Search

Colorectal cancer (CRC) has one of the highest cancer incidences and mortality rates. In stage III, postoperative chemotherapy benefits

Control of Mycobacterium tuberculosis (Mtb) infection requires the generation of T cells that migrate to granulomas, specialized immune structures surrounding sites of bacterial replication. Here we report that persistent influx of neutrophils impedes T-cell recruitment to the lung and limits bacterial control in tuberculosis (TB) granulomas. We found that type I interferons (IFN-1) induced by Mtb infection dampens T-cell production in the bone marrow by driving a granulocyte-biased hematopoietic program. Neutrophil-intrinsic IFN-1 signaling, and bacterial type VII secretion system are critical for this process. Mechanistically, we demonstrate that IFN-1 facilitates neutrophil influx to the lung, where these cells produce granulocyte colony stimulating factor (G-CSF) in response to Mtb and neutrophil extracellular traps (NETs). Augmented and sustained release of G-CSF support granulopoiesis and impair lymphopoiesis, causing sub-optimal T-cell responses in the lung and spleen. Therapeutically, inhibiting IFN-1 signaling or neutralizing G-CSF limits neutrophil-driven immunopathology and bacterial replication by restoring optimal T-cell responses. Thus, we identify a mechanism by which neutrophils limit T-cell immunity during TB by disrupting the hematopoietic equillibrium and propose “IFN-1→NETosis→GCSF” immune axis as targets for host-directed therapy and immunization

Cancer immunotherapy can result in lasting tumor regression, but predictive biomarkers of treatment response remain ill-defined. Here, we performed single-cell proteomics, transcriptomics, and genomics on matched untreated and IL2 injected metastases from patients with melanoma. Lesions that completely regressed following intralesional IL2 harbored increased fractions and densities of nonproliferating CD8+ T cells lacking expression of PD-1, LAG-3, and TIM-3 (PD-1−LAG-3−TIM-3−). Untreated lesions from patients who subsequently responded with complete eradication of all tumor cells in all injected lesions (individuals referred to herein as “extreme responders”) were characterized by proliferating CD8+ T cells with an exhausted phenotype (PD-1+LAG-3+TIM-3+), stromal B-cell aggregates, and expression of IFNγ and IL2 response genes. Loss of membranous MHC class I expression in tumor cells of untreated lesions was associated with resistance to IL2 therapy. We validated this finding in an independent cohort of metastatic melanoma patients treated with intralesional or systemic IL2. Our study suggests that intact tumor-cell antigen presentation is required for melanoma response to IL2 and describes a multidimensional and spatial approach to develop immuno-oncology biomarker hypotheses using routinely collected clinical biospecimens.

Supplementary Data from Tumor MHC Class I Expression Associates with Intralesional IL2 Response in Melanoma

Cancer immunotherapy can result in lasting tumor regression, but predictive biomarkers of treatment response remain ill-defined. Here, we performed single-cell proteomics, transcriptomics, and genomics on matched untreated and IL2 injected metastases from patients with melanoma. Lesions that completely regressed following intralesional IL2 harbored increased fractions and densities of nonproliferating CD8+ T cells lacking expression of PD-1, LAG-3, and TIM-3 (PD-1−LAG-3−TIM-3−). Untreated lesions from patients who subsequently responded with complete eradication of all tumor cells in all injected lesions (individuals referred to herein as “extreme responders”) were characterized by proliferating CD8+ T cells with an exhausted phenotype (PD-1+LAG-3+TIM-3+), stromal B-cell aggregates, and expression of IFNγ and IL2 response genes. Loss of membranous MHC class I expression in tumor cells of untreated lesions was associated with resistance to IL2 therapy. We validated this finding in an independent cohort of metastatic melanoma patients treated with intralesional or systemic IL2. Our study suggests that intact tumor-cell antigen presentation is required for melanoma response to IL2 and describes a multidimensional and spatial approach to develop immuno-oncology biomarker hypotheses using routinely collected clinical biospecimens.

PURPOSE It has been estimated that about 25% of patients with inflammatory bowel disease (IBD) are diagnosed before the age of 18. Food insecurity, limited or lack of access to enough food, has been associated with poor outcomes in childhood health including mental health problems. United States Department of Agriculture (USDA) defines a food desert as a census tract that meets criteria for low access to a grocery store and low income. Depression and anxiety have been well documented in pediatric patients with IBD. Functional Disability Inventory (FDI) is a measure of limitations in both physical and psychosocial functioning secondary to physical health. Our hypothesis is that patients in low access areas to healthy food will have more evidence of depression, anxiety and higher disability scores. METHODS This study is a retrospective chart review of patients diagnosed with Crohn's Disease (CD) or ulcerative colitis (UC) between 2019 to 2022. Patient’s addresses were entered into the USDA food access atlas to determine if they lived in an area of low access. Psychology screening included depression screening (PHQ-9), anxiety screening (GAD-7), and functional disability inventory (FDI). Screening questionnaires including PHQ-9, GAD-7, FDI were completed if patients were 12 years of age or old. If 11 years of old or less, parents completed FDI screening forms. RESULTS Our study consisted of 57 patients with an average age of 12.8 years at diagnosis. Thirty-three patients (58%) were non-Hispanic white, 19 (33%) were African American, and 3 were Hispanic (5%). Depression screening was positive in 22 patients (48%) and anxiety screening was positive in 25 patients (60%). Of those living in low access areas, 45% met criteria for depression, 53% met criteria for anxiety. There were comparable rates of depression between the low access group compared to the adequate access group (45% vs 50%, respectively) and rates of anxiety (53% vs. 61%, respectively). Patients in areas of low access had higher disability scores (9.88±11.9) than those in adequate access (5.88±6.1). When considering those who lived in a food desert (11 total), 8 patients were African American, 2 were non-Hispanic white, and 1 was self-described as other. Patients in the food desert group had higher rates of depression (78%) and anxiety (70%) compared to the low access group. Patients in the food desert group had the highest FDI scores (12.7 ±9.7). CONCLUSION Pediatric IBD patients had higher rate of depression and anxiety; However, there was no relationship between food access and depression or anxiety. When considering those in a food desert, the rate of depression increased to 78% and anxiety increased to 70%. Patients living in a food desert had the highest FDI scores. More information is needed on how multiple psychosocial factors might affect depression in children with IBD.

