Search

HORMAD1 expression is usually restricted to germline cells, but it becomes mis-expressed in epithelial cells in ~60% of triple-negative breast cancers (TNBCs), where it is associated with elevated genomic instability (1). HORMAD1 expression in TNBC is bimodal with HORMAD1-positive TNBC representing a biologically distinct disease group. Identification of HORMAD1-driven genetic dependencies may uncover novel therapies for this disease group. To study HORMAD1-driven genetic dependencies, we generated a SUM159 cell line model with doxycycline-inducible HORMAD1 that replicated genomic instability phenotypes seen in HORMAD1-positive TNBC (1). Using small interfering RNA screens, we identified candidate genes whose depletion selectively inhibited the cellular growth of HORMAD1-expressing cells. We validated five genes (ATR, BRIP1, POLH, TDP1 and XRCC1), depletion of which led to reduced cellular growth or clonogenic survival in cells expressing HORMAD1. In addition to the translesion synthesis (TLS) polymerase POLH, we identified a HORMAD1-driven dependency upon additional TLS polymerases, namely POLK, REV1, REV3L and REV7. Our data confirms that out-of-context somatic expression of HORMAD1 can lead to genomic instability and reveals that HORMAD1 expression induces dependencies upon replication stress tolerance pathways, such as translesion synthesis. Our data also suggest that HORMAD1 expression could be a patient selection biomarker for agents targeting replication stress.

Purpose: To assess the frequency and prognostic relevance of germinal centres (GC) in cancer-free axillary lymph nodes (ALNs), in relation to stromal tumour-infiltrating lymphocytes (sTILs) and the presence of tertiary lymphoid structures (TLS) in the primary carcinoma, in LN-positive Hormone Receptor (HR)-negative invasive breast cancer patients. Patients and methods: A cohort of 161 patients with HR-negative invasive breast cancer of no special type (NST) and LN-positive status treated between 2005-10 at Tianjin Medical University (China) was identified. sTILs and TLS at the primary tumour site and GC in 2,841 involved and cancer-free ALNs were evaluated on H&E stained sections. Markers were tested for prognostic value for invasive Disease-Free Survival (iDFS), distant Disease-Free Survival (dDFS) and Overall Survival (OS), using Cox regression models adjusted for clinico-pathological factors. Results: Among the 161 HR-negative breast cancers, 47% (n=75) had >=20% sTILs and 24% (n=38) peritumoural TLS. 75% (121/161) and 76% (122/161), respectively, displayed GCs in their cancer-free and involved ALNs. Significantly higher numbers of GCs were seen in both cancer-free and involved ALNs when the primary tumours showed >=20% sTILs. The presence of TLS was significantly associated with increased GC numbers in involved but to a lesser extend in cancer-free ALNs (Kruskal-Wallis rank sum test, p Using an iterative process to determine an optimal cut off point by a minimal P value approach, a non-monotonic relationship between the frequency of GC in cancer-free ALNs and all endpoints was observed. Patients with >2 GC but less than the top 5% (= 62 GC) in cancer-free ALNs showed improved iDFS, dDFS and OS, whilst patients with 62 GC across all assessed cancer-free ALNs had poorer iDFS, dDFS and OS (see Table). In the multivariate models, the frequency of GC in cancer-free ALNs added independent prognostic information for all endpoints across all patients. In patients with >=20% sTILS, cancer-free ALNs within the top 5% for GC remained associated with a worse dDFS and iDFS (see Table). Cancer-free ALNs with Five-year iDFS, dDFS and OS in patients with 2 GC whose five-year iDFS, dDFS and OS were 65%, 65% and 69%, respectively. Conclusions: The prognostic importance of GC assessment in cancer-free ALNs in women with LN-positive HR-negative breast cancers is demonstrated. A better outcome in patients with GC formation in their cancer-free ALNs, despite low sTILs in the primary carcinoma, may suggest a systemic anticancer immune response. High sTILs but with extreme GC formation in the cancer-free ALNs could potentially reflect an overdriving but less effective immune response. The assessment of the combination of primary and nodal immune response in HR-negative breast cancers is imperative in these high-risk patients. Table 1. Univariate and Multivariate Cox Regression Analysis of Outcome by GC in Cancer-Free ALNsiDFSUnivariateAll cases=20% sTILTotal GCs number - binnedModel PHRCIModel PHRCIModel PHRCIGC624.631.75 - 12.41.930.26 - 14.5612.793.34 - 48.97MultivariateCorrected for: pNstage, sTILS & TLSCorrected for: pTstage & TLSCorrected for: pNstageTotal GCs number - binnedCovariate PModel PHRCICovariate PModel PHRCICovariate PModel PHRCIGC624.15E-059.583.25 - 28.201.94E-014.070.49 - 33.876.62E-038.861.83 - 42.82dDFSUnivariateAll cases=20% sTILTotal GCs number - binnedModel PHRCIModel PHRCIModel PHRCIGC625.872.14 - 16.082.490.32 - 19.0718.434.04 - 84.03MultivariateCorrected for: pTstage, pNstage, sTILS & TLSCorrected for: Age, pTstage, TLSCorrected for: pNstageTotal GCs number - binnedCovariate PModel PHRCICovariate PModel PHRCICovariate PModel PHRCIGC622.49E-0615.304.91 - 47.635.19E-028.610.98 - 75.553.11E-0313.302.39 - 73.89OSUnivariateAll cases=20% sTILTotal GCs number - binnedModel PHRCIModel PHRCIModel PHRCIGC624.131.18 - 14.42.570.33 - 19.8311.891.65 - 85.45MultivariateCorrected for: Age, pTstage, pNstage, sTILS & TLSCorrected for: Age & TLSCorrected for: pNstage, LVI & TLSTotal GCs number - binnedCovariate PModel PHRCICovariate PModel PHRCICovariate PModel PHRCIGC628.62E-0516.064.02 - 64.247.01E-027.050.85 - 58.342.80E-012.970.41 - 21.43*too few events Citation Format: Fangfang Liu, Thomas Hardiman, Patrycja Gazinska, Jelmar Quist, Sarah Pinder, Anita Grigoriadis. The formation of GCs in cancer-free ALNs, a non-monotonic prognostic factor in HR-negative invasive breast cancer patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS6-36.

Triple-negative breast cancers (TNBC) expressing PD-L1 qualify for checkpoint inhibitor immunotherapy. Cyclin E/CDK2 is a potential target axis in TNBC; however, small-molecule drugs at efficacious doses may be associated with toxicity, and treatment alongside immunotherapy requires investigation. We evaluated CDK inhibition at suboptimal levels and its anti-tumor and immunomodulatory effects. Transcriptomic analyses of primary breast cancers confirmed higher cyclin E/CDK2 expression in TNBC compared with non-TNBC. Out of the three CDK2-targeting inhibitors tested, the CDK 2, 7 and 9 inhibitor SNS-032 was the most potent in reducing TNBC cell viability and exerted cytotoxicity against all eight TNBC cell lines evaluated in vitro. Suboptimal SNS-032 dosing elevated cell surface PD-L1 expression in surviving TNBC cells. In mice engrafted with human immune cells and challenged with human MDA-MB-231 TNBC xenografts in mammary fat pads, suboptimal SNS-032 dosing partially restricted tumor growth, enhanced the tumor infiltration of human CD45+ immune cells and elevated cell surface PD-L1 expression in surviving cancer cells. In tumor-bearing mice engrafted with human immune cells, the anti-PD-L1 antibody avelumab, given sequentially following suboptimal SNS-032 dosing, reduced tumor growth compared with SNS-032 alone or with avelumab without prior SNS-032 priming. CDK inhibition at suboptimal doses promotes immune cell recruitment to tumors, PD-L1 expression by surviving TNBC cells and may complement immunotherapy.

Fatty acid synthase (FASN) is commonly overexpressed in prostate cancer and associated with tumour progression. FASN is responsible for de novo synthesis of the fatty acid palmitate; the building block for protein palmitoylation. Recent work has suggested that alongside its established role in promoting cell proliferation FASN may also promote invasion. We now find depletion of FASN expression increases prostate cancer cell adhesiveness, impairs HGF-mediated cell migration and reduces 3D invasion. These changes in motility suggest that FASN can mediate actin cytoskeletal remodelling; a process known to be downstream of Rho family GTPases. Here, we demonstrate that modulation of FASN expression specifically impacts on the palmitoylation of the atypical GTPase RhoU. Impaired RhoU activity in FASN depleted cells leads to reduced adhesion turnover downstream of paxillin serine phosphorylation, which is rescued by addition of exogenous palmitate. Moreover, canonical Cdc42 expression is dependent on the palmitoylation status of RhoU. Thus we uncover a novel relationship between FASN, RhoU and Cdc42 that directly influences cell migration potential. These results provide compelling evidence that FASN activity directly promotes cell migration and supports FASN as a potential therapeutic target in metastatic prostate cancer.

During the anti-tumour response to breast cancer, the primary tumour, the peripheral blood, and the lymph nodes each play unique roles. Immunological features at each site reveal evidence of continuous immune cross-talk between them before, during and after treatment. As such, immune responses to breast cancer are found to be highly dynamic and truly systemic, integrating three distinct immune sites, complex cell-migration highways, as well as the temporal dimension of disease progression and treatment. In this review, we provide a connective summary of the dynamic immune environment triad of breast cancer. It is critical that future studies seek to establish dynamic immune profiles, constituting multiple sites, that capture the systemic immune response to breast cancer and define patient-selection parameters resulting in more significant overall responses and survival rates for breast cancer patients.

