Objectives Multidisciplinary team (MDT) meetings are key to delivering cancer care. Increasing caseload and limited resources make them less effective and unsustainable. The aim of this quality improvement project was to assess novel artificial intelligence-based clinical decision support (CDS) technology to develop and validate standard of care (SoC) to streamline the breast MDT meetings in a tertiary cancer centre.Methods A clinical governance group of the MDT approved international guidelines used to develop SoC. Deontics CDS was assessed for its suitability to apply the SoC pathway for benign and malignant breast disease with the exclusion of metastatic and recurrent cancer.Results Patients discussed over the preceding 16 months were added to the platform in cohorts of 50 women: two consisting of 50 women each diagnosed with benign disease (benign A and B: n=100) and three consisting of 50 women each diagnosed with malignant disease (cancer A, B and C: n=150). Concordance between the blinded MDT decision outcomes and SoC recommendations was analysed. This stepwise approach identified knowledge gaps in SoC and refined the CDS. Concordance improved from 82% to 100% in benign and from 94% to 100% in malignant cases.Discussion A sequential process of validating the SoC with data derived from the development of evidence-based SoC protocols based on international guidelines resulted in a final 100% concordance rate between the platform and MDT recommendations for both benign and malignant disease.Conclusions CDS technology could be a milestone in using SoC to deliver a sustainable clinical decision pathway.

Background:Annual surveillance mammograms for an unspecified period, after treatment for early breast cancer, are widely practised in the United States of America and Europe. Current UK guidelines recommend annual mammograms for 5 years, then reverts to 3-yearly screening. The aim of this trial was to evaluate whether less than annual mammography was non-inferior in terms of breast cancer-specific survival and cost-effectiveness in women aged 50 years or older at diagnosis and 3 years post curative surgery. Methods:We conducted a multicentre, randomised phase III trial of annual mammography versus less-frequent mammography (2-yearly after conservation surgery or 3-yearly after mastectomy). Women were eligible if aged ≥ 50 years at initial diagnosis of breast cancer (invasive or ductal carcinoma in situ) and recurrence-free 3 years post curative surgery. The trial was conducted at 114 NHS hospitals in the UK. Participants were randomly assigned (1 : 1) to annual or less-frequent mammograms; followed up for 6 years. Coprimary outcomes were breast cancer-specific-survival and cost-effectiveness; secondary outcomes included recurrence-free interval and overall survival. Analyses were by intention to treat, with a pre-planned per-protocol analysis. Planned sample size was 5000. Clinical results are now reported. Results:Five thousand two hundred and thirty-five women were randomised between April 2014 and September 2018. With a median of 5.7-year follow-up, 343 women have died, of whom 116 died of breast cancer (61 on annual arm; 55 on less-frequent arm). Breast cancer-specific-survival at 5 years was 98% on both arms with a hazard ratio of 0.92 (95% confidence interval 0.64 to 1.32), which demonstrated non-inferiority of less-frequent mammograms at the 3% margin (non-inferiority p < 0.0001) and the 1% margin (non-inferiority p = 0.003). Non-inferiority was demonstrated at the 2% level for both recurrence-free interval [hazard ratio 1.00 (95% confidence interval 0.83 to 1.28); non-inferiority p = 0.0024] and overall survival [hazard ratio 1.07 (95% confidence interval 0.87 to 1.33); non-inferiority p = 0.008]. Less-frequent mammograms were associated with a significant cost saving (mean difference £544, 95% confidence interval -£1116 to £26), heavily driven by mammogram costs. Incorporating societal costs resulted in a larger cost-saving (£1543 per person, 95% confidence interval -£2416 to -£669), increasing cost-effectiveness. There was no impact of less-frequent mammograms on patients' quality of life. Conclusion:For patients aged ≥ 50 years and 3 years post diagnosis, less-frequent mammograms were non-inferior and cost-effective compared with annual mammograms, with no detriment to patients' quality of life. Mammo-50 provides evidence to inform guideline development. Limitations:Adherence to the mammographic schedules was 76%, though the per-protocol analysis showed no difference compared to the intention to treat results. The majority of the participants had small lower-grade oestrogen receptor-positive tumours and were from a White ethnic group. Future work:More research is needed for women with ductal carcinoma in situ; women aged under 50 years old at diagnosis and different ethnic groups, especially those women of Black ethnicity who tend to present younger. Funding:This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number 11/25/03.

