Portal de Periódicos da CAPES

Early Dysregulation of Cell Adhesion and Extracellular Matrix Pathways in Breast Cancer Progression

2009; Elsevier BV; Volume: 175; Issue: 3 Linguagem: Inglês

10.2353/ajpath.2009.090115

1525-2191

Lyndsey A. Emery, Anusri Tripathi�, Chialin King, Maureen Kavanah, Jane Mendez, Michael D. Stone, Antonio de las Morenas, Paola Sebastiani, Carol L. Rosenberg,

Cancer Cells and Metastasis

Proliferative breast lesions, such as simple ductal hyperplasia (SH) and atypical ductal hyperplasia (ADH), are candidate precursors to ductal carcinoma in situ (DCIS) and invasive cancer. To better understand the relationship of breast lesions to more advanced disease, we used microdissection and DNA microarrays to profile the gene expression of patient-matched histologically normal (HN), ADH, and DCIS from 12 patients with estrogen receptor positive sporadic breast cancer. SH were profiled from a subset of cases. We found 837 differentially expressed genes between DCIS-HN and 447 between ADH-HN, with >90% of the ADH-HN genes also present among the DCIS-HN genes. Only 61 genes were identified between ADH-DCIS. Expression differences were reproduced in an independent cohort of patient-matched lesions by quantitative real-time PCR. Many breast cancer-related genes and pathways were dysregulated in ADH and maintained in DCIS. Particularly, cell adhesion and extracellular matrix interactions were overrepresented. Focal adhesion was the top pathway in each gene set. We conclude that ADH and DCIS share highly similar gene expression and are distinct from HN. In contrast, SH appear more similar to HN. These data provide genetic evidence that ADH (but not SH) are often precursors to cancer and suggest cancer-related genetic changes, particularly adhesion and extracellular matrix pathways, are dysregulated before invasion and even before malignancy is apparent. These findings could lead to novel risk stratification, prevention, and treatment approaches. Proliferative breast lesions, such as simple ductal hyperplasia (SH) and atypical ductal hyperplasia (ADH), are candidate precursors to ductal carcinoma in situ (DCIS) and invasive cancer. To better understand the relationship of breast lesions to more advanced disease, we used microdissection and DNA microarrays to profile the gene expression of patient-matched histologically normal (HN), ADH, and DCIS from 12 patients with estrogen receptor positive sporadic breast cancer. SH were profiled from a subset of cases. We found 837 differentially expressed genes between DCIS-HN and 447 between ADH-HN, with >90% of the ADH-HN genes also present among the DCIS-HN genes. Only 61 genes were identified between ADH-DCIS. Expression differences were reproduced in an independent cohort of patient-matched lesions by quantitative real-time PCR. Many breast cancer-related genes and pathways were dysregulated in ADH and maintained in DCIS. Particularly, cell adhesion and extracellular matrix interactions were overrepresented. Focal adhesion was the top pathway in each gene set. We conclude that ADH and DCIS share highly similar gene expression and are distinct from HN. In contrast, SH appear more similar to HN. These data provide genetic evidence that ADH (but not SH) are often precursors to cancer and suggest cancer-related genetic changes, particularly adhesion and extracellular matrix pathways, are dysregulated before invasion and even before malignancy is apparent. These findings could lead to novel risk stratification, prevention, and treatment approaches. Terminal ductal lobular units are the functional unit of the breast. Each lobule is composed of a layer of luminal epithelial cells surrounded by a basal layer of myoepithelial cells (MECs), and supported by a heterogeneous stromal microenvironment. One widely accepted model of breast cancer progression proposes that normal epithelial cells of the terminal ductal lobular unit move through a series of distinct precursors that acquire increasingly aberrant histological and molecular abnormalities, resulting in invasive cancer.1Wellings SR Jensen HM Marcum RG An atlas of