Background The TNT trial (NCT00532727) showed no evidence of carboplatin (C) superiority over docetaxel (D) overall in metastatic triple negative breast cancers (TNBC), but a C benefit was observed in the pre-specified sub-group analysis in patients with a gBRCA1/2 mutation (Tutt et al, Nat Med 2018). Given only ~30% of patients have a gBRCA1/2 mutation, broader predictive biomarkers of response are needed. In this cohort we previously found that DNA Damage Response (DDR) signatures were associated with improved C response in chemotherapy (CT) naïve patients only (Tovey et al, ASCO 2020). Since DDR activities influence tumour immune-microenvironment, we explored the predictive ability of immune cell markers and performed integrative analyses on multi-omics features to identify novel TNBC subgroups. Patients and Methods Tumour infiltrating lymphocytes (TILs) were evaluated on haematoxylin and eosin stained primary tumour (PT) slides for 222/376 TNT patients. Formalin-fixed paraffin-embedded PT tissues from 186/376 TNT patients were successfully profiled using total RNA-sequencing. Matched recurrence (REC) was also sequenced for 13 patients. Twenty-five immune signatures were assessed. Logistic regression and restricted mean progression free survival (PFS) were applied to delineate the relationship of these features with treatment outcomes. Random forest clustering of multi-omics DDR and immune biology markers, including gene expression signatures and mutation/methylation status, was applied to identify subgroups. We further molecularly characterised these clusters through supervised clustering of 693 gene expression “modules” (sets of co-expressed genes), immune cell deconvolution and genomic scars. Results Immune gene expression signatures and TILs were highly correlated. Average immune infiltration based on ConsensusTME was lower in mutated/methylated tumours compared with BRCA1 wildtype tumours (p=0.04). Immune signature score markers decreased from PT to REC, demonstrating a dynamic immune microenvironment. In the overall population and when restricting to prior CT treated patients, high immune infiltration (gene expression based & TILs) was associated with response to D while C response rates were not associated with immune scores (interaction p-values< 0.05). This did not translate to a PFS benefit. Multi-omics clustering identified 6 biological subgroups including immune enriched, immune depleted, DDR deficient and proficient clusters as well as 2 small clusters with no obvious distinguishing features. Immune enriched TNBC were predominantly basal-like immune activated with high B-cell/T-cell diversity. Immune depleted TNBC showed higher activity of proliferation and DDR pathway modules. DDR proficient tumours had low expression of immune markers and enrichment for ESR1/PGR expression, markers of extra cellular formation, cell structure, lipid metabolism and proliferation. The DDR deficient cluster was enriched for proliferation and demonstrated high number of TILs despite no apparent enrichment for gene expression-based immune modules. In the prior CT treated cohort, the immune enriched cluster had preferential response to D (62.5% (D) vs. 29.4% (C); p=0.02). The immune depleted cluster had preferential response to C (8.0% (D) vs. 40.0% (C); p=0.01). Numbers were too small to assess differential response within the other clusters or in the CT naïve cohort. Conclusions Tumours with high immune features have high response to D while those with low immune features have preferential response to C in advanced TNBC. Combining multi-omics markers of DDR deficiency and immune biology can identify clusters of patients with distinct biological profiles and differential treatment specific response rates. Citation Format: Holly Tovey, Orsolya Sipos, Katherine A Hoadley, Joel S Parker, Jelmar Quist, Sarah Kernaghan, Lucy Kilburn, Roberto Salgado, Sherene Loi, Richard D Kennedy, Ioannis Roxanis, Patrycja Gazinska, Sarah E. Pinder, Judith Bliss, Charles M. Perou, Syed Haider, Andrew Tutt, Anita Grigoriadis, Maggie Chon U Cheang. Histopathological and molecular immune landscape and DNA damage response signatures to predict response to carboplatin and docetaxel in TNT trial TNBC cohort [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD9-06.

Systemic immune responses in lymph nodes (LN) convey significant prognostic value for breast cancer patients, which can inform disease progression and optimal treatment management. However, have, so far, not been assessed in large patient cohorts. We have previously shown that morphological alterations in axillary LNs, namely the formation of germinal centres (GCs) in cancer-free LNs, add prognostic value to tumour infiltrating lymphocytes (TILs) in triple-negative breast cancer patients (TNBC) for the development of distant metastasis. Extending manual assessment of LNs beyond the detection of cancer requires the integration of robust deep learning pipelines into the digital pathology workflow. Here, we propose a supervised multiscale deep learning framework named smuLymphNet to capture and quantify GCs and sinuses within LNs from digitised Haematoxylin and Eosin-stained (H&E) whole slide images (WSIs) and show good concordance compared with an inter-pathologist Dice coefficient of manual annotations from four pathologists. The smuLymphNet framework consists of (i) a detection algorithm to determine the boundaries of each LN section on the WSI, using an Otsu-based thresholding method and contouring algorithm; (ii) a supervised multiscale deep learning module for the segmentation of GCs and sinuses; and (iii) quantification of the number, size, and shape of the predicted features. We applied smuLymphNet to a total of 1,800 H&E-stained WSI of >4,000 cancer-free and involved LNs from a retrospectively collected breast cancer cohort collected at Guy’s Hospital (London, UK) from 177 patients (122 N+) enriched for the triple-negative phenotype. A subset of 114 WSI and five breast cancer LN WSIs from each Barts Hospital (London, UK) and Tianjin University Hospital (Tianjin, China) were used to train and evaluate the supervised deep learning module. For training Fully Convolutional Networks (FCNs), WSIs manually annotated for both GCs and sinuses formed a ground-truth set and three FCNs were implemented: (i) a standard U-Net architecture; (ii) a U-Net model with an attention gate mechanism; and (iii) a multiscale-U-Net network (MS U-Net) that encodes, in parallel, a feature representation of the image at multiple resolutions. The MS U-Net achieved the best performance with an average dice score of 0.86 for GCs and 0.74 for sinuses. In comparison, the average dice score amongst four pathologists assessing 24 LN WSI for GCs and sinuses was 0.67 and 0.61, respectively, demonstrating the robustness of the smuLymphNet framework. To establish associations between morphometric immune features and patients’ outcomes, we assessed smuLymphNet captured GCs and sinuses from 686 WSIs from 96 TNBC patients with extensive longitudinal outcome data. We found significant morphological differences in involved and cancer-free LNs between N0 and N+ patients, with the latter displaying larger GCs with more irregular shapes, especially in their involved LNs. Moreover, in alignment with our previously published studies, our multiscale smuLymphNet framework recapitulated and extended the prognostic value of the assessment of GC formation in TNBC N0 patients. We further revealed, for the first time, the prognostic significance of the intranodal lymphatic sinuses when measured in their totality in involved LNs, and the association of alterations in subcapsular sinus areas with superior distant metastasis-free survival in cancer-free and involved LNs in TNBC N+ patients. In summary, smuLymphNet presents a robust multiscale deep learning framework to automatically detect, localise and quantify histopathological immune features in WSI of LNs. By applying smuLymphNet to LNs of TNBC patients from clinical trials, and thereby further evaluating its clinical utility, smuLymphNet could be implemented into the diagnostic digital pathology workflow and, as such, aid in informing on a patient’s disease trajectory. Citation Format: Gregory Verghese, Mengyuan Li, Fangfang Liu, Amit Lohan, Nikhil Cherian, Patrycja Gazinska, Aekta Shah, Aasiyah Oozeer, Cheryl Gillett, Elena Alberts, Thomas Hardiman, Roberto Salgado, Samantha Jones, Louise Jones, Selvam Thavaraj, Sarah E. Pinder, Swapni Rane, Amit Sethi, Anita Grigoriadis. Multiscale Deep Learning framework to capture systemic immune features in lymph nodes predictive of triple negative breast cancer outcome [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-01-01.

B cells; Breast cancer; Lymph nodes Cèl·lules B; Càncer de mama; Ganglis limfàtics Células B; Cáncer de mama; Ganglios linfáticos B cells have recently become a focus in breast cancer pathology due to their influence on tumour regression, prognosis, and response to treatment, besides their contribution to antigen presentation, immunoglobulin production, and regulation of adaptive responses. As our understanding of diverse B cell subsets in eliciting both pro- and anti-inflammatory responses in breast cancer patients increases, it has become pertinent to address the molecular and clinical relevance of these immune cell populations within the tumour microenvironment (TME). At the primary tumour site, B cells are either found spatially dispersed or aggregated in so-called tertiary lymphoid structures (TLS). In axillary lymph nodes (LNs), B cell populations, amongst a plethora of activities, undergo germinal centre reactions to ensure humoral immunity. With the recent approval for the addition of immunotherapeutic drugs as a treatment option in the early and metastatic settings for triple-negative breast cancer (TNBC) patients, B cell populations or TLS may resemble valuable biomarkers for immunotherapy responses in certain breast cancer subgroups. New technologies such as spatially defined sequencing techniques, multiplex imaging, and digital technologies have further deciphered the diversity of B cells and the morphological structures in which they appear in the tumour and LNs. Thus, in this review, we comprehensively summarise the current knowledge of B cells in breast cancer. In addition, we provide a user-friendly single-cell RNA-sequencing platform, called “B singLe cEll rna-Seq browSer” (BLESS) platform, with a focus on the B cells in breast cancer patients to interrogate the latest publicly available single-cell RNA-sequencing data collected from diverse breast cancer studies. Finally, we explore their clinical relevance as biomarkers or molecular targets for future interventions. The authors acknowledge support by Breast Cancer Now (KCL-BCN-Q3), the Medical Research Council [MR/X012476/1]; Mengyuan Li was awarded funds by the China Scholarship Council [202008440233]; Elena Alberts was funded by the UK Medical Research Council Doctoral Training Programme; Victoire Boulat was awarded funds by Cancer Research UK (RE18831).

Genomic instability is a key feature of carcinogenesis, giving rise to tumour promoting aberrations. The inherent unpredictability of these processes leads to a high degree of intratumour heterogeneity, previously shown to be advantageous for cancer progression in a series of cancer types. We sought to explore the role of genomic instability and intratumour heterogeneity in the progression of ductal carcinoma in situ (DCIS) to invasive disease, in a cohort of 49 women with pure DCIS that developed subsequent ipsilateral invasive disease and 90 women with pure DCIS with no evidence of recurrence with a median follow up of 5.4 years. All samples were identified from within the Sloane project, a prospective national cohort study of DCIS within the UK National Health Service Breast Screening Programme. DNA extracted from bulk tissue samples was processed on the HumanCytoSNP array, and allele-specific copy number aberration (CNA) data was obtained from ASCAT. We measured genomic instability using Scores Of Chromosomal Instability Scarring (SCINS), as well as by calculating the Shannon diversity index for the different copy number states occurring in the tumour genome. We also attempted to deconvolute the clonal structure of the tumours by applying the Shannon diversity index on the raw copy number estimates provided by ASCAT. Non-integer copy number states were interpreted as indicative of the presence of subclonal CNAs. As a positive control, we compared ER+ and ER- DCIS in a larger cohort of 173 DCIS samples (129 ER+ and 44 ER-), using the same methods. We expect ER- tumours to be more genomically unstable, and consequentially more clonally diverse. Measured using SCINS, both copy-neutral loss of heterozygosity (≥ 4 Mbp) and allelic-imbalanced CNAs (≥ 8 Mbp) were shown to be higher in ER- tumours (Fisher’s exact test, p = 0.014 and p = 0.034, respectively). Overall genomic instability was also shown to be higher in ER- tumours using the Shannon diversity index (Fisher’s exact test, p = 0.0085). Our attempt at clonal structure deconvolution also indicates increased heterogeneity in ER- tumours (Fisher’s exact test, p = 0.027). However, no significant difference was identified in the genomic instability or heterogeneity of DCIS that developed subsequent invasive disease compared to that of indolent DCIS, using any of the aforementioned measures. We were therefore unable to identify evidence of a direct link between DCIS heterogeneity and progression to invasive cancer. Our methods did, however, demonstrate that DCIS in general is highly genomically diverse, with ER- tumours exhibiting increased diversity compared to ER+ ones. We aim to validate these findings using spatial sampling data from our cohort, as well as integrating mutation data to allow the use of established clonal heterogeneity deconvolution pipelines. Citation Format: Anargyros Megalios, Vandna Shah, Rana Shami, Salpie Nowinski, Jelmar Quist, Mathini Sridharan, Carolina Salinas de Souza, Karen Clements, Anita Grigoriadis, Sarah Pinder, Alastair Thompson, Elinor Sawyer. Heterogeneity of DCIS does not appear to be a biomarker for development of subsequent invasive cancer [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P4-06-05.