Ki-67 is a well-established marker of tumour proliferation and an important prognostic and predictive biomarker in breast cancer, particularly in hormone receptor-positive (HR-positive), HER2-negative disease. Despite its biological relevance, clinical implementation has been limited by the reported interobserver and interlaboratory variability. Recent therapeutic advances have created an increased need for accurate and reproducible Ki-67 assessment in clinical practice. This review summarizes the biological basis for the use of Ki-67 as a marker of proliferation, technical requirements for reliable immunohistochemistry and the influence of pre-analytical and analytical variables on staining performance. We evaluate established and emerging scoring approaches and provide scoring recommendations for practising pathologists. A simplified calibrated global assessment method is presented as an alternative to exhaustive visual quantification that preserves its accuracy while substantially reducing scoring time and avoiding the variability of estimated methods. Image analysis/artificial intelligence (AI) using validated algorithms is recommended where available. We also review the role of Ki-67 in predicting response to neoadjuvant endocrine and chemotherapy, its integration into prognostic models such as the PEPI score, and its utility in selecting patients for adjuvant CDK4/6 inhibition. The limitations of fixed cut-off values are discussed, together with the potential advantages of tiered classification and continuous modelling. Finally, we outline the growing role of digital pathology and AI, which have demonstrated improved reproducibility, reduced turnaround time, and prognostic performance superior to manual scoring. Ki-67 is a clinically meaningful biomarker, the value of which can only be fully realized through rigorous standardization, validated scoring approaches and close communication between pathologists and oncologists. This guidance provides a practical framework for high-quality Ki-67 assessment and supports its safe and effective integration into contemporary breast cancer management.

Borderline breast lesions (B3 lesions, also termed lesions of uncertain malignant potential or high-risk lesions) represent a heterogeneous group of entities associated with variable risks of malignancy. While the management of screen-detected B3 lesions has become increasingly standardised, no dedicated international recommendations exist for symptomatic B3 lesions, despite them posing a distinct clinical challenge. Symptomatic lesions differ from screen-detected lesions in their mode of presentation, lesion characteristics, biopsy techniques, and diagnostic objectives; consequently, management strategies derived from screening populations may not be directly applicable. This review summarises the current evidence and proposes a pragmatic management framework for B3 lesions encountered in the symptomatic setting (defined as those presenting with breast symptoms outside population-based screening programmes). This is particularly relevant for patients under 50 years of age, where the primary objective is to exclude malignancy at the index site rather than solely to stratify long-term cancer risk. Clinical-radiological-pathological concordance and multidisciplinary assessment are central to management decisions. Lesions presenting as palpable abnormalities (which are typically larger) or those demonstrating radiological-pathological discordance warrant a lower threshold for additional sampling or excision. Conversely, selected concordant lesions without atypia may be managed conservatively following adequate sampling. The long-term cancer risk associated with epithelial atypia should also be considered with subsequent risk-based surveillance implemented where appropriate. Overall, management requires a risk-adapted, multidisciplinary approach integrating pathological, radiological, and clinical factors alongside patient symptoms and quality-of-life considerations. Given the limited evidence, current recommendations rely on expert consensus and a proposed practical framework. They underscore the need for large-scale, high-quality studies with long-term follow-up to establish formal clinical guidelines.

Introduction: Patients undergoing neoadjuvant chemotherapy (NACT) for breast cancer are monitored using Magnetic Resonance Imaging (MRI) with Dynamic Contrast Enhancement (DCE), the gold-standard imaging technique to assess tumour response. DCE-MRI has a sensitivity of 80-90% but a low specificity of 37-97% when assessing complete response (CR). Although alternative methods to assess CR have been attempted, such as biopsies, no highly accurate method has been identified. Magnetic Resonance Elastography (MRE) is a non-invasive imaging technique that uses biomechanics to identify tissue alterations within the tumour during NACT. With recent discussions regarding the de-escalation of surgical treatment for breast cancer post-NACT, an accurate assessment of CR is essential. Methods: This prospective, first-in-human study includes patients undergoing five MRI-MRE scans at different time points during the course of NACT. Breast MRE is a 7-minute sequence added to the routine clinical breast MRI, with mechanical vibrations applied to the patient’s breast via paddles attached to a gravitational driver incorporated into a Siemens biopsy coil for a 1.5T system. The tissue occupying the tumour at diagnosis on DCE imaging was defined throughout the NACT treatment. Biomechanics within these regions were quantified, primarily observing relative changes in tumour stiffness (elasticity) between pre-NACT and post-NACT (tumour stiffness ratio: TSR). Relative changes in phase angle (the phase lag between viscosity and elasticity) in the specific tumour region after the 1st cycle from pre-NACT were also observed (phase angle ratio: PAR). Post-surgical histopathology was used to determine complete and partial responders. Furthermore, a repeatability analysis was done on nine patients to assess concordance of the scans. Results: The analysis included complete datasets from forty-one patients. After NACT, the TSR significantly decreased for complete pathological responders and increased for partial responders (p<.001). The PAR's post-cycle 1.1 predictive ability for a complete pathological response was also significant (p<.001). When TSR was combined with DCE imaging, the specificity improved considerably compared to DCE alone (43.5%→95.7%), while maintaining the high sensitivity of DCE (94.4%). The repeatability analysis demonstrated excellent agreement for elasticity (ICC=0.969, RC=8.9%) and phase angle (ICC=0.876, RC=12.6%). Conclusion: This novel technology, which uses a combined approach including biomarkers (DCE+MRE), shows great promise as a non-invasive imaging method for assessing complete pathological response at the end of NACT. As a stand-alone method or combined with other investigative techniques to diagnose complete responses accurately and non-invasively at the end of chemotherapy, this can aid consideration in de-escalating surgical treatment. Furthermore, the information after the 1st cycle can help predict if tumours will eventually have a complete response or partial/no response; the latter may need early re-discussion in multi-disciplinary meetings to consider an alteration to their chemotherapy regimens. Further studies are required to strengthen these findings. Citation Format: Aaditya Sinha, Patriek Jurrius, Anne-Sophie van Schelt, Omar Darwish, Giacomo Annio, Belul Shifa, Zhane Peterson, Hannah Jeffery, Karen Welsh, Anna Metafa, John Spence, Ashutosh Kothari, Hisham Hamed, Georgina Bitsakou, Vasileios Karydakis, Mangesh Thorat, Elina Shaari, Ali Sever, Anne Rigg, Tony Ng, Sarah Pinder, Ralph Sinkus, Arnie Purushotham. Magnetic Resonance Elastography: A Novel Imaging tool to predict response in patients undergoing Neo-Adjuvant Chemotherapy for Breast Cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr P4-05-19.

