Introduction Achieving negative margins at the first operation is a central goal of breast-conserving surgery, yet positive margins and re-excision remain common. Conventional preoperative MRI is acquired in the prone position, a position the breast does not assume during surgery, and the resulting prone-to-supine deformation may degrade image transfer to the operative field. Supine MRI (SMRI) has been proposed to overcome this limitation. This systematic review evaluated its effect on positive surgical margin and re-excision rates in breast-conserving surgery. Methods Four databases were searched from inception to May 2026, with backward citation tracking. Absolute rates were summarized narratively by transfer technique; a post-hoc exploratory relative-risk meta-analysis was restricted to studies with a comparator. Results Thirteen studies (1,718 patients) were included. Positive margin rates with SMRI ranged from 0% to 25.9%; heterogeneity in margin definitions precluded pooling of absolute rates. Across the five comparative studies, SMRI was associated with a significantly lower risk of a positive margin (relative risk 0.57, 95% CI 0.36–0.88) and a non-significant reduction in re-excision (relative risk 0.64, 95% CI 0.40 –1.04), without detectable heterogeneity. The evidence base was limited: a single randomized trial and 11 of 12 non-randomized studies at serious risk of bias, with baseline confounding that generally disfavored the intervention. Conclusion SMRI is a biologically coherent and promising approach that may reduce positive margins, but current evidence is preliminary. Randomized comparison against contemporary localization techniques, using harmonized margin definitions, is required.
Breast cancer prognosis prediction remains challenging due to the heterogeneous nature of clinical and multi-omics data and the complex interactions among diverse biological mechanisms. This study proposes a multimodal deep learning framework that integrates clinical variables, gene expression profiles, and copy number variation (CNV) data for breast cancer prognosis prediction. Three modality-specific models were developed, including a one-dimensional convolutional neural network for clinical data, a multilayer perceptron for gene expression data, and a denoising autoencoder-based neural network for CNV data. Snapshot ensemble learning was incorporated into all base models to improve prediction diversity, while bootstrap aggregation was additionally applied to the clinical model to enhance robustness. The probability outputs of the unimodal models were subsequently transformed into engineered interaction features and integrated using a nonlinear deep stacking meta-learner to exploit complementary prognostic information across modalities. The proposed framework was evaluated on the METABRIC breast cancer cohort using an independent test set and comprehensive performance metrics. The multimodal framework achieved a ROC-AUC of 0.878, 84.6% accuracy, 79.4% precision, 95.6% specificity, and an average precision of 72.5%, outperforming the individual unimodal models while producing satisfactory probability calibration. These findings demonstrate that multimodal stacking can effectively leverage complementary clinical and molecular information to improve breast cancer prognosis prediction and provides a flexible framework for multimodal data integration in precision oncology.•A multimodal deep learning framework integrates clinical, gene expression, and CNV data for breast cancer prognosis prediction.•Snapshot ensemble learning, bootstrap aggregation, and nonlinear deep stacking improve model robustness and multimodal integration.•The proposed framework achieved a ROC-AUC of 0.878 on an independent METABRIC test cohort, outperforming individual unimodal models.
Aim: To evaluate whether the ER/Ki67 ratio can predict axillary nonsentinel lymph node metastasis and thereby help avoid unnecessary axillary lymph node dissection (ALND) in breast cancer patients with negative preoperative imaging but positive sentinel lymph node (SLN) findings.Methods: Among 631 patients undergoing breast surgery with SLNB, 87 met the inclusion criteria. All had no axillary involvement on preoperative PET imaging but showed SLN metastasis on intraoperative frozen section. Clinicopathological features, including ER, PR, Ki67, tumor characteristics, and lymph node data, were retrospectively analyzed. An ROC curve was constructed to determine the optimal ER/Ki67 cut-off for predicting nonsentinel lymph node metastasis.Results: Of the 87 patients, 57.4% had no additional metastatic lymph nodes beyond the SLN, whereas 42.6% had further axillary involvement. An ER/Ki67 ratio >3.0833 was significantly associated with a lower risk of nonsentinel lymph node metastasis.Conclusion: The ER/Ki67 ratio was associated with additional NSLN metastasis in this selected cohort of SLNB-positive patients with negative preoperative axillary assessment. However, prediction of additional NSLN metastasis should not be interpreted as a stand-alone indication for completion ALND. The ER/Ki67 ratio may contribute to biological risk stratification, but the findings remain hypothesis-generating and require validation in larger contemporary cohorts. The ER/Ki67 ratio should therefore be interpreted as a supportive risk-stratification marker rather than as a direct criterion for performing or omitting completion ALND.