Simple Summary Breast cancer is a complex disease, and the identification of its underlying molecular mechanisms is critical for the development of treatment strategies. The purpose of this study was to implement a computational framework that is capable of combining many types of data into a meaningful classification. While our approach can be used on many types of data and in many diseases, we applied this framework to breast cancer data and identified six triple-negative breast cancer subtypes with distinct underlying molecular mechanisms. The relevance of our approach is highlighted by the clinical outcome analysis in which a group of patients responding poorly to standard-of-care adjuvant chemotherapy was identified. This study serves as a starting point for our computational framework, which can be extended to different types of data from different diseases. Abstract Advances in high-throughput technologies encourage the generation of large amounts of multiomics data to investigate complex diseases, including breast cancer. Given that the aetiologies of such diseases extend beyond a single biological entity, and that essential biological information can be carried by all data regardless of data type, integrative analyses are needed to identify clinically relevant patterns. To facilitate such analyses, we present a permutation-based framework for random forest methods which simultaneously allows the unbiased integration of mixed-type data and assessment of relative feature importance. Through simulation studies and machine learning datasets, the performance of the approach was evaluated. The results showed minimal multicollinearity and limited overfitting. To further assess the performance, the permutation-based framework was applied to high-dimensional mixed-type data from two independent breast cancer cohorts. Reproducibility and robustness of our approach was demonstrated by the concordance in relative feature importance between the cohorts, along with consistencies in clustering profiles. One of the identified clusters was shown to be prognostic for clinical outcome after standard-of-care adjuvant chemotherapy and outperformed current intrinsic molecular breast cancer classifications.

Despite their role in cancer surveillance, adoptive immunotherapy using γδ T-cells has achieved limited efficacy. To enhance trafficking to bone marrow, circulating γδ T-cells were expanded in serum-free medium containing TGF-β1 and IL-2 (γδ[T2] cells).Unexpectedly, the yield and viability of γδ[T2] cells were also increased by TGF-β1,when compared to γδ[2] controls (IL-2 alone). γδ[T2] cells were less differentiated and yet displayed increased cytolytic activity, cytokine release and anti-tumor activity in several leukemic and solid tumor models. Efficacy was further enhanced by cancer cell sensitization using amino bisphosphonates or cytarabine, or by CAR re-programming. A number of contributory effects of TGF-β1 were identified, including prostaglandinE2 receptor downmodulation, upregulation of CD103 and enhanced IL-9 production by γδ[T2] cells. Biological relevance is supported by the identification of a favorable γδ[T2] signature in acute myeloid leukemia. Given their enhanced therapeutic activity and compatibility with allogeneic use, γδ[T2] cells warrant evaluation in cancer immunotherapy.

Capturing tumor infiltrating leucocytes (TILS) and systemic immune responses in breast cancer informs disease progression and optimal treatment management. We have previously shown that morphological alterations in axillary lymph nodes (LNs), namely the formation of germinal centers in cancer-free LNs, adds prognostic value to TILs in triple negative breast cancer patients (TNBC) for the development of distant metastasis. Extending manual assessment of LNs beyond the detection of cancer requires the integration of robust deep learning pipelines into the digital pathology workflow. In this retrospective study, we used 1,100 Haematoxylin & Eosin-stained (H&E) Whole Slide Images (WSI) from Guy’s Hospital (London, UK) of metastatic and cancer-free LNs from 151 patients (100 N+) enriched for triple-negative or HER2-positive breast cancer to implement a supervised deep learning pipeline. A subset of 114 WSI, along with 5 breast cancer LN WSIs from each of Barts Hospital (London, UK) and Tianjin University Hospital (Tianjin, China), and 5 head and neck squamous cell carcinomas LN WSI (Guy’s Hospital) were used to develop, train and evaluate the segmentation task. For training Fully Convolutional Networks (FCNs), WSIs manually annotated for both germinal centers and sinuses formed a ground-truth set. Three FCNs were implemented: (i) a standard U-Net architecture; (ii) a U-Net model with an attention gate mechanism; and (iii) a multiscale-U-Net network (MSA-U-Net) that encodes, in parallel, a feature representation of the image at multiple resolutions. The MSA-U-Net achieved the best performance with an average dice score of 0.85 for germinal centers and 0.75 for sinuses. In comparison, the average dice score amongst 4 pathologists assessing 25 LN WSI for germinal centers and sinuses, was 0.67 and 0.61 respectively, demonstrating the robustness of the MSA-U-Net model. To quantify germinal centers and sinuses in LNs across the entire cohort, the trained MSA-U-Net was used in an inference step on all 1,100 WSI. The detected morphological features were initially localized within LNs using image thresholding and contouring techniques, and quantitatively assessed based on their number, area, shape, and Shannon diversity. We found significant morphological differences in metastatic and cancer-free LNs between N0 and N+ patients, with the latter displaying larger germinal centers with more irregular shapes especially in their metastatic LNs. In addition, we found differences in the Sinus area between LNs containing GCs and those without. Here, we propose a robust deep learning pipeline based on a multiscale FCN framework to automatically detect, localize and quantify histopathological immune features in WSI of LNs. By applying our pipeline to LNs of cancer patients, such as breast or head and neck, in prospective studies or clinical trials, we will further evaluate their prognostic and predictive values. Citation Format: Gregory Verghese, Mengyuan Li, Amit Lohan, Nikhil Cherian, Swapnil Rane, Fangfang Liu, Aekta Shah, Pat Gazinska, Selvam Thavaraj, Amit Sethi, Anita Grigoriadis. A deep learning pipeline to capture the prognostic immune responses in lymph nodes of breast cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6233.

541 Background: The meiotic DNA break regulator HORMAD1 is aberrantly expressed in many cancers and is associated with increased genomic instability. The susceptibility of HORMAD1 expressing tumors to agents targeting DNA damage repair (DDR) pathways is poorly understood since clinical data within the context of a randomized clinical trial (RCT) is lacking. Here, we retrospectively studied HORMAD1 expression as a putative predictive biomarker in an RCT for benefit of adjuvant high-dose platinum-based chemotherapy (HDCT) with autologous stem cell support in patients with high-risk HER2-negative early breast cancer (BC). Methods: Patients with stage III BC participated in an RCT comparing HDCT to conventional chemotherapy (CDCT; Rodenhuis et all, NEJM, 2003; Steenbruggen et all, JAMA Oncol, 2020). We studied the subgroup with HER2-negative BC for whom tumor BRCA1-like classification was previously determined using a validated DNA comparative genomic hybridization algorithm (Vollebergh et all, BCR, 2014). Tumor HORMAD1 expression was determined on FFPE samples using RNAscope, an RNA in situ hybridization method, and classified as negative (no expression) or positive (any expression detected). Results: For 195/246 (79.3%) HER2-negative patients treated according to protocol, HORMAD1 RNAscope status was available; dropout was due to absence or insufficient quality of tumor specimens. HORMAD1 positivity was enriched in triple-negative breast cancer (TNBC) (23/47; 48.9%). Furthermore, in all HER2-negative BCs, HORMAD1 positivity (45/195; 23.1%) was associated with age ≤40 years, histological grade III, 10%. Such association, although not significant, was also observed within TNBC. During a median follow-up of 20.3 years, 124 (63.6%) recurrences and 115 deaths (59.0%) occurred. The prognostic effect of HORMAD1 positivity on overall survival (OS) varied with follow-up time and was borderline significant at 10 years and significant thereafter (10-year: adjusted (adj.) HR 0.47, 95% CI 0.21-1.04; 15-year: adj. HR 0.25, 95% CI 0.07-0.91). Benefit on RFS from HDCT over CDCT was stronger in patients with HORMAD1-positive tumors (adj. HR 0.18, 95% CI 0.06-0.54) than in patients with HORMAD1-negative tumors (adj. HR 0.69, 95% CI 0.46-1.02) (P-interaction = 0.02). Similar results were observed for OS. Conclusions: In this retrospective sub study of 195 patients with high-risk HER2-negative BC participating in an RCT, tumor HORMAD1 expression is predictive for benefit of high-dose platinum-based chemotherapy. Our observations are consistent with the prior observations that HORMAD1 expression is associated with genomic instability and impaired DDR pathways. Further research is warranted to validate our findings. Clinical trial information: NCT03087409.

In breast cancer, humoral immune responses may contribute to clinical outcomes, especially in more immunogenic subtypes. Here, we investigated B lymphocyte subsets, immunoglobulin expression, and clonal features in breast tumors, focusing on aggressive triple-negative breast cancers (TNBC). In samples from patients with TNBC and healthy volunteers, circulating and tumor-infiltrating B lymphocytes (TIL-B) were evaluated. CD20+CD27+IgD− isotype-switched B lymphocytes were increased in tumors, compared with matched blood. TIL-B frequently formed stromal clusters with T lymphocytes and engaged in bidirectional functional cross-talk, consistent with gene signatures associated with lymphoid assembly, costimulation, cytokine–cytokine receptor interactions, cytotoxic T-cell activation, and T-cell–dependent B-cell activation. TIL-B–upregulated B-cell receptor (BCR) pathway molecules FOS and JUN, germinal center chemokine regulator RGS1, activation marker CD69, and TNFα signal transduction via NFκB, suggesting BCR–immune complex formation. Expression of genes associated with B lymphocyte recruitment and lymphoid assembly, including CXCL13, CXCR4, and DC-LAMP, was elevated in TNBC compared with other subtypes and normal breast. TIL-B–rich tumors showed expansion of IgG but not IgA isotypes, and IgG isotype switching positively associated with survival outcomes in TNBC. Clonal expansion was biased toward IgG, showing expansive clonal families with specific variable region gene combinations and narrow repertoires. Stronger positive selection pressure was present in the complementarity determining regions of IgG compared with their clonally related IgA in tumor samples. Overall, class-switched B lymphocyte lineage traits were conspicuous in TNBC, associated with improved clinical outcomes, and conferred IgG-biased, clonally expanded, and likely antigen-driven humoral responses. Significance: Tumor-infiltrating B lymphocytes assemble in clusters, undergoing B-cell receptor–driven activation, proliferation, and isotype switching. Clonally expanded, IgG isotype-biased humoral immunity associates with favorable prognosis primarily in triple-negative breast cancers.

Innate-like tissue-resident γδ T cell compartments capable of protecting against carcinogenesis are well established in mice. Conversely, the degree to which they exist in humans, their potential properties, and their contributions to host benefit are mostly unresolved. Here, we demonstrate that healthy human breast harbors a distinct γδ T cell compartment, primarily expressing T cell receptor (TCR) Vδ1 chains, by comparison to Vδ2 chains that predominate in peripheral blood. Breast-resident Vδ1+ cells were functionally skewed toward cytolysis and IFN-γ production, but not IL-17, which has been linked with inflammatory pathologies. Breast-resident Vδ1+ cells could be activated innately via the NKG2D receptor, whereas neighboring CD8+ αβ T cells required TCR signaling. A comparable population of Vδ1+ cells was found in human breast tumors, and when paired tumor and nonmalignant samples from 11 patients with triple-negative breast cancer were analyzed, progression-free and overall survival correlated with Vδ1+ cell representation, but not with either total γδ T cells or Vδ2+ T cells. As expected, progression-free survival also correlated with αβ TCRs. However, whereas in most cases TCRαβ repertoires focused, typical of antigen-specific responses, this was not observed for Vδ1+ cells, consistent with their innate-like responsiveness. Thus, maximal patient benefit may accrue from the collaboration of innate-like responses mounted by tissue-resident Vδ1+ compartments and adaptive responses mounted by αβ T cells.

