Abstract Platinum-based chemotherapy remains a cornerstone of treatment for triple-negative breast cancer (TNBC), yet the molecular determinants governing platinum response remain poorly defined. By leveraging the randomized Phase II INFORM trial, which compared neoadjuvant cisplatin to anthracycline-based therapy in BRCA1/2 -mutant breast cancer—we identified miR-362-3p as a specific regulator of cisplatin sensitivity. Higher plasma miR-362-3p expression was exclusively associated with favorable clinical outcome in the cisplatin arm, with no association observed in the AC arm, decoupling platinum-specific vulnerability from general chemotherapy response. We used gain- and loss-of-function TNBC models to establish that miR-362-3p functions as a potent sensitizer to cisplatin in vitro and in vivo . Integrated TCGA analysis and experimental validation identified BCLAF1, a key regulator of DNA damage response, as a direct repression target of miR-362-3p. We uncovered a novel role for the miR-362-3p/ BCLAF1 axis in overcoming platinum resistance in TNBC.

BACKGROUND:Tumor-infiltrating lymphocytes (TILs), assessed by visual examination, are prognostic and predictive in early-stage triple-negative breast cancer. Computational assessment may provide a complementary approach. We evaluated the prognostic value of TILs by visual examination and computational assessment. METHODS:Cisplatin vs Paclitaxel for Triple Negative Breast Cancer (TBCRC030; ClinicalTrials.gov identifier NCT01982448) was a randomized phase 2 trial enrolling patients with BRCA1/2-proficient stage I to III triple-negative breast cancer to receive preoperative cisplatin or paclitaxel. The primary endpoint was pathological response at surgery. The TILs were visually scored on digitized pretreatment biopsies per International TILS Working Group recommendations. Computational assessment used the 4D Path QPOR platform to generate TILs, an immune heterogeneity index, and a combined immune/cell cycle biomarker (CmbI). Predictive performance for residual cancer burden 0/1 was assessed using receiver operating characteristic curves and odds ratios (ORs) with 95% CIs; all statistical tests were 2-sided. RESULTS:Of 139 response-evaluable patients, 121 had matched visual examination and computational assessment data (59 on cisplatin, 62 on paclitaxel). Median visual examination TILs were higher in responders (40.0% vs. 10.0%; P = .002) and predicted response (OR = 1.86, 95% CI = 1.24 to 2.87; area under the curve = 0.69, 95% CI = 0.57 to 0.80). Computational assessment CmbI differed by response group and predicted residual cancer burden 0/1 (OR = 3.20, 95% CI = 1.05 to 11.07; area under the curve = 0.62, 95% CI = 0.51 to 0.73). Computational assessment TILs and immune heterogeneity index were not predictive. Visual examination TILs and computational assessment CmbI predicted response to paclitaxel (OR = 2.91, 95% CI = 1.56 to 6.14; OR = 9.17, 95% CI = 2.01 to 66.39, respectively) but not to cisplatin. CONCLUSION:Visual examination TILs and computational assessment CmbI were each associated with response to neoadjuvant chemotherapy in triple-negative breast cancer in the overall cohort and the paclitaxel arm. Computational assessment CmbI did not outperform visual assessment. Further validation is needed before clinical implementation of computational approaches.

HER2-low expression is associated with hormone receptor (HR) expression in HR-positive breast cancer. We aimed to evaluate its association with androgen receptor (AR) among 196 patients with metastatic triple-negative breast cancer (mTNBC). Central determination of AR showed significant enrichment in HER2-low compared with HER2-0 mTNBC (mean: 33.7% vs. 21.4%, p = 0.038), whereas no significant immunological differences were observed. HER2-low/AR-positive patients trended towards longer overall survival, highlighting the potential relevance of these biomarkers.

