVI status in the cohorts and external validation in NKBC. The fourth nodal status model was developed for 80% (14,906/18,663) of NKBC cases and validated in the remaining 20% (3727/18,663). Three alternatives for imputation of LVI status were compared. The discriminatory capacity was evaluated using the validation area under the receiver operating characteristics curve (AUC) in 3 of the nodal status models. The clinical feasibility of the models was evaluated using calibration and decision curve analyses.

RESULTS

External validation of the original NILS model was performed in NKBC (AUC 0.699, 95% CI 0.690-0.708) with good calibration and the potential of sparing 16% of patients with node-negative disease from SLNB. The LVI model was externally validated (AUC 0.747, 95% CI 0.694-0.799) with good calibration but did not improve the discriminatory performance of the nodal status models. A new nodal status model was developed in NKBC without information on LVI (AUC 0.709, 95% CI: 0.688-0.729), with excellent calibration in the holdout internal validation cohort, resulting in the potential omission of 24% of patients from unnecessary SLNBs.

CONCLUSIONS

The NILS model was externally validated in NKBC, where the imputation of LVI status did not improve the model's discriminatory performance. A new nodal status model demonstrated the feasibility of using register data comprising only the variables available in the preoperative setting for NILS using machine learning. Future steps include ongoing preoperative validation of the NILS model and extending the model with, for example, mammography images.

Multicenter Survival Analysis and Application of an Olfactory Neuroblastoma Staging Modification Incorporating Hyams Grade

Importance

Current olfactory neuroblastoma (ONB) staging systems inadequately delineate locally advanced tumors, do not incorporate tumor grade, and poorly estimate survival and recurrence.

Objective

The primary aims of this study were to (1) examine the clinical covariates associated with survival and recurrence of ONB in a modern-era multicenter cohort and (2) incorporate Hyams tumor grade into existing staging systems to assess its ability to estimate survival and recurrence.

Design, Setting, and Participants

This retrospective, multicenter, case-control study included patients with ONB who underwent treatment between January 1, 2005, and December 31, 2021, at 9 North American academic medical centers.

Intervention

Standard-of-care ONB treatment.

Main Outcome and Measures

The main outcomes were overall survival (OS), disease-free survival (DFS), and disease-specific survival (DSS) as C statistics for model prediction.

Results

A total of 256 patients with ONB (mean [SD] age, 52.0 [15.6] years; 115 female [44.9%]; 141 male [55.1%]) were included. The 5-year rate for OS was 83.5% (95% CI, 78.3%-89.1%); for DFS, 70.8% (95% CI, 64.3%-78.0%); and for DSS, 94.1% (95% CI, 90.5%-97.8%). On multivariable analysis, age, American Joint Committee on Cancer (AJCC) stage, involvement of bilateral maxillary sinuses, and positive margins were associated with OS. Only AJCC stage was associated with DFS. Only N stage was associated with DSS. When assessing the ability of staging systems to estimate OS, the best-performing model was the novel modification of the Dulguerov system (C statistic, 0.66; 95% CI, 0.59-0.76), and the Kadish system performed most poorly (C statistic, 0.57; 95% CI, 0.50-0.63). Regarding estimation of DFS, the modified Kadish system performed most poorly (C statistic, 0.55; 95% CI, 0.51-0.66), while the novel modification of the AJCC system performed the best (C statistic, 0.70; 95% CI, 0.66-0.80). Regarding estimation of DSS, the modified Kadish system was the best-performing model (C statistic, 0.79; 95% CI, 0.70-0.94), and the unmodified Kadish performed the worst (C statistic, 0.56; 95% CI, 0.51-0.68). The ability for novel ONB staging systems to estimate disease progression across stages was also assessed. In the novel Kadish staging system, patients with stage VI disease were approximately 7 times as likely to experience disease progression as patients with stage I disease (hazard ratio [HR], 6.84; 95% CI, 1.60-29.20). Results were similar for the novel modified Kadish system (HR, 8.99; 95% CI, 1.62-49.85) and the novel Dulguerov system (HR, 6.86; 95% CI, 2.74-17.18).