subgross pathology of the human breast with special reference to possible precancerous lesions.J Natl Cancer Inst. 1975; 55: 231-273PubMed Google Scholar, 2Allred DC Wu Y Mao S Nagtegaal ID Lee S Perou CM Mohsin SK O'Connell P Tsimelzon A Medina D Ductal carcinoma in situ and the emergence of diversity during breast cancer evolution.Clin Cancer Res. 2008; 14: 370-378Crossref PubMed Scopus (232) Google Scholar In an attempt to understand these molecular abnormalities, extensive gene expression studies have been done in invasive breast cancers. These studies have revealed that invasive breast cancer is a heterogeneous disease that can be subclassified based on distinct molecular features.3Sotiriou C Pusztai L Gene-expression signatures in breast cancer.N Engl J Med. 2009; 360: 790-800Crossref PubMed Scopus (1143) Google Scholar These classifications have been repeatedly correlated with clinically relevant tumor phenotypes, including hormone receptor status and histological grade.3Sotiriou C Pusztai L Gene-expression signatures in breast cancer.N Engl J Med. 2009; 360: 790-800Crossref PubMed Scopus (1143) Google Scholar Understanding the molecular changes that occur before the development of invasive cancer is important and is the focus of our studies. Ductal carcinoma in situ (DCIS) is a nonobligate precursor to invasive breast cancer.4Kuerer HM Albarracin CT Yang WT Cardiff RD Brewster AM Symmans WF Hylton NM Middleton LP Krishnamurthy S Perkins GH Babiera G Edgerton ME Czerniecki BJ Arun BK Hortobagyi GN Ductal carcinoma in situ: state of the science and roadmap to advance the field.J Clin Oncol. 2009; 27: 279-288Crossref PubMed Scopus (123) Google Scholar Approximately 50% of DCIS will progress to invasive cancer if left untreated, and 12% to 20% will recur within 10 years despite appropriate treatment.5Thompson A Brennan K Cox A Gee J Harcourt D Harris A Harvie M Holen I Howell A Nicholson R Steel M Streuli C Evaluation of the current knowledge limitations in breast cancer research: a gap analysis.Breast Cancer Res. 2008; 10: R26Crossref PubMed Scopus (80) Google Scholar Gene expression studies have revealed similarities between DCIS and invasive ductal carcinoma (IDC).6Ma XJ Salunga R Tuggle JT Gaudet J Enright E McQuary P Payette T Pistone M Stecker K Zhang BM Zhou YX Varnholt H Smith B Gadd M Chatfield E Kessler J Baer TM Erlander MG Sgroi DC Gene expression profiles of human breast cancer progression.Proc Natl Acad Sci USA. 2003; 100: 5974-5979Crossref PubMed Scopus (730) Google Scholar, 7Vincent-Salomon A Lucchesi C Gruel N Raynal V Pierron G Goudefroye R Reyal F Radvanyi F Salmon R Thiery JP Sastre-Garau X Sigal-Zafrani B Fourquet A Delattre O Integrated genomic and transcriptomic analysis of ductal carcinoma in situ of the breast.Clin Cancer Res. 2008; 14: 1956-1965Crossref PubMed Scopus (128) Google Scholar In fact, DCIS have been found to be more similar to their IDC component than to other DCIS, suggesting that DCIS are both direct precursors to invasive cancer and are highly heterogeneous.2Allred DC Wu Y Mao S Nagtegaal ID Lee S Perou CM Mohsin SK O'Connell P Tsimelzon A Medina D Ductal carcinoma in situ and the emergence of diversity during breast cancer evolution.Clin Cancer Res. 2008; 14: 370-378Crossref PubMed Scopus (232) Google Scholar, 6Ma XJ Salunga R Tuggle JT Gaudet J Enright E McQuary P Payette T Pistone M Stecker K Zhang BM Zhou YX Varnholt H Smith B Gadd M Chatfield E Kessler J Baer TM Erlander MG Sgroi DC Gene expression profiles of human breast cancer progression.Proc Natl Acad Sci USA. 2003; 100: 5974-5979Crossref PubMed Scopus (730) Google Scholar, 7Vincent-Salomon A Lucchesi C Gruel N Raynal V Pierron G Goudefroye R Reyal F Radvanyi F Salmon R Thiery JP Sastre-Garau X Sigal-Zafrani B Fourquet A Delattre O Integrated genomic and transcriptomic analysis of ductal carcinoma in situ of the breast.Clin Cancer Res. 2008; 14: 1956-1965Crossref PubMed Scopus (128) Google Scholar, 8Tamimi RM Baer HJ Marotti J Galan M Galaburda L Fu Y Deitz AC Connolly JL Schnitt SJ Colditz GA Collins LC Comparison of molecular phenotypes of ductal carcinoma in situ and invasive breast cancer.Breast Cancer Res. 2008; 10: R67Crossref PubMed Scopus (247) Google Scholar, 9Iakovlev VV Arneson NC Wong V Wang C Leung S Iakovleva G Warren K Pintilie M Done SJ Genomic differences between pure ductal carcinoma in situ of the breast and that associated with invasive disease: a calibrated aCGH study.Clin Cancer Res. 2008; 14: 4446-4454Crossref PubMed Scopus (48) Google Scholar Although what determines the transition from DCIS to IDC remains uncertain, MECs, stromal microenvironment, and extracellular matrix (ECM) may play an important role.10Hu M Yao J Carroll DK Weremowicz S Chen H Carrasco D Richardson A Violette S Nikolskaya T Nikolsky Y Bauerlein EL Hahn WC Gelman RS Allred C Bissell MJ Schnitt S Polyak K Regulation of in situ to invasive breast carcinoma transition.Cancer Cell. 2008; 13: 394-406Abstract Full Text Full Text PDF PubMed Scopus (396) Google Scholar Despite advances in understanding the progression of DCIS to invasive cancer, current knowledge of earlier events in breast tumorigenesis is limited. Proliferative breast lesions, such as simple ductal hyperplasia (SH) and atypical ductal hyperplasia (ADH), may be important precursors to DCIS and IDC. ADH are morphologically heterogeneous and have some, but not all, of the cytologic and architectural features of DCIS.11Moulis S Sgroi DC Re-evaluating early breast neoplasia.Breast Cancer Res. 2008; 10: 302Crossref PubMed Scopus (19) Google Scholar An ADH diagnosis confers up to a fourfold increased risk of breast cancer, but only 15% of women with ADH develop invasive disease.12Hartmann LC Sellers TA Frost MH Lingle WL Degnim AC Ghosh K Vierkant RA Maloney SD Pankratz VS Hillman DW Suman VJ Johnson J Blake C Tlsty T Vachon CM Melton 3rd, LJ Visscher DW Benign breast disease and the risk of breast cancer.N Engl J Med. 2005; 353: 229-237Crossref PubMed Scopus (680) Google Scholar, 13Degnim AC Visscher DW Berman HK Frost MH Sellers TA Vierkant RA Maloney SD Pankratz VS de Groen PC Lingle WL Ghosh K Penheiter L Tlsty T Melton 3rd, LJ Reynolds CA Hartmann LC Stratification of breast cancer risk in women with atypia: a Mayo cohort study.J Clin Oncol. 2007; 25: 2671-2677Crossref PubMed Scopus (209) Google Scholar Little is known about ADH’s genetic abnormalities. Loss of heterozygosity studies reveal that ADH are genetically related, although not identical, to DCIS.11Moulis S Sgroi DC Re-evaluating early breast neoplasia.Breast Cancer Res. 2008; 10: 302Crossref PubMed Scopus (19) Google Scholar, 14Larson PS de las Morenas A Cerda SR Bennett SR Cupples LA Rosenberg CL Quantitative analysis of allele imbalance supports atypical ductal hyperplasia lesions as direct breast cancer precursors.J Pathol. 2006; 209: 307-316Crossref PubMed Scopus (54) Google Scholar, 15Chuaqui RF Zhuang Z Emmert-Buck MR Liotta LA Merino MJ Analysis of loss of heterozygosity on chromosome 11q13 in atypical ductal hyperplasia and in situ carcinoma of the breast.Am J Pathol. 1997; 150: 297-303PubMed Google Scholar, 16O'Connell P Pekkel V Fuqua SA Osborne CK Clark GM Allred DC Analysis of loss of heterozygosity in 399 premalignant breast lesions at 15 genetic loci.J Natl Cancer Inst. 1998; 90: 697-703Crossref PubMed Scopus (294) Google Scholar Gene expression analyses of ADH are severely limited. We are aware of only two studies: one including nine patient-matched ADH (whose data are not publicly available), and another including eight samples limited to only ADH.6Ma XJ Salunga R Tuggle JT Gaudet J Enright E McQuary P Payette T Pistone M Stecker K Zhang BM Zhou YX Varnholt H Smith B Gadd M Chatfield E Kessler J Baer TM Erlander MG Sgroi DC Gene expression profiles of human breast cancer progression.Proc Natl Acad Sci USA. 2003; 100: 5974-5979Crossref PubMed Scopus (730) Google Scholar, 17Poola I DeWitty RL Marshalleck JJ Bhatnagar R Abraham J Leffall LD Identification of MMP-1 as a putative breast cancer predictive marker by global gene expression analysis.Nat Med. 2005; 11: 481-483Crossref PubMed Scopus (149) Google Scholar SH represent a large, heterogeneous group of proliferative breast lesions. While very