Obesity (OB) and overweight (OW) are common in patients with breast cancer and are associated with inferior outcomes, yet their impact on tumor molecular architecture remains incompletely understood. This study aimed to define the genomic and transcriptomic features of breast cancer in OW/OB Chinese patients and to identify clinically relevant biomarkers and prognostic models for this population. We performed targeted sequencing of matched tumor–blood samples from 3924 Chinese female patients with breast cancer to profile somatic and germline alterations, together with comprehensive clinicopathological annotation. In parallel, transcriptomic profiling was conducted in 236 tumors to characterize body mass index (BMI)-associated biological programs and explore potential therapeutic vulnerabilities. OW/OB status was associated with significantly worse survival, particularly in the hormone receptor–positive/human epidermal growth factor receptor 2–negative (HR+/HER2−) subtype. In HR+/HER2− tumors, OW/OB patients showed higher frequencies of FOXA1 and ALK somatic mutations and FANCA germline variants, whereas TP53 and ARID1A somatic alterations were more frequent in underweight/normal-weight patients. Among OW/OB cases, we identified 20 adverse prognostic mutations and developed a survival model with robust discriminatory performance. Notably, NOTCH2 mutation emerged as a key predictor of poor outcome in OW/OB patients, especially in the HR+/HER2− subgroup. Transcriptomic analyses further demonstrated that NOTCH2-mutant tumors exhibited signatures associated with impaired drug response and phospholipid efflux, along with reduced apoptosis-related programs, and were predicted to be less sensitive to tamoxifen and palbociclib, suggesting a potential endocrine-resistant phenotype. This integrative multi-omics study reveals that elevated BMI is associated with distinct molecular features and unfavorable clinical outcomes in breast cancer, especially in HR+/HER2− disease. Our findings identify NOTCH2 as a potential biomarker of endocrine resistance and provide a molecular basis for BMI-adapted prognostication and personalized therapeutic strategies in OW/OB breast cancer patients.
Air pollution is the leading environmental risk factor in the U.S. for premature morbidity and mortality and disproportionately impacts marginalized populations. Although the causal link between air pollution and lung cancer is well established, its impact on breast cancer is less understood. Therefore, in this study, we examined the association between traffic-related air pollution (NO2) and particulate matter (PM2.5) with the development of breast cancer-specific subtypes among non-Hispanic Black (NHB) and non-Hispanic White (NHW) women in Georgia. Using the Georgia Cancer Registry, we identified women with a first primary breast cancer diagnosis 2010–2017 and geocoded their address at diagnosis to the census tract level. Daily PM2.5 and NO2 concentrations modeled on a 1-km grid (SEDAC) were aggregated to census tracts and averaged to annual means; women were assigned the annual mean concentration lagged 5 years prior to diagnosis based on tract of residence at diagnosis. We used logistic regression to compute case-only odds ratios (ORs) and 95
To identify stable and interpretable MRI radiomic features for developing a predictive model of neoadjuvant chemotherapy (NAC) response in HR+/HER2− breast cancer. This retrospective multicenter study included four analytical cohorts representing 686 unique patients with HR+/HER2− breast cancer, including an imaging-development cohort, an external imaging-validation cohort, an imaging–transcriptomic cohort, and an independent transcriptomic cohort. NAC response was evaluated using the Residual Cancer Burden or Miller-Payne grading system. Multiregional radiomics features were extracted from pre-treatment DCE-MRI, including intratumoral subregions, peritumoral regions, kinetic features, and multiregional spatial interaction (MSI). Five feature selection strategies (stepwise regression, LASSO, RFE, GBM-based embedded selection, and a proposed stability-driven strategy) were combined with four classifiers (LR, SVM, RF, XGBoost) to build predictive models. Stable features were defined as those selected in at least 50