Background Histological changes in the cancer-free lymph nodes (LNs) can add to the risk prediction of developing distant metastasis for LN-positive breast cancer patients. We have previously shown that the formation of germinal centres and enlarged sinus are premetastatic features and indicative of superior prognosis. As Whole Slide Images (WSI) demonstrated suitability for detection of LN metastasis with high accuracy based on Convolutional Neural Networks (CNN), we extended the CNN assessments to the detection and segmentation of these premetastatic features. Materials For 136 breast cancer patients, retrospectively collected from Guy’s Hospital, 1,422 Whole Slide Images (WSI) of axillary LNs (containing 1 to 3 LNs per WSI) and primary invasive tumours were 40x scanned by the Hamamatsu Nanozoomer. Two pathologists manually annotated germinal centres and sinuses in 111 of these images facilitating the ground-truth masks for a binary segmentation task using deep learning models. Results Each WSI was split into a series of overlapping patches along with corresponding ground-truth masks at 2.5x, 5x and 10x. Fully Convolutional Neural Networks (FCNs) based on the U-Net architecture were trained on patches from 95 annotated WSI to perform a segmentation for both germinal centres and sinus, with the remaining images used for validation (n= 6) and test sets (n= 10). In addition to a vanilla U-Net encoder-decoder architecture, a model with an attention mechanism to filter feature maps along skip connections was also explored. Additionally, a U-Net with dilated convolutions at each convolution block was used in a multi-scale mechanism that combines features from three different resolutions. To increase the robustness of the FCNs, training data was augmented using random image rotation and different colour perturbation transformations. Predictions were made on regions of WSIs containing whole LNs and were compared with the pathologist annotations on holdout testing images. The Dice coefficient was used to compare the predicted segmentations with the ground-truth annotations. Standard U-Net and attention U-Net architectures achieved an average Dice coefficient of 0.75 across the holdout testing images at 2.5x magnification for germinal centres, with individual images achieving a Dice score above 0.8. Multiscale models achieved the best results at 10x magnification with an average Dice score of 0.85 for germinal centres and 0.75 for sinus. The trained models were then applied to > 1,300 unseen and non-annotated WSIs of axillary LNs. Conclusion Our developed FCN models allow an automated, precise, reproducible and time-efficient approach to annotate germinal centres and sinuses in cancer-free and metastatic LNs. Trained models are currently applied to >1,300 unseen and non-annotated WSIs of axillary LNs, and will demonstrate their comparable performance with manual pathological assessment. Citation Format: Gregory Verghese, Anita Grigoriadis, Amit Sethi, Amit Lohan, Nikhil Cherian, Swati Meena, Harry Chinque, Fangfang Liu, Ed Martin, Tristan Clark, Cheryl Gillet, Julie Owen, Swapnil Rane. Deep learning-based segmentation accurately captures histological features in cancer-free lymph nodes of breast cancer patients [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr PO-014.

Background: Lobular carcinoma in situ (LCIS) is typically clinically undetectable but is being increasingly diagnosed as a result of breast screening mammography and is often found associated with other breast pathologies such as invasive lobular breast cancer (ILC), invasive carcinoma of ductal /no special type (IDC) and ductal carcinoma in situ (DCIS). It is also considered a risk factor for the development of subsequent invasive breast disease. The aim of this study was to understand the genetic relationship between LCIS that presents with synchronous DCIS, IDC and/or ILC in order to ascertain whether the components have common precursors and also to understand the clonal relationship between LCIS and subsequent invasive disease. Methods: 25 cases of LCIS with synchronous ILC, 7 cases of LCIS with synchronous DCIS & IDC, and 8 pure LCIS that developed a subsequent invasive recurrence were identified from the GLACIER study (MREC 06/Q1702/64). DNA was extracted from archival paraffin embedded tissue and underwent copy number analysis using either the Oncoscan™ Array (Affymetrix) or HumanCytoSNP FFPE-12 BeadChip (Illumina). Four of 7 cases of LCIS with synchronous DCIS & IDC also underwent targeted sequencing using a custom 121 breast cancer-associated gene panel (SureSelectXT HS kit, Agilent Technologies). Clonal relatedness was assessed using a novel methodology based on the presence of shared copy number aberration breakpoints and mutations. Results: Of the 25 synchronous LCIS and ILC cases, 17 appeared related, 4 were ambiguous (sharing the typical lobular signature of 1q gain and 16q loss) and 4 demonstrated no evidence of relatedness. Of the 7 cases with synchronous LCIS, DCIS and IDC, all had copy number data available and 4 had mutation data available. In 3 cases the synchronous LCIS, DCIS and IDC were clonally related according to copy number and for two there was mutation data that supported this (one sharing PIK3CA and CDH1 mutations, the other a TP53 mutation). In two cases the LCIS was not related to the DCIS or IDC, but the DCIS and IDC were related to each other; while in one case LCIS was related to IDC but not to the DCIS by copy number but all components shared the same CHEK2 mutation. Finally in one case none of the three components were related to each other by copy number but the LCIS and IDC shared a PIK3CA mutation, albeit at much lower allele frequency in LCIS than in IDC. Of the 8 patients with pure LCIS, 4 developed an ipsilateral invasive recurrence of various combinations of morphologies: 1 ILC & LCIS, 1 ILC & DCIS, 1 IDC & DCIS, 1 ILC & IDC, and 4 a contralateral recurrence (1 tubular, 1 IDC, 2 ILC), with a median time to recurrence of 69 months (range 34-175). The primary LCIS was related to at least one component of the recurrent disease in all four ipsilateral cases; in two the primary LCIS and all components of the recurrent disease were related, and in the remainder we observed a variety of putative evolutionary patterns. Conclusions: The majority (68%) of cases of synchronous LCIS and ILC appeared to be clonally related by copy number. 50% of cases of co-existing LCIS and IDC appeared to have a common clonal origin by either copy number or targeted sequencing. As these are genomically stable tumours, copy number data may also be underestimating relatedness. In the four cases of pure primary LCIS that developed an ipsilateral recurrence, different subtypes of breast cancer were noted as the recurrence morphology, supporting the historical view that LCIS is a risk lesion rather than a true precursor. However, in all cases the preceding LCIS was found to be related to at least one component of the subsequent invasive tumour including DCIS and IDC. This data shows that clonal relatedness between LCIS and both synchronous and asynchronous invasive disease and DCIS is more complex than previously thought, with LCIS acting as a precursor lesion even in some cases of IDC. Citation Format: Elinor Sawyer, Mateja Sborchia, Anargyros Megalios, Vandna Shah, Salpie Nowinski, Cloe Vassart, Anita Grigoriadis, Alastair Thompson, Ian Tomlinson, Rebecca Roylance, Sarah Pinder. Clonal relatedness of LCIS with synchronous and asynchronous invasive disease [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS16-03.

Integrative analysis of diverse high-dimensional molecular, histopathological and clinical data provides an effective way to identify biologically and clinically relevant subclasses across multi-level data. However, unbiased integrative methods to identify distinctive features and group structure in such data remains problematic. We propose a machine learning clustering technique based on random forest methods that enables unbiased integration. Using a permutation-based framework for the tree construction procedure and measuring of feature importance, robust and pure clusters can be produced. The performance of standard, regularised, and conditional inference random forest methods was evaluated using the adjusted Rand index, the Calinski-Harabasz index, and cluster and feature purity. In simulations studies, random forest clustering techniques were able to identify clusters of high purity. Using datasets from the UCI Machine Learning Repository as a proof of concept, all three techniques were able to identify clusters in mixed data, whereby the conditional inference method produced clusters with the highest feature purity. Next, we applied our clustering techniques to high-dimensional data obtained from two independent breast cancer studies: (i) International Cancer Genome Consortium (ICGC), consisting of 560 cases and 147 features, and (ii) Sweden Cancerome Analysis Network - Breast (SCAN-B), incorporating 241 cases and 53 features. Features included rearrangement and mutational signatures, somatic mutation in cancer drivers, germline mutations in BRCA1/2, genomic instability measures, intrinsic molecular breast cancer subtypes, and clinico-pathological characteristics. Despite dissimilarities in the breast cancer subtype composition between these two datasets, the conditional inference random forest method was able to identify concordant subgroups between the studies supported by molecular and histopathological characteristics. Moreover, novel relationships amongst molecular features with potential clinical relevance were revealed. For example, one cluster was enriched for BRCA2-deficient breast cancer cases with MYC amplifications, while another predominantly consisted of non-basal-like triple-negative breast cancers with PIK3CA mutations. Together, these results support the use of our machine learning clustering technique based on random forest methods to identify robust and biologically relevant group structures using complex high-dimensional mixed data. Citation Format: Jelmar Quist, Lawson Taylor, Johan Staaf, Anita Grigoriadis. Application of random forest machine learning techniques on mixed data from breast cancer studies [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2108.

1074 Background: In the Triple Negative Trial we observed no improved response rate (RR) to C over D in aTNBC [Tutt et al, Nat Med 2018], but we did in BRCA1/2 mutated (mut) patients (pts). We hypothesise tumors with other aberrant DNA damage response (DDR) characteristics having higher RR to DNA damage inducing C than D. Methods: We tested the predictive value of DDR process related gene expression signatures (PARPi7, chromosomal instability CIN70, TP53 & DDR Deficiency (DDRD)) on 192 treatment naïve primary tumours (PT) by total RNA-sequencing. Odds ratio (OR) for RR are reported. Paired PT & recurrent (REC) signature scores were compared. Results: Unexpectedly, high DDRD and PARPi7 were associated with higher RR to D than C ( p =0.01 & 0.06). No effect was observed for CIN70 or TP53 signature. To assess whether the unexpected results were due to biological changes 12 PT-REC pairs were available from pts who received chemotherapy (CT) between PT & REC. CIN70 increased from PT to REC, DDRD (non-significantly) & PARPi7 decreased. 4/5 TP53 wildtype classified PT samples classified as mut in REC. The BRCA1/2 & DDRD-treatment interactions only held in pts who received CT before trial entry (table). The PARPi7-treatment interaction only held in CT naïve pts. In CT naïve pts, high CIN70 tumors suggested higher C RR as hypothesized. Restricted to the 149 PAM50 basal-like pts, results were non-significant but similar trends seen. Conclusions: In this trial of aTNBC, DDRD high pts with prior CT had better RR to D than C. A possible explanation for this unexpected result is selective pressure of adjuvant DNA damaging CT and selection for relative taxane sensitivity in those who recur despite a high DDRD score. The hypothesised CIN70 treatment interaction was observed in CT naïve pts. Our results suggest care is required in application of signatures to initial diagnostic material when predicting response to DNA damaging agents at REC particularly in pts with prior CT. [Table: see text]

Purpose: Highly aggressive triple-negative breast cancers (TNBCs) lack validated therapeutic targets and have high risk of metastatic disease. Folate receptor alpha (FRα) is a central mediator of cell growth regulation that could serve as an important target for cancer therapy. Experimental Design: We evaluated FRα expression in breast cancers by genomic (n = 3,414) and IHC (n = 323) analyses and its association with clinical parameters and outcomes. We measured the functional contributions of FRα in TNBC biology by RNA interference and the antitumor functions of an antibody recognizing FRα (MOv18-IgG1), in vitro, and in human TNBC xenograft models. Results: FRα is overexpressed in significant proportions of aggressive basal like/TNBC tumors, and in postneoadjuvant chemotherapy–residual disease associated with a high risk of relapse. Expression is associated with worse overall survival. TNBCs show dysregulated expression of thymidylate synthase, folate hydrolase 1, and methylenetetrahydrofolate reductase, involved in folate metabolism. RNA interference to deplete FRα decreased Src and ERK signaling and resulted in reduction of cell growth. An anti-FRα antibody (MOv18-IgG1) conjugated with a Src inhibitor significantly restricted TNBC xenograft growth. Moreover, MOv18-IgG1 triggered immune-dependent cancer cell death in vitro by human volunteer and breast cancer patient immune cells, and significantly restricted orthotopic and patient-derived xenograft growth. Conclusions: FRα is overexpressed in high-grade TNBC and postchemotherapy residual tumors. It participates in cancer cell signaling and presents a promising target for therapeutic strategies such as ADCs, or passive immunotherapy priming Fc-mediated antitumor immune cell responses. Clin Cancer Res; 24(20); 5098–111. ©2018 AACR.