PURPOSE Inflammatory bowel disease (IBD) is an autoimmune disease that consists of Crohn’s Disease (CD) and ulcerative colitis (UC). Food insecurity is defined as limited or uncertain access to enough food. It is estimated that 1 in 7 children in the United States experience food insecurity or about 13 million children. Louisiana has one of the higher rates of food insecurity in the country with over 249,000 children affected. Food insecurity has been associated with worse health outcomes in children including developmental risks, poor mental health, and poor educational outcomes. Functional Disability Inventory (FDI) is a measure of limitations in both physical and psychosocial functioning secondary to physical health. Our hypothesis is that patients in food insecure homes will have higher FDI scores compared to those in food secure homes. METHODS This study is a prospective study of newly diagnosed IBD patients seen by the Division of Pediatric Gastroenterology at Children’s Hospital of New Orleans. Patients were screened for food insecurity using the United States Department of Agriculture’s food insecurity screening questionnaire and the American Academy of Pediatrics’ Food Insecurity 2-Question Screen. Disability scores were obtained at diagnosis using the FDI self-reporting questionnaire if older than 12 years of age or a parent form was completed for patients less than 12. RESULTS Our study consisted of 20 newly diagnosed pediatric IBD patients with an average age of 14.5 years. Nineteen (95%) patients had CD. Most patients were female (70%), non-Hispanic white (50%) with Medicaid (55%) as their primary insurance. Eight (40%) patients were African American and two (10%) were listed as other. Eight patients or their caregivers identified as living in a food insecure home (40%) and these patients averaged a FDI score of 10.2 The food secure group had an average FDI of 9.3. Non-Hispanic, white patients had an average FDI score of 10.5 while the minority patients had an average FDI of 8.5. Only two patients lived in an area with a median household income below poverty with an average FDI of 15.5. Ten patients lived below the state median income and had an average FDI was 8.2. Those patients living in areas above the state median income level, 8 total, had the highest FDI scores, 10.8. CONCLUSION Food insecurity had higher rates of disability scores compared to families from food secure homes. Families in higher income areas and who were non-Hispanic white had higher FDI scores compared to minority patients and those in lower income areas. Further data will be collected with changes in FDI through the first 6 months of diagnosis. Additional larger studies are needed to determine a relationship between FDI scores and food insecurity in pediatric IBD.

Mucinous rectal cancer is associated with a higher incidence of microsatellite instability, and a poorer response to neoadjuvant chemoradiotherapy compared to other subtypes of rectal adenocarcinoma. Immune checkpoint inhibitors are an emerging family of anti-cancer therapeutics associated with highly variable outcomes in colorectal cancer. Though the immune landscape of mucinous rectal cancer has not been fully explored, the presence of mucin is thought to act as a barrier preventing immune cell infiltration.The aim of this study was to determine the immune properties of mucinous rectal cancer and investigate the degree of lymphocyte infiltration in this cohort.This is a retrospective cohort study which involved, multiplexed immunofluorescence staining of tumor microarrays.Samples originated from a single university teaching hospital.Our cohort included 15 cases of mucinous and 43 cases of non-mucinous rectal cancer.Immune cells were classified and quantified. Immune cell counts were compared between mucinous and non-mucinous cohorts. Immune marker expression within tumor epithelial tissue was evaluated to determine degree of lymphocyte infiltration.Cytotoxic (p = 0.022), and regulatory T-cells (p = 0.010) were found to be overrepresented in the mucinous cohort compared to the non-mucinous group. PD-1 expression was also found to be significantly greater in the mucinous group (p = 0.001). CD3 (p = 0.001) and CD8 (p = 0.054) expression within tumor epithelium was also higher in the mucinous group, suggesting adequate immune infiltration despite the presence of mucin. Microsatellite instability status was not found to be a predictor of immune marker expression in our analysis.The relatively small size of the cohort.Mucinous rectal cancer is associated with an immune rich tumor microenvironment, which was not associated with microsatellite instability status. See Video Abstract at http://links.lww.com/DCR/C65.

There is currently an urgent need to identify factors predictive of immunogenicity in colorectal cancer (CRC). Mucinous CRC is a distinct histological subtype of CRC, associated with a poor response to chemotherapy. Recent evidence suggests the commensal facultative anaerobe Fusobacterium may be especially prevalent in mucinous CRC. The objectives of this study were to assess the impact of Fusobacterium prevalence on immune cell expression and prognosis in mucinous CRC. Our study included two independent colorectal cancer patient cohorts, The Cancer Genome Atlas (TCGA) cohort, and a cohort of rectal cancers from the Beaumont RCSI Cancer Centre (BRCC). Multiplexed immunofluorescence staining of a tumor microarray (TMA) from the BRCC cohort was undertaken using Cell DIVE technology. Our cohorts included 87 cases (13.3%) of mucinous and 565 cases (86.7%) of non-mucinous CRC. Mucinous CRC in the TCGA dataset was associated with increased CD8 + lymphocyte (p = 0.018), regulatory T-cell (p = 0.001) and M2 macrophage (p = 0.001) expression. Similarly in the BRCC cohort, mucinous RC was associated with enhanced CD8 + lymphocyte (p = 0.022), regulatory T-cell (p = 0.047), and B-cell (p = 0.025) counts. Elevated Fusobacterium expression was associated with increased CD4+ (p = 0.031) and M1 macrophage (p = 0.006) expression, whilst M2 macrophages (p = 0.043) were under-expressed in the TCGA cohort. Increased Fusobacterium relative abundance in mucinous CRC was associated with improved clinical outcomes in our TCGA cohort despite having no association with MSI status (DSS: likelihood ratio p = 0.04, logrank p = 0.052). Fusobacterium abundance is associated with improved outcomes in mucinous CRC, possibly due its modulatory effect on the host immune response.

