Weisheng Wang

Primary cancer cell culture: mammary-optimized vs conditional reprogramming

The impact of different culture conditions on biology of primary cancer cells is not always addressed. Here, conditional reprogramming (CRC) was compared with mammary-optimized EpiCult-B (EpiC) for primary mammary epithelial cell isolation and propagation, allograft generation, and genome-wide transcriptional consequences using cancer and non-cancer mammary tissue from mice with different dosages of Brca1 and p53 Selective comparison to DMEM was included. Primary cultures were established with all three media, but CRC was most efficient for initial isolation (P<0.05). Allograft development was faster using cells grown in EpiC compared with CRC (P<0.05). Transcriptome comparison of paired CRC and EpiC cultures revealed 1700 differentially expressed genes by passage 20. CRC promoted Trp53 gene family upregulation and increased expression of epithelial differentiation genes, whereas EpiC elevated expression of epithelial-mesenchymal transition genes. Differences did not persist in allografts where both methods yielded allografts with relatively similar transcriptomes. Restricting passage (<7) reduced numbers of differentially expressed genes below 50. In conclusion, CRC was most efficient for initial cell isolation but EpiC was quicker for allograft generation. The extensive culture-specific gene expression patterns that emerged with longer passage could be limited by reducing passage number when both culture transcriptomes were equally similar to that of the primary tissue. Defining impact of culture condition and passage on the transcriptome of primary cells could assist experimental design and interpretation. For example, differences that appear with passage and culture condition are potentially exploitable for comparative studies targeting specific biological networks in different transcriptional environments.

Responsiveness of Brca1 and Trp53 Deficiency–Induced Mammary Preneoplasia to Selective Estrogen Modulators versus an Aromatase Inhibitor in Mus musculus

An intervention study initiated at age 4 months compared the impact of tamoxifen (25 mg), raloxifene (22.5 mg), and letrozole (2.5 mg) administered by 60-day release subcutaneous pellet on mammary preneoplasia prevalence at age 6 months in conditional genetically engineered mouse models with different Breast cancer 1 (Brca1) gene dosages targeted to mammary epithelial cells and germline Tumor protein P53 (Trp53) haploinsufficiency (10–16/cohort). The proportion of unexposed control mice demonstrating mammary preneoplasia at age 6 months was highest in Brca1fl11/fl11/Cre/p53−/+ (54%) mice followed by Brca1WT/fl11/Cre/p53−/+ mice (30%). By age 12 months, invasive mammary cancers appeared in 80% of Brca1fl11/fl11/Cre/p53−/+and 42% of Brca1WT/fl11/Cre/p53−/+control unexposed mice. The spectrum of cancer histology was similar in both models without somatic mutation of the nongenetically engineered Brca1, Trp53, Brca2, or Death-associated protein kinase 3 (Dapk3) alleles. Two-month exposure to tamoxifen, raloxifene, and letrozole significantly reduced estrogen-mediated tertiary branching by 65%, 71%, and 78%, respectively, in Brca1fl11/fl11/Cre/p53−/+mice at age 6 months. However, only letrozole significantly reduced hyperplastic alveolar nodules (HAN) prevalence (by 52%) and number (by 30%) and invasive cancer appeared despite tamoxifen exposure. In contrast, tamoxifen significantly reduced HAN number by 95% in Brca1WT/fl11/Cre/p53−/+ mice. Control mice with varying combinations of the different genetically modified alleles and MMTV-Cre transgene demonstrated that the combination of Brca1 insufficiency and Trp53 haploinsufficiency was required for appearance of preneoplasia and no individual genetic alteration confounded the response to tamoxifen. In summary, although specific antihormonal approaches showed effectiveness, with Brca1 gene dosage implicated as a possible modifying variable, more effective chemopreventive approaches for Brca1 mutation–induced cancer may require alternative and/or additional agents. Cancer Prev Res; 10(4); 244–54. ©2017 AACR.