Conclusions and Relevance

The study findings indicate that 5-year OS for ONB is favorable and that incorporation of Hyams grade into traditional ONB staging systems is associated with improved estimation of disease progression.

Artificial intelligence in breast cancer histopathology

This is a review on the use of artificial intelligence for digital breast pathology. A systematic search on PubMed was conducted, identifying 17,324 research papers related to breast cancer pathology. Following a semimanual screening, 664 papers were retrieved and pursued. The papers are grouped into six major tasks performed by pathologists-namely, molecular and hormonal analysis, grading, mitotic figure counting, ki-67 indexing, tumour-infiltrating lymphocyte assessment, and lymph node metastases identification. Under each task, open-source datasets for research to build artificial intelligence (AI) tools are also listed. Many AI tools showed promise and demonstrated feasibility in the automation of routine pathology investigations. We expect continued growth of AI in this field as new algorithms mature.

A Nomogram for Predicting Liver Metastasis of Lymph-Node Positive Luminal B HER2 Negative Subtype Breast Cancer by Analyzing the Clinicopathological Characteristics of Patients with Breast Cancer

Background: Luminal B-like human epidermal growth factor receptor 2 negative (Luminal B [HER2-]) is the most common molecular subtype of breast cancer (BC). Since the relationship between Luminal B (HER2-) BC and liver metastasis (LM) is poorly defined, this retrospective study aimed to develop an LM risk nomogram for patients with lymph node-related (N + Luminal B [HER2-]) BC. Methods: Data were obtained for patients initially diagnosed with BC from the Tianjin Medical University Cancer Institute and Hospital. There were 30,975 Chinese female patients with stage I-III BC and follow-up confirming 1217 subsequent patients with LM, and 427 patients with N + Luminal B (HER2-). The LM risk was assessed using Cox proportional hazards regression, histogram, Venn diagram, and Kaplan-Meier survival analysis, with further analysis for patients with N + Luminal B (HER2-) BC. A nomogram was established based on the N + Luminal B (HER2-) BC data, which was validated using calibration plots. Results: The median age of 427 patients with N + Luminal B (HER2-) liver metastasis of breast cancer (BCLM) was 49 years. The largest number of patients with BCLM was diagnosed between the second to the 6th year, the longest interval from initial BC diagnosis to subsequent LM was 145 months. The patients with LM as the first site of distant metastasis which is associated with better survival were analyzed by Kaplan-Meier. The nomogram was constructed for the risk of LM that included age, menstrual status, unilateral oophorectomy, pregnancy, hepatitis B antigen, region of residence, tumor size, lymph node, clavicular lymph nodes, progesterone receptor, and lymph vessel invasion. Conclusion: We described the clinicopathological characteristics of patients with stage I-III BC, and constructed a nomogram for calculating personalized LM probabilities for patients with N + Luminal B (HER2-), which could guide future prolonged or early extensive treatment decisions.

Immune and Genetic Signatures of Breast Carcinomas Triggering Anti-Yo-Associated Paraneoplastic Cerebellar Degeneration

BACKGROUND AND OBJECTIVES

Paraneoplastic cerebellar degeneration (PCD) with anti-Yo antibodies is a cancer-related autoimmune disease directed against neural antigens expressed by tumor cells. A putative trigger of the immune tolerance breakdown is genetic alteration of Yo antigens. We aimed to identify the tumors' genetic and immune specificities involved in Yo-PCD pathogenesis.

METHODS

Using clinicopathologic data, immunofluorescence (IF) imaging, and whole-transcriptome analysis, 22 breast cancers (BCs) associated with Yo-PCD were characterized in terms of oncologic characteristics, genetic alteration of Yo antigens, differential gene expression profiles, and morphofunctional specificities of their in situ antitumor immunity by comparing them with matched control BCs.