Background The frequency of mammographic surveillance for women after diagnosis of breast cancer varies globally. The aim of this study was to evaluate whether less than annual mammography was non-inferior in terms of breast cancer-specific survival in women aged 50 years or older. Methods Mammo-50 was a multicentre, randomised, phase 3 trial of annual versus less frequent mammography (2-yearly after conservation surgery; 3-yearly after a mastectomy) for women aged 50 years or older at initial diagnosis of invasive or non-invasive breast cancer and who were recurrence free 3 years post curative surgery. The trial was conducted at 114 National Health Service hospitals in the UK. Participants were randomly assigned (1:1) to annual or less frequent mammograms at 3 years post curative surgery and were followed up for 6 years. The co-primary outcomes were breast cancer-specific survival and cost-effectiveness. The cost-effectiveness analysis will be reported elsewhere. Breast cancer-specific survival was assessed in the intention-to-treat population. Secondary outcomes were recurrence-free interval, overall survival, and referrals back to the hospital system. 5000 women provided 90% power to detect a 3% absolute non-inferiority margin for breast cancer-specific survival with 25% one-sided significance. The trial was registered with the ISRCTN registry, ISRCTN48534559; recruitment is complete but longer-term followup is ongoing. Findings Between April 22, 2014, and Sept 28, 2018, 5235 women were randomly assigned to annual mammography (n=2618) or less frequent mammography (n=2617). 3858 (736%) women were aged 60 years or older, 4202 (803%) had undergone conservation surgery, 4576 (874%) had invasive disease, 1159 (221%) had node positive disease, and 4330 (827%) had oestrogen receptor-positive tumours. With a median of 57 years follow-up (IQR 50-60; 87 years post curative surgery), 343 women died, including 116 who died of breast cancer (61 in the annual mammography group and 55 in the less frequent mammography group). 5-year breast cancer-specific survival was 981% (95% CI 975-986) in the annual mammography group and 983% (978-988) in the less frequent mammography group (hazard ratio 092, 95% CI 064-132), demonstrating non-inferiority of less frequent mammography at the pre- specified 3% margin (non-inferiority p<00001). 5-year recurrence-free interval was 941% (95% CI 931-949) in the annual mammography group and 945% (935-953) in the less frequent mammography group. Overall survival at 5 years was 947% (95% CI 938-955%) and 945% (935-953), respectively. 224 (649%) of 345 breast cancer events were detected from emergency admissions or symptomatic referrals back to the hospital system, including 108 (617%) of 175 in the annual mammography group and 116 (682%) of 170 in the less frequent mammography group. Interpretation For patients aged 50 years or older and at 3 years post diagnosis, less frequent mammograms were non-inferior compared with annual mammograms for breast cancer-specific survival, recurrence-free interval, and overall survival, and should be considered for this population.

Fibroepithelial lesions (FELs) of the breast represent a diverse group of biphasic tumors with varying morphologies and clinical behavior. The classification of FELs is mainly based on a constellation of diagnostic criteria, and intralesional heterogeneity is not uncommon. Therefore, reporting FELs in a core needle biopsy (CNB) with limited tissue material can be challenging as not all the features may be represented for assessment. Differentiating a classic fibroadenoma from a well-sampled phyllodes tumor (PT) is generally straightforward. However, cellular fibroadenoma, morphologically heterogeneous benign PT, and myoid hamartoma can overlap histologically. Accurate grading of PT is also challenging on CNB and carries significant management implications. In this article, we provide an overview and propose a pragmatic approach to reporting FELs on CNB, particularly for lesions with overlapping features. Guidance using the UK/European "B" classification of FELs alongside descriptive reporting of the various lesions, is also presented to aid in management decisions.

Supplementary table 1 - Biomarkers of interest

BACKGROUND:Ductal Carcinoma In Situ (DCIS) can progress to ipsilateral invasive breast cancer (IBC) but over 75% of DCIS lesions do not progress if untreated. Currently, DCIS that might progress to IBC cannot reliably be identified. Therefore, most patients with DCIS undergo treatment resembling IBC. To facilitate identification of low-risk DCIS, we developed deep learning models using histology whole-slide images (WSIs) and clinico-pathological data. METHODS:We predicted invasive recurrence in patients with primary, pure DCIS treated with breast-conserving surgery using clinical Cox proportional hazards models and deep learning. Deep learning models were trained end-to-end with only WSIs or in combination with clinical data (integrative). We employed nested k-fold cross-validation (k = 5) on a Dutch multicentre dataset (n = 558). Models were also tested on the UK-based Sloane dataset (n = 94). FINDINGS:Evaluated over 20 years on the Dutch dataset, deep learning models using only WSIs effectively stratified patients into low-risk (no recurrence) and high-risk (invasive recurrence) groups (negative predictive value (NPV) = 0.79 (95% CI: 0.74-0.83); hazard ratio (HR) = 4.48 (95% CI: 3.41-5.88, p < 0.0001); area under the receiver operating characteristic curve (AUC) = 0.75 (95% CI: 0.70-0.79)). Integrative models achieved similar results with slightly enhanced hazard ratios compared to the image-only models (NPV = 0.77 (95% CI 0.73-0.82); HR = 4.85 (95% CI 3.65-6.45, p < 0.0001); AUC = 0.75 (95% CI 0.7-0.79)). In contrast, clinical models were borderline significant (NPV = 0.64 (95% CI 0.59-0.69); HR = 1.37 (95% CI 1.03-1.81, p = 0.041); AUC = 0.57 (95% CI 0.52-0.62)). Furthermore, external validation of the models was unsuccessful, limited by the small size and low number of cases (22/94) in our external dataset, WSI quality, as well as the lack of well-annotated datasets that allow robust validation. INTERPRETATION:Deep learning models using routinely processed WSIs hold promise for DCIS risk stratification, while the benefits of integrating clinical data merit further investigation. Obtaining a larger, high-quality external multicentre dataset would be highly valuable, as successful generalisation of these models could demonstrate their potential to reduce overtreatment in DCIS by enabling active surveillance for women at low risk. FUNDING:Cancer Research UK, the Dutch Cancer Society (KWF), and the Dutch Ministry of Health, Welfare and Sport.

This is an example of DDR-deficient case with high TILs and low gene-expression measurements. These cases were confirmed to have high TIL content (black delineation) and are characterized by both high tumour area- stromal area ratio as well as a high tumour cell- stromal cell ratio. Moreover, all these cases were characterized by high grade features, such as necrosis (blue delineation), high mitotic activity (green arrow), and high levels of atypia (blue arrow), and all had a solid growth pattern, with no formation of glands.

Phyllodes tumours (PTs) of the breast present diagnostic challenges due to their complex histological features and potential for malignant behaviour. The World Health Organisation (WHO) classification requires the presence of five adverse histological criteria to categorise PTs as malignant, aiming to avoid overdiagnosis and improve diagnostic consistency. However, emerging evidence suggests that these strict criteria may underdiagnose tumours with metastatic potential and histological features that would otherwise be considered malignant in soft tissue tumours, leading to significant implications for prognosis and treatment. Recent studies have highlighted cases where tumours classified as borderline PT by WHO criteria exhibited metastatic behaviour, emphasising the need to refine the diagnostic framework. Microscopic criteria used to classify PT also vary among reporting pathologists, resulting in suboptimal reproducibility. This review examines the histological parameters utilised in the classification of malignant PT, highlights existing evidence gaps and analyses international breast pathologist survey data to propose a pragmatic diagnostic approach. We recommend redefining malignant PTs to include cases meeting four of the five WHO criteria, supplemented by comprehensive sampling and clinical context. This approach balances the risk of underdiagnosis with the need for standardised, reproducible diagnostic practices. Future collaborative efforts should focus upon developing evidence-based, biologically relevant classification systems and leveraging technological advancements to enhance diagnostic precision. These efforts aim to refine classification, improve prognostic accuracy and optimise patient management strategies.