Expanding primary cells from mucoepidermoid and other salivary gland neoplasms for genetic and chemosensitivity testing

ABSTRACT Restricted availability of cell and animal models is a rate-limiting step for investigation of salivary gland neoplasm pathophysiology and therapeutic response. Conditionally reprogrammed cell (CRC) technology enables establishment of primary epithelial cell cultures from patient material. This study tested a translational workflow for acquisition, expansion and testing of CRC-derived primary cultures of salivary gland neoplasms from patients presenting to an academic surgical practice. Results showed that cultured cells were sufficient for epithelial cell-specific transcriptome characterization to detect candidate therapeutic pathways and fusion genes, and for screening for cancer risk-associated single nucleotide polymorphisms (SNPs) and driver gene mutations through exome sequencing. Focused study of primary cultures of a low-grade mucoepidermoid carcinoma demonstrated amphiregulin-mechanistic target of rapamycin-protein kinase B (AKT; AKT1) pathway activation, identified through bioinformatics and subsequently confirmed as present in primary tissue and preserved through different secondary 2D and 3D culture media and xenografts. Candidate therapeutic testing showed that the allosteric AKT inhibitor MK2206 reproducibly inhibited cell survival across different culture formats. By contrast, the cells appeared resistant to the adenosine triphosphate competitive AKT inhibitor GSK690693. Procedures employed here illustrate an approach for reproducibly obtaining material for pathophysiological studies of salivary gland neoplasms, and other less common epithelial cancer types, that can be executed without compromising pathological examination of patient specimens. The approach permits combined genetic and cell-based physiological and therapeutic investigations in addition to more traditional pathologic studies, and can be used to build sustainable bio-banks for future inquiries. This article has an associated First Person interview with the first author of the paper. Summary: Genetics and therapeutic responses of uncommon neoplasms, including salivary gland cancers, are investigatable through isolation of primary cells from patient material using conditional reprogramming cell technology without compromising pathological diagnosis.

Abstract 656: Histological and 3D morphological evaluation of mammary cancers and primary cells from genetically engineered mice with only one copy of Brca1 disrupted in combination with Trp53 haploinsufficiency

Background: Previously established genetically engineered mouse (GEM) models with spontaneous mammary cancer development have both Brca1 alleles disrupted withTrp53 haploinsufficiency (Brca1f11/f11/MMTV-Cre/Trp53+/-). Cell culture plates with nanoimprinted scaffolds (SCIVAX Life Sciences, Inc) purportedly increase concordance between in vitro/in vivo results but reports are limited to human cancer cell lines. Here we characterized hyperplasia and cancer development in Brca1f11/WT11/MMTV-Cre/Trp53+/- mice and established conditions for 3D culture of primary mammary epithelial cells (MEC) using the imprinted plates. Methods: Time-course of mammary hyperplasia and cancer development was defined in female Brca1f11/WT11/MMTV-Cre/Trp53+/- (n = 23) and Brca1f11/f11/MMTV-Cre/Trp53+/- (n = 13) mice euthanized at age 6 and 12 months (m) or when largest tumor reached 1 cm3. Mammary tissue was taken at necropsy for histology (age 6/12m) and primary MEC culture (age 6m) using EpiCult-B (StemCell Technologies) on nanoimprinted scaffold plates. Primary normal (wild-type, n = 2) and precancerous MEC (Brca1f11/f11/MMTV-Cre/Trp53+/-, n = 2; Brca1f11/WT11/MMTV-Cre/Trp53+/-; n = 2) were cultured. Optimal conditions for spheroid growth were established by comparison of spheroid formation from thoracic vs. inguinal mammary glands with different numbers of plated cells and percentage FBS after 7 days culture. Endpoints included sphere number, size and presence/absence of concurrent monolayer growth. Spheres were harvested for histological examination. Hematoxylin and eosin (HE 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 656.

