Search

Because the incidence of breast cancer increases decades after ionizing radiation exposure, aging has been implicated in the evolution of the tumor microenvironment and tumor progression. Here, we investigated radiation-induced carcinogenesis using a model in which the mammary glands of 10-month-old BALB/c mice were transplanted with Trp53-null mammary tissue 3 days after exposure to low doses of sparsely ionizing γ-radiation or densely ionizing particle radiation. Mammary transplants in aged, irradiated hosts gave rise to significantly more tumors that grew more rapidly than those in sham-irradiated mice, with the most pronounced effects seen in mice irradiated with densely ionizing particle radiation. Tumor transcriptomes identified a characteristic immune signature of these aggressive cancers. Consistent with this, fast-growing tumors exhibited an immunosuppressive tumor microenvironment with few infiltrating lymphocytes, abundant immunosuppressive myeloid cells, and high COX-2 and TGFβ. Only irradiated hosts gave rise to tumors lacking cytotoxic CD8+ lymphocytes (defined here as immune desert), which also occurred in younger irradiated hosts. These data suggest that host irradiation may promote immunosuppression. To test this, young chimera mice were fed chow containing a honeybee-derived compound with anti-inflammatory and immunomodulatory properties, caffeic acid phenethyl ester (CAPE). CAPE prevented the detrimental effects of host irradiation on tumor growth rate, immune signature, and immunosuppression. These data indicated that low-dose radiation, particularly densely ionizing exposure of aged mice, promoted more aggressive cancers by suppressing antitumor immunity. Dietary intervention with a nontoxic immunomodulatory agent could prevent systemic effects of radiation that fuel carcinogenesis, supporting the potential of this strategy for cancer prevention.

Supplemental data tables and analysis

Supplementary Figure 4. Macrophage differentiation from circulating human PBMC.

Supplemental Figure 1. Training a Response Classifier to predict whether treatment-naïve advanced melanoma patients will respond to immunotherapy. (a) Schematic showing our computational workflow. We initially trained the neural network with a 5-fold cross validation. Hyper-parameter selection occurred during this phase. After selecting training conditions, we retrained the model on the entire NYU dataset using those parameters. We then tested the classifier on the independent cohort from Vanderbilt (b) During the 5-fold cross validation, neural network accuracy continued to improve until around 175,000 iterations. ROC curves show the performance of each of the 5 test sets at the optimal threshold. Supplemental Figure 2. Performance of the fully trained neural network and logistic regression classifiers from the NYU training dataset. Supplemental Figure 3. Development of a machine learning classifier that predicts whether treatment-naïve advanced melanoma patients will respond to immunotherapy or suffer disease progression. (a) Schematic showing the computational work-flow that is presented in this study. After development of the Segmentation Classifier, all tiles within the training and validation datasets were labeled for the 3 regions of interest. Hyper-parameter selection was performed during the 5-fold cross validation. Then, the model was retrained on the entire NYU dataset using the optimized foundation architecture and tested on the Vanderbilt cohort. (b) Suggested final pipeline where the slide is first pre-processed (tiled with suppression of the background tiles) and segmented by the first convolutional neural network. Identified tumor tiles are then assessed by the second network to extract a first estimate of the response to immunotherapy. The prediction is then refined by combining the output of the neural network with select clinical data. Supplemental Figure 4. Class activation mapping (CAM) analyses identify regions within the tiles that are used by the DCNN to classify the tile as (a) POD or (b) Response. From left to right, for each label, is an image of the tile image overlaid by the image from CAM analyses, the original tile image alone, and the image of CAM analyses alone. Supplemental Figure 5. CellProfiler was used to segment cell nuclei within tiles classified as having a high probability of POD or Response. The POD tiles analyzed had a prediction probability of 0.75 or higher (136,109 tiles). The Response tiles analyzed had a prediction probability of 0.25 or lower (51,220 tiles). Student's t-test was used to compare basic features of the segmented nuclei including area, density, and eccentricity. Supplemental Figure 6. Value importance of DCNN prediction, ECOG performance status, and treatment category was calculated as the absolute value of their Z-scores.

Supplementary tables 1-3

Supplemental Table 1. Patient, slide, and tile distribution during the 5-fold cross validation for training the neural network on the NYU dataset. Supplemental Table 2. Odds Ratios (OR) and 95% Confidence Intervals (CI) from univariate analyses of baseline clinical and demographic variables. Supplemental Table 3. AUC for each of 5 independent test runs and the average AUC for both classifiers when applied to Vanderbilt slides scanned with the Aperio AT2. Supplemental Table 4. AUC for each of 5 independent test runs and the average AUC for both classifiers when applied to Vanderbilt slides scanned with the Leica SCN400. Supplemental Table 5. AUC with 95% confidence intervals for the DCNN and multivariable classifiers when applied exclusively to lymph node or soft tissue from the Vanderbilt dataset. Supplemental Table 6. Results of 3 independent test runs and the average performance of the DCNN classifier on slides from Vanderbilt scanned at 10x magnification. Shown are area under the curve (AUC) and 95% confidence intervals (CI). Supplemental Table 7. Hazard Ratios (HR) with 95% Confidence Intervals (CI) for the clinical variables used in the multivariable prediction model. Supplemental Table 8. AUC and 95% confidence intervals for the neural network prediction, ECOG performance status, treatment category, and each combination of the variables. Supplemental Table 9. Value importance of treatment category, baseline ECOG, and DCNN are reported as the absolute value of the Z-score. Supplemental Table 10. Confusion matrix for predictions of PFS on the cohort of Vanderbilt patients when slides were scanned with an Aperio AT2 scanner. Supplemental Table 11. Confusion matrix for predictions of PFS on the cohort of Vanderbilt patients when slides were scanned with a Leica SCN400 scanner.

Purpose:Several biomarkers of response to immune checkpoint inhibitors (ICI) show potential but are not yet scalable to the clinic. We developed a pipeline that integrates deep learning on histology specimens with clinical data to predict ICI response in advanced melanoma.Experimental Design:We used a training cohort from New York University (New York, NY) and a validation cohort from Vanderbilt University (Nashville, TN). We built a multivariable classifier that integrates neural network predictions with clinical data. A ROC curve was generated and the optimal threshold was used to stratify patients as high versus low risk for progression. Kaplan–Meier curves compared progression-free survival (PFS) between the groups. The classifier was validated on two slide scanners (Aperio AT2 and Leica SCN400).Results:The multivariable classifier predicted response with AUC 0.800 on images from the Aperio AT2 and AUC 0.805 on images from the Leica SCN400. The classifier accurately stratified patients into high versus low risk for disease progression. Vanderbilt patients classified as high risk for progression had significantly worse PFS than those classified as low risk (P = 0.02 for the Aperio AT2; P = 0.03 for the Leica SCN400).Conclusions:Histology slides and patients' clinicodemographic characteristics are readily available through standard of care and have the potential to predict ICI treatment outcomes. With prospective validation, we believe our approach has potential for integration into clinical practice.