Assessment of axillary lymph nodes in breast cancer patients following neoadjuvant chemotherapy (NACT) is a crucial part of the clinical and pathological assessment of the disease and has prognostic and management implications. This, however, currently lacks standardisation and focuses only on the number of lymph nodes with metastases still present, the largest metastasis, and the presence of pathological complete response. Potential changes in any residual disease or within the lymph node parenchyma are not examined. Novel methods of more nuanced approaches are rare in the literature, even when considering multiple cancer types, but can offer an insight into the potential additional information to be gained and improvement in patient stratification. Given how common NACT is as the backbone of cancer therapy, there is a surprising lack of research into the lymph node response and determination of the biological factors driving what is seen histologically. Furthermore, with NACT now being administered alongside immunotherapy, there is an increasing need to understand the functional and architectural changes induced in the lymph nodes by metastatic tumour and systemic therapies. This review summarises current approaches, with breast cancer as an exemplar, and discusses the literature investigating a possible more granular approach to lymph node assessment after NACT. Translating these multiple carcinoma studies to breast cancer patients may prompt tissue-based research and, with clinical validation studies, changes to the reporting of lymph node response, for example percentage of viable tumour and immunological architectural features such as germinal centres. © 2025 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Introduction Mammographic screening identifies many women with small breast cancers with favourable biological features, which have an excellent prognosis. Some of these may never have become clinically apparent without screening and are commonly described as ‘overdiagnosed’ cancers. Despite this, all patients with screen-detected cancers are currently treated with surgical excision and sentinel lymph node biopsy, although this may represent overtreatment. There is, therefore, a need for less invasive approaches to reduce treatment burden for patients while maintaining current excellent oncological outcomes. Vacuum-assisted excision (VAE) may represent such an alternative treatment approach, and the SMALL (Open Surgery versus Minimally invasive-vacuum Assisted excision for smaLL screen-detected breast cancer) trial aims to investigate the use of VAE for the safe de-escalation of surgical treatment for such excellent prognosis invasive breast cancers.Methods SMALL is a prospective, multicentre, randomised phase III trial of VAE versus surgery in patients with small, biologically favourable screen-detected invasive breast cancer. SMALL has an innovative hybrid design with coprimary endpoints. These include a randomised non-inferiority comparison of surgical re-excision rates following initial treatment, and a single-arm analysis of local recurrence at 5 years following VAE. Secondary outcomes include complication rates, overall survival, quality of life and a health economic analysis. The trial includes a QuinteT Recruitment Intervention to support recruitment.Ethics and dissemination Ethical approval was obtained from the Office for Research Ethics (Northern Ireland) for all UK sites. Results will be submitted for publication in a peer-reviewed journal, presented, shared with patient partners and with relevant professional organisations to inform future guideline development for the management of screen-detected breast cancer.Trial registration number ISRCTN12240119.