In a Chinese cohort of 402 T1-2, N0-1, ER+/HER2- breast cancer patients, low-risk patients identified by Low Risk Prediction (LRP) model derived from a breast cancer subtyping assay MammaTyper® demonstrated a favourable long-term prognosis. These patients could potentially waive chemotherapy without compromising prognosis. Background Although Oncotype DX recurrence score (ODX-RS) has been proved its predictive and prognostic utility in ER+/HER2− breast cancer (BC), cost and accessibility limit its use. The MammaTyper® breast cancer subtyping assay derived Low Risk Prediction (LRP) model has demonstrated to highly predict an ODX-RS≤25. This study aimed to evaluate the prognostic value of the LRP model and its potential to identify patients who might forego chemotherapy. Methods In this retrospective study, women diagnosed with T1-2, N0-1, ER+/HER2- BC between 2014 and 2018 at Peking University First Hospital were included. Patients were stratified with the MammaTyper® LRP model. Distant disease-free survival (DDFS) was compared using Kaplan-Meier curves and Cox proportional hazards models. Results Among 402 eligible patients, 221 (55%) were classified as low-risk by LRP. The 9-year DDFS rate was significantly higher in this group than in the high-risk group (96.2% vs. 87.0%, p=0.0021). LRP low risk status was an independent favourable prognostic factor for DDFS (adjusted HR=0.37, p=0.04). No significant DDFS differences were observed between ET and ET+CT groups among LRP low-risk patients (HR=0.38, p>0.05). Similarly, clinical risk-high/LRP-low patients did not benefit from chemotherapy in DDFS (HR=0.33, p>0.05). Conclusions The MammaTyper® LRP model effectively identified a subset of low-risk patients with excellent long-term prognosis in 9-year DDFS from early BC. No significant DDFS benefits were observed from adjuvant chemotherapy in LRP low-risk patients, even in those with high risk clinical features. This results support the LRP model identified low-risk patients might undergo chemotherapy de-escalation, warranting prospective validation.
Purpose Older women with early-stage luminal A breast cancer have an excellent prognosis with a 5-year relative survival >99%. Research on curtailing overtreatment has challenged historical standards of care by reducing radiotherapy dose and volume, and selectively omitting sentinel lymph node biopsy. This study assessed the utility of Oncotype Dx genomic testing in ultra-low risk luminal A breast cancer. Methods A community hospital cancer registry was queried to identify consecutive breast cancer patients from 2014 through 2022. An ultra-low risk population was defined as age ≥60 years, hormone receptor positive, HER2-negative, pathologic stage T1b-T2N0 without lymphovascular invasion, and Ki-67 ≤13.25%. Analyses examined the distribution of Oncotype Dx scores categorized as low risk (0-18), intermediate risk (19-25), or high risk (26-100), as well as recurrence and overall survival. Results Of 1711 patients, 91 met ultra-low risk eligibility criteria with available Oncotype Dx results. The median age was 70 years with a median follow-up of 5.1 years, and no patients received chemotherapy. Only 1 patient (1.1%) had a high risk Oncotype Dx score of ≥26, while 12 (13.2%) had intermediate scores and 78 (85.7%) were low risk. Five-year local control and overall survival were 100% and 93%, respectively. Ten non-breast cancer deaths occurred and no patients developed distant metastases. Conclusion We describe a pragmatic method using common clinical and pathological features to define an ultra-low risk cohort. Older luminal A patients with favorable pathology have a very low probability of receiving a high risk Oncotype Dx score and demonstrate an excellent prognosis with standard adjuvant therapy.MicroAbstractIn older patients with ultra-low risk luminal A breast cancer and favorable clinicopathologic features, the probability of a high Oncotype Dx score is low. These findings suggest Oncotype testing in an ultra-low risk population could be selectively omitted.