Deidentified patient clinical and demographic characteristics

Purpose:Women treated with radiotherapy before 30 years of age have increased risk of developing breast cancer at an early age. Here, we sought to investigate mechanisms by which radiation promotes aggressive cancer.Experimental Design:The tumor microenvironment (TME) of breast cancers arising in women treated with radiotherapy for Hodgkin lymphoma was compared with that of sporadic breast cancers. To investigate radiation effects on carcinogenesis, we analyzed tumors arising from Trp53-null mammary transplants after irradiation of the target epithelium or host using immunocompetent and incompetent mice, some of which were treated with aspirin.Results:Compared with age-matched specimens of sporadic breast cancer, radiation-preceded breast cancers (RP-BC) were characterized by TME rich in TGFβ, cyclooxygenase 2, and myeloid cells, indicative of greater immunosuppression, even when matched for triple-negative status. The mechanism by which radiation impacts TME construction was investigated in carcinomas arising in mice bearing Trp53-null mammary transplants. Immunosuppressive TMEs (iTME) were recapitulated in mice irradiated before transplantation, which implicated systemic immune effects. In nu/nu mice lacking adaptive immunity irradiated before Trp53-null mammary transplantation, cancers also established an iTME, which pointed to a critical role for myeloid cells. Consistent with this, irradiated mammary glands contained more macrophages and human cells cocultured with polarized macrophages underwent dysplastic morphogenesis mediated by IFNγ. Treating mice with low-dose aspirin for 6 months postirradiation prevented establishment of an iTME and resulted in less aggressive tumors.Conclusions:These data show that radiation acts via nonmutational mechanisms to promote markedly immunosuppressive features of aggressive, RP-BCs.

Supplemental Table 1 and Figures S1-6. Table S1: tumor features. Fig. S1. Tumor growth curves from different irradiation group. Fig. S2. Host irradiation promotes ER-negative tumor growth. Fig. S3. Unsupervised hierarchical clustering of mouse tumors based on SD of 1.0. Fig. S4. Highly expressed genes in mammary stem cell are preferentially underexpressed in K18 compared to K14/18 tumor. Fig. S5. K14/18-IHPsi is negatively associated with MaSC signature. Fig. S6. Host irradiation exerts more robust effects on gene expression profile in ER-negative(ER-neg) than ER-positive(ER-pos) tumors.

Purpose: Several biomarkers of response to immune checkpoint inhibitors (ICI) show potential but are not yet scalable to the clinic. We developed a pipeline that integrates deep learning on histology specimens with clinical data to predict ICI response in advanced melanoma. Experimental Design: We used a training cohort from New York University (New York, NY) and a validation cohort from Vanderbilt University (Nashville, TN). We built a multivariable classifier that integrates neural network predictions with clinical data. A ROC curve was generated and the optimal threshold was used to stratify patients as high versus low risk for progression. Kaplan–Meier curves compared progression-free survival (PFS) between the groups. The classifier was validated on two slide scanners (Aperio AT2 and Leica SCN400). Results: The multivariable classifier predicted response with AUC 0.800 on images from the Aperio AT2 and AUC 0.805 on images from the Leica SCN400. The classifier accurately stratified patients into high versus low risk for disease progression. Vanderbilt patients classified as high risk for progression had significantly worse PFS than those classified as low risk (P = 0.02 for the Aperio AT2; P = 0.03 for the Leica SCN400). Conclusions: Histology slides and patients' clinicodemographic characteristics are readily available through standard of care and have the potential to predict ICI treatment outcomes. With prospective validation, we believe our approach has potential for integration into clinical practice.

Tumor-infiltrating lymphocytes (TIL) have potential prognostic value in melanoma and have been considered for inclusion in the American Joint Committee on Cancer (AJCC) staging criteria. However, interobserver discordance continues to prevent the adoption of TIL into clinical practice. Computational image analysis offers a solution to this obstacle, representing a methodological approach for reproducibly counting TIL. We sought to evaluate the ability of a TIL-quantifying machine learning algorithm to predict survival in primary melanoma. Digitized hematoxylin and eosin (HE) slides from prospectively enrolled patients in the NYU melanoma database were scored for % TIL using machine learning and manually graded by pathologists using Clark's model. We evaluated the association of % TIL with recurrence-free survival (RFS) and overall survival (OS) using Cox proportional hazards modeling and concordance indices. Discordance between algorithmic and manual TIL quantification was assessed with McNemar's test and visually by an attending dermatopathologist. In total, 453 primary melanoma patients were scored using machine learning. Automated % TIL scoring significantly differentiated survival using an estimated cutoff of 16.6% TIL (log-rank P 0.001 for RFS; P = 0.002 for OS). % TIL was associated with significantly longer RFS (adjusted HR = 0.92 [0.84-1.00] per 10% increase in % TIL) and OS (adjusted HR = 0.90 [0.83-0.99] per 10% increase in % TIL). In comparison, a subset of the cohort (n = 240) was graded for TIL by melanoma pathologists. However, TIL did not associate with RFS between groups (P 0.05) when categorized as brisk, nonbrisk, or absent. A standardized and automated % TIL scoring algorithm can improve the prognostic impact of TIL. Incorporation of quantitative TIL scoring into the AJCC staging criteria should be considered.

There is currently no targeted therapy to treat NF1-mutant melanomas. In this study, we compared the genomic and transcriptomic signatures of NF1-mutant and NF1 wild-type melanoma to reveal potential treatment targets for this subset of patients. Genomic alterations were verified using qPCR, and differentially expressed genes were independently validated using The Cancer Genome Atlas data and immunohistochemistry. Digital spatial profiling with multiplex immunohistochemistry and immunofluorescence were used to validate the signatures. The efficacy of combinational regimens driven by these signatures was tested through in vitro assays using low-passage cell lines. Pathogenic NF1 mutations were identified in 27% of cases. NF1-mutant melanoma expressed higher proliferative markers MK167 and CDC20 than NF1 wild-type (P = 0.008), which was independently validated both in The Cancer Genome Atlas dataset (P = 0.01, P = 0.03) and with immunohistochemistry (P = 0.013, P = 0.036), respectively. Digital spatial profiling analysis showed upregulation of LY6E within the tumor cells (false discovery rate0.01, log2 fold change1), confirmed with multiplex immunofluorescence showing colocalization of LY6E in melanoma cells. The combination of MAPK/extracellular signal‒regulated kinase kinase and CDC20 coinhibition induced both cytotoxic and cytostatic effects, decreasing CDC20 expression in multiple NF1-mutant cell lines. In conclusion, NF1-mutant melanoma is associated with a distinct genomic and transcriptomic profile. Our data support investigating CDC20 inhibition with MAPK pathway inhibitors as a targeted regimen in this melanoma subtype.

Purpose: Women treated with radiotherapy before 30 years of age have increased risk of developing breast cancer at an early age. Here, we sought to investigate mechanisms by which radiation promotes aggressive cancer. Experimental Design: The tumor microenvironment (TME) of breast cancers arising in women treated with radiotherapy for Hodgkin lymphoma was compared with that of sporadic breast cancers. To investigate radiation effects on carcinogenesis, we analyzed tumors arising from Trp53-null mammary transplants after irradiation of the target epithelium or host using immunocompetent and incompetent mice, some of which were treated with aspirin. Results: Compared with age-matched specimens of sporadic breast cancer, radiation-preceded breast cancers (RP-BC) were characterized by TME rich in TGFβ, cyclooxygenase 2, and myeloid cells, indicative of greater immunosuppression, even when matched for triple-negative status. The mechanism by which radiation impacts TME construction was investigated in carcinomas arising in mice bearing Trp53-null mammary transplants. Immunosuppressive TMEs (iTME) were recapitulated in mice irradiated before transplantation, which implicated systemic immune effects. In nu/nu mice lacking adaptive immunity irradiated before Trp53-null mammary transplantation, cancers also established an iTME, which pointed to a critical role for myeloid cells. Consistent with this, irradiated mammary glands contained more macrophages and human cells cocultured with polarized macrophages underwent dysplastic morphogenesis mediated by IFNγ. Treating mice with low-dose aspirin for 6 months postirradiation prevented establishment of an iTME and resulted in less aggressive tumors. Conclusions: These data show that radiation acts via nonmutational mechanisms to promote markedly immunosuppressive features of aggressive, RP-BCs.