1024 Background: Invasive lobular carcinoma (ILC) is a distinct breast cancer (BC) subtype with dissemination patterns differing from invasive breast carcinoma of no special type (IBC-NST). While genomic differences between ILC and IBC-NST have been described, the extent to which molecular and immune profiles vary by metastatic organ and histology is poorly characterized. Methods: We conducted a retrospective analysis to evaluate organ- and histology-specific molecular features of metastatic BC (mBC). Patients with metastatic IBC-NST or pure ILC underwent NGS (592, NextSeq; WES/WTS, NovaSeq; Caris Life Sciences). Tumor mutational burden (high ≥10 mut/Mb), PD-L1 expression (22C3), and immune cell fractions (RNAseq deconvolution, quanTIseq) were assessed. Analyses were restricted to sites with ≥10 biopsies per histology. Comparisons used chi-square or Mann–Whitney U tests with multiple testing correction (q < 0.05). Results: A total of 2645 mBC biopsies (605 ILC; 2040 IBC-NST) were analyzed (Table). In ILC, ERBB2 mut frequency differed by metastatic site (q<0.01), with higher prevalence in liver (32.7%) and lower in gastrointestinal (GI) lesions (3.7%). In IBC-NST, ESR1 and GATA3 mut and FGFR1 amplification were enriched in liver metastases, whereas PD-L1 expression was highest in skin (20.7%) and lowest in liver (5.0%). No organ-specific differences were observed for PIK3CA, AKT, PTEN, BRCA1/2 or RB1 . Across metastatic sites, both ILC and IBC-NST showed differences in B cells, macrophages (M1/M2), neutrophils, NK cells, dendritic cells, CD8+ T cells, and Tregs (all q < 0.05); CD4+ T-cell differences were observed only in IBC-NST (q = 0.002). In organ-specific ILC vs IBC-NST comparisons, CDH1 mut were more frequent in ILC across sites, along with higher PIK3CA mut in skin (ILC 51.3% vs IBC-NST 34.4%) and higher ERBB2 mut in liver (32.7% vs 3.3%), whereas a higher TP53 mut frequency was observed in IBC-NST in skin (28.5% vs 52.3%) and lymph nodes (LND) (23.5% vs 55.5%) (all q < 0.01). Conclusions: mBC exhibits marked organ-specific molecular and immune heterogeneity that differs by histology. These findings support histology- and site-aware interpretation of metastatic biopsies and may inform biomarker assessment and treatment decisions in advanced disease. BC-NST ILC Skin LND Bone Breast Liver Perit. CNS q-value Skin LND Bone Breast GI Liver GYN Perit. CNS q-value N (specimens) 785 401 109 361 220 15 44 - 172 84 62 56 56 54 43 37 11 - ERBB2 mut (%) 3.2 3.1 3.2 2.3 3.3 7.1 2.5 1.0 7.1 7.4 16.4 15.4 3.7 32.7 2.4 5.7 11.1 <0.01 ESR1 mut (%) 9.4 5.2 8.3 16.0 28.2 7.1 2.4 0.0 7.8 7.2 8.9 9.8 16.7 26.0 4.9 20.0 0.0 0.9 FGFR1 amp (%) 10.1 6.6 5.6 12.8 16.5 0.0 9.7 0.02 4.4 3.4 6.7 6.1 2.4 5.6 7.7 18.5 0.0 1.0 IHC-PD-L1 (%) 20.7 - 5.3 16.0 5.0 11.1 10.3 0.02 8.9 - 5.0 5.7 6.1 0.0 0.0 0.0 14.3 1.0 TMB-High (%) 10.7 10.1 8.8 8.8 8.5 14.3 21.4 0.8 27.6 11.0 19.6 28.0 11.3 19.6 9.8 32.4 22.2 0.8 Perit: peritoneum; CNS: central nervous system; GYN: genital tract.

Tumors presumed to be metastatic triple-negative breast cancer (TNBC) may occasionally represent misdiagnoses of primary or metastatic tumors from other origins—particularly lung—due to overlapping clinical characteristics and lack of organ-specific morphological features. While immunohistochemical markers can resolve these discrepancies, they are often underutilized without prior clinical suspicion, leading to significant treatment delays. To address this, we employed GPSai, a tissue-of-origin artificial intelligence (AI) model integrated into routine molecular profiling at Caris Life Sciences, to systematically identify differential diagnoses among 2423 presumed TNBC cases. By integrating AI results with clinical and molecular evidence, we identified a misdiagnosis rate of 3.0%. While over half of these cases were reclassified as non-small cell lung cancer, the misdiagnosis cohort also included various other tumor types. These diagnostic shifts profoundly impact staging, prognosis, and therapeutic selection, highlighting the necessity of an unbiased, systematic approach to identifying misdiagnoses among presumed TNBC.