Background There is limited knowledge about DCIS cellular composition and relationship with breast cancer events (BCE). Methods Immunofluorescence multiplexing (MxIF) was used to image and quantify 32 cellular biomarkers in FFPE DCIS tissue microarrays. Over 75,000 DCIS cells from 51 patients (median 9 years follow-up for non-BCE cases) were analysed for profiles predictive of BCE. K-means clustering was used to evaluate cellular co-expression of epithelial markers with ER and HER2. Results Only ER, PR and HER2 significantly correlated with BCE. Cluster analysis identified 6 distinct cell groups with different levels of ER, Her2, cMET and SLC7A5. Clusters 1 and 3 were not significant. Clusters 2 and 4 (high ER/low HER2 and SLC7A5/mixed cMET) significantly correlated with low BCE risk (P = 0.001 and P = 0.034), while cluster 6 (high HER2/low ER, cMET and SLC7A5) correlated with increased risk (P = 0.018). Cluster 5 (similar to cluster 6, except high SLC7A5) trended towards significance (P = 0.072). A continuous expression score (Escore) based on these 4 clusters predicted likelihood of BCE (AUC = 0.79, log-rank test P = 5E–05; LOOCV AUC = 0.74, log-rank test P = 0.006). Conclusion Multiplexed spatial analysis of limited tissue is a novel method for biomarker analysis and predicting BCEs. Further validation of Escore is needed in a larger cohort.

Mycobacterium tuberculosis (Mtb) infection induces persistent influx of neutrophils that associates with poor bacterial control and clinical outcome from tuberculosis (TB). Although implicated in TB pathology, the mechanism by which these cells contribute to pathogenesis is poorly understood. Using Cell-DIVE multiplexed immunofluorescence imaging and spatial analysis of inflammatory TB lesions, we demonstrated that persistent neutrophil infiltration affects the spatiotemporal organization of T-lymphocytes and impairs their function. Instead of directly suppressing T-cells, neutrophils produce granulocyte colony stimulating factor (CSF3/G-CSF) that collaborates with type I interferon (IFN-I) to promote a granulocyte-skewed hematopoiesis impacting T-lymphocyte production. Importantly, neutrophil-intrinsic IFN-I receptor 1 (IFNAR1) is both necessary and sufficient to promote pathologic granulopoiesis. Finally, inhibition of IFNAR1-signaling alone mitigates immunopathogenesis by restoring hematopoietic equilibrium. Collectively, our work uncovers a potential immunevasion strategy by which virulent Mtb strains induce IFN-I to generate pathogen-permissive neutrophils that produce G-CSF and sustain pathogenic hematopoiesis to impair T-cell immunity during TB.

Mapping the human body at single cell resolution in three-dimensions (3D) is an important step toward a “digital twin” model that captures important structure and dynamics of cell-cell interactions. Current 3D imaging methods suffer from low resolution and are limited in their ability to distinguish cell types and their spatial relationships. We present a novel 3D workflow: MATRICS-A (Multiplexed Image Three-D Reconstruction and Integrated Cell Spatial - Analysis) that generates a 3D map of cells from multiplexed images and calculates cell type distance from endothelial cells and other features of interest. We applied this workflow to multiplexed data from sequential skin sections from younger and older donors (n=10; 33-72 years) with biopsies from ten anatomical regions with different sun exposure effects (mild, moderate-marked). Up to 26 sequential sections from each sample underwent multiplexed imaging with 18 biomarkers covering 12 cell types (keratinocytes (granular, spinous, basal), epithelial and myoepithelial cells, fibroblasts, macrophages, T helpers, T killers, T regs, neurons and endothelial cells, markers of DNA damage and repair (p53, DDB2) and cell proliferation (Ki67). Following cell classification, the tissue and classified cells were reconstructed into 3D volumes. A significant inverse correlation between DDB2 positive cells and age was found (corr= -0.78, adj. p=0.047). This suggests reduced capacity for repair in non-cancer older sun-exposed individuals. While absolute immune cell count did not differ by age or sun exposure, the ratio of T Helper/T Killer cells was positively correlated with age (corr=0.82, adj. p=0.048) This is the first such 3D study in skin and paves the way for cataloging more cell types and spatial relationships in aging and disease in skin and other organs.

Standard clinicopathological parameters (age, growth pattern, tumor size, margin status and grade) have been shown to have limited value in predicting recurrence in ductal carcinoma in situ (DCIS) patients. Early and accurate recurrence prediction would facilitate a more aggressive treatment policy for high-risk patients (mastectomy or adjuvant radiation therapy), and simultaneously reduce over-treatment of low-risk patients. Generative adversarial networks (GAN) are a class of DL models in which two adversarial neural networks, generator and discriminator, compete with each other to generate high quality images. In this work, we have developed a deep learning (DL) classification network that predicts breast cancer events (BCEs) in DCIS patients using hematoxylin and eosin (H&E) images. The DL classification model was trained on 67 patients using image patches from the actual DCIS cores and GAN generated image patches to predict breast cancer events (BCEs). The hold-out validation dataset (n= 66) had an AUC of 0.82. Bayesian analysis further confirmed the independence of the model from classical clinico-pathological parameters. DL models of H&E images may be used as a risk stratification strategy for DCIS patients to personalize therapy.Simple SummaryDuctal carcinoma in situ (DCIS) patients have an excellent overall survival rate and over-treatment is always a cause for concern due to potential side-effects. Standard clinicopathological parameters have limited value in predicting breast cancer events (BCEs) and stratification of high and low risk patients. Herein, we have developed a deep learning (DL) classification framework to predict BCEs in DCIS patients. A generative adversarial network (GAN) augmented deep learning (DL) classification of histological features associated with aggressive disease was trained on hematoxylin and eosin (H&E) tissue microarray (TMA) images of DCIS to predict BCEs. The area under the curve (AUC) for BCE’s in the validation set was 0.82. Early and accurate prediction of DCIS BCEs would facilitate a personalized approach to therapy.