Rural breast cancer survivors are less likely to meet physical activity (PA) recommendations than those residing in urban areas. Ecologic frameworks account for multilevel factors related to PA and may be used to contextually adapt interventions to meet the needs of underserved cancer survivors. This study explored ecologic determinants of PA in rural breast cancer survivors to inform future intervention adaptations and implementation in rural communities. An explanatory sequential mixed-methods design was used to explore determinants among breast cancer survivors who resided in a rural area. Surveys included measures of individual, social, and environmental factors that influence PA, and in-depth interviews expanded on multilevel determinants of PA and preferences. Rural cancer survivors (N = 219) completed questionnaires. Of those, 67 were breast cancer survivors and included in quantitative analyses, and 38 completed an interview. Participants were in their early 60s (M age = 62.0 ± 13.1) and were 7.2 ± 8.1 years post-diagnosis. Less than half (46.2
Trophoblast cell-surface antigen 2 (Trop-2) is a key therapeutic target in metastatic triple-negative breast cancer (TNBC), yet its prognostic significance in early-stage disease remains poorly defined. We evaluated Trop-2 expression and its clinical impact within a sub-cohort of the phase III GEICAM_CIBOMA trial (NCT00130533). Trop-2 membrane expression was assessed by immunohistochemistry (H-score) in a pragmatic sub-cohort of 70 patients, representative of the trial’s intention-to-treat population. High/medium (positive) Trop-2 expression (H-score ≥ 100) was observed in 74.3
Natural regulatory T cells (nTregs) are a key immunosuppressive component of the tumor microenvironment (TME), but their assessment typically requires specialized molecular assays. This study aimed to develop and validate a deep learning-based pathomics model to predict nTregs infiltration directly from routine hematoxylin and eosin (H E)-stained whole slide images (WSI) of breast cancer and evaluate its prognostic significance. Data from 1097 breast cancer patients in The Cancer Genome Atlas (TCGA) were analyzed. A cohort of 928 patients with complete RNA-seq data was used to establish the prognostic value of nTregs (estimated by ImmuneCellAI). A subset of 791 patients with matched high-quality H E-stained WSI was randomly split into training (n = 633) and validation (n = 158) cohorts. A total of 1488 quantitative pathomic features were extracted. After feature selection via minimum Redundancy Maximum Relevance (mRMR) and Recursive Feature Elimination (RFE), a Gradient Boosting Machine (GBM) classifier was trained to predict high versus low nTregs status, generating a continuous Pathomics Score (PS). The PS was validated against FOXP3 immunohistochemistry (IHC) and evaluated for its association with overall survival (OS). Multi-omics analyses explored the underlying biology of PS-defined groups. High nTregs infiltration was an independent predictor of poor OS (HR = 1.58, 95
Interval breast cancers (IBC), diagnosed between routine screening rounds, tend to have a worse prognosis than screen-detected breast cancers (SDBC). Identifying risk factors for IBCs is critical for improving early detection and developing risk-stratified screening strategies to reduce their incidence. We evaluated associations of breast density, reproductive, hormonal, lifestyle and medical factors with IBC compared to SDBC in a large UK cohort. Analyses included 1,940 women diagnosed with breast cancer between 2004 and 2018 after enrolment in the Breast Cancer Now Generations Study, a prospective UK cohort linked to the National Health Service Breast Screening Programme. Pre-diagnostic risk factors were collected at enrolment, and breast density was estimated from pre-diagnostic mammograms using Cumulus in a subset of 1,191 cases. Risk factor associations with density were estimated using linear regression models. We identified 1,185 SDBCs and 755 IBCs included in logistic regression models to estimate odds ratios (OR) and 95