1026 Background: Digital pathology (DP), including artificial intelligence (AI)-based image analysis methods, enables IHC biomarker assessment on a continuous scale, potentially improving precision and accuracy versus conventional IHC scoring. DP may also capture features relevant to ADC response that traditional methods miss. As proof of concept, we developed and evaluated an AI-assisted DP image analysis pipeline to explore the association between target antigen expression and clinical outcomes of sac-TMT, a TROP2-directed ADC with a unique bifunctional linker, in TNBC. Methods: We analyzed whole-slide images of IHC-stained tumor samples from participants with TNBC enrolled in the phase 1/2 MK-2870-001 study (NCT04152499) evaluating sac-TMT in pretreated advanced solid tumors. We established a set of prespecified human-interpretable features (HIFs), including IHC signal intensity, subcellular localization, and spatial patterns of cell and signal distribution. To mitigate overfitting, we prioritized a subset of 35 HIFs based on correlation structure and biological hypotheses, and in a blinded fashion, assessed their association with clinical outcomes (BOR; PFS) in a development cohort (DC) (n = 58). After unblinding clinical outcome data, additional HIFs and a multivariate model trained to predict clinical outcomes in the DC were prioritized for validation. Both sets of HIFs (identified from blinded and unblinded analyses) were assessed for their relationship to clinical outcomes in an independent validation cohort (IVC) from the same study (n = 34). In both cohorts, DP HIFs were compared with conventional TROP2 H-scores (on paired slides) for prediction of clinical outcome using the area under the receiver operating characteristic curve (AUROC) and Harrell C-index. Results: The blinded approach identified a HIF that was positively associated with BOR (multiplicity-adjusted P = 0.023), with an AUROC higher than TROP2 H-scores (0.76 vs 0.70). After unblinding, 5 additional HIFs and a multivariate model were selected for their associations with BOR (AUROC, 0.70-0.80). In the IVC, the 6 prioritized HIFs and 1 multivariate model were associated with response to sac-TMT (AUROC, 0.60-0.69); each outperformed TROP2 H-scores with respect to association with BOR (H-score AUROC, 0.57) and demonstrated incrementally better association with PFS than TROP2 H-scores (Harrell C-index, 0.64-0.66 vs 0.62). Conclusions: In this proof-of-concept study, DP-derived HIFs were associated with response to sac-TMT in TNBC and showed incrementally better nominal performance than conventional TROP2 H-scores. While the sample size was small, data from this study suggest that DP-based image analysis can identify novel biomarkers of response to sac-TMT in TNBC.

Abstract Background: PARP inhibitors (PARPi) induce synthetic lethality in BRCA1/2-mutant (BRCA-MUT) tumors and can activate DNA-damage-linked immune pathways. The TALAVE study (NCT03964532) examined the combination of the PARPi (talazoparib) with PD-L1 blockade (avelumab). Here we evaluated spatially resolved immune signaling and remodeling in response to talazoparib alone and with avelumab. Methods: 24 patients with advanced HER2-negative breast cancer (12 BRCA-MUT, 12 BRCA-WT) received talazoparib then talazoparib + avelumab. Serial biopsies (baseline, post-PARPi [BX2], post-combination [BX3]) underwent transcriptomic, spatial protein, and multiplex IF. BRCA-dependent TME remodeling and cellular neighborhood (CN) shifts were assessed. Results: BRCA-MUT tumors showed 83% objective response and 100% clinical benefit, whereas BRCA-WT tumors exhibited minimal activity. BRCA-MUT tumors became fragmented with increased immune activity, while BRCA-WT tumors remained compact and immunosuppressed. γH2AX and pTBK1 were spatially co-expressed and sustained in BRCA-MUT tumors during treatment but declined in BRCA-WT tumors. BRCA-MUT tumors displayed enrichment of CD8+ T cells and CD163+ macrophages after PARPi, whereas CD4+ T cells and CD68+CD163+ macrophages were depleted in BRCA-WT tumors. PD-1+ CD8+ T cells were strongly linked to local CD4+ T cell density, and PD-1+ frequency in CD8+ T cells correlated with longer PFS at baseline and BX3 in BRCA-MUT but not BRCA-WT tumors.CN analysis revealed CD4+ and CD8+ enriched neighborhoods with intermediate PD-1 expression that expanded after therapy in BRCA-MUT but not BRCA-WT tumors, and these CNs lacked PD-L1+ cells. Between BX2 and BX3, BRCA-MUT tumors sustained immune activity but showed no further T cell activation or infiltration. Spatial mapping identified three PD-L1+ CN types: (1) T cell-dense niches with high PD-1/PD-L1 and γH2AX-pTBK1 activity enriched only at baseline, largely lost after PARPi; (2) macrophage-T cell mixtures that were depleted during therapy; and (3) PD-L1+ dying tumor cells lacking pTBK1 activity and T cell engagement. Across contexts, PD-L1 was either lost before PD-L1 blockade or confined to regions isolated from T cells, leaving little opportunity to reinvigorate T cells. Conclusions: PARP inhibition reshaped the TME of BRCA-MUT tumors by inducing tumor fragmentation, sustaining γH2AX-pTBK1 signaling, and restoring CD4+ and CD163+ immune cells, whereas BRCA-WT tumors remained structurally intact and immunosuppressed. PD-1+ T cells localized to PD-L1-negative neighborhoods, and PD-L1+ tumor/myeloid cells were rapidly lost or confined to dying, immune-excluded regions, limiting the impact of PD-L1 blockade. Although PARPi re-engaged T cell programs in BRCA-MUT tumors, strategies beyond PD-L1 inhibition will be required to further enhance T cell infiltration and activation. Citation Format: Kenichi Shimada, Filipa Lynce, Claudine Isaacs, Xue Geng, Edward T. Richardson, Candace Mainor, Mei Wei, Julie M. Collins, Paula R. Pohlmann, Arielle L. Heeke, Kelly F. Zheng, Madeline Townsend, Lauren M. Sloat, Jane Staunton, Stuart J. Schnitt, Hongkun Wang, Joan S. Brugge, Geoffrey I. Shapiro, Jennifer L. Guerriero. Spatially resolved multi-omic profiling reveals BRCA-dependent immune remodeling during PARP inhibition and PD-L1 blockade [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3434.