Recurrence Prediction in Ductal Carcinoma In Situ (DCIS) Patients Using Generative Adversarial Network (GAN) Augmented Deep Learning Model Background: DCIS patients have an excellent overall survival rate and over-treatment is always a cause for concern due to potential side-effects. Standard clinicopathological factors (age, growth pattern, tumor size, margin status and grade) have been shown to have limited value in predicting recurrence and segregation of high and low risk patients. Early and accurate recurrence prediction would facilitate a more aggressive treatment policy for high-risk patients (mastectomy or adjuvant radiation therapy), and simultaneously reduce over-treatment of low-risk patients. In this work, we have developed a deep learning (DL) classification framework that predicts recurrence in DCIS patients from Tissue microarrays (TMAs) hematoxylin and eosin (H&E) images using a generative adversarial network (GAN) augmented deep learning (DL) classification model. A GAN is a class of DL models, in which two adversarial neural networks, generator and discriminator contest among each other to generate high quality images. During the adversarial training process, the generator learns to synthesize realistic images similar to those in the training set while the discriminator learns to distinguish between real and generated images. In recent years, high quality medical images have been generated by GAN models. To the best of our knowledge, this is the first time a GAN model has been used to generate H&E images to train a DL classification model to predict recurrence in DCIS patients. Materials and methods: The cohort was comprised of 68 DCIS patients, aged between 35-89 years, lesion size of 5-90 mm, with a mix of low (15%), intermediate (35%) and 50% high grade cases. Patients were treated with mastectomy and/or a combination of lumpectomy, radiation and hormone therapy. TMAs were constructed from 2mm cores (1-3 cores per patient) in consultation with a breast pathologist to create hematoxylin and eosin (H&E) images for further analysis. The cohort was split into independent training (n=50 patients, 10 with recurrences at 5years) and validation groups (n=18 patients, 6 with recurrences at 5years). TMA (H&E) images were divided into smaller image patches of size 256x256 to train a GAN to generate image patches. A DL classification network (Resnet-Inception v2) was trained using TMA image patches and aggressive image patches generated by GAN to predict recurrence. The ability to generate synthetic image patches of aggressive lesions permitted training of a large DL classification network and predict recurrence in DCIS patients. Importantly, manual annotation was not necessary for the process. Results: The DL classification model trained with both TMA and GAN generated image patches predicted recurrence with an AUC of 0.87, sensitivity of 0.83 and specificity of 0.91 in the validation dataset. The DL classification model trained with image patches from TMAs only predicted recurrence with an AUC of 0.81. Conclusions: The use of a GAN model to generate H&E images circumvents the needs for a large cohort and accurate labor-intensive manual annotation of histopathological images, which is often required for training a large DL classification model. The use of GAN generated aggressive image patches during training significantly improves recurrence prediction accuracy of the DL classification model. Validation in independent larger cohorts is ongoing, and if successful, could provide a novel assay for risk prediction that does not waste precious tissue samples. Citation Format: Ghose Soumya, Yesim Gokmen-Polar, Sanghee Cho, Elizabeth McDonough, Cynthia Davis, Jhimli Mitra, Zhanpan Zhang, Fiona Ginty, Sunil Badve. Recurrence Prediction in Ductal Carcinoma In Situ (DCIS) Patients from Tissue Microarrays (TMAs) [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 P6-04-10.

Tissues and organs are composed of distinct cell types that must operate in concert to perform physiological functions. Efforts to create high-dimensional biomarker catalogs of these cells are largely based on transcriptomic single-cell approaches that lack the spatial context required to understand critical cellular communication and correlated structural organization. To probe in situ biology with sufficient coverage depth, several multiplexed protein imaging methods have recently been developed. Though these antibody-based technologies differ in strategy and mode of immunolabeling and detection tags, they commonly utilize antibodies directed against protein biomarkers to provide detailed spatial and functional maps of complex tissues. As these promising antibody-based multiplexing approaches become more widely adopted, new frameworks and considerations are critical for training future users, generating molecular tools, validating antibody panels, and harmonizing datasets. In this perspective, we provide essential resources and key considerations for obtaining robust and reproducible multiplexed antibody-based imaging data compiling specialized knowledge from domain experts and technology developers., Comment: {\ddag}authors contributed equally to this work

Cancer cells’ ability to inhibit apoptosis is key to malignant transformation and limits response to therapy. Here, we performed multiplexed immunofluorescence analysis on tissue microarrays with 373 cores from 168 patients, segmentation of 2.4 million individual cells, and quantification of 18 cell lineage and apoptosis proteins. We identified an enrichment for BCL2 in immune, and BAK, SMAC, and XIAP in cancer cells. Ordinary differential equation-based modeling of apoptosis sensitivity at single-cell resolution was conducted and an atlas of inter- and intra-tumor heterogeneity in apoptosis susceptibility generated. Systems modeling at single-cell resolution identified an enhanced sensitivity of cancer cells to mitochondrial permeabilization and executioner caspase activation compared to immune and stromal cells, but showed significant inter- and intra-tumor heterogeneity.

Colorectal cancer (CRC) has one of the highest cancer incidences and mortality rates. In stage III, postoperative chemotherapy benefits in situ multiplexed immunofluorescence imaging and single cell analysis technology (Cell DIVE™). Tissue microarrays (TMAs) with up to three 1mm diameter cores per patient were prepared from 117 stage III CRC patients treated with adjuvant fluoropyrimidine/oxaliplatin chemotherapy. Single sections underwent multilplexed immunofluorescence with Cy3- and Cy5-conjugated antibodies for immune cell markers (CD45, CD3, CD4, CD8, FOXP3, PD1) and cell segmentation markers (DAPI, pan-cytokeratin, AE1, NaKATPase and S6). We applied a probabilistic multi-class, multi-label classification algorithm based on multi-parametric models to build statistical models of protein expression to classify immune cells. Expert annotations of immune cell markers were made on a range of images, and Support Vector Machines (SVM) were used to derive a statistical model for cell classification. Images were also manually scored independently by a Pathologist as ‘high’, ‘moderate’ or ‘low’, for stromal and total immune cell content. Excellent agreement was found between manual and total automated scores (p

Colorectal cancer (CRC) has one of the highest cancer incidences and mortality rates. In stage III, postoperative chemotherapy benefits20% of patients, while more than 50% will develop distant metastases. Biomarkers for identification of patients at increased risk of disease recurrence following adjuvant chemotherapy are currently lacking. In this study, we assessed immune signatures in the tumor and tumor microenvironment (TME) using an in situ multiplexed immunofluorescence imaging and single-cell analysis technology (Cell DIVE

Over the last 15 years, advances in immunofluorescence-imaging based cycling methods, antibody conjugation methods, and automated image processing have facilitated the development of a high-resolution, multiplexed tissue immunofluorescence (MxIF) method with single cell-level quantitation termed Cell DIVETM. Originally developed for fixed oncology samples, here it was evaluated in highly fixed (up to 30 days), archived monkeypox virus-induced inflammatory skin lesions from a retrospective study in 11 rhesus monkeys to determine whether MxIF was comparable to manual H-scoring of chromogenic stains. Six protein markers related to immune and cellular response (CD68, CD3, Hsp70, Hsp90, ERK1/2, ERK1/2 pT202_pY204) were manually quantified (H-scores) by a pathologist from chromogenic IHC double stains on serial sections and compared to MxIF automated single cell quantification of the same markers that were multiplexed on a single tissue section. Overall, there was directional consistency between the H-score and the MxIF results for all markers except phosphorylated ERK1/2 (ERK1/2 pT202_pY204), which showed a decrease in the lesion compared to the adjacent non-lesioned skin by MxIF vs an increase via H-score. Improvements to automated segmentation using machine learning and adding additional cell markers for cell viability are future options for improvement. This method could be useful in infectious disease research as it conserves tissue, provides marker colocalization data on thousands of cells, allowing further cell level data mining as well as a reduction in user bias.