Purpose: Unprecedented clinical outcomes have been achieved in a variety of cancers by targeting immune checkpoint molecules. This preclinical study investigates heme oxygenase-1 (HO-1), an immunosuppressive enzyme that is expressed in a wide variety of cancers, as a potential immune checkpoint target in the context of a chemotherapy-elicited antitumor immune response. We evaluate repurposing tin mesoporphyrin (SnMP), which has demonstrated safety and efficacy targeting hepatic HO in the clinic for the treatment of hyperbilirubinemia, as an immune checkpoint blockade therapy for the treatment of cancer. Experimental Design: SnMP and genetic inactivation of myeloid HO-1 were evaluated alongside 5-fluorouracil in an aggressive spontaneous murine model of breast cancer (MMTV-PyMT). Single-cell RNA sequencing analysis, tumor microarray, and clinical survival data from breast cancer patients were used to support the clinical relevance of our observations. Results: We demonstrate that SnMP inhibits immune suppression of chemotherapy-elicited CD8+ T cells by targeting myeloid HO-1 activity in the tumor microenvironment. Microarray and survival data from breast cancer patients reveal that HO-1 is a poor prognostic factor in patients receiving chemotherapy. Single-cell RNA-sequencing analysis suggests that the myeloid lineage is a significant source of HO-1 expression, and is co-expressed with the immune checkpoints PD-L1/2 in human breast tumors. In vivo, we therapeutically compare the efficacy of targeting these two pathways alongside immune-stimulating chemotherapy, and demonstrate that the efficacy of SnMP compares favorably with PD-1 blockade in preclinical models. Conclusions: SnMP could represent a novel immune checkpoint therapy, which may improve the immunological response to chemotherapy. Clin Cancer Res; 24(7); 1617–28. ©2018 AACR.

The molecular complexity of triple-negative breast cancers (TNBCs) provides a challenge for patient management. We set out to characterize this heterogeneous disease by combining transcriptomics and genomics data, with the aim of revealing convergent pathway dependencies with the potential for treatment intervention. A Bayesian algorithm was used to integrate molecular profiles in two TNBC cohorts, followed by validation using five independent cohorts (n = 1,168), including three clinical trials. A four-gene decision tree signature was identified, which robustly classified TNBCs into six subtypes. All four genes in the signature (EXO1, TP53BP2, FOXM1, and RSU1) are associated with either genomic instability, malignant growth, or treatment response. One of the six subtypes, MC6, encompassed the largest proportion of tumors (∼50%) in early diagnosed TNBCs. In TNBC patients with metastatic disease, the MC6 proportion was reduced to 25%, and was independently associated with a higher response rate to platinum-based chemotherapy. In TNBC cell line data, platinum sensitivity was recapitulated, and a sensitivity to the inhibition of the phosphatase PPM1D was revealed. Molecularly, MC6-TNBCs displayed high levels of telomeric allelic imbalances, enrichment of CD4+ and CD8+ immune signatures, and reduced expression of genes negatively regulating the MAPK signaling pathway. These observations suggest that our integrative classification approach may identify TNBC patients with discernible and theoretically pharmacologically tractable features that merit further studies in prospective trials.

Cancer cells tend to metastasize first to tumor-draining lymph nodes, but the mechanisms mediating cancer cell invasion into the lymphatic vasculature remain little understood. Here, we show that in the human breast tumor microenvironment (TME), the presence of increased numbers of RORγt+ group 3 innate lymphoid cells (ILC3) correlates with an increased likelihood of lymph node metastasis. In a preclinical mouse model of breast cancer, CCL21-mediated recruitment of ILC3 to tumors stimulated the production of the CXCL13 by TME stromal cells, which in turn promoted ILC3–stromal interactions and production of the cancer cell motile factor RANKL. Depleting ILC3 or neutralizing CCL21, CXCL13, or RANKL was sufficient to decrease lymph node metastasis. Our findings establish a role for RORγt+ILC3 in promoting lymphatic metastasis by modulating the local chemokine milieu of cancer cells in the TME. Cancer Res; 77(5); 1083–96. ©2017 AACR.

Triple-negative breast cancers (TNBCs) have poor prognosis and lack targeted therapies. Here we identified increased copy number and expression of the PIM1 proto-oncogene in genomic data sets of patients with TNBC. TNBC cells, but not nonmalignant mammary epithelial cells, were dependent on PIM1 for proliferation and protection from apoptosis. PIM1 knockdown reduced expression of the anti-apoptotic factor BCL2, and dynamic BH3 profiling of apoptotic priming revealed that PIM1 prevents mitochondrial-mediated apoptosis in TNBC cell lines. In TNBC tumors and their cellular models, PIM1 expression was associated with several transcriptional signatures involving the transcription factor MYC, and PIM1 depletion in TNBC cell lines decreased, in a MYC-dependent manner, cell population growth and expression of the MYC target gene MCL1. Treatment with the pan-PIM kinase inhibitor AZD1208 impaired the growth of both cell line and patient-derived xenografts and sensitized them to standard-of-care chemotherapy. This work identifies PIM1 as a malignant-cell-selective target in TNBC and the potential use of PIM1 inhibitors for sensitizing TNBC to chemotherapy-induced apoptotic cell death.

Somatic mutations in isocitrate dehydrogenase 1 (IDH1) and IDH2 occur in gliomas and acute myeloid leukaemia (AML). Since patients with multiple enchondromas have occasionally been reported to have these conditions, we hypothesized that the same mutations would occur in cartilaginous neoplasms. Approximately 1200 mesenchymal tumours, including 220 cartilaginous tumours, 222 osteosarcomas and another ∼750 bone and soft tissue tumours, were screened for IDH1 R132 mutations, using Sequenom(®) mass spectrometry. Cartilaginous tumours and chondroblastic osteosarcomas, wild-type for IDH1 R132, were analysed for IDH2 (R172, R140) mutations. Validation was performed by capillary sequencing and restriction enzyme digestion. Heterozygous somatic IDH1/IDH2 mutations, which result in the production of a potential oncometabolite, 2-hydroxyglutarate, were only detected in central and periosteal cartilaginous tumours, and were found in at least 56% of these, ∼40% of which were represented by R132C. IDH1 R132H mutations were confirmed by immunoreactivity for this mutant allele. The ratio of IDH1:IDH2 mutation was 10.6 : 1. No IDH2 R140 mutations were detected. Mutations were detected in enchondromas through to conventional central and dedifferentiated chondrosarcomas, in patients with both solitary and multiple neoplasms. No germline mutations were detected. No mutations were detected in peripheral chondrosarcomas and osteochondromas. In conclusion, IDH1 and IDH2 mutations represent the first common genetic abnormalities to be identified in conventional central and periosteal cartilaginous tumours. As in gliomas and AML, the mutations appear to occur early in tumourigenesis. We speculate that a mosaic pattern of IDH-mutation-bearing cells explains the reports of diverse tumours (gliomas, AML, multiple cartilaginous neoplasms, haemangiomas) occurring in the same patient.

Background Breast cancers can be classified by hierarchical clustering using an "intrinsic" gene list into one of at least five molecular subtypes: basal-like, HER2, luminal A, luminal B, and normal breast-like. Five different intrinsic gene lists composed of varying numbers of genes have been used for molecular subtype identification and classification of breast cancers. The aim of this study was to determine the objectivity and interobserver reproducibility of the assignment of molecular subtype classes by hierarchical cluster analysis. Methods Three publicly available breast cancer datasets (n = 779) were subjected to two-way average-linkage hierarchical cluster analysis using five distinct intrinsic gene lists. We used free-marginal Kappa statistics to analyze interobserver agreement among five breast cancer researchers for the whole classification and for each molecular subtype separately according to each intrinsic gene list for each breast cancer dataset. Results None of the classification systems tested produced almost perfect agreement (Kappa ≥ 0.81) among observers. However, substantial interobserver agreement (70.8% to 76.1% of the samples and free-marginal Kappa scores from 0.635 to 0.701) was consistently observed in all datasets for four molecular subtypes (luminal, basal-like, HER2, and normal breast-like). When luminal cancers were subdivided (luminal A, B, and C), none of the classification systems produced substantial agreement (Kappa ≥ 0.61) in all the datasets analyzed. Analysis of each subtype separately revealed that only two (basal-like and HER2) could be reproducibly identified by independent observers (Kappa ≥ 0.81). Conclusions Assignment of molecular subtype classes of breast cancer based on the analysis of dendrograms obtained with hierarchical cluster analysis is subjective and shows modest interobserver reproducibility. For the development of a molecular taxonomy, objective definitions for each molecular subtype and standardized methods for their identification are required.

SummaryBreast cancers in BRCA1 mutation carriers frequently have a distinctive basal-like phenotype. It has been suggested that this results from an origin in basal breast epithelial stem cells. Here, we demonstrate that deleting Brca1 in mouse mammary epithelial luminal progenitors produces tumors that phenocopy human BRCA1 breast cancers. They also resemble the majority of sporadic basal-like breast tumors. However, directing Brca1 deficiency to basal cells generates tumors that express molecular markers of basal breast cancers but do not histologically resemble either human BRCA1 or the majority of sporadic basal-like breast tumors. These findings support a derivation of the majority of human BRCA1-associated and sporadic basal-like tumors from luminal progenitors rather than from basal stem cells. They also demonstrate that when target cells for transformation have the potential for phenotypic plasticity, tumor phenotypes may not directly reflect histogenesis. This has important implications for cancer prevention strategies.