Because the incidence of breast cancer increases decades after ionizing radiation exposure, aging has been implicated in the evolution of the tumor microenvironment and tumor progression. Here, we investigated radiation-induced carcinogenesis using a model in which the mammary glands of 10-month-old BALB/c mice were transplanted with Trp53-null mammary tissue 3 days after exposure to low doses of sparsely ionizing γ-radiation or densely ionizing particle radiation. Mammary transplants in aged, irradiated hosts gave rise to significantly more tumors that grew more rapidly than those in sham-irradiated mice, with the most pronounced effects seen in mice irradiated with densely ionizing particle radiation. Tumor transcriptomes identified a characteristic immune signature of these aggressive cancers. Consistent with this, fast-growing tumors exhibited an immunosuppressive tumor microenvironment with few infiltrating lymphocytes, abundant immunosuppressive myeloid cells, and high COX-2 and TGFβ. Only irradiated hosts gave rise to tumors lacking cytotoxic CD8+ lymphocytes (defined here as immune desert), which also occurred in younger irradiated hosts. These data suggest that host irradiation may promote immunosuppression. To test this, young chimera mice were fed chow containing a honeybee-derived compound with anti-inflammatory and immunomodulatory properties, caffeic acid phenethyl ester (CAPE). CAPE prevented the detrimental effects of host irradiation on tumor growth rate, immune signature, and immunosuppression. These data indicated that low-dose radiation, particularly densely ionizing exposure of aged mice, promoted more aggressive cancers by suppressing antitumor immunity. Dietary intervention with a nontoxic immunomodulatory agent could prevent systemic effects of radiation that fuel carcinogenesis, supporting the potential of this strategy for cancer prevention.

Dermal invasion is a hallmark of malignant melanoma. Although the molecular alterations that drive the progression of primary melanoma to metastatic disease have been studied extensively, the early progression of noninvasive primary melanoma to an invasive state is poorly understood. To elucidate the mechanisms underlying the transition from radial to vertical growth, the first step in melanoma invasion, we developed a zebrafish melanoma model in which constitutive activation of ribosomal protein S6 kinase A1 drives tumor invasion. Transcriptomic analysis of ribosomal protein S6 kinase A1–activated tumors identified metabolic changes, including up-regulation of genes associated with oxidative phosphorylation. Vertical growth phase human melanoma cells show higher oxygen consumption and preferential utilization of glutamine compared to radial growth phase melanoma cells. Peroxisome proliferator activated receptor γ coactivator (PGC)-1α, has been proposed as a master regulator of tumor oxidative phosphorylation. In human primary melanoma specimens, PGC1α protein expression was found to be positively associated with increased tumor thickness and expression of the proliferative marker Ki-67 and the reactive oxygen species scavenger receptor class A member 3. PGC1α depletion modulated cellular processes associated with primary melanoma growth and invasion, including oxidative stress. These results support a role for PGC1α in mediating glutamine-driven oxidative phosphorylation to facilitate the invasive growth of primary melanoma.

Background: Data from the Consortium of Investigators of Modifiers of BRCA1/2 (CIMBA) indicates that common variation in the BRCA1 locus modifies the penetrance of mutations. The rs5820483 variation is one such polymorphism in which patients homozygous for the major allele have a greater risk of breast cancer. The BRCA1 gene contains 24 exons that generate a full-length (FL) protein. Major alternatively spliced isoforms include skipping of exon 11 (Δ11) and a truncated form (IRIS), each of which appear to have distinct roles in cellular processes. rs5820483 is the only variant that maps to intron 10, flanking exon 11. Experimental models suggest that deregulation of the FL/D11 isoform ratio can have profound functional consequences. We have shown that cells from patients with the major rs5820483 allele express more FL relative to Δ11 compared to cells from patients with the minor allele. Breast cancer exhibit increased IGF-1 and IGF-1 activity in patients with BRCA1 mutations relative to breast cancer in women without BRCA1 mutations. IGF-1 produced in the mammary stroma mediates estrogen dependent proliferation. Here we test whether the rs5820483 allele affects IGF-1 activity in breast tissue of BRCA1 mutation carriers. Design: We assessed the zygosity of the rs5820483 alleles in 28 BRCA1 mutation carriers. Phosphorylated IGF-1 receptor (p-IGF-1R rabbit polyclonal Ab39398, Abcam, MA) was measured by immunofluorescence and Ki-67 was measured by immunohistochemistry in sections from histologically normal breast tissue from these cases were compared to heterozygous controls. Image analysis was used to assess the intensity of the p-IGF-1R immunofluorescence at the epithelial cell membrane with appropriate controls. Ki67 mmunohistochemistry was also performed on these specimens. Results:The rs5820483 allele was heterozygous in 10 specimens, homozygous for the major allele in 11 and homozygous for the minor allele in 7. The intensity of p-IGF-1R was higher in major homozygous cases (140±54 SD) than in either the minor homozygous (99±24) or heterozygous cases (95±44). The frequency of Ki-67+ cells was higher in the major homozygous case (3.1±2.9 SD) than in either the minor homozygous (1.5±1.9 SD) or heterozygous cases (2.2±2.5 SD). However, neither p-IGF-IR immunoreactivity nor frequency of Ki-67+ cells was statistically different between groups. Conclusion: Breast cancer risk in BRCA1 mutation carriers is modified by common genetic variation at the corresponding locus. We have identified a variation at rs5820483 that affects the isoform ratio. Our preliminary analysis suggests that IGF-1 activity increases in those women homozygous for the major allele, concomitant with increased Ki-67. Corroboration of this analysis in a larger series is ongoing. A statistically significant difference might have fundamental implications for cancer prevention in those carriers. Citation Format: Singh B, Bochaca II, Ruiz de Garibay G, Damiola F, Mazoyer S, Antoniou A, Chenevix-Trench G, Pujana MA, Barcellos-Hoff MH, Kleinberg D. The BRCA-1 polymorphism (major homozygous rs5820483) is associated with higher expression of phosphorylated -IGF-1 receptor. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P2-10-01.