PURPOSE Inflammatory bowel disease (IBD), consisting of both Crohn’s Disease (CD) and ulcerative colitis (UC), affects nearly 80,000 children in the United States. African American patients have more perirectal phenotype which might indicate more severe disease activity however it is unclear why. There is no published data on transportation access and how it relates to pediatric IBD. The purpose of this study is to compare pediatric IBD patients with low access to care to those with more access to care and their severity of disease at diagnosis. Our hypothesis is that those patients who live farther from the children’s hospital and bus stops will have more of a lag time in their diagnosis as defined by two or more visits with another provider for complaints concerning for IBD compared to those patients with more ease of access to care. We aim to assess the impact of proximity to care facilities, bus stops and primary children’s hospital and disease severity at the time of diagnosis for pediatric IBD population. METHODS This study is a retrospective chart review of patients seen by the Louisiana State University Health Science Center Pediatric Gastroenterology division at Children’s Hospital New Orleans from 2019 until 2021. A list of patients with the diagnosis of either CD or UC after 2019 was generated. Both qualitative and quantitative data was collected. Patient’s addresses were used to determine the distance to their primary gastroenterologist clinic. Data collected also included their proximity location to bus stops, number of visits prior to diagnosis of IBD. RESULTS There were 71 pediatric patients diagnosed during the study period. The patients were predominantly male (51%) and white (65%) with average age at diagnosis of 12 years old (± 3.3 years). Forty-eight (68%) patients were diagnosed with CD. More Crohn’s Disease patients had 2 or more visits before diagnosis compared to those with UC (48% versus 30%). The average distance for African American patients to the children’s hospital was 50.5 miles (± 55.9 miles) which was similar to the white patients (53.0 ± 59.8 miles). Of the 23 African American patients included in the study 14 (61%) had at least 2 visits before being diagnosed with IBD. In comparison, 15 of the 46 (33%) white patients had at least 2 visits before receiving a diagnosis of IBD. Most patients had a pediatrician listed as their primary care provider for both African American and white patients (70% versus 80%). CONCLUSION The significance of a delay in diagnosis and pediatric IBD is unclear, however further studies to determine if less access of care is predictive of greater disease severity and activity at presentation may be warranted. Additional studies will be completed to determine the relationship between proximity to bus stops and disease burden at diagnosis.

Glioma is recognized to be a highly heterogeneous CNS malignancy, whose diverse cellular composition and cellular interactions have not been well characterized. To gain new clinical- and biological-insights into the genetically-bifurcated IDH1 mutant (mt) vs wildtype (wt) forms of glioma, we integrated multiplexed immunofluorescence single cell data for 43 protein markers across cancer hallmarks, in addition to cell spatial metrics, genomic sequencing and magnetic resonance imaging (MRI) quantitative features. Molecular and spatial heterogeneity scores for angiogenesis and cell invasion differ between IDHmt and wt gliomas irrespective of prior treatment and tumor grade; these differences also persisted in the MR imaging features of peritumoral edema and contrast enhancement volumes. Longer overall survival for IDH1mt glioma patients may reflect generalized altered cellular, molecular, spatial heterogeneity which manifest in discernable radiological manifestations.

Colorectal cancer (CRC) has one of the highest Worldwide incidences and mortality rates. Genotoxic chemotherapy following surgery in stage III patients confers treatment benefit to less than 20% of the patients, with more than 50% of stage III patients going on to develop distant metastases. Currently, there are no predictive biomarkers that can identify which stage III patients will recur, which patients will benefit from chemotherapy and which should be redirected towards alternative therapeutic interventions. A major challenge in identifying such a universal biomarker is that CRC is a heterogeneous disease with multiple subtypes. In the current study, we assessed clinically-relevant immune cell populations in the tumour microenvironment (TME) of stage III tumours using a novel hyperplex in situ immunofluorescence imaging technology (Cell DIVETM, GE Healthcare, Issaquah, WA). Tissue microarrays (TMAs) with up to three 1mm diameter cores per patient were prepared from 139 stage III CRC patients treated with adjuvant FOLFOX chemotherapy. Single sections (5 µm) were iteratively stained with Cy3- and Cy5-conjugated antibodies for immune cell markers as well as markers of cell death and metabolism. The images underwent illumination correction, DAPI-based registration and autofluorescence removal. After image quality control corrections, single cell segmentation was performed using a combination of DAPI [nuclear], pan-cytokeratin [epithelial], NaKATPase [membrane] and S6 [cytoplasmic] segmentation markers and an average of ~3,000 stromal cells and ~ 4,000 epithelial were segmented per tumour core. A machine learning-based algorithm for immune cell classification and quantification was used to analyse the immune markers CD45, CD3, CD4, CD8, FOXP3 and PD1 to identify: cytotoxic T cells, T helper cells, regulatory T cells and potential relevance of immune checkpoint therapy. In the tumour tissues, the median proportion of CD3+ segmented cells was ~8%. Classified immune cells were counted within epithelial and stromal regions, with patients categorised as Low, Intermediate and High (based on 75th percentile, respectively) for each cell type. Preliminary survival analyses show that patients with ‘CD8 High' intratumoural cytotoxic T cells have better Disease-Free Survival compared to ‘CD8 Low' patients in this FOLFOX-treated cohort. By combining single-cell data with clinicopathological patient data, we aim to identify immune-, cell death- and metabolism-related signatures that can predict benefit from adjuvant FOLFOX chemotherapy for Stage III CRC patients. Citation Format: Xanthi Stachtea, Andreas Lindner, Manuela Salvucci, Sanghee Cho, Anup Sood, Elizabeth McDonough, Alberto Santamaria-Pang, John Graf, Philip Dunne, Mark Lawler, Jochen Prehn, Fiona Ginty, Daniel Longley. Hyperplexed immunofluorescence analysis (Cell DIVETM) of immune-related tumor heterogeneity in stage III colorectal cancer [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 2676.

