Search Results (3,565)

Breast cancer imaging is central to screening, diagnosis, staging, treatment monitoring, and post-treatment surveillance, but image interpretation remains limited by variable image quality, interobserver variability, false-positive findings and heterogeneous tumor biology. This narrative review summarizes current applications of deep learning in multimodal breast cancer imaging, with emphasis on image reconstruction, image enhancement, lesion detection, segmentation, classification, biomarker prediction, treatment response assessment, prognosis and clinical implementation. A structured literature search was performed across major biomedical and technical databases, focusing on studies involving mammography, digital breast tomosynthesis, ultrasound, MRI, PET/CT, digital pathology and multimodal fusion approaches. Current evidence indicates that deep learning can support image denoising; super-resolution, low-dose, and accelerated reconstruction; lesion localization; tumor segmentation; benign–malignant classification; and molecular or biomarker-related prediction. Multimodal models integrating radiological imaging, histopathology, clinical variables, and molecular markers show particular promise for treatment response prediction, recurrence risk estimation, and personalized decision support. However, clinical translation remains limited by retrospective study designs, small and imbalanced datasets, domain shift, inconsistent annotations, limited explainability, bias, lack of prospective validation, and regulatory challenges. Deep learning should therefore be viewed as a decision-support framework that may improve breast cancer imaging workflows if validated in diverse, prospective, and clinically representative settings. Full article

Background: S100A11, a calcium-binding S100 family protein, is increasingly implicated in carcinogenesis, yet its molecular regulation and clinical relevance across cancers remain unclear. Hepatocellular carcinoma (HCC) carries a dismal prognosis, in part due to a lack of reliable biomarkers for risk stratification of established disease. Methods: We conducted a pan-cancer analysis of S100A11 genomic alterations across 31 studies (10,767 samples) obtained from TCGA, encompassing copy number alterations, somatic mutations, and DNA methylation. HCC-specific analyses evaluated S100A11 expression, its potential as a diagnostic/prognostic marker, co-expression networks, and pathway enrichment using TCGA-LIHC data, with univariate and multivariate Cox regression to assess survival associations. Results: S100A11 alterations were predominantly driven by copy number amplification, with the highest frequencies in hepatobiliary cancers, lung and breast cancers. Copy number amplification showed a consistent inverse relationship with promoter methylation, indicating amplification-driven transcriptional activation. In HCC, S100A11 was markedly overexpressed compared with normal liver tissue, with strong diagnostic discriminatory capacity. High S100A11 expression was significantly associated with inferior overall survival (log-rank p = 0.032; HR = 1.46, 95% CI 1.03–2.06) and remained an independent predictor of overall survival after adjustment for age, sex, and AJCC pathologic stage (HR = 1.27, 95% CI 1.01–1.60, p = 0.038). Co-expression and pathway analyses demonstrated an association between S100A11 and metabolic reprogramming, extracellular matrix remodelling, and immune dysregulation. Conclusions: These findings identify S100A11 as a candidate diagnostic and prognostic biomarker in HCC whose overexpression is associated with metabolic reprogramming, ECM remodelling, and immune dysregulation, warranting experimental validation of a mechanistic role. Full article

(1) Background: Neoadjuvant chemotherapy (NAC) is important for breast cancer, but prognosis varies widely. Ki67 and apparent diffusion coefficient (ADC) reflect proliferation and cellularity, respectively. This study evaluated the prognostic value of the Ki67/ADC ratio (KA) and post-treatment ADC change (δADC) in breast cancer patients receiving NAC, and developed a survival prediction model incorporating these indicators. (2) Methods: Two cohorts of breast cancer patients treated with NAC were collected. Pre- and post-treatment breast MRI with diffusion-weighted imaging were obtained; ADC values were measured by two blinded radiologists. KA was calculated as pre-treatment Ki67 divided by pre-treatment ADC, and δADC as post-ADC minus pre-ADC. Disease-free survival (DFS) was the primary outcome. Cox regression and a predictive Cox model were used. (3) Results: A total of 419 patients were analyzed. Both KA and δADC were associated with survival. In multivariable analysis, KA remained an independent prognostic factor (HR 0.40, 95% CI 0.19–0.84, p = 0.015). High KA was associated with worse prognosis, particularly in patients without pathological complete response. The model incorporating KA showed better predictive performance than clinicopathological variables alone and effectively stratified high- vs. low-risk patients. (4) Conclusion: KA is a promising complementary biomarker for prognosis in breast cancer patients undergoing NAC. Its integration into a prognostic model improved survival risk prediction and may aid individualized post-treatment management. Full article