The 6th edition of the WHO Classification of Breast Tumours introduces both major and minor changes based on recent advances in our understanding of breast biology, developments in diagnostic modalities, identification of specific molecular targets and new treatment regimens necessitating modifications to pathology reporting and tumour biomarker categorisation. This review summarises the main changes that strive towards a classification of global relevance. In invasive carcinoma, predictive factors increasingly inform modern breast cancer treatment. The 6th edition provides an update on HER2 reporting categories following the DESTINY-Breast 04 and 06 trials. Terminologies used to classify invasive tumours are clarified, with the term 'variant' now reserved for molecular/genetic alterations. Invasive lobular carcinoma (ILC) with extracellular mucin is recognised as a new diagnostic entity with prognostic implications. The diagnosis of mucinous carcinoma (MC) is reserved for mucin secreting carcinomas with grade 1 or 2 morphology and a favourable biomarker profile. The diagnosis of malignant phyllodes tumours requires only four of the original five adverse histological criteria. Classification of neuroendocrine tumours (NETs) is revised, recognising that the unified model, promoted in the 5th edition, is difficult to apply to the breast. New approaches to the classification of adenomyoepithelioma are discussed but the 5th edition system is broadly retained. A new section on 'Small Diagnostic Samples' outlines the merits of non-operative biopsy diagnosis, the B coding system and the importance of multidisciplinary review. Changes to diagnostic practice and the emerging role of artificial intelligence, with advantages and challenges, are discussed in a new section on 'Digital Pathology'.

Abstract The INFORM trial (NCT01670500) was a randomized, two-arm Phase II neoadjuvant study comparing the efficacy of a platinum-based regimen (cisplatin) versus an anthracycline (AC)-based regimen in participants with germline BRCA1/2 mutations and early-stage, HER2-negative breast cancer. MicroRNAs (miRNAs) are emerging as promising non-invasive biomarkers for disease detection and treatment monitoring. The initial aim of this study was to evaluate whether pretreatment plasma miRNA profiles were associated with treatment outcomes in the INFORM trial cohort. Pretreatment plasma samples from 97 INFORM participants (n=53 cisplatin and n=44 AC) were screened for 352 miRNAs commonly implicated in cancer using the qPCR-based ID3EALTM Cancer Panel. Twenty out of 53 participants achieved residual cancer burden (RCB) score of 0 or 1 with cisplatin and 20/44 with AC. Higher plasma miR-362-3p expression was associated with a favorable response to cisplatin (1.7-fold; p<0.01), but not to AC. MiR-362-3p expression in paired pretreatment tumor biopsies (n=79) did not significantly correlate with plasma expression and was not associated with RCB 0/1 (p>0.05). In TCGA, miR-362-3p expression was higher in breast tumors than in adjacent normal tissue, and higher in triple negative breast cancers (TNBCs) compared with hormone receptor-positive tumors (both FDR<0.001), independent of BRCA mutation or Single Base Substitution Signature 3 (SBS3) status. The miR-362-3p findings in INFORM did not replicate in a sister trial, TBCRC 030 (NCT01982448; 11 out of 46 cisplatin-treated patients achieved RCB 0/1, p=0.46). These results suggest that while miR-362-3p is unsuitable as a circulating biomarker for cisplatin response, it may play an important biological role in mediating cisplatin sensitivity in TNBCs, independent of BRCA1/2 mutations. Functional studies demonstrated that miR-362-3p is expressed and secreted by TNBC cell lines, operates as a tumor suppressor, and generally correlates with cisplatin sensitivity. Overexpression of miR-362-3p rendered MDA-MB-231 and CAL-51 cells cisplatin sensitive, while knockdown induced resistance in MDA-MB-436 cells. Ongoing studies are testing whether these effects translate in xenograft models. In silico and functional assays identified BCLAF1, a DNA damage response (DDR) regulator, as a direct target of miR-362-3p. Overexpression of miR-362-3p suppresses BCLAF1 mRNA and protein levels. Our findings demonstrate that miR-362-3p enhances cisplatin responsiveness by targeting BCLAF1. A miR-362-3p-based therapeutic may be useful as a co-agent to enhance tumor responses to platinum-based chemotherapy. Citation Format: Zhaoji Liu, Shizhong Ke, Xiaohui Li, Catherine Wu, Madison M. Uyemura, Brian R. Sardella, Erica S. Massicott, Lin Wang, Emily K. Aronson, Dimitra Karagkouni, Nikolas Kalavros, Ioannis S. Vlachos, Felipe Batalini, Cristina S. Bogsan, Jit Kong Cheong, Lihan Zhou, He Cheng, Phillip Munson, Erica L. Mayer, Judy E. Garber, Stuart J. Schnitt, Nadine M. Tung, Andrea L. Kasinski, Frank J. Slack, Gerburg M. Wulf, Yujing J. Heng. MiR-362-3p enhances platinum response in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7903.