Glioma is recognized to be a highly heterogeneous CNS malignancy, whose diverse cellular composition and cellular interactions have not been well characterized. To gain new clinical- and biological-insights into the genetically-bifurcated IDH1 mutant (mt) vs wildtype (wt) forms of glioma, we integrated data from protein, genomic and MR imaging from 20 treatment-naïve glioma cases and 16 recurrent GBM cases. Multiplexed immunofluorescence (MxIF) was used to generate single cell data for 43 protein markers representing all cancer hallmarks, Genomic sequencing (exome and RNA (normal and tumor) and magnetic resonance imaging (MRI) quantitative features (protocols were T1-post, FLAIR and ADC) from whole tumor, peritumoral edema and enhancing core vs equivalent normal region were also collected from patients. Based on MxIF analysis, 85,767 cells (glioma cases) and 56,304 cells (GBM cases) were used to generate cell-level data for 24 biomarkers. K-means clustering was used to generate 7 distinct groups of cells with divergent biomarker profiles and deconvolution was used to assign RNA data into three classes. Spatial and molecular heterogeneity metrics were generated for the cell data. All features were compared between IDH mt and IDHwt patients and were finally combined to provide a holistic/integrated comparison. Protein expression by hallmark was generally lower in the IDHmt vs wt patients. Molecular and spatial heterogeneity scores for angiogenesis and cell invasion also differed between IDHmt and wt gliomas irrespective of prior treatment and tumor grade; these differences also persisted in the MR imaging features of peritumoral edema and contrast enhancement volumes. A coherent picture of enhanced angiogenesis in IDHwt tumors was derived from multiple platforms (genomic, proteomic and imaging) and scales from individual proteins to cell clusters and heterogeneity, as well as bulk tumor RNA and imaging features. Longer overall survival for IDH1mt glioma patients may reflect mutation-driven alterations in cellular, molecular, and spatial heterogeneity which manifest in discernable radiological manifestations.

Colorectal cancer (CRC) has one of the highest worldwide incidences and mortality rates. About 25% of stage II patients develop recurrence within 5 years. However, use of adjuvant chemotherapy yields minimal therapeutic benefit. Clinical decision-making in this population would be informed by prognostic and predictive biomarkers that help indicate more targeted interventions and/or intensified disease monitoring. Higher cMET expression has been shown to be a prognostic marker in CRC and has been evaluated as a drug target in several cancers, including CRC. cMET acts as the receptor for hepatocyte growth factor (HGF), and is associated with increased proliferation, migration and morphogenesis of epithelial cells via activation of multiple downstream pathways including PI3K/AKT, MAPK and the NF-kB pathways. Here, we investigated whether increased cellular expression of multiple markers in the cMET/adjacent pathways was correlated with clinical outcome. The study population consisted of 283 patients with stage II CRC, without neoadjuvant treatment. Using a multiplexed immunofluorescence method (MxIF) with cell-level quantification (Cell DIVETM, GE Healthcare), 41 biomarkers in the cMET, mTOR, MAPK and associated pathways, and lymphocyte markers were analyzed in an iterative sequence of staining, imaging and dye inactivation, followed by image registration, cell segmentation and biomarker intensity-quantitation. Poor quality images and cells were filtered based on poor segmentation and tissue quality following manual review. Images that did not perfectly register were also excluded. For epithelial cell analysis, all stromal cells were filtered, resulting in a total of 559,952 epithelial cells. K-means clustering was conducted on cMET and related pathways (cMET, pGSK3β, 4EBP1, p4EBP1, S6, pS6, pStat3, pp38MAPK, pNFkBp65, pNFkBp105, EGFR, pERK1/2, HER2, IGF1R, CA-IX, Glut1, SLC7A5 and Ki67). Cluster analysis identified 6 cell clusters with varying expression levels of cMET and other pathway markers. A cluster corresponding to low cMET network activation (cMETLow) was correlated with reduced risk of recurrence (cox pH model Likelihood ratio test p-value = 0.056). Furthermore, MMR proficient patients (who had a higher recurrence rate) had a significantly greater proportion of cells with elevated cMET and related pathway marker (cMETHigh) s(p= 0.007). They also had a significantly lower fraction of infiltrating helper T cells (CD3+CD4+, pvalue 0.076) and CD68+ cells in the epithelial region, compared to MMR deficient patients (p=0.005). Further research will include continuing in-depth analysis of MMR status, cMET pathway and lymphocyte response as well as the role of cellular heterogeneity. In summary, comprehensive cMET pathway analysis using a multiplexed single cell approach indicates for the first time an association between low cMET network activation and superior clinical outcomes. Citation Format: Elizabeth McDonough, Anup Sood, Fiona Ginty, Sanghee Cho, John Graf, Jochen Prehn, Philip Dunne, Andreas Lindner, Manuela Salvucci, Daniel Longley, Mark Lawler, Travis Hollmann, Jinru Shia. Lower cellular activation of cMET signaling network is associated with reduced recurrence risk in stage II colorectal cancer [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr A039. doi:10.1158/1535-7163.TARG-19-A039