RESULTS

Yo-PCD BCs were invasive carcinoma of no special type, which early metastasized to lymph nodes. They overexpressed human epidermal growth factor receptor 2 (HER2) but were hormone receptor negative. All Yo-PCD BCs carried at least 1 genetic alteration (variation or gain in copy number) on CDR2L, encoding the main Yo antigen that was found aberrantly overexpressed in Yo-PCD BCs. Analysis of the differentially expressed genes found 615 upregulated and 54 downregulated genes in Yo-PCD BCs compared with HER2-driven control BCs without PCD. Ontology enrichment analysis found significantly upregulated adaptive immune response pathways in Yo-PCD BCs. IF imaging confirmed an intense immune infiltration with an overwhelming predominance of immunoglobulin G-plasma cells.

DISCUSSION

These data confirm the role of genetic alterations of Yo antigens in triggering the immune tolerance breakdown but also outline a specific biomolecular profile in Yo-PCD BCs, suggesting a cancer-specific pathogenesis.

Predicting lymph node metastasis and prognosis of individual cancer patients based on miRNA-mediated RNA interactions

Background

Lymph node metastasis is usually detected based on the images obtained from clinical examinations. Detecting lymph node metastasis from clinical examinations is a direct way of diagnosing metastasis, but the diagnosis is done after lymph node metastasis occurs.

Results

We developed a new method for predicting lymph node metastasis based on differential correlations of miRNA-mediated RNA interactions in cancer. The types of RNAs considered in this study include mRNAs, lncRNAs, miRNAs, and pseudogenes. We constructed cancer patient-specific networks of miRNA mediated RNA interactions and identified key miRNA–RNA pairs from the network. A prediction model using differential correlations of the miRNA–RNA pairs of a patient as features showed a much higher performance than other methods which use gene expression data. The key miRNA–RNA pairs were also powerful in predicting prognosis of an individual patient in several types of cancer.

Conclusions

Differential correlations of miRNA–RNA pairs identified from patient-specific networks of miRNA mediated RNA interactions are powerful in predicting lymph node metastasis in cancer patients. The key miRNA–RNA pairs were also powerful in predicting prognosis of an individual patient of solid cancer.

Fast Segmentation of Metastatic Foci in H&E Whole-Slide Images for Breast Cancer Diagnosis

Breast cancer is the leading cause of death for women globally. In clinical practice, pathologists visually scan over enormous amounts of gigapixel microscopic tissue slide images, which is a tedious and challenging task. In breast cancer diagnosis, micro-metastases and especially isolated tumor cells are extremely difficult to detect and are easily neglected because tiny metastatic foci might be missed in visual examinations by medical doctors. However, the literature poorly explores the detection of isolated tumor cells, which could be recognized as a viable marker to determine the prognosis for T1NoMo breast cancer patients. To address these issues, we present a deep learning-based framework for efficient and robust lymph node metastasis segmentation in routinely used histopathological hematoxylin−eosin-stained (H−E) whole-slide images (WSI) in minutes, and a quantitative evaluation is conducted using 188 WSIs, containing 94 pairs of H−E-stained WSIs and immunohistochemical CK(AE1/AE3)-stained WSIs, which are used to produce a reliable and objective reference standard. The quantitative results demonstrate that the proposed method achieves 89.6% precision, 83.8% recall, 84.4% F1-score, and 74.9% mIoU, and that it performs significantly better than eight deep learning approaches, including two recently published models (v3_DCNN and Xception-65), and three variants of Deeplabv3+ with three different backbones, namely, U-Net, SegNet, and FCN, in precision, recall, F1-score, and mIoU (p<0.001). Importantly, the proposed system is shown to be capable of identifying tiny metastatic foci in challenging cases, for which there are high probabilities of misdiagnosis in visual inspection, while the baseline approaches tend to fail in detecting tiny metastatic foci. For computational time comparison, the proposed method takes 2.4 min for processing a WSI utilizing four NVIDIA Geforce GTX 1080Ti GPU cards and 9.6 min using a single NVIDIA Geforce GTX 1080Ti GPU card, and is notably faster than the baseline methods (4-times faster than U-Net and SegNet, 5-times faster than FCN, 2-times faster than the 3 different variants of Deeplabv3+, 1.4-times faster than v3_DCNN, and 41-times faster than Xception-65).