TPS656 Background: Male breast cancer is a rare disease, and most cases are hormone receptor-positive (HR+). Due to a lack of clinical trials, male breast cancer has historically been treated based on data extrapolated from women. However, there are substantial knowledge gaps in the comparative efficacy, safety, and patient-reported outcomes of endocrine therapies for men. While tamoxifen is the current standard of care, additional data are warranted for aromatase inhibitors (AI), gonadal suppression, and cyclin-dependent kinase 4/6 (CDK4/6) inhibitors. Methods: This is an open-label, multicenter, randomized, phase II trial designed to evaluate different endocrine therapies in men with HR+ and human epidermal growth factor receptor 2 (HER2)-negative breast cancer. A total of 60 men will be enrolled across 9 sites within the Translational Breast Cancer Research Consortium (TBCRC). Key eligibility criteria include male sex and stage I, II, or III HR+/HER2- breast cancer before surgical resection of the primary tumor and axillary nodes. Key exclusion criteria include prior anti-cancer therapy within the past 12 months, and inflammatory breast cancer. The trial has two phases. The first phase is a window of opportunity in which newly diagnosed men are randomized 1:1:1 to either tamoxifen (Arm A), anastrozole (Arm B), or anastrozole plus degarelix (Arm C) given for 3 weeks. The primary endpoint for the window phase is Ki-67 reduction from the baseline diagnostic biopsy to the research biopsy at the end of the window phase. The second phase consists of neoadjuvant treatment, in which the tamoxifen group is randomized 1:1 to tamoxifen (Arm D) vs tamoxifen plus abemaciclib (Arm E), and the anastrozole alone (Arm B) and anastrozole plus degarelix (Arm C) groups are merged and then randomized 1:1 to anastrozole plus degarelix (Arm F) vs anastrozole plus degarelix plus abemaciclib (Arm G). The duration of the neoadjuvant phase is 4 months, and the primary endpoint of this phase is residual cancer burden (RCB) index at time of surgery. The trial is powered for the RCB endpoint in a 2 x 2 factorial design to detect a 0.6 unit decrease in RCB index with 80% power and alpha = 10%. For the Ki-67 endpoint, we assume an approximately 50% reduction in Arms A and B; Arm C will be of interest if it leads to ≥80% reduction in Ki-67. Secondary endpoints include: change in estradiol and testosterone levels from baseline, preoperative endocrine prognostic index (PEPI) score at surgery, adverse events, and patient-reported outcomes (including quality of life). Tumor tissue will be collected for correlative analyses. The study opened at Dana-Farber in October 2023, and is also open at Mayo Clinic, MD Anderson Cancer Center, Georgetown University, University of North Carolina, University of Pennsylvania, University of Pittsburgh, and Vanderbilt University. One more site will open in 2026. Clinical trial information: NCT05501704 .

Context.—:The Nottingham grading system, developed by Elston and Ellis, is the recommended method for grading invasive breast carcinoma. A previous study demonstrated the mean concordance for 35 breast pathologists in classifying 58 images as glandular (acinar)/tubule formation (G/TF) based on the World Health Organization definition was only 64%. Objective.—:To determine if an expanded description of G/TF according to the original definition and current use of the Nottingham grading system would improve recognition of G/TF among breast pathologists and pathologists in training. Design.—:Fifty-eight images with one structure circled were classified as G/TF or non-G/TF by Dr Ian Ellis. Images were sent as a PowerPoint (Microsoft) file to the breast pathologists who participated in the original study and to 21 trainees. Participants were asked to classify the structures based on the expanded description and were also provided with the 58 images from the first study with annotation. Results.—:Among the participating 28 of the original 35 breast pathologists, the mean concordance increased from 64% (range, 40%-97%) to 94% (range, 86%-100%). Trainees had a mean concordance of 90% (range, 52%-100%). Conclusions.—:The expanded description assisted in the recognition of G/TF for both breast pathologists and trainees. The most important impact on grading will likely be for carcinomas with complex cribriform patterns or micropapillary patterns with "inverted tubules." Participants endorsed that the expanded description of G/TF and the annotated images would be helpful reference material for pathologists.