Selective estrogen receptor modulators (SERMs) decrease the risk of breast cancer in high-risk women. We sought to determine the association between germline DNA methylation and single nucleotide polymorphism (SNP) biomarkers and benefit from SERM prevention. DNA from blood at time of entry on NSABP P-1 or P-2 was utilized from a matched case–control study of women receiving a SERM. Methylation was determined with Illumina Infinium Methylation EPIC BeadChipv2 that targets 937,055 sites. Differential methylation analysis was performed utilizing the Wilcoxon signed-rank test. Analysis was performed on 22 potentially relevant genes. Five modeling strategies for predicting case–control status were performed. Polygenic risk scores (PRS) were integrated with methylation data. Risk SNPs and methylation data were jointly analyzed. DNA methylation data were available for 1706 participants (587 cases, 1119 controls). Wilcoxon signed-rank analysis was performed on 887,318 CpG sites and 25,746 methylated regions. Six CpGs were significantly associated with breast cancer risk with smallest p-value of 3.70E−08 for the top CpG that was related to AFF3, which has been associated with resistance to tamoxifen. Dfferences in beta (methylation) values between cases and controls were small. No CpG regions were significantly associated. CYP1A1 and CYP3A7 were significantly associated with case–control status. Five prediction modeling strategies revealed median AUCs of 0.524–0.590. Integration of PRS with methylation did not improve predictive performance. Joint analysis of two previously published SNPs (related to ZNF423 and CTSO) and cg11423397 revealed an odds ratio of 16.5 for differences in breast cancer risk for those with all protective factors versus all risk factors. We identified differentially methylated CpG sites that achieved statistical significance but minimal differences in beta values. No CpG regions were significant. Differential methylation in two CYP genes was identified but of unclear importance. No improvement in performance from prediction models or integrating PRS and methylation was identified. Joint analysis of cg11423397 and SNPs from ZNF423 and CTSO suggest further discrimination of breast cancer risk. While current methylation approaches utilizing germline DNA show limited predictive utility for breast cancer risk in women receiving a SERM, our results point to specific methylation and genetic biomarkers that warrant further study.
Immunotherapy efficacy in triple-negative breast cancer (TNBC) remains limited, partly due to mechanisms of immune suppression within the tumor microenvironment (TME). Ectonucleoside triphosphate diphosphohydrolase 1 (ENTPD1, encoding cluster of differentiation 39, CD39) is a key immunoregulatory ectonucleotidase, but its treatment-associated distribution and coordination across myeloid and T-cell states in TNBC remain incompletely characterized. We re-analyzed a single-cell RNA sequencing (scRNA-seq) dataset comprising 156120 immune cells from 27 tumor and metastatic samples across 14 patients with TNBC. Trajectory and cell-cell communication analyses were complemented by NicheNet, independent single-cell and bulk transcriptomic validation, CosMx spatial transcriptomics, and multiplex immunofluorescence. CD39 protein expression and its association with survival were evaluated in an independent cohort of 112 patients with TNBC using immunohistochemistry (IHC), Kaplan-Meier analysis, and multivariable Cox proportional hazards regression. Three signed-R2-based summary indices-treatment contribution (TC), ENTPD1 dynamic index (EDI), and tumor response index (TRI)-were applied to summarize treatment-associated immune dynamics. ENTPD1 expression was observed across macrophage and regulatory T-cell states, including CD4+CD39+ Tregs. Higher macrophage ENTPD1 expression was associated with M2-related programs, attenuated stimulator of interferon genes (STING)-type I interferon signaling, and reduced antigen-presentation programs. Cell-cell communication analysis revealed preferential interactions between myeloid cells and CD4+CD39+ T cells. Inferred lymphotoxin (LT) signaling involving CD4+CD39+ T cells and myeloid populations was higher in non-responders. Cross-cohort and patient-level analyses supported coordinated CD39-associated immune programs, while spatial transcriptomics and representative multiplex immunofluorescence imaging visualized the implicated populations within shared tumor regions. In the independent IHC cohort, high tissue-level CD39 expression remained independently associated with shorter overall survival (OS) after adjustment for age group and American Joint Committee on Cancer (AJCC) stage. ENTPD1/CD39 was associated with coordinated myeloid and T-cell states, an adenosine-related immunoregulatory context, and reduced response to anti-programmed death-ligand 1 (PD-L1) therapy plus chemotherapy in TNBC. These findings support further evaluation of ENTPD1/CD39-associated immune states in the context of immune checkpoint blockade (ICB). High tissue-level CD39 expression was independently associated with shorter OS in an independent clinical cohort.