common, only 5% of women diagnosed with SH eventually develop carcinomas.12Hartmann LC Sellers TA Frost MH Lingle WL Degnim AC Ghosh K Vierkant RA Maloney SD Pankratz VS Hillman DW Suman VJ Johnson J Blake C Tlsty T Vachon CM Melton 3rd, LJ Visscher DW Benign breast disease and the risk of breast cancer.N Engl J Med. 2005; 353: 229-237Crossref PubMed Scopus (680) Google Scholar There are relatively few studies of these lesions’ genetic alterations, and interpretation of their results is complicated because some report few chromosomal abnormalities and others report many.18Reis-Filho JS Lakhani SR The diagnosis and management of pre-invasive breast disease: genetic alterations in pre-invasive lesions.Breast Cancer Res. 2003; 5: 313-319Crossref PubMed Scopus (116) Google Scholar Little is known about the genetic relationship of these lesions with cancer. While a small subset of SH may be clonal precursors to invasive cancer, supporting evidence remains weak.18Reis-Filho JS Lakhani SR The diagnosis and management of pre-invasive breast disease: genetic alterations in pre-invasive lesions.Breast Cancer Res. 2003; 5: 313-319Crossref PubMed Scopus (116) Google Scholar Taken together, there is a paucity of information regarding the gene expression abnormalities associated with breast cancer initiation and progression. Using microdissection to isolate co-existing, or patient-matched, histologically normal (HN), SH (when available), ADH, and DCIS epithelium from estrogen receptor positive (ER+) sporadic breast cancers, we characterized each lesion’s gene expression profile. This design accounts for many patient-specific variables such as age, medication use, and individual genomic variation. Our goal is to reveal the genetic relationships between normal epithelium, proliferative lesions, and preinvasive breast cancer (and by extension invasive disease). Ultimately, we aim to better understand breast cancer progression, improve precursor risk stratification, and identify novel prevention and treatment strategies. Primary breast tissue not needed for diagnosis was obtained with appropriate institutional review board approval from surgical oncology patients at Boston University Medical Center. Once surgical specimen were procured, tissue was snap frozen in liquid nitrogen and immediately embedded in frozen tissue blocks by using TissueTek OCT media (VWR Scientific Products Corporation, San Diego, CA) to ensure RNA integrity. Twelve cases were chosen based on a diagnosis of ER+ DCIS or IDC and the absence of chemo- or radiation therapy before surgery. Frozen tissue blocks were cryosectioned into serial 10 μm sections. H&E-stained sections were reviewed by a breast pathologist who identified lesions of interest. Each case contained HN, ADH, and DCIS (some cases also contained SH). DCIS were assigned a grade and histology according to the Van Nuys classification scheme.19Silverstein MJ Poller DN Waisman JR Colburn WJ Barth A Gierson ED Lewinsky B Gamagami P Slamon DJ Prognostic classification of breast ductal carcinoma-in-situ.Lancet. 1995; 345: 1154-1157Crossref PubMed Scopus (619) Google Scholar When present, concurrent IDC was classified according to the Nottingham combined histological grade scheme.20Dalton LW Pinder SE Elston CE Ellis IO Page DL Dupont WD Blamey RW Histologic grading of breast cancer: linkage of patient outcome with level of pathologist agreement.Mod Pathol. 2000; 13: 730-735Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar Estrogen Receptor and Progesterone Receptor status was determined by immunohistochemistry and human epidermal growth factor receptor 2 (Her2) status was determined by a combination of immunohistochemistry and fluorescence in situ hybridization. Without an IDC component present, DCIS Her2 status was unavailable. HN, SH (when available), ADH, and DCIS lesions were isolated via laser capture microdissection (LCM) from consecutive sections by using the PixCell IIe LCM system (Arcturus Engineering, Mountain View, CA) according to our standard procedures.21King C Guo N Frampton GM Gerry NP Lenburg ME Rosenberg CL Reliability and reproducibility of gene expression measurements using amplified RNA from laser-microdissected primary breast tissue with oligonucleotide arrays.J Mol Diagn. 