Background: Increased physical activity (PA) is associated with lower risk of all subtypes of breast cancer in both pre- and post-menopausal women, but the biological mechanisms through which PA impacts breast carcinogenesis are not well understood. Methods: We conducted a single-arm pilot study to evaluate the impact of an aerobic and strength training PA intervention on tissue biomarkers in inactive women at increased risk of breast cancer due to the presence of increased mammographic breast density (MBD). Eligibility criteria included having at least 50% MBD (BIRADS C or D); engaging in <90 minutes per week of moderate or vigorous PA (MVPA); age <60 years; no concurrent use of hormone replacement therapy, oral/implantable birth control agents, or chemoprevention for breast cancer; and no prior history of breast cancer. Participants (pts) took part in a 12-week aerobic and strength-training intervention including 2 supervised exercise sessions and 1+ session of home-based aerobic exercise. Pts underwent biopsy of benign breast tissue at baseline and 12-weeks of mirror-image sites in opposite breasts. Biopsies were timed to coincide with the luteal phase of the menstrual cycle in pre-menopausal participants. Areas of interest (AOIs) were marked from terminal ductal lobular units (TDLU’s), ductal epithelium, and stroma by a breast pathologist and by staining for pan-keratin and CD45. RNA-sequencing of the AOIs was performed using GeoMx platform. Changes over time between pre- and post-intervention biopsies were evaluated using paired t-tests on average (per patient) immune cell type abundances that were estimated from gene expressions using the GeomxTools R package. All p-values were adjusted for multiple testing using the Benjamini-Hochberg procedure. Single cell spatial analysis targeting 20 immune and proliferative markers was performed using Co-detection by indexing (CODEX), a multiplexed imaging method, in 11 paired biopsies with sufficient tissue remaining after gene expression analysis. Analysis included cell segmentation and clustering with deconvolution of cellular populations followed by comparisons between matched samples pre and post exercise. Comparisons differences in number and density of cellular populations, cell-cell interactions and cell neighborhoods. Results: Thirty pts were enrolled between 10/2018 and 11/2019. Due to the COVID-19 pandemic, 3 participants did not undergo a week-12 biopsy and 1 additional patient was lost to follow up for non-COVID related reasons, leaving 26 participants with paired pre- and post-intervention tissue. Average age of study pts was 47.6 (±5.7) years, average BMI was 29.2 (±6.3) kg/m2, and median minutes of MVPA at baseline was 72.5 (range 0-360). Pts attended an average of 20.4 of 24 exercise training sessions (85% adherence) and increased exercise by a median of 92.5 minutes/week (range -140 to 700). Cell abundance based on gene expression data showed increased CD8+ memory T cells (mean difference (MD)=1.34, standard error (SE)=0.85), neutrophils (MD=2.25, SE=0.83), and NK cells (MD=1.18,SE=0.57) in the epithelium (N=11 patients) and increased neutrophils (MD=1.00, SE=0.33) in the stroma (N=10 patients) after the 12-week exercise intervention at a 25% FDR threshold. Details of the CODEX analysis will also be presented. Conclusions: In this single-arm pilot study exploring the impact of an exercise intervention on biomarkers in benign breast tissue in women at increased risk of breast cancer due to elevated MBD, there was evidence of immune changes in post-exercise tissue samples, providing some of the first evidence that exercise could impact gene expression in benign breast tissue. This work provides a foundation for further work exploring the potential pathways through which exercise could prevent breast cancer in at-risk populations. Citation Format: Jennifer Ligibel, Douglas Russo, Kun Huang, Anita Giobbie-Hurder, Anna Tanasijevic, Tari A. King, Judy Garber, Myles Brown, Stuart Schnitt, Kornelia Polyak, Rinath Jeselsohn. Impact of Exercise on Immune Biomarkers in Benign Breast Tissue from Women with High Mammographic Breast Density (MBD) [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 P2-01-20.