10066 Background: Inclusion of tumor-infiltrating lymphocytes (TIL) into AJCC staging criteria has been proposed due to evidence suggesting its prognostic significance. However, subjective inter-observer discordance prevents adoption of semi-quantitative TIL grading (e.g. absent, non-brisk, brisk) into clinical practice. We hypothesize that digital-image analysis (DIA) of TIL can provide a standardized, quantitative scoring system that more accurately predicts survival compared to currently used semi-quantitative grading methods. Methods: Clinical data and tumor specimens were analyzed from prospectively enrolled primary melanoma patients in the New York University Interdisciplinary Melanoma Cooperative Group with median follow-up of 5 years. H&E-stained slides were digitized using an Aperio ScanScope at 20X magnification. QuPath software was used for automated TIL quantification. Cox regression analysis was used to assess the improved prognostic value of TIL on recurrence-free (RFS) and overall survival (OS). Patients were separated into high- and low-TIL groups using a score threshold determined by the Youden Index. Results: 453 patients (18% stage I, 42% stage II, 40% stage III) were scored using automated TIL assessment and scores were significantly correlated with better RFS and OS per 10% increase in TIL (stage adjusted hazard ratio [aHR] = 0.92 [0.84-1.00] for RFS and aHR = 0.90 [0.83-0.99] for OS). A model combining TIL score with stage increased prognostic ability for both RFS (0.68 to 0.70, P = 0.02) and OS (0.62 to 0.64, P = 0.01), as assessed by concordance indices (C-index). Kaplan-Meier curves of high- ( > 16.6%) versus low-TIL (≤16.6%) patients showed clear separation in RFS and OS (median RFS = 155 vs 48 months, P < 0.001; median OS = 155 vs 89 months, P = 0.002). For comparison, a subset of the cohort (n = 250) was semi-quantitatively graded (absent, non-brisk, brisk) by an attending melanoma pathologist; however, this did not significantly differentiate RFS between groups (P > 0.05). Conclusions: A standardized, quantitative TIL scoring system significantly improved prediction of RFS and OS in primary melanoma patients compared with semi-quantitative TIL grading. Incorporation of quantitative TIL scoring into prognostic algorithms, such as AJCC criteria, should be considered.

10010 Background: Several predictors of response to checkpoint inhibitors show potential, but use pathological assays and/or molecular characterization, which limits their clinical utility outside of the academic setting. We aimed to develop a streamlined approach by leveraging information immediately available through standard care. Here, we present a computational method that integrates deep learning on histology specimens with clinicodemographic variables to predict treatment outcomes in advanced melanoma. Methods: We used hematoxylin and eosin (H&E) sections from 72 patients ( n= 153 slides) from New York University (NYU) to build a Segmentation Classifier that distinguishes tumor, lymphocyte, and connective tissue. Using pre-treatment H&E slides from 121 NYU patients ( n =302 slides), we trained a Response Classifier to predict response by selectively analyzing tumor regions. We then developed a logistic regression classifier that combines neural network output with clinicodemographic variables. The classifiers were tested on an independent cohort of 32 patients from Vanderbilt University ( n =42 slides). Area under the curve (AUC) was calculated as a measure of prediction accuracy. Results: The Segmentation Classifier distinguished tumor, lymphocyte, and connective tissue with AUCs 0.886-0.984. For the Response Classifier, optimal learning conditions were identified through training on NYU patients and testing on Vanderbilt patients (AUCs 0.685 and 0.728, respectively). The fully trained Response Classifier performed with AUC 0.711 on Vanderbilt patients. The logistic regression model performed with enhanced prediction accuracy with AUC 0.803 on NYU patients and AUC 0.793 on Vanderbilt patients. Conclusions: Histology slides and patients’ clinicodemographic characteristics are readily available through routine standard of care and have the potential to predict immunotherapy response. Our approach is time-efficient, reproducible, and requires minimal resource allocation, thus overcoming multiple common barriers to generalizability for contemporary biomarkers.

10060 Background: Preclinical data suggest that melanoma angiogenesis promotes resistance to MAPK-pathway (MAPKi) and immune checkpoint inhibitors (ICIs). However, phase II clinical trials of anti-angiogenic therapy in melanoma were disappointing. We previously identified a pathogenic germline variant Q472H in the kinase insert domain receptor [KDR Q472H; vascular endothelial growth factor receptor-2 (VEGFR-2)] in 35% of primary melanoma patients. We hypothesize that KDR Q472H promotes resistance to MAPKi or ICIs, and that combined MAPKi or ICI and VEGF pathway inhibition may improve outcomes in patients harboring the variant. Methods: Metastatic melanoma (MM) patient clinical data and biospecimens enrolled in the NYU Langone Medical Center Melanoma program were studied. KDR status was determined by TaqMan assays. Tumor microvessel density (MVD) was assessed by CD34 immunohistochemistry. The impact of KDR Q472H on the tumor microenvironment was determined by RNA-seq and Nanostring. Synergy between BRAF (dabrafenib) and VEGFR-2 (lenvatinib) inhibitors in KDR-genotyped MM cell lines was assessed using cell proliferation assays and the Chou-Talalay method. Synergy between ICIs and anti-VEGFR-2 was evaluated in vivo using a B16 melanoma model. Results: We studied 221 MM patients (38% KDR Q472H variants). KDR Q472H variant was significantly associated with higher tumor MVD (P = 0.002). Among the MAPKi-treated patients, KDR Q472H homozygotes had shorter median progression-free survival (PFS, 3.3 vs 9.7 months, P = 0.009) than KDR wild type (WT). In patients treated with anti-PD-1-based therapies, response rates were lower in KDR Q472H variant patients compared to WT (P = 0.012), with shorter median PFS (8.4 months vs not reached, P = 0.0443). Transcriptomic analyses identified an immunosuppressive phenotype in KDR Q472H tumors, with reduced expression of genes associated with chemotaxis, inflammation, T cell activity, and antigen presentation. Consistent with this finding, VEGFR-2 blockade in a KDR Q472H B16 mouse melanoma model augmented the anti-melanoma immune response. KDR Q472H cell lines displayed synergistic cytotoxicity with dabrafenib and lenvatinib, compared to KDR WT cells. Conclusions: Our data demonstrate that melanoma patients with pathogenic germline variant KDR Q472H may be more resistant to both ICIs and MAPKi. Anti-angiogenic therapy should be reconsidered within this specific subset of patients in prospective clinical trials.

The extent of PTEN loss that confers clinical and biological impact in melanoma is unclear. We evaluated the clinical and biologic relevance of PTEN dosage in melanoma and tested the postulate that partial PTEN loss is due to epigenetic mechanisms. PTEN expression was assessed by immunohistochemistry in a stage III melanoma cohort (n = 190) with prospective follow up. Overall, 21 of 190 (11%) tumors had strong PTEN expression, 51 of 190 (27%) had intermediate PTEN, 44 of 190 (23%) had weak PTEN, and 74 of 190 (39%) had absent PTEN. Both weak and absent PTEN expression predicted shorter survival in multivariate analyses (hazard ratio = 2.13, P 0.01). We show a continuous negative correlation between PTEN and activated Akt in melanoma cells with titrated PTEN expression and in two additional independent tumor datasets. PTEN genomic alterations (deletion, mutation), promoter methylation, and protein destabilization did not fully explain PTEN loss in melanoma, whereas PTEN levels increased with treatment of melanoma cells with the histone deacetylase inhibitor LBH589. Our data indicate that partial PTEN loss is due to modifiable epigenetic mechanisms and drives Akt activation and worse prognosis, suggesting a potential approach to improve the clinical outcome for a subset of patients with advanced melanoma.