Triple-Negative Breast Cancers (TNBC) are characterized by high levels of structural chromosome alterations and harbor mutational scars of defects in homologous recombination. Importantly, by acting either as drivers, or as tumor specific dependencies mutated genes in these regions represent potential therapeutic targets In search of such genes we identified that knockdown of the PUM3 gene, which is contained in a 9p24 recurrent ampliicon, impaired the growth of a subset of breast cancer but not mammary epithelial cell lines. Furthermore, the growth defects observed upon PUM3 knockdown in SUM149, a BRCA1 deficient cell line, were reversed by reactivation of BRCA1's function suggesting a BRCA1-deficiency associated dependency. PUM3 knockdown led to an increase in gH2Ax and 53BP1 nuclear foci, impaired restart of stalled replication forks, and decreased RAD51 mediated repair of collapsed replication forks. As a result PUM3 depleted cells have reduced DNA replication rate and slowed cell cycle progression. Knockdown of PUM3 also impaired replication stress induced PARylation and PAR foci formation and PUM3 and PARP1 knockdown showed phenotypic epistasis with respect to cell growth suggesting a potential mechanism for PUM3s replication fork related functions. PUM3 forms a complex with gH2Ax and the replication stalling agent hyroxyurea and PARP-inhibitors increases the nuclear concentration of PUM3. Furthermore PUM3 knockdown increased the sensitivity of breast cancer cells to these agents indicating a role for PUM3 in the repair of lesions generating by these DNA damaging agents. By studying PUM3 we have identified a novel player in replication fork restart and repair whose activity may be particularly relevant in homologous recombination deficient TNBCs. A greater understanding of PUM3's function and the pathways that regulate it may identify novel therapeutic targets for TNBCs. Importantly, direct or indirect modulation of PUM's functions may potentiate the effects of replication stress inducing therapies such as PARP-inhibitors. Citation Format: Daniel Weekes, Elodie Noel, Callum Walker, Nick Balan, Vandna Shah, Bhavna Sidhu, Anna Pardix, Stephen Pettitt, Christopher J. Lord, Anita Grigoriadis, Tutt N. Andrew. PUM3 is a triple-negative breast cancer dependency gene that functions in replication fork restart and repair [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3363.

In order to identify novel tumor drivers and gene addictions in breast cancer we have recently completed an integrated in silico/in vitro analysis of gene copy-number alterations and gene-expression profiles in a Triple Negative Breast Cancer-enriched cohort. An amplified and overexpressed gene that appeared to be a driver required for breast cancer cell proliferation is GPHR/GPR89. Several lines of evidence have recently suggested GPHR to be an anion channel, resident in the Golgi, where it controls intraluminal pH. Perturbation of GPHR function leads to altered secretory protein modification and impairment of intracellular transport. Additional data obtained in non-mammalian systems suggest that the localization of GPHR is wider and includes the endoplasmic reticulum (ER). Here, we show that GPHR is amplified and highly expressed across the breast cancer subtypes in our cohort, which we validate in the publicly available TCGA and METABRIC datasets. Using orthogonal characterization by both siRNA-shRNA and CRISPR/Cas9 gene editing, we demonstrate that GPHR down regulation impairs breast cancer cell proliferation and clonogenic ability. To further exclude the possibility that the observed phenotype was due to an “off target” effect, we also performed a rescue of function experiment overexpressing a GPHR protein resistant to GPHR shRNA that confirmed specificity. Importantly, breast cancer spheroids inducibly overexpressing GPHR are shown to have an increased area compared to control spheroids, demonstrating an important role of GPHR in breast cancer cell proliferation. Furthermore, by gradient sucrose subcellular fractionations and immunofluorescence analysis, we show that GPHR is localized in the ER of breast cancer cells. Unfolded protein response (UPR) signatures based on first-degree protein-protein interactions indicate that GPHR is positively correlated with IRE1α and negatively correlated with ATF4/CHOP networks. GPHR is also positively correlated with the XBP1 gene signature. We then further demonstrate that over expression or down regulation of GPHR are able to modulate the IRE1α/XBP1 network of proteins. In summary, in order to sustain uncontrolled growth, breast cancer cells undergo significant stress due to accelerated metabolism, deficient nutrient provision and oxidative stress. UPR signaling and GPHR expression are then essential to protect cells and to allow tumor growth by reducing protein synthesis and favoring chaperonine functions. Here we demonstrate that GPHR is over expressed across the breast cancer subtypes where it controls IRE1α/XBP1 network, and, therefore, can be a novel target for breast cancer therapy as malignant breast cells overexpressing GPHR are selectively dependent on GPHR expression. Citation Format: Riccardo Ferro, Pierfrancesco Marra, Ana Mendes Pereira, Virinder Reen, Cynthia Prince, Sumi Mathew, Alessandra Facchetti, Priyanka Ghongane-Salpe, Sedigeh Kareemaghay, Patrycja Gazinska, David Robertson, Farzana Noor, Fara Braso' Maristany, Daniel Weekes, Dragomir Krastev, Tencho Tenev, Hasan Mirza, Nirmesh Patel, Anita Grigoriadis, Andrew Tutt. The anion channel GPHR/GPR89 regulates protein-folding homeostasis by regulating the IRE1α pathway in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 443.

Introduction Triple-negative breast cancers (TNBCs) are characterised by increased tumour-infiltrating lymphocytes and the presence of complex compositions of structural and single nucleotide variations. We previously described HORMAD1 , a Cancer/Testis (CT) antigen, as a novel driver of homologous recombination (HR) deficiency. Our ultimate aim was to identify biomarkers in HORMAD1 positive TNBCs that could be used to stratify patients for immunomodulatory and/or DNA damage response (DDR) targeted agents. In order to do this, we first defined the transcriptomics and genomics features of HORMAD1 positive TNBCs. Material and methods HORMAD1 positive TNBCs were identified using either microarray or RNA-sequencing data. Gene set enrichment analysis was used to identify hyper- or hypo-activated pathways. Whole-exome sequencing (WES) from TCGA TNBCs was used to identify genes exclusively mutated in HORMAD1 positive TNBCs. Mutational signatures in tumours were delineated using whole-genome sequencing from ICGC TNBCs. Results and discussions In six independent TNBC cohorts (total n=719) HORMAD1 expression was bimodal. More than 50% of the TNBCs were identified as HORMAD1 positive and tended to display PAM50 basal-like, intClust 10, or a basal-like 1 TNBCtype-4 subtype (P value HORMAD1 positive TNBCs, genes involved in DNA repair by HR or DNA mismatch repair displayed increased expression (Q value PIK3CA were found exclusively in HORMAD1 negative TNBCs (P value=0.002). Mutational signature 3 and rearrangement signature 5, previously associated with HR deficiency in BRCA1- and BRCA2 -deficient tumours, were more prevalent in HORMAD1 positive ICGC TNBCs (n=72, Q value=0.049 and 0.028 respectively), as was an increase in substitution burden (P value=0.032). HORMAD1 positive TNBCs also displayed a transcriptomic signature reminiscent of activated CD4 +T cells (Q value HORMAD1 expression, or its role as a CT antigen, or both, could drive an adaptive immune response, thus explaining this CD4 +T cell signature. Conclusion Taken together, the molecular profiles of HORMAD1 positive TNBCs indicate not only the underlying biology of this TNBC subset, but also opportunities for therapeutic exploitation, including agents that target the DDR and immune system.

Introduction Active surveillance (AS) for men with low-risk prostate cancer (PCa) currently depends on repeated biopsies for prognostication. We aim to identify molecular patterns that are present in initial PCa biopsies that inform future histopathological upgrading of PCa for men on AS. Material and methods TAPS2.0 and Patchwork were respectively used to establish somatic copy number aberrations (SCNAs) for 44 Gleason score (GS)6 (3+3) TCGA PRAD Affymetrix SNP6.0 and 18 GS6 (3+3) ICGC WGS samples. Elastic net regularisation was used to identify SCNA predictors of biochemical recurrence (BCR). In an internal cohort, SCNAs were established using QDNAseq from initial and repeat biopsies of 5 men on AS, using low-pass WGS. Four of the 5 men upgraded from a GS6 to GS7 (3+4) upon repeat biopsy whilst one remained GS6. Results and discussions SCNAs were robustly established for 62 GS6 TCGA and ICGC PCas. The most frequent copy number losses were at 8 p (50%), 13q (29%) and 6q (26%). Applying a Shannon-diversity index on SCNAs provided a measure of genomic heterogeneity and indicated a trend towards increased diversity in PCa of men with BCR (p-value=0.087, t-test). Six cytobands with recurrent SCNAs were identified as predictive of BCR, including a loss on 10q11 and a gain on 2 p13. These predictive SCNAs were interrogated in genomes from our internal cohort of initial and repeat biopsies. Copy number losses were identified at 8 p (46%), 13q (23%), and 6q (46%), and were comparable in frequency to losses in TCGA and ICGC. None of the six predictors were detected in the initial PCa biopsy of the stable patient, while two predictors were found in men who had upgraded to GS7: 14q13 (gain), present in two men, and 6q11 (loss), in another man. Patients that upgraded had the highest similarities (81%) of SCNAs between patient-matched initial and repeat biopsies. Given that these biopsies were taken furthest apart in the prostate gland, missampling and GS is likely not to affect prediction. Conclusion SCNAs indicative of upgrading were identified in GS6 TCGA-ICGC data using BCR as a surrogate for histopathological upgrading whilst on AS. These were observed to some extent in an internal cohort of men on AS. Validation of these predictors in men on AS is currently on-going in a further 8 men and may provide the basis for developing a prognostication tool to guide therapy for men with early prostate cancer.

e15129Background: Unprecedented clinical outcomes have been achieved in a variety of cancers by targeting the immune checkpoint molecules PD-1 and CTLA-4. However, as a significant number of patien...

Cancer/testis (CT) genes are predominantly expressed in human germ line cells, but not somatic tissues, and frequently become activated in different cancer types. Several CT antigens have already proved to be useful biomarkers and are promising targets for therapeutic cancer vaccines. The aim of the present study was to investigate the expression of CT antigens in breast cancer. Using previously generated massively parallel signature sequencing (MPSS) data, together with 9 publicly available gene expression datasets, the expression pattern of CT antigens located on the X chromosome (CT-X) was interrogated. Whereas a minority of unselected breast cancers was found to contain CT-X transcripts, a significantly higher expression frequency was detected in estrogen and progesterone receptor (ER) negative breast cancer cell lines and primary breast carcinomas. A coordinated pattern of CT-X antigen expression was observed, with MAGEA and NY-ESO-1/CTAG1B being the most prevalent antigens. Immunohistochemical staining confirmed the correlation of CT-X antigen expression and ER negativity in breast tumors and demonstrated a trend for their coexpression with basal cell markers. Because of the limited therapeutic options for ER-negative breast cancers, vaccines based on CT-X antigens might prove to be useful.