2005; 7: 57-64Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 22Tripathi A King C de la Morenas A Perry VK Burke B Antoine GA Hirsch EF Kavanah M Mendez J Stone M Gerry NP Lenburg ME Rosenberg CL Gene expression abnormalities in histologically normal breast epithelium of breast cancer patients.Int J Cancer. 2008; 122: 1557-1566Crossref PubMed Scopus (101) Google Scholar Briefly, serial sections were mounted on uncoated glass slides and immediately stored at −80°C. Slides were lightly stained with diluted H&E to enable visualization and to prevent nucleotide alterations introduced from H&E. Lesions were isolated via LCM from 30 to 50 consecutive slides, depending on the size of the lesion and its continuity throughout the block. In one case, (380) HN microdissected from different blocks was pooled. The presence of any IDC, lymphocyte infiltration, and/or necrosis was avoided, and extreme care was taken to keep every sample separate to avoid contamination. Following the manufacturer’s protocol, total RNA was extracted and purified by using the Picopure RNA Isolation Kit (Arcturus Engineering). RNA concentration was quantitated by using the NanoDrop ND-100 Spectrophotometer (NanoDrop Technologies, Wilmington, DE) and its quality was assessed by using the RNA Pico Chips and the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). To convert the RNA to cDNA, the purified total RNA was linearly amplified for two rounds by using the MessageAMP aRNA Kit (Ambion, Austin, TX), in which the first round is based on an oligo-dT primer and T7 RNA polymerase. Our lab has successfully used amplified RNA from microdissected samples for microarray analysis from samples obtained by LCM with a high degree of reliability and reproducibility.21King C Guo N Frampton GM Gerry NP Lenburg ME Rosenberg CL Reliability and reproducibility of gene expression measurements using amplified RNA from laser-microdissected primary breast tissue with oligonucleotide arrays.J Mol Diagn. 2005; 7: 57-64Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 22Tripathi A King C de la Morenas A Perry VK Burke B Antoine GA Hirsch EF Kavanah M Mendez J Stone M Gerry NP Lenburg ME Rosenberg CL Gene expression abnormalities in histologically normal breast epithelium of breast cancer patients.Int J Cancer. 2008; 122: 1557-1566Crossref PubMed Scopus (101) Google Scholar For second round amplification, cDNA for each sample was in vitro transcribed to biotin-labeled cRNA with the in vitro transcribed labeling kit (Affymetrix, Santa Clara, CA). Gene expression was determined by using the U133A GeneChip microarray (Affymetrix) using standard procedures for quality control, hybridization, staining, and scanning.22Tripathi A King C de la Morenas A Perry VK Burke B Antoine GA Hirsch EF Kavanah M Mendez J Stone M Gerry NP Lenburg ME Rosenberg CL Gene expression abnormalities in histologically normal breast epithelium of breast cancer patients.Int J Cancer. 2008; 122: 1557-1566Crossref PubMed Scopus (101) Google Scholar The CEL files for each sample (40 total) were processed with MAS5.0. The unnormalized data were then rescaled to a target of 100 to remove bias. The microarray data from these samples are available from the National Center for Biotechnology Information Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo) under accession GSE16873. The data were analyzed in the following three comparisons: DCIS-HN; ADH-HN; and ADH-DCIS. An initial filter of the microarray output removed all probesets with 1, so that probesets in which this probability is either very small ( 0.975) can be taken as differentially expressed. This method has a very large sensitivity but lower specificity because it employs large sample approximation. We have shown that with 10 samples per group the sensitivity to detect a fold-change of two or larger can be 100%, but the specificity can be 1 was either >0.90 or <0.10. These relaxed thresholds were chosen to also take into account the fact that