Characteristics of metastasis and survival between male and female breast cancer with different molecular subtypes: A population-based observational study

Objective

Male breast cancer (BC) is a rare disease, having different clinicopathological features and survival outcomes from female patients. The aim of this research was to, combine with molecular subtypes, analyze the metastatic patterns, and prognosis between male and female patients, and to determine whether the gender was the independent prognostic factor for BC.

Methods

Data used in this study were acquired from the SEER database from 2010 to 2016. The clinicopathology features and metastatic patterns were compared by the Chi‐square test and Fisher's exact test. Kaplan–Meier method was performed to compare overall survival (OS) and factors correlated with OS were determined by Cox regression models. Competing risk models were used to ascertain factors related to breast cancer‐specific death (BCSD).

Results

Compared with female BC, the incidence of regional LN (HR 1.849, 95% CI 1.674–2.043, p < 0.001) and distant metastasis (HR 1.421, 95%CI: 1.157–1.744, p < 0.001) was higher in male BC. For regional LN metastasis, hormone receptor (HoR)−/HER2+ subtype occupied the majority in both male (55.56%) and female (36.86%) groups. For distant metastasis, HoR−/HER2− subtype (21.26%), and HoR−/HER2+ (7.67%) were in major in male and female group separately. Male patients shared similar combinations of metastases with female groups as for single‐site, bi‐site, and tri‐site metastasis. Gender was an independent prognostic factor for OS (p < 0.001) but not for BCSD(p = 0.620). In subgroup of patients with HoR+/HER2−(OS: p = 0.003; BCSD: p = 0.606), HoR+/HER2+(OS: p = 0.003; BCSD: p = 0.277), regional LN positive(OS: p = 0.005; BCSD: p = 0.379), or bone metastasis (OS: p = 0.030; BCSD: p = 0.862), the male cohort had poorer OS but similar BCSD with female cohort.

Conclusions

Compared with female patients, male BC had different metastasis patterns and prognostic outcomes, and the affection of breast subtypes on metastasis and survivorship was also different. More attention needs to be paid for specific molecular subtype and more personalized therapeutic strategies should be customized while treating male patients.

Predicting Axillary Lymph Node Metastasis in Early Breast Cancer Using Deep Learning on Primary Tumor Biopsy Slides

Objectives

To develop and validate a deep learning (DL)-based primary tumor biopsy signature for predicting axillary lymph node (ALN) metastasis preoperatively in early breast cancer (EBC) patients with clinically negative ALN.

Methods

A total of 1,058 EBC patients with pathologically confirmed ALN status were enrolled from May 2010 to August 2020. A DL core-needle biopsy (DL-CNB) model was built on the attention-based multiple instance-learning (AMIL) framework to predict ALN status utilizing the DL features, which were extracted from the cancer areas of digitized whole-slide images (WSIs) of breast CNB specimens annotated by two pathologists. Accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curves, and areas under the ROC curve (AUCs) were analyzed to evaluate our model.

Results

The best-performing DL-CNB model with VGG16_BN as the feature extractor achieved an AUC of 0.816 (95% confidence interval (CI): 0.758, 0.865) in predicting positive ALN metastasis in the independent test cohort. Furthermore, our model incorporating the clinical data, which was called DL-CNB+C, yielded the best accuracy of 0.831 (95%CI: 0.775, 0.878), especially for patients younger than 50 years (AUC: 0.918, 95%CI: 0.825, 0.971). The interpretation of DL-CNB model showed that the top signatures most predictive of ALN metastasis were characterized by the nucleus features including density (p = 0.015), circumference (p = 0.009), circularity (p = 0.010), and orientation (p = 0.012).

Conclusion

Our study provides a novel DL-based biomarker on primary tumor CNB slides to predict the metastatic status of ALN preoperatively for patients with EBC.