Objectives: Platinum-based neoadjuvant therapy can improve pathological response in triple-negative breast cancer (TNBC), but severe hematotoxicity may compromise treatment delivery. This study developed and validated a multitask learning framework to jointly predict pathological complete response (pCR) and severe hematotoxicity and to support benefit–hematotoxicity stratification. Methods: This multicenter retrospective study included 2060 consecutive patients with TNBC receiving platinum-based neoadjuvant therapy at three institutions. Patients were assigned to a training cohort (n = 1406), an internal validation cohort (n = 351), or an external validation cohort (n = 303). A multitask TabNet (MT-TabNet) model was developed using pretreatment clinical, pathological, imaging, laboratory, electrocardiographic, and planned treatment exposure variables to jointly estimate pCR and severe hematotoxicity. Model performance was evaluated using discrimination, calibration, decision curve analysis, and interpretability analyses. The predicted probabilities were further integrated into a utility-based framework for benefit–hematotoxicity stratification. Results: Overall, 639 patients (31.0%) achieved pCR, and 651 (31.6%) developed severe hematotoxicity. MT-TabNet achieved AUCs of 0.874, 0.842, and 0.819 for pCR prediction and 0.859, 0.846, and 0.818 for severe hematotoxicity prediction in the training, internal validation, and external validation cohorts, respectively. The model showed generally acceptable calibration and clinical net benefit. pCR prediction was predominantly associated with tumor-related characteristics, whereas severe hematotoxicity prediction was more strongly associated with planned treatment exposure and host-related laboratory indicators. Utility-based stratification showed progressively higher pCR rates and lower severe hematotoxicity rates from the low- to high-benefit groups across all three cohorts. Survival differed significantly among benefit groups for both progression-free survival (PFS) and overall survival (OS) in the training and internal validation cohorts; in the external validation cohort, OS differed significantly, whereas PFS showed a similar but nonsignificant trend. Conclusions: MT-TabNet enabled joint estimation of pCR and severe hematotoxicity in patients with TNBC receiving platinum-based neoadjuvant therapy. The utility-based framework integrated efficacy and toxicity predictions into clinically interpretable benefit–hematotoxicity stratification and warrants further prospective evaluation for individualized risk assessment and toxicity monitoring. Full article

Background/Objectives: Predicting sentinel lymph node (SLN) involvement in breast cancer can reduce unnecessary axillary lymph node dissections (ALND) and facilitate clinical decision-making. This study aimed to: (i) identify independent predictors for SLN metastasis, and (ii) assess the predictive performance of the MSKCC nomogram for SLN positivity in early-stage breast cancer patients. Methods: A retrospective cohort study spanning a six-year period was conducted at a tertiary care oncology department in Belgrade, Serbia. The study included 684 consecutive patients with histologically confirmed primary breast carcinoma, without distant metastasis, who underwent surgical treatment with sentinel lymph node biopsy (SLNB). Results: Out of 684 patients, 198 (29%) were SLN-positive. Among the positive cases, 178 (89.9%) had one positive SLN, 19 (9.6%) had two, and 1 (0.5%) had three. Multivariate logistic regression revealed that lymphovascular invasion and tumor sizes >20 mm significantly increased the risk for SLN involvement. Patients with SLN metastasis had significantly higher MSKCC-predicted probabilities compared to SLN-negative patients (52.1 ± 18.5% vs. 33.1 ± 16.9%, p < 0.001). The area under the receiver operating characteristic curve (AUC) for the nomogram was 0.778 (95% CI: 0.738–0.818; p < 0.001). The nomogram demonstrated acceptable discrimination, with an AUC of 0.778 (95% CI: 0.738–0.818). However, calibration-in-the-large was −0.526 (95% CI: −0.709 to −0.342), indicating systematic overestimation of absolute risk. The calibration slope was 1.152 (95% CI: 0.921–1.383), and the Brier score was 0.174. Conclusions: Lymphovascular invasion and primary tumor size >20 mm are the only independent predictors of SLN metastasis. Furthermore, The MSKCC nomogram demonstrated acceptable discrimination but systematically overestimated absolute SLN-metastasis risk. Recalibration and further validation are required before clinical application in this population. Full article