patient-matching was not included in the analysis and could impact the evidence for differential expression. These probesets were further analyzed by using a regression model with log-normal errors, covariates to represent the biological conditions, and a random effect per patient to take into account patient-matching across all probesets. The model was fitted by using a stochastic algorithm known as the Gibbs sampling that is implemented in WinBugs 1.4. The Gibbs sampling was run for 10,000 iterations that were sufficient to reach convergence. The values generated in the 10,000 iterations were used to estimate the fold-change (in the real scale), the 95% credible intervals (CI), and the SE evidence of differential expression was based on 95% CI nonoverlapping one. Heatmaps were generated by using the HeatPlus software package (Bioconductor, Seattle, WA). Simple hierarchical clustering was used to cluster samples based on their expression profiles. Gene annotation and pathway analyses were conducted by using several approaches including the following: Gene Ontology (www.geneontology.org); the Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov); and the Ingenuity Pathway Analysis software, version 5.5 (Ingenuity Systems, Redwood City, CA). P values reported from DAVID were calculated by using a modified Fisher Exact test (EASE Score) to measure the gene enrichment in annotation terms.24Dennis Jr, G Sherman BT Hosack DA Yang J Gao W Lane HC Lempicki RA DAVID: Database for Annotation, Visualization, and Integrated Discovery.Genome Biol. 2003; 4: P3Crossref PubMed Google Scholar, 25Huang da W Sherman BT Lempicki RA Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources.Nat Protoc. 2009; 4: 44-57Crossref PubMed Scopus (25848) Google Scholar Genes were selected for validation based on the degree, consistency, and direction of differential gene expression between HN, ADH, and/or DCIS; whether there was overlap between several comparisons; potential pathway involvement; and potential role in tumorigenesis. Using these criteria, we chose six genes for validation: fibronectin 1 (FN1), G-protein coupled receptor 5A (GPCR5A), calpain 6 (CAPN6), membrane metallo-endopeptidase (MME), oxytocin receptor (OXTR), and cyclin-dependent kinase inhibitor 1C/p57KIP2 (CDKN1C). 5-methyltetrahydrofolate-homocysteine methyltransferase (MTR) was used as the endogenous control gene, because the microarray indicated stable expression in all samples.26Dydensborg AB Herring E Auclair J Tremblay E Beaulieu JF Normalizing genes for quantitative RT-PCR in differentiating human intestinal epithelial cells and adenocarcinomas of the colon.Am J Physiol Gastrointest Liver Physiol. 2006; 290: G1067-G1074Crossref PubMed Scopus (142) Google Scholar For initial validation, we used the original, unamplified RNA from 8 of the 12 cases analyzed by microarray. For prospective validation, 12 independent breast cancer cases were chosen based on the same criteria used for initial case selection. Following the manufacturer’s protocol, total RNA was extracted and purified from these independent cases by using a different kit (RNAqueous-Micro Kit, Ambion). Reverse transcription (RT) was conducted in 50 μl volumes by using the appropriate amount of unamplified total RNA required for a given reaction plate (2 ng cDNA/reaction well for quantitative real-time PCR). RT was performed by using random hexamer primers from the TaqMan Reverse Transcription Reagent Kit (Applied Biosystems, Foster City, CA) and standard thermocycler conditions.22Tripathi A King C de la Morenas A Perry VK Burke B Antoine GA Hirsch EF Kavanah M Mendez J Stone M Gerry NP Lenburg ME Rosenberg CL Gene expression abnormalities in histologically normal breast epithelium of breast cancer patients.Int J Cancer. 