Apoptosis is essential for chemotherapy responses. We previously developed the mathematical models APOPTO-CELL and DR_MOMP to estimate the magnitude of tumor cells resilience to undergo and execute apoptosis in response to chemotherapy, and demonstrated that these models are prognostic markers of clinical responses in stage 3 colorectal cancer patients [Salvucci M et al., Clin Cancer Res 2017; Lindner AU et al., Gut 2018]. However, the heterogeneity of colorectal cancer tissue represents a hurdle that we have yet to overcome on our way to entirely understand differences in response and clinical outcome in patients. Quantification of protein levels as input for systems studies routinely reflects a local cell population average rather than protein levels in individual cells. In this study we aim to gain a better understanding of apoptosis resilience in tumors with respect to tumor cell heterogeneity and cell composition and local aspects of the tumor microenvironment. Two to three core slices of formalin-fixed and paraffin-embedded primary colorectal cancer tissue were collected from 165 stage II-IV patients from a previously used cohort [Salvucci M et al. 2017; Lindner AU et al. 2018]. The levels of 25 protein markers for cell type and immune status, apoptosis, metabolism and cell proliferation, as well as cell segmentation markers, were measured on TMA slides using a multiplexed tissue immunofluorescence imaging platform (Cell DIVE [Gerdes et al., Proc Natl Acad Sci USA 2013]). Batch correction between slides was performed using linear regression (using markers for regulatory T cell in patient tissues as housekeeping cells. Formalin-fixed gene deficient cells were used in parallel for antibody validation. The layout allowed us to create protein profiles for on average 13,079 (± 2,282 SD) cells per core per patient, and to categorize these as epithelial-like cells, regulatory T cells, T helper cells, cytotoxic T cells, other leukocytes and other cells. The data will be used to identify cells contributing most to the signal measured with traditional methods (such as reverse protein phase arrays) that were used to predict patients clinical response to therapy in our previous studies. Further, we will assess spatial differences in distribution and frequency of different cells and their contribution to clinical outcome in colorectal cancer patients. Specifically, we are investigating metrics to characterize the spatial features and layout found in each core in terms of textures, degree of lacunarity, neighboring cells composition and patterns of protein expression (diffuse vs. patchy) and co-expression among cell types. Further, we will characterize the degree of intra- and inter-patient spatial heterogeneity and will examine the association between the identified spatial motifs with macroscopical characteristics of the tumor and outcome. Citation Format: Andreas Urlich Lindner, Manuela Salvucci, Xanthi Stachtea, Steven Carberry, Philip D. Dunne, Anup Sood, Elizabeth McDonough, Sanghee Cho, Pierre Laurent-Puig, Sandra Van Schaeybroeck, Manuel Salto-Tellez, John F. Graf, Markus Rehm, Mark Lawler, Daniel B. Longley, Fiona Ginty Fiona, Jochen H. Prehn. Exploratory multiplex tissue image analysis of the impact of heterogeneity in the microenvironment of primary colorectal cancer on apoptosis markers in patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-088.

e14289 Background: Immune checkpoint inhibitors have changed the way we treat most malignant tumors especially lung carcinomas. While PD-L1 is the only FDA approved companion diagnostic used to predict response, its accuracy and specificity is lacking. Thus, new biomarkers and mechanisms of checkpoint inhibitor need to be elucidated. A major limitation with the current approved method, immunohistochemistry, is that it can typically only measure one parameter per tissue section. Currently new multiplexing microscopy techniques are being developed that allow us to see multiple parameters at the cell level, which can answer new questions regarding tumor immune microenvironments (TIME). In this study we used a novel multiplexed immunofluorescent (MxIF) microscopy technique, Cell DIVE™ (GE Healthcare, Issaquah, WA), to analyze formalin fixed paraffin embedded (FFPE) tissue samples from small cell lung carcinoma (SCLC) patients treated with a PD-L1 inhibitor, durvalumab (MEDI4736), in combination with a PARP inhibitor, olaparib. Methods: Biopsies from patients enrolled in a phase II expansion cohort clinical trial (NCT02484404) treated with durvalumab and olaparib were analyzed. Tumor immune response from a cohort of 3 responders and 3 non-responders, before and after treatment, underwent MxIF analysis using biomarkers including CD3, CD4, CD8, CD20, CD39, CD45RA, CD45RO, CD56, CD57, CD103, FOXP3, GATA3, LAMP1, PD-L1 and T-bet. Our panel was able to effectively discern T-cells (including subsets, maturation, and function), B-cells, and NK cells and their interaction with tumor cells. The results were validated for accuracy and consistency using immunohistochemistry and Vectra Polaris Imaging System. Results: Spatial analysis of TIME showed responders (3/3) were associated with an inflamed CD8+ T-cell tumor microenvironment while non-responders (3/3) displayed an excluded or desert CD8+ T-cell tumor microenvironment. MxIF/Cell DIVE™ allowed us to further characterize the immune response. Conclusions: Analyzing the type and pattern of immune response using multiplexed immunofluorescent and single cell characterization microscopy may better predict prognosis than PD-L1 immunohistochemistry in SCLC patients treated with immune checkpoint inhibitors. Clinical trial information: NCT02484404.

Cytomegalovirus (CMV) duodenitis is a rare occurrence, especially in pediatric patients. A thirteen-month-old female presented to the Emergency Department for a febrile seizure. She was incidentally admitted for severe malnutrition with an initial workup remarkable for only a slight elevation in her ALT at 48. The patient was found to have an oral aversion requiring nasogastric tube feeds for adequate caloric intake. She continued to fail to gain weight and underwent an EGD that demonstrated a duodenal ulcer. She was consequently started on sucralfate and omeprazole. Post-EGD lab work demonstrated a pronounced increase in AST and ALT. Pathology from the EGD biopsies later demonstrated viral inclusion bodies consistent with CMV duodenitis. Apart from malnutrition, other causes of immune deficiency were eliminated from the differential diagnosis due to negative HIV PCR and normal immunoglobulins. While on antiviral treatment, her viral load of 1080 IU/mL trended to resolution and her liver enzymes normalized. The patient was ultimately discharged home demonstrating adequate weight gain via gastrostomy tube feeds. This case advocates for pediatricians to include immunodeficiency and infectious etiologies in their differential for malnourished patients in order to lead to earlier diagnosis and management of this treatable condition.

The phenotypic diversity of cancer results from genetic and nongenetic factors. Most studies of cancer heterogeneity have focused on DNA alterations, as technologies for proteomic measurements in clinical specimen are currently less advanced. Here, we used a multiplexed immunofluorescence staining platform to measure the expression of 27 proteins at the single-cell level in formalin-fixed and paraffin-embedded samples from treatment-naive stage II/III human breast cancer. Unsupervised clustering of protein expression data from 638,577 tumor cells in 26 breast cancers identified 8 clusters of protein coexpression. In about one-third of breast cancers, over 95% of all neoplastic cells expressed a single protein coexpression cluster. The remaining tumors harbored tumor cells representing multiple protein coexpression clusters, either in a regional distribution or intermingled throughout the tumor. Tumor uptake of the radiotracer 18F-fluorodeoxyglucose was associated with protein expression clusters characterized by hormone receptor loss, PTEN alteration, and HER2 gene amplification. Our study demonstrates an approach to generate cellular heterogeneity metrics in routinely collected solid tumor specimens and integrate them with in vivo cancer phenotypes.