Background/Objectives: Endometrial polyps are a prevalent pathological condition within the uterine cavity. While the majority of lesions are classified as benign, histopathological examinations indicate a potential for malignant transformation occurring in 0.5–5% of cases. The present study aimed to evaluate the prevalence of endometrial carcinoma and pre-malignant histopathological findings (glandular hyperplasia without atypia and atypical glandular hyperplasia) associated with endometrial polyps and to identify pertinent risk factors among Polish women undergoing hysteroscopic polypectomy under local anesthesia. Methods: A retrospective cohort study was conducted at the Outpatient Hysteroscopy Center of the Heliodor Swiecicki University Hospital of Gynecology and Obstetrics, affiliated with Poznan University of Medical Sciences. The study included patients treated using the GUBBINI mini-resectoscope under local anesthesia with the Hystero-Block system (Tontarra Medizintechnik GmbH, Wurmlingen, Germany) between December 2022 and July 2026. Statistical analysis of risk factors was performed using the Kruskal–Wallis H test and Fisher’s exact test. Odds ratios for potential risk factors were also calculated. Results: The study comprised 2588 patients. Endometrial carcinoma associated with endometrial polyps was identified in 16 women (0.62%). Additional histopathological findings included glandular hyperplasia in 360 patients (13.91%) and atypical hyperplasia in 35 patients (1.35%), with the remaining patients diagnosed with benign uterine polyps (84.12%). Patients with malignant and pre-malignant lesions were significantly older than those with benign polyps (p < 0.001). Regarding polyp size, patients diagnosed with glandular hyperplasia without atypia had significantly larger polyps than those with benign lesions (p < 0.001). Estimated postmenopausal status was associated with significantly higher odds of endometrial carcinoma (Odds Ratio [OR] = 6.56, 95% Confidence Interval [CI]: 2.12–20.30, p = 0.001) and atypical glandular hyperplasia (OR = 3.13, 95% CI: 1.41–6.96, p = 0.005) compared to benign polyps. Similarly, obesity increased the odds of glandular hyperplasia without atypia (OR = 1.32, 95% CI: 1.01–1.74, p = 0.045) and atypical glandular hyperplasia (OR = 3.33, 95% CI: 1.67–6.65, p < 0.001). Furthermore, hypertension was associated with higher odds of glandular hyperplasia without atypia compared to benign polyps (OR = 1.41, 95% CI: 1.02–1.97, p = 0.040). Other evaluated risk factors, including type II diabetes mellitus, infertility, abnormal uterine bleeding, breast cancer history, hypothyroidism and polyendocrine metabolic ovarian syndrome (PMOS), did not reach statistical significance (p > 0.05). This large cohort provides important data regarding the prevalence and risk factors of carcinoma associated with endometrial polyps and pre-malignant uterine lesions in Polish women. Conclusions: Endometrial carcinoma associated with endometrial polyps is a rare entity among Polish women undergoing outpatient hysteroscopy (0.62%). Nevertheless, the primary diagnostic value of this large-scale study lies in demonstrating that despite the low prevalence of malignancy in standard “see-and-treat” outpatient settings, systematic histopathological evaluation remains an absolute diagnostic mandate, as relying solely on visual or clinical parameters is insufficient to exclude malignancy. Furthermore, identified risk factors, such as age, polyp size, estimated postmenopausal status, obesity and hypertension, serve as critical diagnostic red flags, associated with increased odds of concurrent premalignant lesions or carcinoma. Full article