Background: Triple-Negative Breast Cancer (TNBC) is a highly aggressive subtype of breast cancer and only ∼40% of TNBC patients achieve pathological complete response (pCR) following neoadjuvant chemotherapy (NACT). Since molecular heterogeneity of TNBC contributes to NACT response, we investigated their histology and spatial transcriptomics (ST) patterns. Methods: In total, 97 core biopsies, taken prior to (n = 92) or during (n = 5) treatment, and 44 post-treatment surgical resections were collected from 96 women with TNBC receiving NACT from archival samples and the ongoing FORCE (NCT03238144) clinical trial at Guy’s Hospital, London, UK. Each tissue section was H&E stained, imaged, and annotated by a pathologist. A 6-digit classifier was designed to capture the histology of each region of interest (ROI), characterising the tumour region, epithelial type and immune cell localisation, population, distribution, and abundance. The annotations were used to guide ST profiling of serial sections, using the GeoMx Digital Spatial Profiler. Tissue sections were stained with fluorescent antibodies against PanCK and CD45. Tumour and immune-enriched compartments, defined as PanCK+/CD45- and PanCK-/CD45+, respectively, were acquired. Data analysis was performed in R 4.3.1, including quality control, batch correction, differential and topographical gene expression, immune cell deconvolution and gene set enrichment analysis (GSEA). Epithelial states (ES) were defined by identifying gene modules of co-expressed genes and annotating these with enriched pathways. RESULTS: To date, we have performed ST profiling of 120 TNBCs from 85 patients (40 Non-Responders; 45 Responders), totalling 4506 tumour or immune-cell enriched regions. Genes signatures known to predict pCR of NACT in women with TNBC were first assessed by generating a pseudo-bulk dataset produced in silico by combining gene counts from all regions. Two gene signatures, namely the PAM50 proliferation signature and the GeparSixto Immune signature, were found to be predictive (PAM50: OR = 0.12, p-value = 0.06; GeparSixto: OR= 0.19, p-value = 0.03) in our data. Seven transcriptionally distinct epithelial states (ES), each underpinning unique molecular functions, were defined by identifying gene modules of tumour-enriched regions. The relative abundance of ES2 (DNA repair mechanisms) and ES4 (deregulated ECM, increased epithelial-mesenchymal transition and TGF-b signaling) in pre-NACT core biopsies was significantly associated with pCR (OR=1.03, p-value = 0.02) and residual disease (OR = 0.98, p-value = 0.05), respectively. Within the immune compartment, seven clusters of tumour-immune microenvironments (TIMEs), were obtained by unsupervised hierarchical clustering of immune cell proportions. A high relative abundance of TIME1 (naïve CD4 T cells and plasma cells) and TIME5 (naïve CD4 T cells, naïve B cells, memory B cells and plasma cells) were significantly associated with pCR (TIME1 OR = 1.05, p-value = 0.01; TIME5 OR = 1.04, p-value = 0.05). We further investigated rates of co-occurrence between each ES and TIME, identifying combinations that were statistically most probable to occur and identified TIME as a confounding factor in the relative risk associated with the presence of individual ES. Moreover, we identify potential ligand-receptor interactions between distinct ES and TIMEs which may provide new insights into the mechanisms driving differential responses to NACT in TNBC. Conclusion: To date, we have collected one of the largest spatial transcriptomics data sets with detailed histological annotation. The abundance of epithelial states and tumour-immune microenvironments is predictive for pCR in NACT TNBC. The spatial location, colocalization and ligand-receptor interactions in ES and TIME provide novel insights into the molecular heterogeneity of TNBCs and mechanisms regulating responses to NACT. Citation Format: Isobelle Wall, Jelmar Quist, Sarah Pinder, KCL Biobank, Sheeba Irshad, Cheryl Gillet, Victoria Seewaldt, David Frankhouser, Shankar Subramanium, Maddy Parsons, and Anita Grigoriadis. Spatial transcriptomics identifies molecular patterns predictive of response to neoadjuvant chemotherapy in triple-negative breast cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr PS17-01.

Background Cancer research emphasises early detection, yet quantitative methods for analysing normal tissue remain limited. Hematoxylin and eosin (H&E)-stained tissues in digitised whole slide images (WSIs) enable computational histopathology; however, artificial intelligence (AI)-based analyses of normal breast tissue (NBT) remain scarce. Methods We curated 70 WSIs of NBTs from multiple sources with pathologist-guided manual annotations of epithelium, stroma, and adipocytes, and developed robust convolutional neural network (CNN)-based, patch-level classification models, named NBT-Classifiers , to tessellate and classify NBTs at different scales. Data and code are available at and . Findings Across three external cohorts, NBT-Classifiers trained on 128 x 128 µm and 256 x 256 µm patches achieved AUCs of 0·98–1·00. Two explainable artificial-intelligence (AI)- visualisation techniques confirmed the biological relevance of tissue class predictions. An end-to-end WSI pre-processing framework was then integrated, capable of localising lobules and peri-lobular stroma. The outputs are compatible with the QuPath v0.3.0 platform and enable downstream digital image analysis, such as texture and nuclei morphology assessment, at the patch level. Interpretation NBT-Classifiers represent a robust, generalisable, end-to-end deep learning framework, which will enable broader application in studies of normal tissues, in the context of breast and breast cancer. Funding The Breast Cancer Research Trust, Breast Cancer Now (and their legacy charity Breakthrough Breast Cancer), the Medical Research Council (MRC) [MR/X012476/1], Cancer Research UK [CRUK/07/012, KCL-BCN-Q3], and CRUK City of London Centre Award [CTRQQR-2021/100004]. Evidence before this study We searched PubMed from Jan 1, 2013, to Mar 20, 2025, using the query (("normal breast" OR "terminal duct lobular unit" OR "TDLU" OR "acini") AND ("artificial intelligence" OR "deep learning" OR "convolutional neural network" OR "CNN" OR "machine learning" OR "automation" OR "classification" OR "segmentation" OR "detection")) AND (journal article[pt] NOT review[pt]) NOT ("computed tomography" OR "CT" OR "ultrasound" OR "sonography" OR "US" OR "mammography" OR "mammogram" OR "Raman" OR "single-cell" OR "MRI"). This yielded 264 initial hits and after a thorough manual review, only eight studies were found to be relevant for using deep learning or machine learning to automate tissue recognition or quantification of normal breast tissues (NBTs) on digitised histological images. All these methods are highly specialised for detecting and quantifying lobules or quantifying tissue composition, and none of them are specifically tailored for providing patch-level tissue classifications on WSIs of NBTs. Moreover, NBTs are heavily underrepresented in large annotated WSI databases. A recent literature review summarised publicly available breast hematoxylin and eosin (H&E) WSI datasets from 2015 to 2023, identifying 17 datasets comprising a total of 10,385 breast H&E WSIs. Of these, two datasets contain female normal breast WSIs with pathology-guided manual annotations. Added value of this study Motivated by the research gap, we collected a comprehensive WSI dataset that captures real-world variability present across NBTs and provided by expert ground-truth manual annotations. Based on extensive cross-validation and external testing, we developed robust patch-level classification models to classify three major tissue compartments within NBTs, at different scales. Implications of all the available evidence With high AUCs of 0·98–1·00 across three external cohorts, our artificial-intelligence (AI) tool can be used to automatically classify tissue compartments and localise lobular regions for downstream normal breast research. These approaches have the potential to enhance our understanding of how various NBT components contribute to both benign and malignant breast pathology and lay the groundwork for the development of more advanced deep learning models and spatial-defined molecular large-scale analyses in the future. ### Competing Interest Statement Anita Grigoriadis, Louise J. Jones and Greg Verghese are Co-Founders of PharosAI. Salim Arslan and Pahini Pandya and employed by Panakeia Technology, UK. All other authors declare no conflict of interest. WSIs involved in this study are stored at the OASIS repository: , which currently can be accessed upon request. The Breast Cancer Research TrustThe Breast Cancer Research Trust, , Breast Cancer NowBreast Cancer Now, , the Medical Research Council (MRC)the Medical Research Council (MRC), , MR/X012476/1 Cancer Research UKCancer Research UK, , CRUK/07/012, KCL-BCN-Q3 CRUK City of London Centre AwardCRUK City of London Centre Award, , CTRQQR-2021/100004