Abstract 4280: Paralleled workflow for expansion of limited patient material using CRC for in vitro/PDX/biological/genetic studies of a low-grade mucoepidermoid carcinoma

Acquiring live cancer cells preserving original characteristics is a critical step for precision medicine. Here we illustrate a paralleled workflow using a low-grade sublingual salivary mucoepidermoid carcinoma (MEC) as an example. Methods: Working under IRB approval, tissue from two distinct regions of a MEC were obtained and primary epithelial cell cultures established using CRC (F medium with Rho Kinase (ROCK) inhibitor Y-27632 with irradiated Swiss 3T3-J2 mouse fibroblast feeder cells). Epithelial cells were separated from feeders to test under non-CRC conditions including 2D (MammoCult™), colony formation in Matrigel (F medium, Y-27632), patient derived xenograft (PDX) formation (10*6 cells/50:50 matrigel/PBS/mammary fatpad), and RNA and DNA extracted for RNAseq, exome sequencing and PCR/RT-PCR for a CRTC1-MAML2 fusion gene (reported in MEC). Sequences were analyzed for relative transcript abundance, differentially expressed genes (DEGs), and single nucleotide polymorphisms (SNPs) (TopHat/CuffLinks, SAMTOOLS, variant database: dbSNP & 1000G, UCSC hg19) between the two MEC regions and fusion genes (FusionCatcher). Potentially pathophysiological SNPs were identified (OMIM and SNPedia). Gene ontology was performed on transcriptome data (PARTEK Genomic Suite, Pathway Studio). Immunohistochemistry (IHC) was used to compare protein expression of selected DEG and downstream effectors in original cancer and surrounding normal tissue, cell pellets, Matrigel colonies and PDX (confirmed as human using MAB1273). Results: CRC cultures established from both sites grew similarly under all in vitro and in vivo conditions. PDX showed well-differentiated histology comparable to original cancer. One of the 11 DEG was amphiregulin but protein expression was equivalent in the two cell cultures. Amphiregulin, EGFR, p-EGFR, p-AKT, and mTOR were expressed in vitro (CRC and non-CRC), in vivo (PDX) and in original cancer, where they were up-regulated as compared to surrounding normal tissue. Five potentially pathophysiologically significant SNPs were detected (BRCA2 (rs144848), TP53 (rs1042522), AURKA (rs2273535), DBYD (rs1801265), and (SOD2 rs4880)) but no CRCTC1-MAML2 fusion gene. FusionCatcher detected a possible novel fusion product currently under investigation. Conclusion: The paralleled approach provided sufficient material to identify amphiregulin as an upregulated growth factor pathway in a CRTC1-MAML2 fusion gene negative MEC. Previously, amphiregulin upregulation was pathophysiologically linked to the CRTC1-MAML2 fusion gene. Results from the two MEC sites were biologically concordant and genetically similar. In summary, CRC can be used to expand limited patient derived material for more extensive studies under non-CRC conditions. Support: R56DE023259 (PAF) Citation Format: Ahmad M. Alamri, Xuefeng Liu, Weisheng Wang, Xiaogang Zhong, Bhaskar Kallakury, Bruce Davidson, Priscilla A. Furth. Paralleled workflow for expansion of limited patient material using CRC for in vitro/PDX/biological/genetic studies of a low-grade mucoepidermoid carcinoma. [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 4280.

Mammary Gland-Endocrinology

Abstract The field of endocrinology spans basic physiology through clinical medical application. It focuses on organs and cell types that generate hormones and their signaling through receptors in target organs. The mammary gland both produces hormones and receives exogenous hormonal stimulation. In the mammary gland these include both steroid hormones such as estrogen and progesterone linked to sexual development and cytokine-like molecules including prolactin and related molecules that mediate lactation. Both types of hormones have been associated with breast cancer. Here we will review endogenous and exogenous mammary gland hormones, cognate receptor expression, and define influences on physiology and disease.

Abstract 5576: Tamoxifen fails to induce regression of mammary preneoplasia in mice lacking either one or two intact BRCA1 genes in combination with p53 haploinsufficiency but without evidence of in vivo agonist activity