Cellular senescence (CS) is increasingly recognized as an important cell-state programme involved in breast cancer progression and therapeutic response, but its context-dependent molecular heterogeneity limits its application in prognostic assessment. In this study, GeneCompass-based all-gene in silico perturbation analysis was performed to identify candidate genes predicted to induce senescence or rejuvenation, thereby expanding the known senescence-related gene set. Machine learning further established a seven-gene prognostic signature that may serve as an adjunctive tool for prognostic assessment across multiple cohorts. The time-dependent AUCs at 1, 3, and 5 years were 0.707, 0.700, and 0.684 in the training cohort; 0.657, 0.661, and 0.629 in the test cohort; and 0.611, 0.646, and 0.637 in the external validation cohort, respectively. Single-cell and spatial transcriptomic analyses suggested that the risk component of the prognostic signature reflects not only malignant epithelial cell states but also stromal–vascular remodelling in the tumor microenvironment. Among the signature genes, ADGRF5 exhibited the most pronounced expression alteration, and its knockdown suppressed malignant phenotypes in breast cancer cells. These findings provide an AI-assisted strategy for senescence biomarker discovery and highlight ADGRF5 as a candidate functional risk gene associated with breast cancer progression. Full article

Breast cancer remains one of the leading causes of cancer-related mortality among women worldwide, underscoring the importance of early and accurate diagnosis to improve patient outcomes. Magnetic resonance imaging (MRI) is a highly sensitive imaging modality for detecting breast malignancies, particularly in patients with dense breast tissue or those at high risk, where conventional imaging techniques may have limited sensitivity. Recent advances in deep learning (DL) have demonstrated considerable potential for improving the automated analysis of breast MRI, including tumour classification, prediction, and segmentation. This systematic review synthesises peer-reviewed studies published between 2014 and 2025 that exclusively applied DL techniques to breast MRI for cancer classification, prediction, or segmentation. The included studies were critically evaluated with respect to model architectures, dataset characteristics, image preprocessing methods, validation strategies, and reported performance metrics. The reviewed literature demonstrates that DL models consistently achieve high diagnostic performance and have the potential to enhance radiological workflows by supporting automated lesion detection and clinical decision-making. However, several challenges continue to limit their translation into routine clinical practice, including limited access to large, diverse, and well-annotated datasets, inadequate external validation, variability in MRI acquisition protocols, and concerns regarding model interpretability and generalisability. Future research should prioritise the development of robust, explainable, and clinically validated DL models trained on multicentre datasets using standardised evaluation frameworks. Addressing these challenges will be essential to improve the reliability, reproducibility, and clinical applicability of AI-assisted breast cancer diagnosis using MRI. Full article

Breast cancer [BC] is the most diagnosed cancer among women worldwide. Thanks to advancements in early detection, systemic therapies, and supportive care, survival rates have significantly improved. As more patients survive BC, cardiovascular disease [CVD] has emerged as a leading cause of long-term morbidity and mortality in this population. This is largely due to a combination of shared risk factors, such as obesity, diabetes and metabolic syndrome, combined to cardiotoxic effects of certain anticancer therapies, particularly anthracyclines, HER2-targeted therapies and radiotherapy. This review explores the cardiovascular [CV] implications of modern breast cancer treatments, including chemotherapy, endocrine therapy, targeted agents, radiotherapy, and emerging modalities such as immunotherapy. It also highlights the impact of patient-specific factors—such as diabetes, lipid profile, and treatment duration—on CVD risk. Current data suggest that breast cancer survivors are at increased risk of CVD compared to the general population, and that risk persists for years after treatment completion. CVD risk profile of each patient is different because of patient age, previous risk factors, and type of oncological treatment, and must be carefully delineated and keep well in mind during and after cancer treatment. Full article