Recent advances in endocrine therapy have led to the development of novel oral agents that target estrogen receptor signaling in estrogen receptor-positive, HER2-negative (ER+/HER2-) breast cancer. This review outlines key findings from clinical trials of Food and Drug Administration (FDA)-approved selective estrogen receptor degraders (SERDS), including elacestrant and imlunestrant, the newly FDA-approved proteolysis-targeting chimera (PROTAC) vepdegestrant, as well as emerging oral SERDs, PROTACs, complete estrogen receptor antagonists, selective estrogen receptor covalent antagonists (SERCAs), and other targeted therapies. We explore their mechanisms, efficacy, tolerability, and potential to improve treatment options and outcomes in both metastatic and early-stage disease.
BACKGROUND:Metastatic (m) HR-positive/HER2-negative special-type breast cancer (STBC) exhibits distinct biological behaviors compared with no special-type BC (NSTBC), yet their management is largely extrapolated from trials dominated by NSTBC. To improve understanding of real-world (rw) clinical outcomes, this study evaluated the effectiveness of CDK4/6 inhibitors (i) specifically in patients with mSTBC, providing clinical relevant insights into their therapeutic impact. PATIENTS AND METHODS:This retrospective cohort study included patients with mSTBC treated with first- or second-line CDK4/6i plus endocrine therapy (ET). Real-world progression-free survival (rwPFS) and overall survival (rwOS) were estimated using the Kaplan-Meier method, and differences between subgroups were evaluated using log-rank tests. RESULTS:From April 2016 to December 2024, 181 patients received CDK4/6i-based therapy. Median rwPFS was 18.2 months (95% CI, 16-22.1) and median rwOS was 58.1 months (95% CI, 49.4-77.4). A statistically significant difference in rwOS was observed between patients treated with aromatase inhibitors + CDK4/6i versus fulvestrant + CDK4/6i (77.4 vs. 42.4 months; unadjusted HR = 2.2; P = .002). No other subgroup analyses demonstrated statistically significant differences. Common adverse events included neutropenia, anemia, and diarrhea; 32% of patients required dose reductions, which did not adversely affect survival outcomes. Following CDK4/6i progression, overall rwPFS2 was 29.7 months (95% CI, 25.9-32.2). Chemotherapy was the most frequently used post-CDK4/6i treatment, while ET combined with an mTOR-inhibitor provided the longest rwPFS2. CONCLUSION:CDK4/6i represent a safe and effective therapeutic option for mSTBC. Nevertheless, dedicated studies and tailored treatment strategies are warranted to optimize post-CDK4/6i treatment in these subgroups.
Background Neoadjuvant chemotherapy (NACT) combined with pembrolizumab is now standard care for early-stage, triple-negative breast cancer (TNBC), based on KEYNOTE-522 demonstrating improved pathological complete response (pCR), event-free survival, and overall survival. However, the real-world impact of treatment modifications due to toxicity remains unclear. We evaluated clinical outcomes and treatment delivery in routine practice. Materials and Methods This multicenter retrospective study included patients with stage II to III TNBC treated with NACT plus pembrolizumab across 5 UK centers (May 2022-June 2025). Data on immune-related adverse events (irAES), corticosteroid use, immunotherapy omissions, chemotherapy dose reductions and recurrence were collected. pCR rates were compared according to treatment modifications. Results One hundred patients were analyzed (median age 50 years; 54% node-positive). Overall pCR rate was 54.0%, with no significant difference by pembrolizumab omissions (0: 55.0%; 1-3: 55.6%; ≥ 4: 46.2%; P = .83) or corticosteroid use (48.3% vs. 56.3%; P = .61). Grade ≥ 3 toxicities occurred in 37%, and 29% required corticosteroids, most commonly between cycles 2 to 4. Pembrolizumab omissions occurred in 40%, chemotherapy dose reductions in 57%; 23% did not complete planned chemotherapy. Fourteen recurrences were observed (median 11 months), predominantly in baseline node-positive disease; 4 occurred despite pCR. Conclusions In real-world practice, pembrolizumab-based NACT is associated with frequent treatment modification, yet pCR rates remain comparable to trial data. Early recurrences—often in node-positive patients and occasionally despite pCR—highlight the need for improved risk stratification, optimized adjuvant strategies, and proactive toxicity management to ameliorate long-term outcomes.