Abstract Background Sacituzumab govitecan (SG) is an antibody drug conjugate targeting trophoblast cell surface antigen 2 (Trop2) that is approved for treatment of patients with metastatic triple negative breast cancer (TNBC). Tumor necrosis factor-related apoptosis inducing ligand (TRAIL) agonists are antitumor agents that interact with death receptors on the cell surface to induce apoptosis in cancer cells, often sparing normal cells. In this study, we investigated whether combination treatment with SG and TRAIL agonists inhibit TNBC cell growth. Methods In vitro, 10 human TNBC cell lines and 4 non-modified, low passage patient derived TNBC cells were treated with SG and the TRAIL agonist Apo2L. Cell death was assessed via a propidium iodide-based assay while cell proliferation was measured using an ATP cell viability assay. The mode of cell death elicited by combination treatment was investigated by using inhibitors of apoptosis, necroptosis and ferroptosis. The drug effect on cell cycle was analyzed with flow cytometry. In vivo, NCr athymic nude (nu/nu) female mice bearing HCC1806 TNBC xenografts were treated with SG and TRAIL agonists after which the effect of treatment on tumor growth and survival was evaluated. Results SG and TRAIL acted synergistically in all 10 TNBC cell lines as well as all 4 patient derived TNBC cells tested. Synergistic cell death and growth inhibition was observed in both Trop2 high-expressing and low-expressing cells, with data suggesting that the linker deconjugation and bystander effects of SG contribute to the efficacy in Trop2 low-expressing cells. Caspase-3/7 assays as well as cell death assays using cell death inhibitors demonstrated that the lethality of dual treatment is primarily mediated by apoptosis. Additionally, SG alone and in combination with Apo2L induced cycle arrest at the G2/M phase. Finally, xenograft models using HCC1806 bearing mice showed that dual treatment with SG and TRAIL agonists significantly inhibited tumor growth and caused a borderline significant benefit on survival with no toxicities noted for both drug treatments. Conclusions Combination treatment with SG and TRAIL agonists induces synergistic lethality in TNBC cells in vitro and inhibits tumor growth in vivo, suggesting a potential clinical benefit of the combination therapy in TNBC.
Neoadjuvant chemotherapy (NAC) is the standard of care for aggressive early-stage and locally advanced hormone receptor-positive, HER2-negative (HR+/HER2−) breast cancer (BC). However, this subtype exhibits a high rate of treatment resistance, necessitating improved predictive tools for response assessment. In this study, we developed predictive models for pathological complete response (pCR) and distant relapse (DR) by integrating clinicopathological features with gene expression data from a retrospective cohort of 44 HR+ /HER2− BC patients treated with anthracycline- and taxane-based NAC. The model combining molecular and clinical variables achieved an AUC of 0.86 in both the discovery and validation cohorts, suggesting promising discriminatory performance for pCR prediction. For DR prediction, seven genes were differentially expressed between relapsed and non-relapsed patients. A combined model incorporating clinical variables and molecular features, including COL2A1 and MAPK8IP1 expression, achieved an AUC of 0.73 in residual disease samples. Notably, COL2A1 overexpression correlated with improved distant relapse-free and overall survival in external datasets. Longitudinal transcriptional analysis of matched primary, residual, and metastatic samples identified immune pathway activation in NAC-resistant tumors, contrasting with cell cycle pathway inhibition. Tumor progression to metastasis was associated with upregulation of TFEB, CD3E, and TNFRSF9, while metastatic samples exhibited increased cell cycle-related signaling and FAK pathway inhibition. Our findings support the potential value of integrating clinical and molecular features to improve response stratification in HR+/HER2− BC patients treated with NAC.