Glioblastoma (GBM) stem cells (GSCs) reside in both hypoxic and vascular microenvironments within tumors. The molecular mechanisms that allow GSCs to occupy such contrasting niches are not understood. We used patient-derived GBM cultures to identify GSC subtypes with differential activation of Notch signaling, which co-exist in tumors but occupy distinct niches and match their metabolism accordingly. Multipotent GSCs with Notch pathway activation reside in perivascular niches, and are unable to entrain anaerobic glycolysis during hypoxia. In contrast, most CD133-expressing GSCs do not depend on canonical Notch signaling, populate tumors regardless of local vascularity and selectively utilize anaerobic glycolysis to expand in hypoxia. Ectopic activation of Notch signaling in CD133-expressing GSCs is sufficient to suppress anaerobic glycolysis and resistance to hypoxia. These findings demonstrate a novel role for Notch signaling in regulating GSC metabolism and suggest intratumoral GSC heterogeneity ensures metabolic adaptations to support tumor growth in diverse tumor microenvironments.

SUMMARYA better knowledge of the regulatory mechanisms involved in stem cell proliferation and/or differentiation could reveal new methods for the treatment of some diseases. Most previous studies in the field of stem cell biology have been carried out on cultured isolated cells. In the case of adult tissue stem cells, mesenchymal bone marrow derived cells have been most widely studied, while the undifferentiated stem cells present in the epithelial tissues are less known. In order to advance further our understanding of epithelial tissue stem cells, new in vivo models are required. The present study focuses on the dynamics of a new and simple model of intestinal epithelial regeneration found in the midgut of the migratory locust, Locusta migratoria (Linnaeus 1758). The locust midgut consists of three cell types: columnar cells, endocrine cells and undifferentiated regenerative clustered cells. The undifferentiated epithelial midgut cells give rise to two other cell types and are located in a nest of regenerative cells known as regenerative niche. We have performed single and continuous bromodeoxyuridine (BrdU) administration experiments to study regeneration niches and their cellular dynamics. Immunocytochemistry and immunofluorescence techniques were used to detect the incorporation of BrdU into regenerative niches and the presence of FMRFamide-like immunoreactivity, as a marker for endocrine cell differentiation. Some isolated label retaining cells (LRC) were observed at the niche base 10-15 days after the final BrdU administration. We propose that these cells are the stem cells of this epithelial tissue. We also calculated the length of the cell cycle phases for a subpopulation of transit amplifying cells within the regenerative niche: G1, 2.5±0.5 h; S,5.5±0.5 h; G2, 0.75±0.25 h; M, 2.5±0.5 h. These amplifying cells will give rise to the columnar epithelial non-endocrine lineage. The differentiation of an endocrine cell from a niche stem cell occurs less frequently and thus leads to a lower proportion of endocrine cells as compared with epithelial columnar digestive cells (up to three endocrine cells per niche). Endocrine cell commitment seems to occur very early in the differentiation process within the niche, as double-labelled BrdU and FMRF endocrine cells have never been found. The only exception is the endocrine cells located in the ampullar region of the midgut, some of which show double immunostaining after long-term chronic BrdU injection. In summary, we have characterized a new and simple animal model of epithelial stem cell regeneration that may be useful for understanding the complex biological process that drives tissue renewal from undifferentiated and uncommitted progenitor cells.

DCIS is a non-obligate precursor in which epithelial cells with features of cancer reside in and expand otherwise normal appearing mammary ducts and lobules. DCIS increases the relative risk of developing invasive breast cancer (IBC) by up to 10 fold. Treatment of DCIS includes breast conserving surgery (BCS), tumor margins permitting, followed by radiation, or mastectomy. Recently we found that pasireotide, a somatostatin analog inhibitor of IGF-I, was effective in reducing cell proliferation and increasing apoptosis in patients with atypical hyperplasia (AH) and or proliferative disease of the breast. At present there are no recognized medical treatments for DCIS. Here we raise the possibility that pasireotide might also be an effective therapy for some DCIS. To make this determination, 8 patients previously diagnosed with DCIS were treated with pasireotide, 600µg twice daily subcutaneously for a period of 20 days. Lesion size was assessed on DCE-MRIs in 3 planes before and after treatment. Of the eight, two had lesions characterized as low grade DCIS on core biopsy, 4 had intermediate or intermediate to high grade lesions, and two had high grade lesions. After treatment they all had surgical excision biopsies. We also measured proliferation and apoptosis to assess effects of pasireotide by expression of Ki67 and caspase 3. To confirm that the effect of pasireotide was mediated via the IGF-I pathway, p-AKT and p-ERK 1/2 were evaluated. Core and excision samples from 14 untreated patients where used as control. Overall, treatment with pasireotide reduced DCIS lesion size by ≥30% according the RECIST criteria in 4 of 8 patients, as measured by DCE-MRI. Of those, pathology of the surgical excision confirmed that there was no remaining DCIS tissue in the two patients with low grade disease. The two additional lesions that were reduced in size one was intermediate to high grade and the other a high grade DCIS. The remaining 4 patients had tumors that were stable in size. Pasireotide reduced cell proliferation and significantly increased apoptosis in the DCIS lesions. While DCIS is less responsive to pasireotide than AH and proliferative disease, it may nevertheless provide a window of opportunity for treatment and possibly eradication of some low grade or higher grade DCIS as well. The main side effect included hyperglycemia, intestinal discomfort and rash or redness at the injection area. Our data show that hyperglycemia disappears relatively fast after ending treatment with pasireotide. In conclusion, treatment with pasireotide eliminated 2 low grade DCIS lesions, shrunk two higher grade ones and also decreased proliferation and increased apoptosis. Additional evidence in a larger number of patients will be necessary. Citation Format: Irineu Illa-Bochaca, Linda Moy, Julia Smith, Debora Axelrod, Baljit Singh, David L. Kleinberg. Treatment of ductal carcinoma in situ (DCIS) with pasireotide, an IGF-I inhibitor [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT086. doi:10.1158/1538-7445.AM2017-CT086

Densely ionizing radiation, which is present in the space radiation environment and used in radiation oncology, has potentially greater carcinogenic effect compared with sparsely ionizing radiation that is prevalent on earth. Here, we used a radiation chimera in which mice were exposed to densely ionizing 350 MeV/amu Si-particles, γ-radiation, or sham-irradiated and transplanted 3 days later with syngeneic Trp53-null mammary fragments. Trp53-null tumors arising in mice irradiated with Si-particles had a shorter median time to appearance and grew faster once detected compared with those in sham-irradiated or γ-irradiated mice. Tumors were further classified by markers keratin 8/18 (K18, KRT18), keratin 14 (K14, KRT14) and estrogen receptor (ER, ESR1), and expression profiling. Most tumors arising in sham-irradiated hosts were comprised of both K18- and K14-positive cells (K14/18) while those tumors arising in irradiated hosts were mostly K18. Keratin staining was significantly associated with ER status: K14/18 tumors were predominantly ER-positive, whereas K18 tumors were predominantly ER-negative. Genes differentially expressed in K18 tumors compared with K14/18 tumor were associated with ERBB2 and KRAS, metastasis, and loss of E-cadherin. Consistent with this, K18 tumors tended to grow faster and be more metastatic than K14/18 tumors, however, K18 tumors in particle-irradiated mice grew significantly larger and were more metastatic compared with sham-irradiated mice. An expression profile that distinguished K18 tumors arising in particle-irradiated mice compared with sham-irradiated mice was enriched in mammary stem cell, stroma, and Notch signaling genes. These data suggest that carcinogenic effects of densely ionizing radiation are mediated by the microenvironment, which elicits more aggressive tumors compared with similar tumors arising in sham-irradiated hosts. Cancer Res; 74(23); 7137–48. ©2014 AACR.