The lymphatic vasculature: An active and dynamic player in cancer progression

The lymphatic vasculature has been widely described and explored for its key functions in fluid homeostasis and in the organization and modulation of the immune response. Besides transporting immune cells, lymphatic vessels play relevant roles in tumor growth and tumor cell dissemination. Cancer cells that have invaded into afferent lymphatics are propagated to tumor‐draining lymph nodes (LNs), which represent an important hub for metastatic cell arrest and growth, immune modulation, and secondary dissemination to distant sites. In recent years many studies have reported new mechanisms by which the lymphatic vasculature affects cancer progression, ranging from induction of lymphangiogenesis to metastatic niche preconditioning or immune modulation. In this review, we provide an up‐to‐date description of lymphatic organization and function in peripheral tissues and in LNs and the changes induced to this system by tumor growth and progression. We will specifically focus on the reported interactions that occur between tumor cells and lymphatic endothelial cells (LECs), as well as on interactions between immune cells and LECs, both in the tumor microenvironment and in tumor‐draining LNs. Moreover, the most recent prognostic and therapeutic implications of lymphatics in cancer will be reported and discussed in light of the new immune‐modulatory roles that have been ascribed to LECs.

Artificial Intelligence Algorithm-Based Ultrasound Image Segmentation Technology in the Diagnosis of Breast Cancer Axillary Lymph Node Metastasis

This paper aimed to investigate the application of ultrasound image segmentation technology based on the back propagation neural network (BPNN) artificial intelligence algorithm in the diagnosis of breast cancer axillary lymph node metastasis, thereby providing a theoretical basis for clinical diagnosis. In this study, 90 breast cancer patients with axillary lymph node metastasis were selected as the research objects and rolled randomly into an experimental group and a control group. Besides, all of them were examined by ultrasound. The BPNN algorithm for the ultrasound image segmentation diagnosis method was applied to the patiens from the experimental group, while the control group was given routine ultrasound diagnosis. Thus, the value of this algorithm in ultrasonic diagnosis was compared and explored. The results showed that when the number of hidden layer nodes based on the BPNN artificial intelligence algorithm was 2, 3, 4, 5, 6, 7, and 8, the corresponding segmentation accuracy was 97.3%, 96.5%, 94.8%, 94.8%, and 94.1% in turn. Among them, the segmentation accuracy was the highest when the number of hidden layer nodes was 2. The correlation of independent variable bubble plot analysis showed that the presence or absence of capsules, the presence of crab feet or burrs in breast cancer lesions was critical influencing factors for the occurrence of axillary lymph node metastasis, and the standardized importance was 99.7% and 70.8%, respectively. Besides, the area under the two-dimensional receiver operating characteristic (ROC) curve of the BPNN artificial intelligence algorithm model classification was always greater than the area under the curve of manual segmentation, and the segmentation accuracy was 90.31%, 94.88%, 95.48%, 95.44%, and 97.65% in sequence. In addition, the segmentation specificity of different running times was higher than that of manual segmentation. In conclusion, the BPNN artificial intelligence algorithm had high accuracy, sensitivity, and specificity for ultrasound image segmentation, with a better segmentation effect. Therefore, it had a better diagnostic effect for breast cancer axillary lymph node metastasis.

Validation of the Skåne University Hospital nomogram for the preoperative prediction of a disease-free axilla in patients with breast cancer

BACKGROUND

Axillary staging via sentinel lymph node biopsy (SLNB) is performed for clinically node-negative (N0) breast cancer patients. The Skåne University Hospital (SUS) nomogram was developed to assess the possibility of omitting SLNB for patients with a low risk of nodal metastasis. Area under the receiver operating characteristic curve (AUC) was 0.74. The aim was to validate the SUS nomogram using only routinely collected data from the Swedish National Quality Registry for Breast Cancer at two breast cancer centres during different time periods.

METHOD

This retrospective study included patients with primary breast cancer who were treated at centres in Lund and Malmö during 2008-2013. Clinicopathological predictors in the SUS nomogram were age, mode of detection, tumour size, multifocality, lymphovascular invasion and surrogate molecular subtype. Multiple imputation was used for missing data. Validation performance was assessed using AUC and calibration.