Hormone receptor–positive/Human epidermal growth factor receptor 2–negative (HR+/HER2–) early breast cancer accounts for the majority of breast cancer cases. Although chemotherapy can reduce recurrence risk, its use in HR+/HER2– high-risk disease remains complex in clinical practice. We aimed to examine regional variation in chemotherapy use in Sweden and its association with prognosis. In this population-based cohort study, we included 61,935 women diagnosed with HR+/HER2– early breast cancer between 2007 and 2023, identified through the Swedish National Quality Register for Breast Cancer. Multivariable logistic regression was applied to identify factors that were associated with chemotherapy use across the Swedish healthcare regions. Overall survival was analysed using Kaplan–Meier estimates and regression standardisation based on Cox models, adjusting for clinicopathological and socioeconomic factors. Chemotherapy was administered to 28·0
The immunosuppressive microenvironment of breast cancer drives a ‘cold’ tumor phenotype characterized by limited immune cell infiltration and poor response to immunotherapy. While intratumoral microbiota and circulating lipopolysaccharide (LPS) have the potential to remodel immunity, the specific mechanisms and prognostic impact of LPS-driven immunomodulation in breast cancer remain unclear. We utilized the Comparative Toxicogenomics Database (CTD) to systematically identify LPS-related genes. Prognosis-associated LPS-related genes (PALRGs) were identified via Cox regression analyses. Subsequently, we applied LASSO-Cox regression to construct an LPS-related risk score (LPSRS) and evaluated its predictive performance and associations with immune microenvironment characteristics in TCGA-BRCA (n = 1,196) and GEO (n = 525) cohorts. For further validation, we recruited an independent clinical cohort of 56 breast cancer patients and confirmed differential expression of LPSRS genes and tumor microenvironment (TME) heterogeneity between high- and low-risk groups. We integrated RNA sequencing with 2bRAD-M sequencing to explore potential interactions between the tumor microbiome and host gene expression. Finally, we conducted mechanistic validation using 4T1 tumor-bearing mice, with treatment groups receiving either Pseudomonas. putida or purified LPS. We identified 26 PALRGs and constructed an 8-gene LPSRS signature as Risk Score which demonstrated strong prognostic performance (C-index = 0.750, 95
This study evaluates a predictive model using quantitative ultrasound features from intratumoral and peritumoral habitats to non-invasively determine HER2 status in invasive breast cancer patients. This retrospective study included 669 patients. The intra-tumoral ROI (Intra-ROI) was manually delineated, and the peri-tumoral ROI (Peri-ROI) was created by expanding 5 mm outward from the tumor boundary. Intra- and peri-tumoral subregions were established via K-Means clustering. Models were developed based on features from the intra-tumoral microenvironment, peri-tumoral microenvironment, entire intra-tumoral area, and entire peri-tumoral area. A habitat integration model was constructed by integrating two subregions, and a comprehensive model was built by combining clinical factors. The study aimed to distinguish HER2-positive and -negative cases and classify IHC 0 and 1 + subgroups among HER2-negative patients. Model effectiveness was evaluated using AUC, accuracy, sensitivity, specificity, DeLong test, Hosmer-Lemeshow test, decision curve analysis (DCA), Precision-Recall Curve and SHAP value analysis. For Task 1 (distinguishing HER2-positive from negative), the habitat + clinical combined model achieved superior performance with a test AUC of 0.757, ACC of 0.692, sensitivity of 0.692, specificity of 0.691, F1-score of 0.466, and MCC of 0.337. The intra- and peritumor habitat ensemble model also demonstrated strong performance with AUC of 0.669, ACC of 0.697, sensitivity of 0.436, specificity of 0.759, F1-score of 0.358, and MCC of 0.251. Both models significantly outperformed single-region models (Intratumor Habitat Model: AUC 0.642, F1-score 0.372; Peritumor Habitat Model: AUC 0.593, F1-score 0.335; Whole Intratumor Model: AUC 0.609, F1-score 0.344; Whole Peritumor Model: AUC 0.577, F1-score 0.326). For Task 2 (distinguishing IHC 0 from 1 + in HER2-negative patients), both the habitat + clinical combined model (test AUC 0.786, ACC 0.745, sensitivity 0.793, specificity 0.640, F1-score 0.811, MCC 0.449) and the intra- and peritumor habitat ensemble model (test AUC 0.778, ACC 0.752, sensitivity 0.838, specificity 0.560, F1-score 0.823, MCC 0.458) demonstrated robust predictive power, significantly outperforming single-region models (Intratumor Habitat Model: AUC 0.693, F1-score 0.751; Peritumor Habitat Model: AUC 0.713, F1-score 0.763; Whole Intratumor Model: AUC 0.671, F1-score 0.697; Whole Peritumor Model: AUC 0.630, F1-score 0.687). SHAP analysis highlighted progesterone receptor (PR), estrogen receptor (ER), and ensemble model probabilities as key predictive factors. Integrating intratumoral and peritumoral ultrasound features with clinical data demonstrates promising potential for non-invasive HER2 status prediction, with the ability to capture intratumoral heterogeneity. However, these findings should be interpreted with caution given the limitations of the current study, including its retrospective single-center design. Further validation through multi-center prospective studies is warranted before any clinical application. This approach may contribute to improving personalized breast cancer treatment strategies pending robust clinical validation.