Purpose: Perilobar nephrogenic rests (PLNRs) are abnormally persistent foci of embryonal immature blastema that have been associated with dysregulation at the 11p15 locus by genetic/epigenetic means and are thought to be precursor lesions of Wilms tumor. The precise genomic events are, however, largely unknown. Experimental Design: We used array comparative genomic hybridization to analyze a series of 50 PLNRs and 25 corresponding Wilms tumors characterized for 11p15 genetic/epigenetic alterations and insulin-like growth factor-II expression. Results: The genomic profiles of PLNRs could be subdivided into three categories: those with no copy number changes (22 of 50, 44%); those with single, whole chromosome alterations (8 of 50, 16%); and those with multiple gains/losses (20 of 50, 40%). The most frequent aberrations included 1p- (7 of 50, 14%) +18 (6 of 50, 12%), +13 (5 of 50, 10%), and +12 (3 of 50, 6%). For the majority (19 of 25, 76%) of cases, the rest harbored a subset of the copy number changes in the associated Wilms tumor. We identified a temporal order of genomic changes, which occur during the insulin-like growth factor-II/PLNR pathway of Wilms tumorigenesis, with large-scale chromosomal alterations such as 1p-, +12, +13, and +18 regarded as “early” events. In some of the cases (24%), the PLNRs harbored large-scale copy number changes not observed in the concurrent Wilms tumor, including +10p, +14q, and +18. Conclusions: These data suggest that although the evidence for PLNRs as precursors is compelling, not all lesions must necessarily undergo malignant transformation.

Background Immunohistochemical assessment of proliferation may provide additional prognostic information in early breast cancer. However, due to a lack of methodological standards proliferation markers are still not routinely used for determining therapy. Even for Ki67, one of the most widely-studied markers, disagreements over the optimal cutoff exist. Improvements in digital microscopy may provide new avenues to standardise and make data more reproducible. Methods We studied the immunohistochemical expression of three markers of proliferation: Ki67, Mini-Chromosome Maintenance protein 2 and Geminin, by conventional light microscope and digital imaging on triplicate TMAs from 309 consecutive cases of primary breast cancers. Differences between the average and the maximum percentage reactivity in tumour cell nuclei from the three TMA cores were investigated to assess the validity of the approach. Time-dependent Receiver Operating Characteristic curves were utilized to obtain optimal expression level cut-offs, which were then correlated with clinico-pathological features and survival. Results High concordance between conventional and digital scores was observed for all 3 markers (Ki67: rs = 0.87, P

Breast cancers with BRCA1 germline mutation have a characteristic DNA copy number (CN) pattern. We developed a test that assigns CN profiles to be ‘BRCA1‐like’ or ‘non‐BRCA1‐like’, which refers to resembling a BRCA1‐mutated tumor or resembling a tumor without a BRCA1 mutation, respectively. Approximately one third of the BRCA1‐like breast cancers have a BRCA1 mutation, one third has hypermethylation of the BRCA1 promoter and one third has an unknown reason for being BRCA1‐like. This classification is indicative of patients' response to high dose alkylating and platinum containing chemotherapy regimens, which targets the inability of BRCA1 deficient cells to repair DNA double strand breaks. We investigated whether this classification can be reliably obtained with next generation sequencing and copy number platforms other than the bacterial artificial chromosome (BAC) array Comparative Genomic Hybridization (aCGH) on which it was originally developed. We investigated samples from 230 breast cancer patients for which a CN profile had been generated on two to five platforms, comprising low coverage CN sequencing, CN extraction from targeted sequencing panels (CopywriteR), Affymetrix SNP6.0, 135K/720K oligonucleotide aCGH, Affymetrix Oncoscan FFPE (MIP) technology, 3K BAC and 32K BAC aCGH. Pairwise comparison of genomic position‐mapped profiles from the original aCGH platform and other platforms revealed concordance. For most cases, biological differences between samples exceeded the differences between platforms within one sample. We observed the same classification across different platforms in over 80% of the patients and kappa values of at least 0.36. Differential classification could be attributed to CN profiles that were not strongly associated to one class. In conclusion, we have shown that the genomic regions that define our BRCA1‐like classifier are robustly measured by different CN profiling technologies, providing the possibility to retro‐ and prospectively investigate BRCA1‐like classification across a wide range of CN platforms., Highlights We investigated concordance of BRCA1‐like classification of copy number profiles across technologies.We included array‐based and (targeted) next generation sequencing profiling techniques.BRCA1‐like classification was highly concordant across techniques and datasets.10% misclassification can be attributed to poorer quality, weak class association and experimental variation.We concluded that BRCA1‐like classification of breast cancer is robust across techniques and datasets.

The normal duct-lobular system of the breast is lined by two epithelial cell types, inner luminal secretory cells and outer contractile myoepithelial cells. We have generated comprehensive expression profiles of the two normal cell types, using immunomagnetic cell separation and gene expression microarray analysis. The cell-type specificity was confirmed at the protein level by immunohistochemistry in normal breast tissue. New prognostic markers for survival were identified when the luminal- and myoepithelial-specific molecules were evaluated on breast tumor tissue microarrays. Nuclear expression of luminal epithelial marker galectin 3 correlated with a shorter overall survival in these patients, and the expression of SPARC (osteonectin), a myoepithelial marker, was an independent marker of poor prognosis in breast cancers as a whole. These data provide a framework for the interpretation of breast cancer molecular profiling experiments, the identification of potential new diagnostic markers, and development of novel indicators of prognosis.

Triple negative breast cancers (TNBCs) lack oestrogen (ER), progesterone (PR) and human epidermal growth factor 2 (HER2) receptors, and have limited targeted treatment options. Large scale genomic and transcriptomic studies have advanced our understanding of the changes which occur in TNBCs. However, the substantial number of copy number and gene expression alterations present in TNBCs makes it difficult to identify putative drivers, biomarkers and/or therapeutic targets of the disease. To overcome this problem, we have carried out an integrative computational and RNAi based approach to identify genes required for proliferation of TNBC. Copy number and gene expression alterations were analysed using Affymetrix Human Exon HTS1.0 and SNP6.0 data of 152 primary breast tumours enriched for a TNBC phenotype and 9 normal breast epithelium. These analyses revealed 141 candidate genes whose upregulated gene expression is copy number driven in TNBC. The functional dependence on each of these genes was subsequently examined using RNAi in an array of 17 breast cancer and non-malignant cell lines using 6-8 cell lines per gene covering the widest possible range of expression levels for that gene. We validated a malignant cell specific functional dependence on 37 of the 141 genes using this method. STRING analysis of validated hits reveals a subset of genes involved in the process of cell division and mitosis including the previously characterised mitotic kinase TTK. Of these, we further validated KIFC1 (HSET) which is known to play a role in clustering supernumerary centrosomes, a common occurrence in breast cancer. We show that siRNA and shRNA mediated depletion of KIFC1 decreases cell viability and clonogenic ability specifically in centrosome amplified cell lines, which can be rescued upon introduction of an si/shRNA resistant KIFC1. KIFC1 depletion also produces a high level of catastrophic multipolar mitoses in centrosome amplified but not in non-amplified cell lines. Furthermore, in-vivo studies show that inducible depletion of KIFC1 suppresses tumour growth of centrosome amplified cell line xenografts. Our work has identified and functionally validated novel drivers, and potential therapeutic targets in TNBC. The data presented here shows KIFC1, a druggable kinesin motor protein is a promising target for therapeutic intervention being expressed through gene copy gain in a significant proportion of TNBCs. We validate its role in cancer-specific amplified centrosome clustering showing KIFC1 plays an essential role in aiding the survival of breast cancer cells that have supernumerary centrosomes in both in-vitro and in-vivo contexts. Citation Format: Nirmesh S. Patel, Konstantinos Drosopoulos, Daniel Weekes, Elodie Noel, Hasan Mirza, Mamunur Rashid, Emanuele de Rinaldis, Fara Brasó Maristany, Sumi Mathew, Erika Francesch Domenech, Patrycja Gazinska, Farzana Noor, Jelmar Quist, Rebecca Marlow, Anita Grigoriadis, Spiros Linardopoulos, Andrew N. Tutt. An integrated copy number and gene expression genomics analysis and RNAi approach identifies and validates the KIFC1 kinesin as a malignant cell selective target in triple negative breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 67.

Triple-Negative Breast Cancers (TNBCs) are aggressive, associated with poor prognosis and lack targeted therapies, mainly due to the absence of suitable molecular targets. The genomic region 6p21-p25 is recurrently amplified in TNBCs and encompasses the PIM1 oncogene. This study interrogated genomic breast cancer datasets and identified gene copy number-driven increase of PIM1 expression in TNBC. To understand the role of PIM1 in malignant phenotypes of TNBCs, functional studies were carried out in breast cancer and non-malignant mammary epithelial cell line models. RNA interference and rescue-of-function experiments revealed addiction to PIM1 kinase for cell population growth and apoptosis protection in TNBC cells, absent in non-malignant cells. In cells sensitive to PIM1 knockdown, we observed a subsequent reduction of the expression of the anti-apoptotic proteins BCL2 and MCL1. Moreover, exogenous expression of BCL2 prevented apoptosis caused by PIM1 knockdown. BH3-profiling analysis further confirmed that PIM1 blocks apoptosis elicited through the mitochondrial pathway in TNBC cell lines. The activity of PIM1 in other cancers has been closely linked to the regulation of c-MYC, an “un-targetable” oncogene that drives malignancy and that is frequently amplified and highly expressed in primary TNBCs and post-chemotherapy residual disease. Importantly, we found that PIM1 expression associates with several MYC-transcriptional signatures in TNBCs and mediates phosphorylation of c-MYC Ser62 and Histone-H3 Ser10, key targets for MYC-driven transcription. Exogenous expression of MYC rescued the phenotypes caused by PIM1 knockdown, providing mechanistic insights for the observed addiction. Finally, the therapeutic potential of targeting PIM1 was assessed using AZD1208, a small molecule inhibitor of the PIM kinase family. AZD1208 inhibited growth and sensitized TNBC derived cell lines, cell line xenografts and PDXs to chemotherapy by increasing apoptosis. We therefore identify PIM1 as a driver of malignancy in TNBC, illustrating relationships with MYC-activation and regulation of anti-apoptotic BCL2 and MCL1 proteins. PIM1 inhibitors could provide a potential therapeutic to abrogate chemotherapy resistance in TNBCs. Citation Format: Fara Braso-Maristany, Simone Filosto, Steven Catchpole, Rebecca Marlow, Jelmar Quist, Erika Francesch Domenech, Gianmaria Liccardi, Leila Zakka, Violeta Serra, Albert Gris, Maggie Cheang, Nirmesh Patel, Anna Perdrix Rosell, Patrycja Gazinska, Elodie Noel, Jonathan Watkins, Pierfrancesco Marra, Anita Grigoriadis, Andrew Tutt. PIM1, a novel target in chemotherapy-resistant triple-negative breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 178.