Background and Aim: Approximately 10% of breast cancer patients present with distant organ metastases at initial diagnosis. This study aimed to evaluate the efficacy of immune-inflammatory markers and histopathological variables in identifying de novo metastatic breast cancer. Methods: Patients with breast cancer referred to a tertiary surgical oncology clinic between January 2020 and December 2022 were retrospectively screened. A total of 412 patients met the strict inclusion criteria. Laboratory parameters and radiological findings obtained during the initial diagnostic workup, prior to the initiation of any therapeutic intervention, were comprehensively evaluated. Results: Among the screened cohort, 56 patients presented with synchronous distant organ metastases at diagnosis. Significant differences (p < 0.05) were observed between the metastatic and non-metastatic groups in serum hemoglobin, albumin, and alkaline phosphatase (ALP) levels, lymphocyte and neutrophil counts, hemoglobin-albumin-lymphocyte-platelet score, prognostic nutritional index (PNI), systemic immune-inflammation index (SII), monocyte-to-lymphocyte ratio, platelet-to-lymphocyte ratio, neutrophil-to-lymphocyte ratio, and pan-immune-inflammation value. Multivariate logistic regression analysis identified Ki-67, T stage, N stage, ALP, SII, and PNI as independent predictors of metastasis (p < 0.05). In the receiver operating characteristic (ROC) analysis, the prognostic score formulated from this predictive model demonstrated an area under the curve (AUC) of 0.821 (95% CI: 0.781–0.857, p < 0.001, and Z-score = 9.34). At an optimal cut-off value of >0.233, the score yielded a sensitivity of 64.3% and a specificity of 92.1%. Furthermore, Decision Curve Analysis (DCA) confirmed a positive net clinical benefit across the decision-making threshold. Conclusions: The developed prognostic score may be a promising clinical tool for differentiating de novo metastatic breast cancer from non-metastatic disease at initial staging. This non-invasive approach may help clinicians risk-stratify patients, reduce diagnostic delays, and optimize early intervention strategies. Nonetheless, larger prospective multicenter studies are warranted for robust validation. Full article

Immunohistochemistry (IHC) of ER, PR, HER2, and Ki67 are established prognostic and predictive markers in breast cancer. Gene expression assays such as PAM50 and Oncotype DX are increasingly used for molecular subtyping and recurrence risk calculation. The APIS Breast Cancer Subtyping Kit (BCSK) quantitatively measures the mRNA expression of these markers (ER, PR, HER2, and Ki67) and includes a novel four-gene BCSK Proliferation Signature (PS) designed for subtype classification and risk stratification. This exploratory retrospective cohort study compares the BCSK with established molecular assays in an age-stratified cohort. We aimed to compare BCSK subtype classification in distinguishing Luminal A versus Luminal B subtypes with IHC and with PAM50. We have also aimed to assess correlations between the BCSK PS and the Oncotype DX Recurrence Score (RS) as well as the PAM50 Risk of Recurrence (ROR) score in patients stratified by age (<65 versus ≥65 years). Formalin-fixed, paraffin-embedded (FFPE) tumor specimens from 59 patients with ER+/HER2− breast cancer (33 < 65 years, 26 ≥ 65 years), diagnosed between 2020 and 2022 at the Cantonal Hospital Baselland and University Hospital Basel, were analyzed using IHC, BCSK, and PAM50 (Prosigna^®). All patients received adjuvant therapy and had ODx scores available. Patient, disease and treatment characteristics were compared between age groups using the Fisher exact test. We assessed concordance in luminal subtype classification across IHC, BCSK, and PAM50 assays pairwise and stratified by age groups descriptively. We used Spearman’s rank correlation to examine associations between BCSK PS, RS, and ROR. Differences between age groups regarding PS were analyzed using the Mann–Whitney U test. In the ≥65-year cohort, subtype classification concordance between the BCSK and PAM50 was higher (84.6%) than in those aged <65-years (60.6%). In patients aged <65 years, the PS was significantly correlated with both RS (ρ = 0.5745, p < 0.0005) and ROR (ρ = 0.391, p = 0.024), whereas RS and ROR were not significantly correlated. In patients ≥65 years, PS correlated significantly with ROR (ρ = 0.603, p = 0.001), while there were no significant correlations between PS and RS (ρ = 0.134, p = 0.514) and between RS and ROR (ρ = 0.149, p = 0.466). Median PS values did not significantly differ between age groups (0.589 versus 0.602, p = 0.97). In this exploratory retrospective age-stratified analysis, the BCSK demonstrated descriptive concordance with PAM50 subtype classifications, especially in patients aged ≥65 years. PS showed significant but moderate correlation with the established molecular recurrence scores RS and ROR, however these differed across age groups. Our findings should be considered hypothesis-generating because this study did not include outcome metrics and was based on a relatively small cohort. Larger prospective studies incorporating clinical outcomes are necessary to determine the diagnostic and prognostic utility as well as comparative cost-effectiveness of the BCSK and its utility across age groups. Full article