RESULTS

The study included 2939 patients (1318 patients treated in Lund and 1621 treated in Malmö). Node-positive disease was detected in 1008 patients. The overall validation AUC was 0.74 (Lund cohort AUC: 0.75, Malmö cohort AUC: 0.73), and the calibration was satisfactory. Accepting a false-negative rate of 5 per cent for predicting N0, a possible SLNB reduction rate of 15 per cent was obtained in the overall cohort.

CONCLUSION

The SUS nomogram provided acceptable power for predicting a disease-free axilla in the validation cohort. This tool may assist surgeons in identifying and counselling patients with a low risk of nodal metastasis on the omission of SLNB staging.

Machine Learning and Surgical Outcomes Prediction: A Systematic Review

BACKGROUND

Machine learning (ML) has garnered increasing attention as a means to quantitatively analyze the growing and complex medical data to improve individualized patient care. We herein aim to critically examine the current state of ML in predicting surgical outcomes, evaluate the quality of currently available research, and propose areas of improvement for future uses of ML in surgery.

METHODS

A systematic review was conducted in accordance with the Preferred Reporting Items for a Systematic Review and Meta-Analysis (PRISMA) checklist. PubMed, MEDLINE, and Embase databases were reviewed under search syntax "machine learning" and "surgery" for papers published between 2015 and 2020.

RESULTS

Of the initial 2677 studies, 45 papers met inclusion and exclusion criteria. Fourteen different subspecialties were represented with neurosurgery being most common. The most frequently used ML algorithms were random forest (n = 19), artificial neural network (n = 17), and logistic regression (n = 17). Common outcomes included postoperative mortality, complications, patient reported quality of life and pain improvement. All studies which compared ML algorithms to conventional studies which used area under the curve (AUC) to measure accuracy found improved outcome prediction with ML models.

CONCLUSIONS

While still in its early stages, ML models offer surgeons an opportunity to capitalize on the myriad of clinical data available and improve individualized patient care. Limitations included heterogeneous outcome and imperfect quality of some of the papers. We therefore urge future research to agree upon methods of outcome reporting and require basic quality standards.

Increased Expression of VANGL1 is Predictive of Lymph Node Metastasis in Colorectal Cancer: Results from a 20-Gene Expression Signature

This study aimed at building a prognostic signature based on a candidate gene panel whose expression may be associated with lymph node metastasis (LNM), thus potentially able to predict colorectal cancer (CRC) progression and patient survival. The mRNA expression levels of 20 candidate genes were evaluated by RT-qPCR in cancer and normal mucosa formalin-fixed paraffin-embedded (FFPE) tissues of CRC patients. Receiver operating characteristic curves were used to evaluate the prognosis performance of our model by calculating the area under the curve (AUC) values corresponding to stage and metastasis. A total of 100 FFPE primary tumor tissues from stage I-IV CRC patients were collected and analyzed. Among the 20 candidate genes we studied, only the expression levels of VANGL1 significantly varied between patients with and without LNMs (p = 0.02). Additionally, the AUC value of the 20-gene panel was found to have the highest predictive performance (i.e., AUC = 79.84%) for LNMs compared with that of two subpanels including 5 and 10 genes. According to our results, VANGL1 gene expression levels are able to estimate LNMs in different stages of CRC. After a proper validation in a wider case series, the evaluation of VANGL1 gene expression and that of the 20-gene panel signature could help in the future in the prediction of CRC progression.

An Immune-Related Gene Panel for Preoperative Lymph Node Status Evaluation in Advanced Gastric Cancer

Background and Aim: Gastric cancer (GC) is the common leading cause of cancer-related death worldwide. Immune-related genes (IRGs) may potentially predict lymph node metastasis (LNM). We aimed to develop a preoperative model to predict LNM based on these IRGs. Methods: In this paper, we compared and evaluated three machine learning models to predict LNM based on publicly available gene expression data from TCGA-STAD. The Pearson correlation coefficient (PCC) method was utilized to feature selection according to its relationships with LN status. The performance of the model was assessed using the area under the curve (AUC) and F1 score. Results: The Naive Bayesian model showed better performance and was constructed based on 26 selected gene features, with AUCs of 0.741 in the training set and 0.688 in the test set. The F1 score in the training set and test set was 0.652 and 0.597, respectively. Furthermore, Naive Bayesian model based on 26 IRGs is the first diagnostic tool for the identification of LNM in advanced GC. Conclusion: These results indicate that our new methods have the value of auxiliary diagnosis with promising clinical potential.