In clinical breast imaging, microcalcifications (MCs) are routinely interpreted as small radiological signs, typically analyzed in clusters. They guide radiologists in assessing breast lesions and determining the likelihood of malignancy. Yet, increasing biological evidence indicates that active processes occur not only within the calcified core but also in the microcalcification-surrounding tissue microenvironment (MCST). This raises the possibility that current assessments may overlook critical diagnostic information embedded in the MCST. However, conventional mammography and digital breast tomosynthesis (DBT) lack the spatial resolution required to determine where diagnostic information truly resides. We used a high-resolution ( ≈ 8 μ m) 3D micro-CT scanner to scan 94 paraffin-embedded breast biopsy blocks, from which 3504 individual MCs were segmented to obtain a binary 3D mask ( M_0 ) for each MC. Unlike previous studies analyzing clustered MCs at mammographic resolution, our analysis operates at the level of individual MCs. Radiomic features were extracted from each MC and used to train machine-learning classifiers to predict the histopathological label (benign or malignant) of the lesion in which each MC occurred. To determine (i) whether discriminative information arises predominantly from the calcified core or from the MCST and (ii) to separate genuine tissue signal from preprocessing or segmentation effects, we conducted three complementary analyses. First, we held M_0 constant and varied only the size of the rectangular preprocessing window around M_0 , i.e. the 3D region of grayscale data considered for feature computation. Second, we assessed robustness to segmentation by making small, incremental changes to M_0 (erosions to restrict the calcified core; dilations to extend into immediately adjacent tissue). Third, to test MCST-only signal, we classified MCs using features computed exclusively from concentric shells defined at incremental offsets from the M_0 : outer shells (thin layers of MCST just outside M_0 ) and inner shells (thin layers within the calcified core, inside M_0 ). Increasing the size of the preprocessing window around a constant M_0 improved classification performance (AUC ≈ 0.688 → 0.811), revealing strong contextual effects in radiomics feature extraction. Moderate mask dilations likewise improved performance (AUC ≈ 0.689 → 0.813), indicating that the diagnostic signal extends beyond the calcified core into the MCST. Remarkably, even when using only concentric shell features, classification performance remained high (AUC ≈ 0.81). This study provides the first imaging-based evidence - at micrometer scale and in 3D - pinpointing where the discriminative information of breast MCs arises: within the calcified core and/or the MCST? From a radiomics and technical standpoint, the preprocessing window matters: it should be controlled and reported, as it directly affects extracted features and can influence diagnostic performance. We find that MCST is the dominant source of discriminative information for classifying individual MCs as benign or malignant: classification using features computed exclusively from concentric shells performs on par with maximally dilated mask (encompassing both the core and MCST) but clearly exceeds calcified-core-only. These findings shift the interpretative focus of MC analysis from the calcified deposits themselves to their immediate microenvironment, suggesting that future biomarkers and quantitative features should target MCST.
CCL5, abbreviated from C–C motif chemokine ligand 5, exerts diverse regulatory effects on both the initiation and progression of breast cancer (BC), making it a focal point of extensive research within the field. By enhancing the growth, invasive ability, and metastatic potential of BC cells, simultaneously recruiting and polarizing immunosuppressive cell populations, CCL5 actively reshapes the tumor immune microenvironment. Furthermore, elevated CCL5 levels are strongly linked to particular clinicopathological parameters, worse clinical outcomes, and a higher likelihood of recurrence in BC, suggesting its value in early diagnosis and prognosis. Current preclinical studies targeting the CCL5/CCR5 axis have demonstrated notable efficacy in inhibiting the progression of BC. This article explores how CCL5 contributes to BC progression and elucidates its regulatory mechanisms, providing theoretical support and potential targets for innovative therapeutic approaches.