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Background: Reduced BRCA1 and/or p53 function in mammary tissue is associated with tamoxifen resistance in vivo and in vitro. A shift towards tamoxifen agonist activity with reduced BRCA1 function is found in vitro. The goal of this study was to identify which combinations of disrupted Brca1 and p53 gene(s) result in preneoplasia and test for tamoxifen resistance and activity in vivo. Methods: Brca1 homozygous and heterozygous conditional exon 11 knockout and wildtype mice all carrying MMTV-Cre on a p53 wildtype or haploinsufficient background received a 25 mg 60-day constant release tamoxifen pellet (Innovative Research of America, Sarasota, FL) or sham surgery at age 4m, euthanasia/necropsy at age 6m and mammary glands taken for whole mount, histological and western blot (pERK / ERK and pHistone/Histone ratios, cleaved PARP). Cohorts of untreated Brca1f11/WT11/MMTV-Cre/p53+/- and Brca1f11/f11/MMTV-Cre/p53+/- mice were observed to age 12m for mammary cancer. Fishers exact and Kruskal-Wallis tests were used for statistical analyses. Results: Hyperplastic alveolar nodules at age 6m and cancer development at age 12m were limited to Brca1f11/WT11/MMTV-Cre/p53+/- and Brca1f11/f11/MMTV-Cre/p53+/- mice. Ductal hyperplasia was detected in Brca1f11/WT11/MMTV-Cre/p53+/+, Brca1f11/f11/MMTV-Cre/p53+/+, Brca1f11/WT11/MMTV-Cre/p53+/-, Brca1f11/f11/MMTV-Cre/p53+/- and Brca1WT11/WT11/MMTV-Cre/p53+/- but not wildtype mice. Only Brca1f11/WT11/MMTV-Cre/p53+/+ and Brca1f11/f11/MMTV-Cre/p53+/+ mice showed significant reductions in ductal hyperplasia on tamoxifen. In contrast prevalence of ductal hyperplasia and/or hyperplastic alveolar nodules was not altered by tamoxifen in Brca1WT11/WT11/MMTV-Cre/p53+/-, Brca1f11/WT11/MMTV-Cre/p53+/- or Brca1f11/f11/MMTV-Cre/p53+/- mice. No evidence of tamoxifen activity was found as proliferation measures were either reduced (percentage mammary epithelial cells with nuclear-localized PCNA) or unchanged (pERK / ERK and pHistone/Histone ratios) with tamoxifen treatment in all genotypes. Cleaved PARP was detected in tamoxifen-treated but not control mice. Conclusions: p53 haploinsufficiency in combination with loss of either one or two copies of intact Brca1 was sufficient to induce hyperplastic alveolar nodules and cancer. Tamoxifen resistance was found in all p53 haploinsufficient mice. There was no indication of a shift towards tamoxifen agonist activity, that is, no increase in expression of any proliferative marker with Brca1 deficiency, and apoptosis was activated. Alternative pathways for tamoxifen resistance involving p53 may be responsible for the tamoxifen resistance observed in Brca1 deficient preneoplastic mammary epithelial cells. Citation Format: Sahar J. Alothman, Weisheng Wang, Priscilla A. Furth. Tamoxifen fails to induce regression of mammary preneoplasia in mice lacking either one or two intact BRCA1 genes in combination with p53 haploinsufficiency but without evidence of in vivo agonist activity. [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 5576. doi:10.1158/1538-7445.AM2014-5576

Abstract 3918: Characterizing growth features, allograft generation and transcriptomes of cultured conditionally reprogrammed cells (CRC) prepared from primary triple negative cancer from Brca1-mutant mice