Objective: HER2 expression is a key prognostic and treatment-influencing factor in breast cancer and is assessed for all invasive breast carcinoma (BC). As with all immunohistochemistry (IHC) staining, visual interpretation of HER2 expression is subjective, which leads to intra- and inter-pathologist variability. This study aims to evaluate the clinical utility (concordance, accuracy, and user feedback) of artificial intelligence (AI)-aided HER2 scoring solution on whole slide digital images of HER2 IHCs of breast samples. Methods: The cohort included biopsies and excisions from 1,997 patients from 12 US, EU, and UK clinical laboratories, including academic medical centers and reference/private laboratories. HER2 slides of diverse BC subtypes from primary and metastatic tumors were stained with anti-HER2 antibody (4B5, VENTANA) at each laboratory and scanned with different scanners (Leica GT450DX, Philips UFS, Aperio AT2). This observational two-arm multi-reader study compared the performance of 26 pathologists (“readers”) on HER2 scoring (each reviewed 50-200 slides) unassisted vs. aided by AI HER2 solution (Ibex Breast HER2®), which detects the invasive tumor area and on slide control, classifies tumor cells based on their staining pattern, and derives a slide-level HER2 score by applying 2018 ASCO/CAP guidelines. Both study arms were compared to ground truth (GT), established as majority score of three breast pathologists (“experts”) who reviewed the slides manually. Results: Experts’ overall inter-observer agreement on all HER2 scores was 73.9% (95%CI: 72.6%,75.2%) and for 0/1+/2+/3+ was 80.8%/ 66.2%/ 63.7%/ 94.3%, respectively. Readers’ overall inter-observer agreement was significantly higher when assisted by AI, 87.5% (85.9%,89.0%) vs. 74.3% (72.3%,76.3%) without AI, p <0.05. Moreover, reader's accuracy for all HER2 scores (agreement with GT) was significantly higher with AI 80.9% (79.7%,82.2%) vs without AI, 76.6% (75.2%,77.9%) (p <0.05). For 0/1+ vs 2+/3+ cutoff, readers with AI showed significantly higher inter-observer agreement 93.1% (91.8%,94.2%) vs. without AI 86.8% (85.2%,88.3%), p <0.05, and significantly higher accuracy, 91.9% (91.0%,92.8%) with AI vs 88.8% (87.7%,89.8%) (p <0.05) without AI. For the HER2-low cutoff of 0 vs. 1+/2+/3+, readers with AI showed significantly higher inter-observer agreement of 95.0% (93.9%,95.9%) vs. 88.8% (87.3%,90.2%), p <0.05, with slightly higher accuracy (89.8% (88.8%,90.7%) with AI vs 88.5% (87.4%,89.4%) without AI). The standalone automatic AI solution demonstrated high accuracy for HER2 scoring of 89.4% (88.0%,90.7%), and 91.2% (89.8%,92.3%) for the respective clinical cutoffs of 0 vs. 1+/2+/3+, 0/1+ vs. 2+/3+. Feedback from reader pathologists' user survey indicates an increased confidence in their HER2 scoring accuracy and consistency. Additionally, 77% of the pathologists expressed a preference for HER2 scoring supported by AI over manual scoring. Conclusions: This study reports a large multi-site validation of a fully automated AI solution for HER2 scoring in BC. Pathologists supported by AI showed significant improvements in HER2 scoring consistency, evidenced by inter-reader agreement, and accuracy overall and for other clinical cutoffs. The AI solution demonstrated high accuracy and generalizability to multiple different laboratories (pre-analytics and staining protocols) and scanners. AI solutions, such as the one investigated here, could be used as decision-support tools for pathologists in routine clinical practice, enhancing the reproducibility and consistency of HER2 scoring, thus enabling optimal treatment pathways and improved patient outcomes. Citation Format: Savitri Krishnamurthy, Stuart Schnitt, Anne Vincent-Salomon, Elena Provenzano, Rita Canas-Marques, Laurent Arnould, Gaetan Mac Grogan, Elisabeth Shearon, Derek Welch, Pranil Chandra, Piotr Borkowski, Sabine Declercq, Joseph Loane, Anu Gunavardhan, Luca Di Tommaso, Vitor Krauss, Jeanne Thomassin, Marie Brevet, Maya Grinwald, Dana Mevorach, Sevde Etoz, Raz Ziv, Shai stein, Giuseppe Mallel, Judith Sandbank, Chaim Linhart, Manuela Vecsler. Improved Standardization and Accuracy of HER2 Score with AI support in Breast Cancer: Large Multicenter Study [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 P1-07-03.