e15162 Background: Invasive lobular carcinoma (ILC) is associated with a unique pattern of dissemination and higher risk of late relapse ( > 10 years) compared with other invasive breast cancer types. However, the predictors of relapse in ILC are largely based on clinicopathological data. In numerous cancers, the tumor microenvironment (TME) is thought to play a significant role in the ability for tumor cells to spread. Several studies have shown that changes in the TME can assist in tumor dissemination and metastasis. However, there is a paucity in literature exploring the potential role of the TME in predicting clinical outcomes. Human replacement therapy (HRT) exposure is a known risk factor for ILC. Yet, the mechanisms by which HRT influences the TME and how these changes may predispose patients to ILC remain unclear. We sought to explore the association between tumor extracellular matrix (ECM), recurrence, and HRT exposure in ILC. Methods: Sections from a tissue microarray, constructed from 115 Formalin-Fixed Paraffin Embedded ILC patients samples from the GLACIER study cohort (REC: 18/SW/0052), were stained with PicroSirius Red solution and Weigert's Iron Hematoxylin. Tumor cell density (TCD) of the cores was extracted using QuPath. High density matrix (HDM) values were calculated as the proportion of image pixels corresponding to the matrix, using the ImageJ plugin ‘The Workflow Of Matrix BioLogy Informatics’ (TWOMBLI) (Wershof et al.). Logistic regression was used for both univariate and multivariate recurrence analysis. Covariates were included based on clinical relevance and previous literature. Cox Proportional Hazards model was employed to assess relapse free survival (RFS). Results: A total of 562 cores from 115 patients with ILC were included in the analysis (Invasive = 439, Normal = 80, lobular carcinoma in situ = 43). Among low TCD cores, the median HDM was higher in recurrence cases (0.78 vs. 0.75, p = 0.044), regression analysis yielded an odds ratio (OR) 3.89 [95%CI (1.26-7.50), (p = 0.018)]. When adjusting for stage and HER2 status, HDM had an even greater OR of 3.03 [95%CI (1.33-7.83), (p = 0.0012)]. There was also a borderline association with shorter RFS and higher HDM, hazard ratio (HR) 1.82 [95%CI (0.97- 3.43), (p = 0.064)] which became significant when adjusting for age, stage, HER2 status, ER status and histological subtype, HR 2.03 [95%CI (1.03-4.03), (p = 0.042). These associations with recurrence were not observed in high TCD cores (0.65 vs 0.67, p = 0.8). Among the low TCD cores, there was no statistical difference in HDM between women exposed to HRT compared to those without (0.75 vs 0.76) (p = 0.9). Nor was there a difference in the high TCD cores (0.66 vs 0.69, p = 0.5). Conclusions: This retrospective analysis of ILC samples reveals the potential role of stromal TME in ILC risk prognostication. Spatial profiling of these samples will be performed to explore these findings further.