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Background: Efficient acquisition of primary cancer cells for genetic and pharmacological testing is a means towards individualized cancer treatment. Genetically engineered mouse models of triple negative mammary cancer were used to compare culture conditions conditionally reprogrammed cell culture CRC with a rho kinase inhibitor (Y-27632) and irradiated J2 feeder cells, EpiCult™, DMEM for establishment efficiency, differentiation, fibroblast contamination, and allograft generation. Goals: 1) Evaluate mammary cancer progenitor cells by allograft and secondary culture 2) Compare RNA-seq transcriptomes in cells and allografts with different Brca1 gene dosages (100%, 50%, 0%) Methods: Triple negative invasive adenocarcinomas from 4 mouse genotypes, all carrying a p53-null allele, were analyzed: A) 1 mutant Brca1 allele; B) 2 mutant Brca1 alleles; C) 1 mutant Brca1 allele and 1 human aromatase transgene; D) human aromatase transgene. Cancers were divided and growth assessed in CRC, EpiCult™ and DMEM cultures. Illumina (V2scriptseq) RNA-seq libraries were prepared from cell pellets and allografts. One million cells injected into mammary gland fat pads of nude mice generated allografts divided for histology/DNA analyses, secondary culture and RNAseq. Transcript abundance and differentially expressed genes were determined after assembling transcriptomes with Cufflinks. Results: CRC was the most efficient methodology based on rapidity and percentage cancer specimens successfully cultured and paucity of fibroblasts. Brca1 gene dosage made no difference in culture efficiency. Differentiation markers were expressed at significantly higher levels in CRC whereas markers of epithelial mesenchymal transition were expressed higher in EpiCult™. Allografts were derived from CRC and EpiCult™ but not DMEM. Genetic testing confirmed allografts were derived from original endogenous cancer cells. Some individual allografts derived from the same culture exhibited different growth curves and histology, suggesting tumor cell heterogeneity within cultures. Overall, palpable EpiCult™ allografts appeared significantly earlier than CRC allografts (p<0.05 Chi Square). RNAseq analyses distinguished gene expression patterns dependent upon culture condition as distinct from those determined by underlying genetics. Summary: CRC technology was an efficient means of generating primary cancer cell cultures, maintained epithelial cell differentiation better than other methods, had no fibroblast contamination, and preserved cancer progenitor cells. Citation Format: Ahmad M. Alamri, Svenja Groeneveld, Keunsoo Kang, Sarah Dabydeen, Weisheng Wang, Lothar Hennighausen, Bhaskar Kallakury, Xuefeng Liu, Priscilla A. Furth. Characterizing growth features, allograft generation and transcriptomes of cultured conditionally reprogrammed cells (CRC) prepared from primary triple negative cancer from Brca1-mutant mice. [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 3918. doi:10.1158/1538-7445.AM2014-3918

Abstract 2972: Conditionally reprogrammed cells (CRCs): A new model for cancer research and personalized medicine

Background: Current cancer cell lines often fail to reflect the genotypes and phenotypes of the tumors from which they were derived due to the accumulation of genetic and epigenetic alterations during passage in vitro. This obviously limits the ability to use such cell lines for predicting responses to drug-, radiation-, or immuno-therapies. Until recently, it remained a challenge to rapidly and efficiently generate cell cultures from individual patients in a time-limited manner that allowed for choosing appropriate therapies. Last year we described the use feeder cells and a ROCK inhibitor to induce the conditional reprogramming of adult epithelial cells into a basal or stem-like (AJP, 2012, 2013; NEJM, 2012, PNAS 2012). Cultures generated from normal tissue, referred to as conditionally reprogrammed cells (CRCs), do not express high levels of proteins characteristic of iPSCs or ESCs such as Sox2, Oct4, Nanog, or Klf4. More important, the induction of CRCs is reversible, and the removal of feeders and ROCK inhibitor allows cells to differentiate normally to tissue origin. The CRC technology can generate 2×10 6 cells in 5 to 6 days from needle biopsies, and can generate cultures from cryopreserved tissue and from fewer than four viable cells. Primary goal: The clinical utility of CRCs for patient care needs to be established. We therefore initiated studies to examine whether CRCs established from tumors would reflect the biology and genotype of the original tumor and whether the tumor CRCs could be used to predict clinical responses. Procedures: We used CRC methods to generate matched cultures from both tumor cells and adjacent normal cells from cancer patients and relevant mouse models. We characterized these CRC lines for their growth properties, induction of tumors in immunodeficient mice, karyotype, differentiation, and transcriptome profile. We also compared the sensitivity of the matched CRCs to commonly used chemotherapy drugs. Results: The CRC technique efficiently established cell cultures from human and rodent tumors. For example, we established matched normal and tumor CRCs for a patient with a squamous carcinoma of the tongue. The tumor CRC from this patient exhibited a highly abnormal karyotype and harbored a mutation of the p53 gene. The matched normal CRC exhibited a normal karyotype and wild p53. The tumor CRCs, but not the normal CRCs, efficiently induced squamous cell carcinomas when injected subcutaneously into nude mice. Similarly, we generated 4 cultures from pancreatic cancers, 3 of which exhibited mutations in the Ras gene. We were also able to utilize the CRC method to define an effective therapy for patient with an aggressive lung papillomatosis. Finally, our studies indicate that we are able to generate micro-heterogenous tumor CRCs from a small biopsy. Conclusion: CRCs show promise for evaluating tumor responses to selected therapies and for defining the functional heterogeneity of their respective primary tumors. Citation Format: Xuefeng Liu, Ewa Krawczyk, Nancy Palechor-Ceron, Weisheng Wang, Hang Yuan, Aleksandra Dakic, Vera Simic, Bhaskar Kullakury, Priscilla Furth, Richard Schlegel. Conditionally reprogrammed cells (CRCs): A new model for cancer research and personalized medicine. [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 2972. doi:10.1158/1538-7445.AM2014-2972