2008; 122: 1557-1566Crossref PubMed Scopus (101) Google Scholar Each real-time PCR reaction was performed in 20 μl volumes by using PCR Universal Master Mix and TaqMan Gene Expression Assays (Applied Biosystems) for the gene to be validated. Using the criteria described, we selected the following primers: FN1 (Hs00415008_m1); GPCR5A (Hs01551896_m1); CAPN6 (Hs01020899_m1); MME (Hs01115449_g1); OXTR (Hs00168573_m1); CDKN1C (Hs00908986_g1); and MTR (Hs01090032_g1). TaqMan Gene Expression Assays (Applied Biosystems) used for the validation studies are indicated in parentheses. The Prism 7500 Sequence Detector System (Applied Biosystems) was used to measure fluorescence under standard cycling parameters.22Tripathi A King C de la Morenas A Perry VK Burke B Antoine GA Hirsch EF Kavanah M Mendez J Stone M Gerry NP Lenburg ME Rosenberg CL Gene expression abnormalities in histologically normal breast epithelium of breast cancer patients.Int J Cancer. 2008; 122: 1557-1566Crossref PubMed Scopus (101) Google Scholar Each condition was analyzed in duplicate with appropriate controls (-RT, -RNA, and no template) to conserve RNA. Relative quantification by quantitative real-time PCR (qRT-PCR) was assessed by using the delta delta Ct (ddCt) method, and relative gene expression was calculated after normalization to MTR.27Livak KJ Schmittgen TD Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method.Methods. 2001; 25: 402-408Crossref PubMed Scopus (127054) Google Scholar First, for both the technical and prospective validation studies, we compared changes in gene expression among the lesions as a group by using the same regression model of the ddCt data with a random effect per patient (to take into account patient-matching) as in the analysis of the microarray data. The Gibbs sampling was used to estimate the ddCt, the fold-changes as 2^(-ddCt), 95% CI, and SEs. Gene expression differences were considered to validate if the direction of change was the same and if the fold-change estimated from the microarray data were within two SEs from that estimated by qRT-PCR. Next, we compared the expression of each gene in each patient-matched pair used for validation by qRT-PCR. Gene expression differences were considered to validate if the direction of change was the same (as estimated by microarray) and if the fold-change was ≥1.5 (as measured by the ddCt method). Summary patient information for the 12 cases used for microarray studies can be seen in Table 1. Patient age ranged from 48 to 92 years. A total of 40 samples (12 HN, 4 SH, 12 ADH, and 12 DCIS) were identified and analyzed by microarray. All cancers were ER+ and most were also progesterone receptor positive (PR+) and Her2 negative (Her2−). The only exception is case 226, which was believed to be ER+ PR+ at the time of microarray, but later determined to be ER− PR−. DCIS histology and grade reflects the specific lesion microdissected. Representative lesions are depicted in Figure 1.Table 1Clinical and Histopathological Features of Patient Samples Used in Microarray StudiesCase numberPatient ageSamplesER/PR/Her2 statusDCIS histologyDCIS gradeIDC grade44448HN*Indicates samples with sufficient RNA remaining and inclusion in initial validation., SH, ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS*Indicates samples with sufficient RNA remaining and inclusion in initial validation.+ / + / NASolid2 and 3Not present24849HN*Indicates samples with sufficient RNA remaining and inclusion in initial validation., ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS+ / + / −Micropapillary2227449HN*Indicates samples with sufficient RNA remaining and inclusion in initial validation., ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS*Indicates samples with sufficient RNA remaining and inclusion in initial validation.+ / + / −Cribriform and micropapillary2138053HN*Indicates samples with sufficient RNA remaining and inclusion in initial validation., SH, ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS*Indicates samples with sufficient RNA remaining and inclusion in initial validation.+ / + / +Micropapillary2223755HN, ADH, DCIS+ / − / −Comedo2223259HN, ADH, DCIS+ / + / NAMicropapillary2Not present22661HN*Indicates samples with sufficient RNA remaining and inclusion in initial validation., ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS− / − / NAMicropapillary2325865HN, ADH*Indicates samples with sufficient RNA remaining and inclusion in initial validation., DCIS*Ind