This review focuses on recent experimental and modeling studies that attempt to define the physiological context in which high linear energy transfer (LET) radiation increases epithelial cancer risk and the efficiency with which it does so. Radiation carcinogenesis is a two-compartment problem: ionizing radiation can alter genomic sequence as a result of damage due to targeted effects (TE) from the interaction of energy and DNA; it can also alter phenotype and multicellular interactions that contribute to cancer by poorly understood non-targeted effects (NTE). Rather than being secondary to DNA damage and mutations that can initiate cancer, radiation NTE create the critical context in which to promote cancer. Systems biology modeling using comprehensive experimental data that integrates different levels of biological organization and time-scales is a means of identifying the key processes underlying the carcinogenic potential of high-LET radiation. We hypothesize that inflammation is a key process, and thus cancer susceptibility will depend on specific genetic predisposition to the type and duration of this response. Systems genetics using novel mouse models can be used to identify such determinants of susceptibility to cancer in radiation sensitive tissues following high-LET radiation. Improved understanding of radiation carcinogenesis achieved by defining the relative contribution of NTE carcinogenic effects and identifying the genetic determinants of the high-LET cancer susceptibility will help reduce uncertainties in radiation risk assessment.

Introduction Estrogen inhibition is effective in preventing breast cancer in only up to 50% of women with precancerous lesions and many experience side effects that are poorly tolerated. As insulin-like growth factor I (IGF-I) underlies both estrogen and progesterone actions and has other direct effects on mammary development and carcinogenesis, we hypothesized that IGF-I inhibition might provide a novel approach for breast cancer chemoprevention. Methods In total, 13 women with core breast biopsies diagnostic of atypical hyperplasia (AH) were treated for 10 days with pasireotide, a somatostatin analog which uniquely inhibits IGF-I action in the mammary gland. They then had excision biopsies. 12 patients also had proliferative lesions and one a ductal carcinoma in situ (DCIS). Primary outcomes were changes in cell proliferation and apoptosis after treatment. Expression of estrogen receptor (ER), progesterone receptor (PR), and phosphorylated Insulin-like growth factor I receptor (IGF-1R), protein kinase B (AKT) and extracellular signal-regulated kinases 1/2 (ERK1/2) were also assessed. Core and excision biopsies from 14 untreated patients served as non-blinded controls. Hyperglycemia and other side effects were carefully monitored. Results Pasireotide decreased proliferation and increased apoptosis in all AH (from 3.6 ± 2.6% to 1.3 ± 1.2% and from 0.3 ± 0.2% to 1.5 ± 1.6%, respectively) and proliferative lesions (from 3.8 ± 2.5% to 1.8 ± 1.8% and from 0.3 ± 0.2% to 1.3 ± 0.6%, respectively). The DCIS responded similarly. ER and PR were not affected by pasireotide, while IGF-1R, ERK1/2 and AKT phosphorylation decreased significantly. In contrast, tissue from untreated controls showed no change in cell proliferation or phosphorylation of IGF-1R, AKT or ERK 1/2. Mild to moderate hyperglycemia associated with reduced insulin levels was found. Glucose fell into the normal range after discontinuing treatment. Pasireotide was well tolerated and did not cause symptoms of estrogen deprivation. Conclusions IGF-I inhibition by pasireotide, acting through the IGF-1R, was associated with decreased proliferation and increased apoptosis in pre-malignant breast lesions and one DCIS. Assuming hyperglycemia can be controlled, these data suggest that inhibiting the IGF-I pathway may prove an effective alternative for breast cancer chemoprevention. Trial registration NCT01372644 Trial date: July 1, 2007. Electronic supplementary material The online version of this article (doi:10.1186/s13058-014-0463-1) contains supplementary material, which is available to authorized users.

Children exposed to ionizing radiation have a substantially greater breast cancer risk than adults; the mechanism for this strong age dependence is not known. Here we show that pubertal murine mammary glands exposed to sparsely or densely ionizing radiation exhibit enrichment of mammary stem cell and Notch pathways, increased mammary repopulating activity indicative of more stem cells, and propensity to develop estrogen receptor (ER) negative tumors thought to arise from stem cells. We developed a mammary lineage agent-based model (ABM) to evaluate cell inactivation, self-renewal, or dedifferentiation via epithelial-mesenchymal transition (EMT) as mechanisms by which radiation could increase stem cells. ABM rejected cell inactivation and predicted increased self-renewal would only affect juveniles while dedifferentiation could act in both juveniles and adults. To further test self-renewal versus dedifferentiation, we used the MCF10A human mammary epithelial cell line, which recapitulates ductal morphogenesis in humanized fat pads, undergoes EMT in response to radiation and transforming growth factor β (TGFβ) and contains rare stem-like cells that are Let-7c negative or express both basal and luminal cytokeratins. ABM simulation of population dynamics of double cytokeratin cells supported increased self-renewal in irradiated MCF10A treated with TGFβ. Radiation-induced Notch concomitant with TGFβ was necessary for increased self-renewal of Let-7c negative MCF10A cells but not for EMT, indicating that these are independent processes. Consistent with these data, irradiating adult mice did not increase mammary repopulating activity or ER-negative tumors. These studies suggest that irradiation during puberty transiently increases stem cell self-renewal, which increases susceptibility to developing ER-negative breast cancer. Stem Cells 2014;32:649–661

CAPE is the major active component of propolis, a widely available, safe, honeybee natural product with anti-inflammatory, antioxidant, and antitumor properties. We have previously shown diverse effects of CAPE in breast cancer. We postulated that CAPE may be useful in chemoprevention for women at high risk for triple-negative breast cancers (TNBC) and evaluated this in a radiation chimera mouse model in which the tumor spectrum is shifted to TNBC. The radiation chimera model consists of surgically clearing the mammary epithelium from the inguinal glands of 3-week old BALB/c mice, the mice were irradiated with 1 Gy or sham irradiated at 10-12 weeks of age, and bilaterally transplanted 3 days later with syngeneic Trp53 null mammary fragments. Mice were placed on a CAPE diet or a control diet at 1 month post transplantation that was maintained for the course of the experiment. Mammary tumor development was monitored by palpation for up to 18 months. The first tumor was resected at 1cm3 and mice monitored for tumor recurrence or second tumor formation. Tumors were immunostained for estrogen receptor (ER) status and ER negative tumors were selected from all the groups for RNA sequencing. No difference in body weight or tumor incidence was observed between mice on the CAPE versus control diet. Host irradiation significantly accelerated tumor growth rate compared to the control sham irradiated mice. CAPE treatment blocked this effect, but did not affect the control tumor growth rate. Resected tumors in CAPE treated mice recurred at significantly longer intervals (average of 40 days in the control group versus 90 days with CAPE treatment) and much less frequently than tumors from mice on a control diet. Mean expression analysis showed that CAPE induces a distinct gene expression pattern in ER negative tumors from irradiated mice. As previously described, the transcriptional profile of ER negative tumors was enriched in immune response genes in tumors from irradiated mice on a control diet. Interestingly, this difference was abrogated in mice on the CAPE diet. These findings support the potential use of CAPE to modify the aggressive behavior of TNBC. This may be due to effects on the immune system in which CAPE acts to re-establish anti tumor immunity. Thus, CAPE may be useful in chemoprevention both for women at high risk for TNBC and to prevent or delay TNBC breast cancer recurrence. Citation Format: Coral O. Omene, Manan Patel, Irineu IllaBochaca, Mary Helen Barcellos-Hoff. Caffeic acid phenethyl ester (CAPE) reverses aggressive breast cancer in the radiation chimera model. [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 5247.