A major challenge in treating Triple-Negative Breast Cancer (TNBC) lies in its molecular, morphological and clinical heterogeneity, which hampers accurate prediction of responses to neoadjuvant treatment. To address this, we introduce SMART: Spatio-Molecular Atlas of Response Trajectories, a comprehensive, multimodal resource compiled from 129 TNBC samples across 89 patients, obtained before, during, and after neoadjuvant chemotherapy (NACT). SMART comprises of 5,096 high quality manually selected spatial transcriptomic profiles enriched for epithelial, immune, or stromal compartments; paralleled with histological annotations, imagebased network analysis and protein expression. Seven novel spatial epithelial archetypes (EAs), seven tumour-immune microenvironments (TIMEs) and their co-localisation patterns were defined, revealing an opposing prevalence of functionally divergent EAs between response groups and the prognostic significance of B-cell enriched TIMEs, in particular those surrounding histologically normal epithelium adjacent to the tumour. The SMART dataset and analytical tools are publicly available via the PharosAI platform, providing the research community with the most comprehensive, manually annotated spatio-molecular transcriptomics atlas of NACT-treated TNBC to date. ### Competing Interest Statement The authors have declared no competing interest. CRUK City of London Centre, CTRQQR-2021/100004 NIHR, NIHR30340 Pathological Society of Great Britain and Ireland, JSPS CPF 1023 05 Jean Shanks Foundation, JSPS CPF 1023 05

Ductal carcinoma in situ (DCIS) is a potential precursor to invasive breast cancer (IBC). The trajectory of an individual’s DCIS, if it will progress to IBC or remain as DCIS is difficult to predict. Currently >80% of DCIS is detected through mammographic screening of breast calcifications. Despite the close association of calcifications with DCIS, their role in the development of DCIS and/or its progression to IBC remains largely unexplored. In this study, we present a spectroscopy based analytical approach to probe chemical compositional changes of both DCIS associated breast calcifications and surrounding soft tissue aiming to identify a cohort at increased risk for invasive progression. Tissue samples from 316 DCIS patients without invasive cancer were obtained from multiple centres as part of the PRECISION consortium (The Netherlands, UK and USA). Three consecutive tissue sections were obtained each of the tissue biopsies. Mid-infrared (mid-IR) and Raman hyperspectral imaging was performed independently on two sections that were left unstained. The third H&E-stained section was used to annotate calcifications and histopathological features. All DCIS samples had known outcome (i) ‘pure DCIS as controls’ (DCIS without progression to invasion) (n=193), (ii) ‘DCIS with progression to invasion as cases’ (DCIS from patients who subsequently developed invasive disease after initial treatment) (n=123). Spectral features of DCIS calcifications and surrounding soft tissue were used as inputs for analysis. Data was divided into a discovery and a validation set. Cluster analysis followed by Principal component analysis fed linear discriminant analysis was carried out on the discovery set to develop DCIS prediction models. For the Raman data, a mean area under the receiver operating characteristic curve (AUROC) value of 0.85 was obtained using calcification spectral features, and 0.75 using soft tissue spectral features in distinguishing controls from cases (N=118 vs 52). Similar analysis on the IR data showed a mean AUROC value of 0.68 for calcification, 0.78 for epithelial and 0.80 for stromal components (N=97 vs 61). Preliminary analysis shows changes in phosphate to carbonate ratio and variations in magnesium whitlockite content in calcifications, and protein secondary structural changes in soft tissue, between the two groups. The models will be tested independently on the validation set and the outcomes will be presented at the AACR conference. Spectroscopic chemical analysis of breast calcifications and soft tissue show promise in predicting the likely progression of DCIS to IBC. Pending further independent validation, these techniques appear to be novel image-based risk assessment tools that can potentially be utilised to inform DCIS prognosis and treatment options. Jayakrupakar Nallala, Doriana Calabrese, Sarah Gosling, Esther Lips, Ihssane Bouybayoune, Rachel Factor, Sarah Pinder, Lorraine King, Jeffrey Marks, Thomas Lynch, Donna Pinto, Alastair Thompson, Elinor Sawyer, Jelle Wesseling, Shelley Hwang, Keith Rogers, Nick Stone, Grand Challenge PRECISION consortium. Predicting the prognosis of ductal carcinoma in situ through chemical analysis of breast microcalcifications and soft tissue using infrared and Raman spectroscopy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3353.