Invasive micropapillary carcinoma of the breast is a rare subtype of breast cancer with a unique "inside out" polarity reversal pattern. Despite exhibiting extremely high rates of lymphatic vessel invasion (LVI) and regional lymph node metastasis, large clinical cohorts often observe long-term survival outcomes that are not inferior or even superior to those of ordinary non-specific invasive cancer (IBC-NST), leading to a long-standing controversy over the "clinical prognosis paradox." This article comprehensively summarizes the multidimensional factors that have led to this controversy, pointing out that the enormous survival benefits brought by systemic treatment largely mask its invasive morphological features. At the same time, this article deeply analyzes the multi-omics molecular mechanisms that drive its early metastasis, and systematically evaluates the application of predictive models in precise risk stratification of invasive micropapillary carcinoma (IMPC). In the future, unifying the pathological diagnostic threshold of "pure type" IMPC, integrating multimodal omics data, and constructing a rigorously validated artificial intelligence decision support system will be the key direction for achieving personalized and precise diagnosis and treatment of IMPC.
PURPOSE:In 2018, the American Joint Committee on Cancer (AJCC) revised the TNM staging system for breast cancer (BC), incorporating biological factors into anatomical staging. This study evaluated the clinical impact of this integration and the relative prognostic contribution of anatomical and biological factors to 5-year overall survival (OS). METHODS:A retrospective analysis was conducted on a cohort of 1075 patients diagnosed with BC at the Breast Centre of São João University Hospital (2015-2020), using clinicopathological data from a EUSOMA-certified database. The primary endpoint was 5-year OS and was analyzed using Kaplan-Meier estimates, log-rank tests, and Cox proportional hazards regression. RESULTS:Stage migration occurred in 42.1% of patients when switching from anatomical to pathological prognostic staging, with 37.5% downstaging. Downstaging was observed across all stages and molecular subtypes except triple-negative BC (TNBC). Consistent with previous studies, 5-year OS rates were equivalent between staging systems, with no significant survival differences between patients who remained in a stage and those who migrated to it. On multivariable analysis, tumor size showed the strongest prognostic impact on 5-year OS, particularly for T3/T4 disease, while N3 status remained independently associated with poorer outcomes. Histological grade and molecular subtype showed more limited prognostic significance; only TNBC was independently associated with worse prognosis. CONCLUSION:The AJCC pathological prognostic staging system promotes substantial stage migration while refining survival prediction. Nevertheless, anatomical tumor burden remains the predominant determinant of 5-year OS, reinforcing the continued clinical relevance of traditional TNM staging despite the incorporation of tissue-based biomarkers.
Purpose To determine the incidence and characteristics of synchronous contralateral breast cancer (SCBC) detected at elective bilateral mastectomy for unilateral breast cancer using the National Cancer Database (NCDB), and to identify factors associated with contralateral disease. Materials and Methods We performed a retrospective analysis of the NCDB Participant Use File (2022–2023) using newly introduced breast-specific surgery codes identifying bilateral mastectomy with contralateral removal. Patients with unilateral breast cancer who underwent bilateral mastectomy were identified (n=13,931). SCBC was defined conservatively as SEQUENCE_NUMBER=01. Multivariable logistic regression with stepwise selection identified independent predictors. Probabilistic matching enabled concordance analysis of 336 paired tumors. Results Among 13,931 patients, 709 had SCBC (incidence 5.09%). Mean age was 57.7±12.3 years. Most primary tumors were invasive (81.4%), HR+/HER2− (64.5%), node-negative (64.0%), and stage 0–IA (56.0%). Age was the strongest predictor, with 6.7-fold higher odds in patients ≥70 versus <40 (OR 6.68, 95% CI 4.58–9.76, p<0.0001). Compared with HR+/HER2− tumors, other subtypes carried lower odds, with triple-negative disease lowest (OR 0.48, 95% CI 0.35–0.65, p<0.0001). Primary tumor size was not associated with SCBC (OR 1.001, p=0.20). Discrimination was modest (c-statistic 0.669). In 336 matched pairs, contralateral tumors were smaller (median 11.5 vs. 18.5 mm) and predominantly early stage (61% stage 0–IA; 3.3% stage II+). Concordance was 86.8% for ER, 78.8% for HER2, 71.8% for PR, 67.0% for histology, and 42.6% for grade, with full concordance in 40.8%. Conclusion SCBC is found in approximately 1 in 20 patients undergoing bilateral mastectomy for unilateral disease. Risk is driven by patient-level factors, particularly age and receptor subtype, rather than primary tumor burden. High biomarker discordance supports independent tumor origins and highlights the diagnostic value of contralateral mastectomy specimens in select populations.