SummaryTissue microenvironment is an important determinant of carcinogenesis. We demonstrate that ionizing radiation, a known carcinogen, affects cancer frequency and characteristics by acting on the microenvironment. Using a mammary chimera model in which an irradiated host is transplanted with oncogenic Trp53 null epithelium, we show accelerated development of aggressive tumors whose molecular signatures were distinct from tumors arising in nonirradiated hosts. Molecular and genetic approaches show that TGFβ mediated tumor acceleration. Tumor molecular signatures implicated TGFβ, and genetically reducing TGFβ abrogated the effect on latency. Surprisingly, tumors from irradiated hosts were predominantly estrogen receptor negative. This effect was TGFβ independent and linked to mammary stem cell activity. Thus, the irradiated microenvironment affects latency and clinically relevant features of cancer through distinct and unexpected mechanisms.

Ionizing radiation is a well-established breast carcinogen in women and has been a widely studied cause of cancer in experimental models. Clinical data and experimental models are discussed to highlight radiation effects on the mammary gland and breast cancer as a system of heterotypic interactions among multiple cell types. Epidemiological data from different populations of women who were exposed to radiation during various stages of physiological development have offered insight into potential biological mechanisms that mediate radiation-induced breast cancer. This chapter discusses classic and current views of radiation’s action within the mammary gland and how these effects interact biologically to inform a more comprehensive conceptualization of radiation as a carcinogen, including new emphasis on a defined adult stem cell population within the mammary gland.

The regenerative potential of mammary epithelium facilitates assessment of the "stemness" of any epithelial subpopulation in transplantation assays. Thus, mammary tissue can be dissociated into single cells, stained for cell surface markers of interest and classified using fluorescence-activated cell sorting. The selected cells can then be transplanted into epithelium-devoided fat pads from recipient hosts. Recent publications have described markers that enrich for mammary repopulating potential. Here, we describe the materials and methods necessary to sort cells according to these markers. This approach can be used interchangeably with other cell surface markers with slight variation to the protocol.

Mammary reconstitution assays can be used to measure the stem cell frequency within an epithelial population by transplanting increasingly diluted single-cell preparations of the population of interest. There are fundamental steps in the single-cell isolation protocol which are directly related to the number of single epithelial cells obtained. Once single-cell suspensions have been obtained, serial dilutions are prepared and transplanted into the cleared fat pads of the host mice. After 8-10 weeks, the transplanted fat pads are reevaluated for the presence of epithelial outgrowths. Based on the frequency of no outgrowth for each one of the transplanted dilutions, it is possible to estimate the frequency of mammary repopulating cells present in a given cell population. Here, we give details on how to carry out all these steps.

The mammary gland consists of an epithelial ductal tree embedded in a fat pad. Adult mammary epithelium has been demonstrated to have outstanding regenerative potential, consistent with the presence of resident, adult stem cells. However, there are currently no bona fide markers to identify these cells within their tissue context. Here, we introduce long-term label retention as a method to investigate the location of quiescent cells (a property attributed to adult stem cells) in situ. Long-term label retaining cells divide actively during tissue development and remain quiescent at homeostasis. These two properties have been attributed to adult stem cells. Therefore, label-retaining cells can be used to identify populations that contain stem cells. We describe the materials and methods necessary to identify and image mammary label-retaining cells, to carry out morphometric analysis on these cells and to map their distribution of the mammary epithelium. The morphometric and spatial analyses described here are generally applicable to any mammary cell populations, and will therefore be useful to characterize mammary stem cells once bona fide mammary stem cell markers become available.

Mammary reconstitution assays can be used to measure the stem cell frequency within an epithelial population by transplanting increasingly diluted single-cell preparations of the population of interest. There are fundamental steps in the single-cell isolation protocol which are directly related to the number of single epithelial cells obtained. Once single-cell suspensions have been obtained, serial dilutions are prepared and transplanted into the cleared fat pads of the host mice. After 8-10 weeks, the transplanted fat pads are reevaluated for the presence of epithelial outgrowths. Based on the frequency of no outgrowth for each one of the transplanted dilutions, it is possible to estimate the frequency of mammary repopulating cells present in a given cell population. Here, we give details on how to carry out all these steps.

The regenerative potential of mammary epithelium facilitates assessment of the "stemness" of any epithelial subpopulation in transplantation assays. Thus, mammary tissue can be dissociated into single cells, stained for cell surface markers of interest and classified using fluorescence-activated cell sorting. The selected cells can then be transplanted into epithelium-devoided fat pads from recipient hosts. Recent publications have described markers that enrich for mammary repopulating potential. Here, we describe the materials and methods necessary to sort cells according to these markers. This approach can be used interchangeably with other cell surface markers with slight variation to the protocol.

We have built a novel computational microscopy platform that integrates image acquisition, storage, processing and analysis to study cell populations in situ. This platform allows high-content studies where multiple features are measured and linked at multiple scales. We used this approach to study the cellular composition and architecture of the mouse mammary gland by quantitatively tracking the distribution and type, position, proliferative state, and hormone receptor status of epithelial cells that incorporated bromodeoxyuridine while undergoing DNA synthesis during puberty and retained this label in the adult gland as a function of tissue structure. Immunofluorescence was used to identify label-retaining cells, as well as epithelial cells expressing the proteins progesterone receptor and P63. Only 3.6% of luminal cells were label-retaining cells, the majority of which did not express the progesterone receptor. Multi-scale in situ analysis revealed that luminal label-retaining cells have a distinct nuclear morphology, are enriched 3.4-fold in large ducts, and are distributed asymmetrically across the tissue. We postulated that LRC enriched in the ventral mammary gland represent progenitor cells. Epithelial cells isolated from the ventral versus the dorsal portion of the gland were enriched for the putative stem cell markers CD24 and CD49f as measured by fluorescence activated cell sorting. Thus, quantitative analysis of the cellular composition of the mammary epithelium across spatial scales identified a previously unrecognized architecture in which the ventral-most, large ducts contain a reservoir of undifferentiated, putative stem cells.

The Transforming Growth Factor beta (TGF-) superfamily includes three isoforms designated TGF-1, 2 and 3. All three isoforms are secreted as latent complex where the TGF- cytokine is non-covalently associated with an isoform specific latency-associated peptide (LAP). Mature cytokine binds cell surface receptors only after release from its LAP making extracellular activation a critical regulatory point for TGF- bioavailability. Proposed activation mechanisms include proteolysis and conformational changes. Previous work from our laboratory showed that latent TGF-1 (LTGF-1) is efficiently activated upon exposure to reactive oxygen species (ROS). ROS activation is restricted to the LTGF-1 isoform. Because of the amino acid sequence differences between the three LAPs, we postulate that the specificity of this activation mechanism lies within the LAP. Furthermore, we hypothesize that the presence of a metal in the latent complex could provide a redox active center for this process. Redox mediated activation provides a novel mechanism for TGF- participation in tissues undergoing oxidative stress. Moreover, this would allow TGF-1 to act both as a sensor of oxidative stress within tissues as well as a transducer of that signal by binding to its cellular receptors.