The glycome acts as an essential interface between cells and the surrounding microenvironment. However, changes in glycosylation occur in nearly all breast cancers, which can alter this interaction. Here, we report that profiles of glycosylation vary between ER-positive and ER-negative breast cancers. We found that genes involved in the synthesis of sialyl-Lewis x (sLex; FUT3, FUT4, and ST3GAL6) are significantly increased in estrogen receptor alpha-negative (ER-negative) tumors compared with ER-positive ones. SLex expression had no influence on the survival of patients whether they had ER-negative or ER-positive tumors. However, high expression of sLex in ER-positive tumors was correlated with metastasis to the bone where sLex receptor E-selectin is constitutively expressed. The ER-positive ZR-75-1 and the ER-negative BT20 cell lines both express sLex but only ZR-75-1 cells could adhere to activated endothelial cells under dynamic flow conditions in a sLex and E-selectin–dependent manner. Moreover, L/P-selectins bound strongly to ER-negative MDA-MB-231 and BT-20 cell lines in a heparan sulfate (HS)–dependent manner that was independent of sLex expression. Expression of glycosylation genes involved in heparan biosynthesis (EXT1 and HS3ST1) was increased in ER-negative tumors. Taken together, our results suggest that the context of sLex expression is important in determining its functional significance and that selectins may promote metastasis in breast cancer through protein-associated sLex and HS glycosaminoglycans. Cancer Res; 71(24); 7683–93. ©2011 AACR.

The disialoganglioside GD3 is overexpressed in ∼50% of invasive ductal breast carcinoma, and the GD3 synthase gene (ST8SIA1) displays higher expression among estrogen receptor–negative breast cancer tumors, associated with a decreased overall survival of breast cancer patients. However, no relationship between ganglioside expression and breast cancer development and aggressiveness has been reported. We have previously shown that overexpression of GD3 synthase induces the accumulation of b- and c-series gangliosides (GD3, GD2, and GT3) at the cell surface of MDA-MB-231 breast cancer cells together with the acquisition of a proliferative phenotype in the absence of serum. Here, we show that phosphoinositide 3-kinase/Akt and mitogen-activated protein kinase/extracellular signal-regulated kinase pathways are constitutively activated in GD3 synthase–expressing cells. Analysis of phosphorylation of tyrosine kinase receptors shows a specific c-Met constitutive activation in GD3 synthase–expressing cells, in the absence of its ligand, hepatocyte growth factor/scatter factor. In addition, inhibition of c-Met or downstream signaling pathways reverses the proliferative phenotype. We also show that GD3 synthase expression enhances tumor growth in severe combined immunodeficient mice. Finally, a higher expression of ST8SIA1 and MET in the basal subtype of human breast tumors are observed. Altogether, our results show that GD3 synthase expression is sufficient to enhance the tumorigenicity of MDA-MB-231 breast cancer cells through a ganglioside-dependent activation of the c-Met receptor. Mol Cancer Res; 8(11); 1526–35. ©2010 AACR.

As genome-scale technologies begin to unravel the complexity of the equivalent tumors in adults, we can attempt detailed characterization of high-grade gliomas in children, that have until recently been lacking. Toward this end, we sought to validate and extend investigations of the differences between pediatric and adult tumors. Purpose: As genome-scale technologies begin to unravel the complexity of the equivalent tumors in adults, we can attempt detailed characterization of high-grade gliomas in children, that have until recently been lacking. Toward this end, we sought to validate and extend investigations of the differences between pediatric and adult tumors. Experimental Design: We carried out copy number profiling by array comparative genomic hybridization using a 32K bacterial artificial chromosome platform on 63 formalin-fixed paraffin-embedded cases of high-grade glioma arising in children and young people (, National Health Service funding to the NIHR Biomedical Research Centre. This work was supported by The Royal Marsden Children's Department Fund, Fundação para a Ciência e Tecnologia, Portugal, and Breakthrough Breast Cancer

Amplification of fibroblast growth factor receptor 1 (FGFR1) occurs in ∼10% of breast cancers and is associated with poor prognosis. However, it is uncertain whether overexpression of FGFR1 is causally linked to the poor prognosis of amplified cancers. Here, we show that FGFR1 overexpression is robustly associated with FGFR1 amplification in two independent series of breast cancers. Breast cancer cell lines with FGFR1 overexpression and amplification show enhanced ligand-dependent signaling, with increased activation of the mitogen-activated protein kinase and phosphoinositide 3-kinase–AKT signaling pathways in response to FGF2, but also show basal ligand-independent signaling, and are dependent on FGFR signaling for anchorage-independent growth. FGFR1-amplified cell lines show resistance to 4-hydroxytamoxifen, which is reversed by small interfering RNA silencing of FGFR1, suggesting that FGFR1 overexpression also promotes endocrine therapy resistance. FGFR1 signaling suppresses progesterone receptor (PR) expression in vitro, and likewise, amplified cancers are frequently PR negative, identifying a potential biomarker for FGFR1 activity. Furthermore, we show that amplified cancers have a high proliferative rate assessed by Ki67 staining and that FGFR1 amplification is found in 16% to 27% of luminal B–type breast cancers. Our data suggest that amplification and overexpression of FGFR1 may be a major contributor to poor prognosis in luminal-type breast cancers, driving anchorage-independent proliferation and endocrine therapy resistance. Cancer Res; 70(5); 2085–94

International audience; Breast cancer is a heterogeneous disease caused by the accumulation of genetic changes in neoplastic cells. We hypothesised that molecular subtypes of breast cancer may be driven by specific constellations of genes whose expression is regulated by gene copy number aberrations. To address this question, we analysed a series of 48 microdissected grade III ductal carcinomas using high resolution microarray comparative genomic hybridisation and mRNA expression arrays. There were 5,931 genes whose expression significantly correlates with copy number identified; out of these, 1,897 genes were significantly differentially expressed between basal-like, HER2 and luminal tumours. Ingenuity Pathway Analysis (IPA) revealed that 'G1/S cell cycle regulation' and 'BRCA1 in DNA damage control' pathways were significantly enriched for genes whose expression correlates with copy number and are differentially expressed between the molecular subtypes of breast cancer. IPA of genes whose expression significantly correlates with copy number in each molecular subtype individually revealed that canonical pathways involved in oestrogen receptor (ER) signalling and DNA repair are enriched for these genes. We also identified 32, 157 and 265 genes significantly overexpressed when amplified in basal-like, HER2 and luminal cancers, respectively. These lists include known and novel potential therapeutic targets (e.g. and in HER2 cancers). Our results provide strong circumstantial evidence that different patterns of genetic aberrations in distinct molecular subtypes of breast cancer contribute to their specific transcriptomic profiles and that biological phenomena characteristic of each subtype (e.g. proliferation, HER2 and ER signalling) may be driven by specific patterns of copy number aberrations.

Background: Although paediatric high grade gliomas resemble their adult counterparts in many ways, there appear to be distinct clinical and biological differences. One important factor hampering the development of new targeted therapies is the relative lack of cell lines derived from childhood glioma patients, as it is unclear whether the well-established adult lines commonly used are representative of the underlying molecular genetics of childhood tumours. We have carried out a detailed molecular and phenotypic characterisation of a series of paediatric high grade glioma cell lines in comparison to routinely used adult lines. Principal Findings: All lines proliferate as adherent monolayers and express glial markers. Copy number profiling revealed complex genomes including amplification and deletions of genes known to be pivotal in core glioblastoma signalling pathways. Expression profiling identified 93 differentially expressed genes which were able to distinguish between the adult and paediatric high grade cell lines, including a number of kinases and co-ordinated sets of genes associated with DNA integrity and the immune response. Significance: These data demonstrate that glioma cell lines derived from paediatric patients show key molecular differences to those from adults, some of which are well known, whilst others may provide novel targets for evaluation in primary tumours. We thus provide the rationale and demonstrate the practicability of using paediatric glioma cell lines for preclinical and mechanistic studies., (undefined)

Purpose: To characterize the molecular genetic profiles of grade 3 invasive ductal carcinomas of no special type using high-resolution microarray-based comparative genomic hybridization (aCGH) and to identify recurrent amplicons harboring putative therapeutic targets associated with luminal, HER-2, and basal-like tumor phenotypes. Experimental Design: Ninety-five grade 3 invasive ductal carcinomas of no special type were classified into luminal, HER-2, and basal-like subgroups using a previously validated immunohistochemical panel. Tumor samples were microdissected and subjected to aCGH using a tiling path 32K BAC array platform. Selected regions of recurrent amplification were validated by means of in situ hybridization. Expression of genes pertaining to selected amplicons was investigated using quantitative real-time PCR and gene silencing was done using previously validated short hairpin RNA constructs. Results: We show that basal-like and HER-2 tumors are characterized by “sawtooth” and “firestorm” genetic patterns, respectively, whereas luminal cancers were more heterogeneous. Apart from confirming known amplifications associated with basal-like (1q21, 10p, and 12p), luminal (8p12, 11q13, and 11q14), and HER-2 (17q12) cancers, we identified previously unreported recurrent amplifications associated with each molecular subgroup: 19q12 in basal-like, 1q32.1 in luminal, and 14q12 in HER-2 cancers. PPM1D gene amplification (17q23.2) was found in 20% and 8% of HER-2 and luminal cancers, respectively. Silencing of PPM1D by short hairpin RNA resulted in selective loss of viability in tumor cell lines harboring the 17q23.2 amplification. Conclusions: Our results show the power of aCGH analysis in unraveling the genetic profiles of specific subgroups of cancer and for the identification of novel therapeutic targets.

Tumour cell lines derived from breast cancer patients constitute one of the cornerstones of breast cancer research. To characterise breast cancer cell lines at the genetic level, we have developed a full tiling path bacterial artificial chromosome (BAC) array collection for comparative genomic hybridisation (aCGH). This aCGH BAC collection covers 98% of the entire human genome at a resolution of 40-60 kbp. We have used this platform alongside an in-house produced 17 K cDNA microarray set to characterise the genetic and transcriptomic profiles of 24 breast cancer cell lines, as well as cell types derived from non-diseased breast. We demonstrate that breast cancer cell lines have genomic and transcriptomic features that recapitulate those of primary breast cancers and can be reliably subclassified into basal-like and luminal subgroups. By overlaying aCGH and transcriptomic data, we have identified 753 genes whose expression correlate with copy number; this list comprised numerous oncogenes recurrently amplified and overexpressed in breast cancer (e.g., HER2, MYC, CCND1 and AURKA). Finally, we demonstrate that although breast cancer cell lines have genomic features usually found in grade III breast cancers (i.e., gains of 1q, 8q and 20q), basal-like and luminal cell lines are characterised by distinct genomic aberrations.

Wilms tumor karyotypes frequently exhibit recurrent, large-scale chromosomal imbalances, among the most common of which are concurrent loss of 1p and gain of 1q. We have previously identified a novel breakpoint at 1p13 by 1 Mb-spaced array CGH, and have now undertaken a fine-tiling oligonucleotide array approach to map the region accurately in four tumors exhibiting rearrangements at this locus. The use of a 10 bp-spaced platform revealed that all four tumors in fact harbored different breakpoints, which targeted intragenic sequences in PHTF1, DCLRE1B, and NRAS, and an intergenic region immediately downstream of TRIM33. All four genes and breakpoints were within the 1.78 Mb intervals identified by the genome-wide BAC arrays. The precise breakpoint interval was in each case mapped to a 200-1,200 bp region and was confirmed for one case to lie within intron 3 of DCLRE1B by quantitative PCR. Analysis of local genome architecture revealed no convincing conservation of repetitive sequences or specific translocation/recombination-associated elements within the breakpoint regions. This study highlights the power of fine-tiling oligonucleotide arrays to delineate breakpoint regions identified by genome-wide screens.