BACKGROUND:Chronological age alone poorly reflects physiologic vulnerability in hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer (mBC) treated with cyclin-dependent kinase 4/6 inhibitors (CDK4/6i). We therefore evaluated phenotypic age acceleration (PhenoAgeAccel), a biomarker of biological aging, as a predictor of survival outcomes. METHODS:This retrospective cohort included 560 patients with HR+/HER2-negative mBC treated with CDK4/6i plus endocrine therapy between 2017 and 2025. PhenoAge was calculated using clinical biomarkers obtained within 1 month prior to treatment initiation, and PhenoAgeAccel was defined as the residual of PhenoAge regressed on chronological age. Patients were categorized as PhenoAgeAccel ≤ 0 or > 0. RESULTS:PhenoAgeAccel was evaluable in 530 patients; 38.3% exhibited PhenoAgeAccel > 0. Median follow-up for the entire cohort was 42.2 months (95% confidence interval [CI], 39.0-45.5). Median progression-free survival (PFS) was 21.1 months (95% CI, 18.6-23.5) and median overall survival (OS) was 49.9 months (95% CI, 41.8-58.0). Patients with PhenoAgeAccel > 0 experienced significantly shorter PFS and OS compared with those without acceleration (median PFS 14.0 vs. 27.6 months, hazard ratio [HR] 1.53, 95% CI, 1.17-2.01; P = .002; median OS 28.9 vs. 68.0 months, HR 1.81, 95% CI, 1.26-2.60; P = .001). The adverse prognostic impact of PhenoAgeAccel was consistent across predefined clinical subgroups. Patients with accelerated biological aging also demonstrated lower objective response (ORR) and disease control rates (DCR) (objective response rate 25.6% vs. 33.9%, P = .043; DCR 72.4% vs. 89.0%, P < .001). CONCLUSIONS:PhenoAgeAccel is an independent prognostic marker associated with inferior survival and reduced treatment response, suggesting that biological age may be more informative than chronological age in HR+/HER2-negative mBC treated with CDK4/6 inhibitors.
AIM:This descriptive study was conducted to determine body perceptions and preferences regarding breast reconstruction among women who had undergone mastectomy. METHODS:The research sample consisted of 138 patients who applied to the Breast Endocrine Surgery Outpatient Clinic of a public training and research hospital, had previously undergone mastectomy, and had not received breast reconstruction or oncoplastic breast surgery. Data were collected using a questionnaire developed by the researcher and the Body-Cathexis Scale. RESULTS:Among the participants, 47.9% had been diagnosed at stage II breast cancer, and 76.1% had undergone modified radical mastectomy with axillary dissection. It was found that 44.9% of the patients used an external breast prosthesis, and although 65.2% had received information about breast reconstruction, 37.0% stated that they did not prefer to undergo reconstruction. The mean Body-Cathexis Scale score was 180.1 ± 18.6, indicating a relatively positive body perception. However, no statistically significant relationship was found between body perception and breast reconstruction preference (P > .05). Preferences for breast reconstruction were significantly associated with place of residence, prior information received, body mass index, and nutritional habits (P < .05). CONCLUSION:The findings suggest that breast reconstruction preference rates among patients are low and that the level of awareness regarding reconstruction remains inadequate. Therefore, it is recommended that patients and their families receive education and counseling about breast reconstruction starting from the time of diagnosis, and that multidisciplinary community-based educational initiatives be implemented to raise awareness of reconstruction options.