Background: Over 50,000 women in the United States will be diagnosed with ductal carcinoma in situ (DCIS) this year alone. Almost all of these diagnoses will be made in completely asymptomatic individuals with a highly variable risk of progression to invasive cancer. In some low-risk malignancies, “watchful waiting,” is offered as a treatment option. Such an approach is likely reasonable for some DCIS and could reduce the harms of treatment while helping to identify those most likely to benefit from more aggressive therapy. To date, this approach has not been tested in a clinical trial setting. Methods: The COMET study (Comparing an Operation to Monitoring, with or without Endocrine Therapy for low risk DCIS; AFT-25) is a large pragmatic randomized non-inferiority trial that compares oncologic outcomes between patients randomized to guideline concordant care (GCC; surgery +/- radiation therapy) or active monitoring (AM). The study population were women seeking treatment for DCIS at one of the Alliance Clinical Trial sites. Eligible participants were age>40 with low-intermediate grade estrogen and/or progesterone receptor positive, HER2 receptor negative (if HER2 tested) DCIS on core biopsy without microinvasive or invasive cancer. The choice for endocrine therapy was offered in both groups. Participants in the AM group had surgical intervention only upon diagnosis of invasive progression. All study endpoints were collected prospectively. Results: This is the first planned interim Intention-to-Treat (ITT) analysis of the COMET trial primary endpoints at a median follow up of XX months. We will present patient characteristics for the 997 participants who enrolled in the study and were randomized to either GCC or AM. The primary endpoint to be presented is whether the ipsilateral invasive cancer rate for AM is non-inferior to that for GCC. Characteristics of invasive cancer events in the two groups will be compared. Secondary endpoints (rates of mastectomy, radiation, chemotherapy) and survival endpoints between groups will also be presented. Conclusion: These data will provide the first randomized trial evidence of whether an active monitoring strategy is a safe alternative for women with low-risk DCIS. Longer-term data could support practice changing guidance as to how DCIS is managed and treated and will have future implications for treatment guidelines for these excellent prognosis patients. Citation Format: Eun-Sil Hwang, Terry Hyslop, Thomas Lynch, Marc D Ryser, Anna Weiss, Anna Wolf, Kelsey Norris, Meredith Witten, Lars Grimm, Stuart Schnitt, Sunil Badve, Rachel Factor, Elizabeth Frank, Deborah Collyar, Desiree Basila, Donna Pinto, Mark A Watson, Robert West, Louise Davies, Jenny Donovan, Ayako Shimada, Yutong Li, Yan Li, Antonia V Bennett, Shoshana Rosenberg, Jeff Marks, Eric Winer, Marc Boisvert, Armando Giuliano, Kelsey Larson, Kathleen Yost, Priscilla McAuliffe, Lisa Carey, Alastair Thompson, Ann H Partridge. Early Oncologic Outcomes Following Active Monitoring or Surgery (+/- Radiation) for Low Risk DCIS: the Comparing an Operation to Monitoring, with or without Endocrine Therapy (COMET) Study (AFT-25) [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 GS2-05.

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.

Novel anti-HER2 antibody-drug conjugates (ADCs), such as trastuzumab deruxtecan (T-DXd), have shown efficacy in tumors with varying HER2 expression, including HER2-low and even tumors with minimal HER2 presence. This has sparked interest in the biology underlying the HER2 expression spectrum. Using molecular and multiplexed imaging, we revealed distinct immune and stromal features in treatment-naive, hormone receptor-positive (HR+) HER2-low versus HER2-0 tumors. HER2-0 tumors exhibit inflammatory and tissue remodeling gene signatures, with enrichment of APOE⁺ tumor-associated macrophages (TAMs) and DOCK4⁺ CD4 T cells. In contrast, HER2-low tumors are more immunosuppressed, with elevated cell cycle, metabolic, and estrogen signaling pathways, suggesting increased proliferative activity. These findings underscore key biological differences between HR+ HER2-low and HER2-0 breast cancers, and may inform more tailored therapeutic strategies. Statement of significance:This study revealed the distinct biological profiles of HR+ HER2-low and HER2-0 breast tumors. HER2-0 tumors exhibit inflammatory and tissue remodeling signatures, whereas HER2-low tumors have elevated cell cycle, metabolic, and estrogen signaling. These insights may help refine therapeutic approaches to improve outcomes for breast cancer patients.

Invasive lobular carcinoma (ILC) is the second most frequent histological type of breast cancer and the most frequent special type. Disruption of cell-to-cell adhesion, caused most often by E-cadherin loss of function, results in the distinctive histomorphology of ILC, which is characterized by single threads of monotonous, dyscohesive neoplastic epithelial cells infiltrating the breast parenchyma with little or no stromal reaction, referred to as classic ILC. In the past 4 decades, ILC variants that differ from classic ILC with regard to architectural, cytological, and/or nuclear features have been described. The recognition and correct characterization of ILC, including its variant forms, is essential to avoid misdiagnosis and its possible treatment implications. Some ILC variants may be associated with more aggressive clinical behavior compared with classic ILC, independent of standard predictive and prognostic parameters. Additionally, the distinctive biological and clinical features of ILC are increasingly being investigated as therapeutic targets in ILC-tailored clinical trials. In this manuscript, we have undertaken an in-depth review of the current state of knowledge about ILC variants. Evidence gained from molecular analysis of ILC and its microenvironment suggests that ILC variants are biologically distinct from classic ILC. However, this conclusion is undermined by the imprecise histopathological identification of ILC variants. In the absence of standardized and simplified criteria for the diagnosis of ILC, underrecognition of ILC variants may translate into missed opportunities for tailored treatment of ILC patients. Therefore, we propose steps toward the development of a roadmap that will ultimately lead to a more reproducible classification of ILC variants and improve our knowledge of these challenging tumors.

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.