Microplastics (MPs) and nanoplastics (NPs) have emerged as pervasive environmental contaminants with increasing evidence of human exposure and biological accumulation. Recent studies have confirmed their presence in multiple human reproductive tissues and fluids, including semen, testicular tissue, ovarian follicular fluid, cervicovaginal secretions, placenta, and breast milk, raising concerns regarding their potential implications for reproductive health. Beyond their widespread distribution, MPs have been shown in experimental studies to interact with cellular and molecular processes, including oxidative stress, inflammatory responses, mitochondrial dysfunction, and DNA damage, which are pathways commonly implicated in carcinogenesis. This review provides a comprehensive and critical synthesis of current evidence linking microplastic exposure to reproductive cancers, including prostate, testicular, ovarian, endometrial, cervical, and vaginal malignancies. Available mechanistic studies suggest that MPs may influence cancer-related biological processes through dysregulation of programmed cell death, genotoxicity, endocrine disruption, and modulation of signaling pathways such as PI3K/AKT and MAPK. Experimental findings also indicate that MPs may alter the tumor microenvironment and affect cellular behaviors associated with proliferation, migration, and invasion. However, the majority of current evidence is derived from in vitro studies, animal models, and indirect mechanistic observations, while direct epidemiological evidence in humans remain limited. Furthermore, methodological heterogeneity in microplastic detection and characterization complicates comparisons across studies and hinders causal inference. Overall, current evidence supports the biological plausibility of an association between microplastic exposure and reproductive cancer-related processes, while highlighting the need for standardized methodologies and well-designed longitudinal human studies to clarify potential health risks. Full article

Background and Objectives: Antibody–drug conjugates (ADCs) have become a major therapeutic platform in breast cancer and other solid tumors. Trastuzumab deruxtecan (T-DXd), trastuzumab emtansine (T-DM1), and sacituzumab govitecan (SG) differ substantially in antibody target, linker, payload, drug-to-antibody ratio, and bystander effect, resulting in heterogeneous pulmonary and cardiac toxicity profiles. This narrative review critically compares interstitial lung disease (ILD)/pneumonitis and cardiotoxicity associated with these three agents, aiming to prevent inappropriate extrapolation of toxicity algorithms and to provide a practical, agent-specific framework for multidisciplinary care. Materials and Methods: A targeted narrative search of PubMed/MEDLINE, Google Scholar, ClinicalTrials.gov, regulatory product information, and oncology/cardio-oncology guidance was performed and updated on 24 August 2026. Priority was given to regulatory documents, pivotal trials, pooled safety analyses, real-world cohorts, systematic reviews, and multidisciplinary recommendations. Pharmacovigilance data and case reports were included only to characterize rare events. Results: T-DXd is associated with a clinically important ILD/pneumonitis risk (approximately 12–15% in pooled analyses), predominantly grade 1–2 but occasionally fatal, requiring proactive surveillance, immediate interruption for suspected disease, and grade-directed corticosteroid therapy. T-DM1 shows a low but established pneumonitis incidence of approximately 1%, with permanent discontinuation recommended upon diagnosis. SG-related pneumonitis is rare and incompletely defined, without a T-DXd-like surveillance mandate. Both T-DM1 and T-DXd retain trastuzumab-derived cardiac monitoring requirements; symptomatic heart failure remains uncommon, although protocol-defined LVEF declines appear more frequent with T-DXd. SG lacks an established cardiomyopathy signal. Conclusions: Cardiopulmonary toxicity of ADCs is agent-specific rather than a class effect. Monitoring intensity, diagnostic thresholds, and management pathways must be tailored to the individual drug, regimen, indication, dose, patient comorbidity, and prior therapy. Close collaboration among oncology, radiology, pulmonology, and cardio-oncology is essential to preserve both treatment efficacy and patient safety. Full article