A major impeding factor in designing effective therapies against glioblastoma (GBM) is its extensive molecular heterogeneity and the diversity of microenvironmental conditions within any given tumor. To test whether heterogeneity with the GBM stem cell (GSC) population is required to ensure tumor growth in such diverse microenvironments, we used human GBM biospecimens to examine the identity of cells marked by two established GSC markers: CD133 and activation of the Notch pathway. Using primary GBM cultures engineered to express GFP upon activation of Notch signaling, we observed only partial overlap between cells expressing cell surface CD133 and cells with Notch activation (n = 3 specimens), contrary to expectations based on prior literature. To further investigate this finding, we FACS-isolated these cell populations and characterized them. While both CD133+ (CD133 + /Notch-) and Notch+ (CD133-/Notch+) cells fulfill GSC criteria, they differ vastly in their transcriptome, metabolic preferences and differentiation capacity, thus giving rise to histologically distinct tumors. CD133+ GSCs have increased expression of hypoxia-regulated and glycolytic genes, and are able to expand under hypoxia by activating anaerobic glycolysis. In contrast, Notch+ GSCs are unable to utilize anaerobic glycolysis under hypoxia, leading to decreased tumorsphere formation ability. While CD133+ GSCs give rise to histologically homogeneous tumors devoid of large tumor vessels, tumors initiated by Notch+ GSCs are marked by large perfusing vessels enveloped by pericytes. Using a lineage tracing system, we showed that pericytes are derived from Notch+ GSCs. In addition, Notch+ cells are able to give rise to all tumor lineages in vitro and in vivo, including CD133 + /Notch- cells, as opposed to Notch- populations, which have restricted differentiation capacity and do not generate Notch+ lineages. Our findings demonstrate that GSC heterogeneity is a mechanism used by tumors to sustain growth in diverse microenvironmental conditions.

SERMs are effective in breast cancer prevention in women at high risk, but they are not without side effects. We assessed a potential alternative approach, based on the knowledge that IGF-I is essential for E2 action in the mammary gland. Inhibitors of IGF-I block both IGF-I and E2 actions. We employed a somatostatin receptor ligand (SRL), pasireotide, which targets the mammary gland to stimulate local IGFBP5, and, as a result, blocks IGF-I binding to IGF-1R. Unlike octreotide that mainly targets SSTR2 receptors, pasireotide targets 4 of the 5 somatostatin receptors, which accounts for its unique action in potently preventing IGF-I from activating IGF-IR. Like other SRLs, pasireotide also inhibits GH secretion from the pituitary, lowering serum IGF-I. We enrolled 15 women with atypical hyperplasia (AH) diagnosed on core biopsy. They were treated with pasireotide for 10 days. 13 patients completed the trial. They had wire placement prior to surgical excision to identify the exact location of the core lesion, before surgical excision. Core and excision biopsies were stained for H&E, Ki67, TUNEL, ER and PR. Phosphorylated AKT and ERK 1/2 were assessed in a subset. Twelve of thirteen patients also had proliferative lesions (PL) on both core and excision specimens, and one had DCIS. Pasireotide decreased proliferation and increased apoptosis in all 6 AH lesions (from a mean of 3.6±2.6% to 1.3±1.2% and from 0.3±0.2% to 1.5±1.6%, respectively) and 12 proliferative lesions (from 3.8±2.5% to 1.8±1.8% and from 0.3±0.2% to 1.3±0.6%, respectively). The DCIS responded similarly. ER and PR were not affected by pasireotide, while phosphorylated ERK 1/2 and AKT decreased significantly. We also examined core and excision biopsies from untreated patients and found no change in cell proliferation or phosphorylation of AKT or ERK 1/2. In depth assessment of side effects including mild to moderate hyperglycemia associated with reduced insulin levels were studied. Glucose fell into the normal range after discontinuing treatment. Pasireotide was generally well tolerated and did not cause symptoms of estrogen deprivation. In conclusion, IGF-I inhibition was found to significantly reduce cell proliferation and increase apoptosis via effects on the downstream IGF-IR pathway in women with AH and PL. ER and PR were not affected by pasireotide. Proliferation was also decreased and apoptosis increased in one patient with DCIS. That IGF-I is known to affect DNA damage, genomic stability and stem or progenitor cells, raises the possibility that inhibition of IGF-I could have effects on other aspects of breast cancer chemoprevention, in addition to inhibiting proliferation and increasing apoptosis. Citation Format: David L. Kleinberg, Deborah Axelrod, Julia Smith, Baljit Singh, Martin Lesser, Pietro Ameri, Ann Danoff, Irineu Illa Bochaca, Cristina de Angelis. Breast cancer chemoprevention by IGF-I inhibition in women with atypical hyperplasia of the breast: A phase 1/2 proof of principle trial. [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 CT306. doi:10.1158/1538-7445.AM2014-CT306

What determines the prevalence of ERα in breast cancer is not well-understood. ER-negative breast cancer is most frequent in young women and certain racial groups, particularly African-American women. We have used a model in which p53 null mammary epithelium is orthotopically transplanted to an irradiated mouse to study radiation effects on host interactions during the development of breast cancer. We observed that host irradiation increased the rate of tumor development, which were significantly more likely to be ERα-negative and PR-negative tumors. These data implicate radiation-induced heterotypic signaling in determining critical clinical features of breast cancer. We then asked how different the expression profiles of ER-negative tumors were in sham and irradiated hosts as a means to infer whether they develop via similar paths. SAM-tandem-bootstrap identified 115 genes that cluster ER-negative tumors from irradiated versus sham-irradiated hosts, but not ER-positive tumors. Mouse Trp53 null tumors are similar to claudin-low breast cancer and both are enriched in the mammary stem cell (MaSC) signature reported by Visvader and colleagues. Since the mammary stem cell is ER-negative, as are tumors that are enriched in the MaSC signature, we asked how the MaSC profile related to the ER-115 profile that clusters tumors from irradiated hosts. Genes up-regulated in the ER-115 signature showed a highly significant (p=0.01) enrichment for genes up-regulated in the MaSC profile using ConceptGen. We also found that MaSC genes were significantly enriched in irradiated mammary glands. Together these data suggested the hypothesis that low dose radiation might affect the mammary lineage hierarchy by altering self-renewal in mammary stem cells. To test this idea, mice were irradiated with 10 or 100 cGy at 3 weeks of age and mammary cells isolated from fully mature 12 week old mice were analyzed by FACS using Cd24med/Cd49hi mammary repopulation markers. The proportion of lin-/ Cd24med/Cd49hi cells in irradiated mice was significantly increased (p Supported by DOE Low Dose Radiation Program, NIEHS Breast Cancer and the Environment Research Center, and DOD Breast Cancer Research Fellowship to DHN. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3326. doi:10.1158/1538-7445.AM2011-3326