Cancer Stem Cell-Associated Pathways in the Metabolic Reprogramming of Breast Cancer

Metabolic reprogramming of cancer is now considered a hallmark of many malignant tumors, including breast cancer, which remains the most commonly diagnosed cancer in women all over the world. One of the main challenges for the effective treatment of breast cancer emanates from the existence of a subpopulation of tumor-initiating cells, known as cancer stem cells (CSCs). Over the years, several pathways involved in the regulation of CSCs have been identified and characterized. Recent research has also shown that CSCs are capable of adopting a metabolic flexibility to survive under various stressors, contributing to chemo-resistance, metastasis, and disease relapse. This review summarizes the links between the metabolic adaptations of breast cancer cells and CSC-associated pathways. Identification of the drivers capable of the metabolic rewiring in breast cancer cells and CSCs and the signaling pathways contributing to metabolic flexibility may lead to the development of effective therapeutic strategies. This review also covers the role of these metabolic adaptation in conferring drug resistance and metastasis in breast CSCs.

The mutational landscape of the SCAN-B real-world primary breast cancer transcriptome

Breast cancer is a disease of genomic alterations, of which the panorama of somatic mutations and how these relate to subtypes and therapy response is incompletely understood. Within SCAN-B (ClinicalTrials.gov: NCT02306096), a prospective study elucidating the transcriptomic profiles for thousands of breast cancers, we developed a RNA-seq pipeline for detection of SNVs/indels and profiled a real-world cohort of 3,217 breast tumors. We describe the mutational landscape of primary breast cancer viewed through the transcriptome of a large population-based cohort and relate it to patient survival. We demonstrate that RNA-seq can be used to call mutations in genes such as PIK3CA, TP53, and ERBB2, as well as the status of molecular pathways and mutational burden, and identify potentially druggable mutations in 86.8% of tumors. To make this rich dataset available for the research community, we developed an open source web application, the SCAN-B MutationExplorer (http://oncogenomics.bmc.lu.se/MutationExplorer). These results add another dimension to the use of RNA-seq as a clinical tool, where both gene expression- and mutation-based biomarkers can be interrogated in real-time within 1 week of tumor sampling.

Preoperative Prediction of Lymph Node Metastasis in Patients With Early-T-Stage Non-small Cell Lung Cancer by Machine Learning Algorithms

Background: Lymph node metastasis (LNM) is difficult to precisely predict before surgery in patients with early-T-stage non-small cell lung cancer (NSCLC). This study aimed to develop machine learning (ML)-based predictive models for LNM.

Methods: Clinical characteristics and imaging features were retrospectively collected from 1,102 NSCLC ≤ 2 cm patients. A total of 23 variables were included to develop predictive models for LNM by multiple ML algorithms. The models were evaluated by the receiver operating characteristic (ROC) curve for predictive performance and decision curve analysis (DCA) for clinical values. A feature selection approach was used to identify optimal predictive factors.

Results: The areas under the ROC curve (AUCs) of the 8 models ranged from 0.784 to 0.899. Some ML-based models performed better than models using conventional statistical methods in both ROC curves and decision curves. The random forest classifier (RFC) model with 9 variables introduced was identified as the best predictive model. The feature selection indicated the top five predictors were tumor size, imaging density, carcinoembryonic antigen (CEA), maximal standardized uptake value (SUV_max), and age.

Conclusions: By incorporating clinical characteristics and radiographical features, it is feasible to develop ML-based models for the preoperative prediction of LNM in early-T-stage NSCLC, and the RFC model performed best.