Although mammography remains the cornerstone of breast cancer screening, its diagnostic performance is influenced by factors such as breast density and tumor characteristics. Blood-based biomarkers may therefore provide complementary biological information to imaging-based detection strategies. While circulating microRNAs (miRNAs) have emerged as promising minimally invasive biomarkers, their diagnostic performance and associations with clinicopathological characteristics remain incompletely understood. This multicenter study aimed to develop and validate a serum miRNA–based diagnostic model for breast cancer and to evaluate its associations with clinical stage, histological grade, and imaging-related factors. In this multicenter observational case-control study, serum samples from 345 patients with breast cancer and 373 cancer-free controls were analyzed using next-generation sequencing–based miRNA profiling. After normalization and batch correction, 114 stably expressed miRNAs were selected for model construction. An ensemble machine learning model integrating five algorithms (Lasso, Ridge, support vector machine, Nu-SVM, and histogram-based gradient boosting) was developed using a training cohort (n = 569) and evaluated in a held-out test cohort (n = 149). Model performance was assessed using receiver operating characteristic curve analysis and the area under the curve (AUC) with 95
Anti-Müllerian hormone (AMH) is widely used to assess ovarian reserve in young women with breast cancer. However, its interpretation during gonadotropin-releasing hormone agonist (GnRHa)-based endocrine therapy remains challenging because pharmacologic ovarian suppression may alter AMH levels over time. We aimed to characterize longitudinal AMH dynamics according to chemotherapy exposure and subsequent GnRHa-based endocrine therapy. We conducted a retrospective longitudinal cohort study of women diagnosed with early breast cancer before age 40 at Sun Yat-sen University Cancer Center between August 2019 and December 2024. Patients were categorized into four groups according to chemotherapy exposure and GnRHa-based endocrine therapy. Longitudinal trajectories of log-transformed AMH were modeled using linear mixed-effects models with restricted cubic splines, adjusting for prespecified clinical covariates. Sensitivity analyses included baseline-AMH adjustment and exclusion of observations with extreme standardized residuals. The full cohort included 685 women contributing 1758 AMH measurements. The adjusted complete-case model included 613 women contributing 1602 measurements. The group-by-time interaction was significant (likelihood ratio χ2 = 249.33, df = 9, P < 0.001), demonstrating distinct AMH trajectories across treatment groups. Among women without chemotherapy, predicted AMH did not differ significantly between those with and without GnRHa-based endocrine therapy. Among chemotherapy-treated patients, the GnRHa-based endocrine therapy group was associated with lower predicted AMH at 12 and 36 months, but the differences attenuated at later follow-up. Similar trajectory patterns were observed in sensitivity analyses. AMH trajectories during GnRHa-based endocrine therapy are time-dependent and influenced by treatment context. AMH measurements during GnRHa exposure should be interpreted in relation to treatment timing and longitudinal changes rather than as isolated indicators of ovarian reserve.
The phase 3 NATALEE trial demonstrated sustained invasive disease–free survival benefit for patients with stage II or III HR+/HER2− early breast cancer (EBC) treated with ribociclib + nonsteroidal aromatase inhibitor (NSAI) vs. NSAI alone. Here we report detailed safety and tolerability data from NATALEE to further inform clinical decision-making. In NATALEE, men and pre- and postmenopausal women with HR+/HER2− EBC were randomized 1:1 to ribociclib 400 mg/day (3 weeks on/1 week off; 36 months) + NSAI (anastrozole or letrozole; ≥60 months) or NSAI alone. Men and premenopausal women also received goserelin. Safety analyses included incidence of treatment-emergent adverse events (AEs) as well as severity, timing, management, and outcomes. Safety parameters among older vs. younger patients were also assessed. At this 45.7-month median follow-up (data cutoff: April 29, 2024), the safety analysis set comprised 4967 patients receiving ribociclib + NSAI (n = 2526) or NSAI alone (n = 2441). The most common grade ≥ 3 AEs occurring with ribociclib + NSAI vs. NSAI alone, respectively, were neutropenia (grouped term; 44.4