INTRODUCTION:With better responses to neoadjuvant therapy (NST), clinicians are increasingly inclined to spare axillary surgery. Yet it is not well defined whether tumor regression in the breast can serve as a reliable surrogate for nodal clearance across diverse nodal burdens at baseline and intrinsic tumor biology. PATIENTS AND METHODS:We retrospectively analyzed 864 patients treated with NST across 5 Chinese institutions. Subgroup analyses were performed according to initial nodal stage (cN0, cN1, cN2/3), molecular subtype, breast pathological response (ypT0, ypTis, or non-pCR), and final preoperative imaging. The endpoint was ypN status. RESULTS:Breast pCR was observed in 34.4%. Among cN0 patients with breast pCR, 95.6% achieved axillary pCR, reaching 100% in TNBC or HER2-positive tumors, irrespective of ypT0 or ypTis. In cN1 disease, breast pCR correlated with a 93.0% axillary pCR; notably, all ypT0 cases in TNBC/HER2-positive were node-negative. For cN2/3 patients with breast pCR, the false-negative rate of imaging in predicting nodal clearance remained below 5% for both ypT0 and ypTis. CONCLUSIONS:Selected patients may be candidates for axillary surgery omission-specifically, TNBC/HER2-positive cN0 patients with breast pCR (ypT0/is), and cN1 patients who achieve ypT0 (but not ypTis). For cN2/3 patients with breast pCR and negative post-treatment imaging, foregoing SLNB appears worth considering, though prospective data are needed to confirm safety.
BACKGROUND:Following neoadjuvant chemotherapy (NAC), a subset of breast cancer patients becomes candidates for breast conserving surgery (BCS) after tumor downstaging. In the literature, there is little analysis of patient choice after NAC. The goal of this study is to investigate clinical and demographic predictors of patients' surgical choices after NAC to better inform shared decision-making. PATIENTS AND METHODS:A retrospective chart review was conducted on breast cancer patients at our institution who became BCS candidates after receiving NAC from January 1, 2010 to September 30, 2020. Patient demographics, clinical characteristics, choice of mastectomy vs. BCS, and outcomes were compared. Exclusion criteria included inflammatory, recurrent, bilateral breast cancers, and patients with genetic predisposition. Statistical analysis utilized R software version 4.2.3 with a significance threshold P < .05. RESULTS:In total, 244 patients met study criteria, of which 181 patients (74.2%) chose BCS, and 63 patients (25.8%) elected mastectomy. On average, mastectomy patients were younger at the time of diagnosis than BCS patients (48 vs. 53 years old, P = .003). Although breast tumor palpability was not associated with choice of surgery (P = .271), palpable axillary lymphadenopathy was associated with a higher likelihood of choosing mastectomy (44.4% vs. 29.3%, P = .041). Patient race, pretreatment clinical stage, and doxorubicin-based chemotherapy were not associated with patients' decisions for type of surgery. CONCLUSION:Most breast cancer patients elect BCS after downstaging from NAC. Factors that may predict choosing mastectomy include younger age at diagnosis and pretreatment palpable axillary lymphadenopathy. These considerations should be integrated into shared surgical decision-making after NAC.
With improved survival rates in breast cancer, managing long-term morbidity of breast cancer patients is becoming increasingly relevant. Chronic pain after breast cancer surgery affects 20% to 60% of patients, of which 30% to 68% are neuropathic in nature. However, there is an existing gap in the literature concerning the exact prevalence of neuropathic pain, specifically regarding surgical axillary procedures in breast cancer patients. This systematic review and meta-analysis aimed to investigate the prevalence of neuropathic pain after axillary procedures, including axillary lymph node dissection (ALND) and sentinel lymph node biopsy (SLNB). The electronic bibliographic databases Embase, Medline, and Web of Science were searched from inception until February 2025. Inclusion and exclusion criteria were defined to include studies reporting on neuropathic pain prevalence following axillary procedures in breast cancer patients. Meta-analyses were performed to quantify pooled prevalences. Overall, 15 studies met the inclusion criteria, of which 14 were included in the meta-analyses, encompassing 2725 patients. A pooled neuropathic pain prevalence after axillary procedures of 22.6% (95% confidence interval [CI], 15.5-30.6) was estimated in the meta-analysis. The meta-analyses on ALND (10 studies, 1253 patients) and SLNB (5 studies, 860 patients) revealed an estimated pooled prevalence of 26.9% (95% CI, 17.1-37.9) and 18.8% (95% CI, 10.5-28.5), respectively. Neuropathic pain affects approximately 1 in 5 breast cancer patients following axillary surgery, with higher rates after ALND than SLNB.