Background/Objectives: Alternative splicing (AS) contributes substantially to transcriptomic diversity and has emerged as an important regulator of cancer progression. However, the genome-wide characterization of AS events specific to estrogen receptor (ER)-positive breast cancer remains limited. This study aims to comprehensively profile AS in ER-positive breast cancer and identify a prognostic AS signature associated with patient outcome. Methods: Clinical and splicing data (Percent Spliced In values) were obtained from The Cancer Genome Atlas (TCGA) for 737 ER-positive samples. Prognostic AS events were identified using Cox regression analysis. The Least Absolute Shrinkage Selection Operator (LASSO) model was used to construct an AS-based prognostic signature, and a standardized risk score was calculated for each sample. The signature was then evaluated by Kaplan–Meier (KM) analysis and receiver operating characteristic (ROC) curves, in addition to other methods to validate model performance. Furthermore, transcript-level annotation and RNA expression correlation were performed to evaluate biological relevance. Results: Profiling identified 6276 AS events across 4457 genes, with exon skipping (ES) representing the most prevalent class (34.4%). Model analysis established a novel five-event AS prognostic signature (comprising DNAJC14, BAZ2B, PCDHAC1, PCDHA7, and DAPL1). The signature significantly stratified patients into high-risk and low-risk groups for both disease-free survival (DFS; p < 0.001) and overall survival (OS; p < 0.001). HER2-specific analysis demonstrated more consistent performance in HER2-negative patients for both DFS (p < 0.001) and OS (p = 0.018). The model achieved area under the curve (AUC) of 0.804 for 60-month follow-up supporting long-term prognostic performance. Additionally, the signature demonstrated stable and reliable discrimination with a concordance index (C-index) of approximately 0.73 across multiple validation methods. Conclusions: The study identified AS signature with promising prognostic value in ER-positive breast cancer. This highlights the potential of splicing-based models to refine risk stratification beyond conventional gene expression analysis. Full article

Background/Objectives: Accurate preoperative detection of locally advanced breast cancer is essential for neoadjuvant therapy planning. We developed and validated gradient-boosting models using structured BI-RADS mammographic features to simultaneously predict locally advanced breast cancer (LABC), DCIS component, and multifocality in a multi-center cohort. Methods: This retrospective study enrolled 2295 patients from three university-affiliated hospitals; features were coded according to BI-RADS. CatBoost and logistic regression models were built using stratified 60/20/20 splits, with performance assessed via bootstrap resampling, nested cross-validation, and sensitivity analyses. AUROC, AUPRC, Brier score, and calibration metrics assessed discrimination and clinical utility; a leakage audit and SHAP analysis supported interpretation. Results: CatBoost achieved an AUROC of 0.906 (95% CI: 0.876–0.932) for LABC. Because several top predictors overlap with the anatomical criteria defining this outcome, we repeated the analysis excluding them; the reduced model retained a mean AUROC of 0.739, indicating genuine predictive signal beyond the staging overlap. Net benefit was positive across all relevant thresholds, with calibration error of 0.053. DCIS prediction was highly accurate (AUROC 0.979; nested AUROC 0.9707), with no evidence of leakage. Multifocality prediction was more modest (AUROC 0.810), reflecting known limits of two-dimensional mammography. Sensitivity analyses confirmed stable performance across splits, training sizes, and class-weighting schemes. Conclusions: Structured mammographic features combined with gradient-boosting support clinically meaningful, though partly overlapping, risk stratification for LABC; once accounted for, the model still retains independent value. The DCIS model performed very well; multifocality prediction remains more limited, and external validation is needed before clinical use. Full article