Over the last three decades, the incidence of early-stage breast cancer has doubled while the incidence of late stage breast cancer has slightly decreased; however, the standard treatment for both has not changed significantly in the last 10 years. Of those diagnosed with early-stage breast cancer, 30% will develop recurrent disease, though there is currently no diagnostic available to identify these cases. Conversely, identifying late stage patients with a low- risk for recurrence could spare them from post-surgical chemotherapy. We have developed a technology that allows in situ staining and analysis of up to 60 proteins on a single tissue slide. In this exploratory study we evaluated 25 known biomarkers on a cohort of more than 800 breast cancer cases assembled onto three tissue arrays. The cohort spans clinical grades and breast cancer subtypes, and is thus useful in novel associations of biomarker expression across a variety of clinical features. Twenty-five markers were chosen based on literature reports for marker association with disease recurrence and represent a range of biological features including signaling pathways and cell phenotypes. The slides were sequentially stained with antibodies that were directly conjugated to Cy3 and Cy5 dyes, imaged, evaluated for quality and subject inclusion, and subsequently analyzed for statistical correlations of expression values to clinical features. The images were then processed using a single cell analysis algorithm which allows for quantification of individual tumor cells and signal intensities for the markers to be extracted from sub-cellular domains (nucleus, cytoplasm, and membrane of each cell) allowing unique patterns of protein expression to be determined. Univariate Cox proportional hazards analysis was first applied to determine those features with strongest association with ‘death due to disease’. Of the 25 markers analyzed, several markers (e.g. API3, AMPH2, beta-catenin, CD44, CEACAM5, CK15, CK19, cMET, Her2, Ki67, TRIM29) were found to have strong association with poor outcome (FDR < 0.05). After adjusting for clinical variables (stage, number of positive nodes, grade), a number of markers (e.g. CD44, CEACAM5, cMET, TRIM29) still show a strong association with poor clinical outcome (FDR < 0.05, Cox proportional hazards model). CEACAM5 is part of the MammostratTM panel of markers used for determining prognosis in ER positive breast cancers after Tamoxifen® treatment. CD44, a cancer stem cell marker, and TRIM 29, a transcriptional regulatory factor involved in carcinogenesis, have been associated with basal-like breast cancers which have the worst prognosis of all the breast cancer subtypes. High levels of cMET have been associated with poor prognosis for all breast cancer subtypes and cMET is being explored as a therapeutic target. The MultiomyxTM methodology enabled us to verify a subset of biomarkers that are clinically relevant to breast cancer outcome. Using this technology could lead to identification of novel biosignatures that stratify patients and enable precision medicine that results in better treatment decisions and prognosis for patients. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-05-12.

Background: While routinely collected human tumor specimen can now readily be examined for genetic alterations on a genome wide scale, multi-parameter measurements of protein expression remain challenging. We developed a multiplexed fluorescence microscopy method (MultiOmyxTM) for the quantitative characterization of multiple analytes in formalin-fixed paraffin-embedded tissue (PMID23818604). We now applied this platform to human breast cancer samples to determine the expression of 25 proteins at a single-cell level and examine their relationship to tumor uptake of the Positron Emission Tomography (PET) radiotracer (18)F-fluorodeoxy-glucose ((18)FDG). Methodology: We stained a single 5 µm section of breast carcinoma from each of 18 patients with antibodies against members of growth factor signaling pathways (HER2, IGF1R, PTEN, p-EGFR, p-PDK1, p-ERK1/2, p-S6 Ribosomal Protein, p-4EBP1, p-eIF4E), the glycolysis pathway (Glut-1, HK2, LDH-A), hormone receptors (ER, PR, AR), tumor cell proliferation markers (KI-67 and phospho-histone H3), and markers of hypoxia (HIF-1α, CA IX), and angiogenesis (CD31). 28-30 representative fields of view (FOVs) were randomly selected in each tumor, each comprising 300-500 cells. Images were automatically separated into subcellular and histopathological compartments based on the staining of tumor cells with a panel of segmentation markers (NaK-ATPase, pan-Cadherin, pan-cytokeratin, S6, and DAPI). A breast pathologist assessed and annotated the histologic composition of each FOV. Further analysis focused on FOVs (n=390) containing only invasive ductal carcinoma without admixed DCIS or normal breast tissue. Multivariate analysis between marker expression and FDG uptake was performed using logistic regression and Cox proportional hazard models. All patients had (18)-FDG PET within four weeks prior to collection of the tumor specimen. Results: Staining results with our MultiOmyxTM platform correlated closely with the results from CLIA-certified single marker biomarker assays (e.g., ER IHC and HER2 FISH assay) performed independently on the same set of samples (PMID: 21646475). Nuclear ER staining was associated with low FDG uptake (p=0.02). KI-67 was higher in tumors with high FDG uptake (p=0.04). K-median clustering identified molecular breast cancer subtypes. Conclusions: Our study illustrates the feasibility of quantitative proteomic measurements using a new in-situ multiplexed fluorescence microscopy platform on only a single routinely collected FFPE tissue section. Quantitative staining results from many thousand cells per tumor support the previously reported relationship between hormone receptor status, tumor cell proliferation, and FDG-uptake in breast cancer. Multiple protein markers showed a high degree of intra- and inter-tumoral heterogeneity. Citation Format: Anup Sood, Alexandra M. Miller, Fiona Ginty, Elizabeth McDonough, Yunxia Sui, Alexander Bordwell, Qing Li, Sireesha Kaanumalle, Zhengyu Pang, Franklin Torres, Edi Brogi, Steven Larson, Ingo Mellinghoff. Application of a multiplexed fluorescence microscopy method (MultiOmyxTM) to dissect proteomic biomarkers of (18)F-fluorodeoxy-glucose ((18)FDG) uptake in breast cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2499. doi:10.1158/1538-7445.AM2014-2499