Abstract 197: Reduction in lobular but not ductal hyperplasia by the PGR inhibitor ORG33628 in a CYP19A1 overexpressing mouse model of breast cancer risk.

Background: The Women9s Health Initiative demonstrated increased breast cancer risk in women exposed to estrogen with progesterone. Genetically engineered mice with conditional mammary epithelial cell targeted Cytochrome P45019A1 (CYP19A1/aromatase) over-expression (tet-op-CYP19A1/MMTV-rtTA mice) tested the impact of a selective progesterone receptor (PGR) inhibitor, ORG33628, on mammary hyperplasia and cancer development. This model demonstrates increased percentages of PGR+ mammary epithelial cells (MECs) cells at the hyperplastic stage and triple negative (Estrogen Receptor (ER)-, PGR-, Receptor tyrosine-protein kinase erbB-2 (HER2/neu)-) adenocarcinomas. Purpose: Test if ORG33628 reduces mammary hyperplasia and blocks cancer formation in tet-op-CYP19A1/MMTV-rtTA mice. Methods: Female tet-op-CYP19A1/MMTV-rtTA mice were implanted subcutaneously with a pellet containing either ORG33628 7.5mg (60-day release) or placebo at age 10m. A smaller study tested ORG33628 in CYP19A1/MMTV-rtTA/trp53+/- mice. Mice were necropsied at age 12m and mammary tissue examined to determine if ORG33628 altered lobular/ductal hyperplasia prevalence or blocked progression to hyperplastic alveolar nodules (HANs) or cancer. Immunohistochemistry (IHC) was employed to compare MEC proliferative index (PI) (Ki-67) and percentage of MECs expressing nuclear-localized ER/PGR. An uterotropic assay tested if ORG33628 blocked estrogen-induced changes in the uterus/cervix. Ovariectomized 6w old mice were implanted subcutaneously with pellets of 0.72g 17β-estradiol and either ORG33628 or placebo, necropsied 3 days later, and uterine weight, PI and ER/PGR expression compared in uterine and cervical epithelium and stroma. Results: ORG33628 treatment reduced lobular hyperplasia prevalence (Fisher9s exact, p Conclusions: ORG33628 reduced lobular but not ductal hyperplasia and did not prevent progression to HANs or cancer. In normal development, estrogen plays the more important role in ductal development and progesterone in lobular development paralleling the isolated impact of ORG33628 on lobular disease here. These results suggest that reducing ductal hyperplasia may be a prerequisite for cancer prevention or that reducing lobular hyperplasia at 10m in this model is too late due to the presence of triple negative cancer progenitor cells. Support: P30CA051008, RO1CA112176, T32CA009686-15, KG080359. Content solely responsibility of authors and does not necessarily represent official NCI/NIH views. Citation Format: Sarah A. Dabydeen, Weisheng Wang, Edgar S. Diaz-Cruz, Levy Kopelovich, Robert I. Glazer, Priscilla A. Furth. Reduction in lobular but not ductal hyperplasia by the PGR inhibitor ORG33628 in a CYP19A1 overexpressing mouse model of breast cancer risk. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 197. doi:10.1158/1538-7445.AM2013-197