Different gene expression patterns in invasive lobular and ductal carcinomas of the breast

Evaluation of MetriGenix custom 4D arrays applied for detection of breast cancer subtypes

BACKGROUND

Previously, a total of five breast cancer subtypes have been identified based on variation in gene expression patterns. These expression profiles were also shown to be associated with different prognostic value. In this study tumour samples from 27 breast cancer patients, previously subtyped by expression analysis using DNA microarrays, and four controls from normal breast tissue were included. A new MetriGenix 4D array proposed for diagnostic use was evaluated.

METHODS

We applied MetriGenix custom 4D arrays for the detection of previously defined molecular subtypes of breast cancer. MetriGenix 4D arrays have special features including probe immobilization in microchannels with chemiluminescence detection that enable shorter hybridization time.

RESULTS

The MetriGenix 4D array platform was evaluated with respect to both the accuracy in classifying the samples as well as the performance of the system itself. In a cross validation analysis using "Nearest Shrunken Centroid classifier" and the PAM software, 77% of the samples were classified correctly according to earlier classification results.

CONCLUSION

The system shows potential for fast screening; however, improvements are needed.

HPtaa database-potential target genes for clinical diagnosis and immunotherapy of human carcinoma

Tumor-associated antigens (TAAs) have been the most actively employed targets in the clinical diagnosis and treatment of human carcinoma, such as PSA in the diagnosis of prostate cancer and NY-ESO-1 in the immunotherapy of melanoma and other cancers. However, identification of TAAs has often been hampered by the complicated and laborsome laboratory procedures. In order to accelerate the process of tumor antigen discovery, and thereby improve diagnosis and treatment of human carcinoma, we have made an effort to establish a publicly available Human Potential Tumor Associated Antigen database (HPtaa) with potential TAAs identified by in silico computing (). Tumor specificity was chosen as the core of tumor antigen evaluation, together with other relevant clues. Various platforms of gene expression, including microarray, expressed sequence tag and SAGE data, were processed and integrated by several penalty algorithms. A total of 3518 potential TAAs have been included in the database, which is freely available to academic users. As far as we know, this database is the first one addressing human potential TAAs, and the first one integrating various kinds of expression platforms for one purpose.

Non-linear tests for identifying differentially expressed genes or genetic networks

MOTIVATION

One of the recently developed statistics for identifying differentially expressed genetic networks is Hotelling T2 statistic, which is a quadratic form of difference in linear functions of means of gene expressions between two types of tissue samples, and so their power is limited.

RESULTS

To improve the power of test statistics, a general statistical framework for construction of non-linear tests is presented, and two specific non-linear test statistics that use non-linear transformations of means are developed. Asymptotical distributions of the non-linear test statistics under the null and alternative hypothesis are derived. It has been proved that under some conditions the power of the non-linear test statistics is higher than that of the T2 statistic. Besides theory, to evaluate in practice the performance of the non-linear test statistics, they are applied to two real datasets. The preliminary results demonstrate that the P-values of the non-linear statistics for testing differential expressions of the genetic networks are much smaller than those of the T2 statistic. And furthermore simulations show the Type I errors of the non-linear statistics agree with the threshold used and the statistics fit the chi2 distribution.

SUPPLEMENTARY INFORMATION

Supplementary data are available on Bioinformatics online.

High-Resolution Genomic Profiling Reveals Association of Chromosomal Aberrations on 1q and 16p with Histologic and Genetic Subgroups of Invasive Breast Cancer

PURPOSE

Invasive ductal carcinoma and invasive lobular carcinoma (ILC) represent the major histologic subtypes of invasive breast cancer. They differ with regard to presentation, metastatic spread, and epidemiologic features. To elucidate the genetic basis of these differences, we analyzed copy number imbalances that differentiate the histologic subtypes.

EXPERIMENTAL DESIGN

High-resolution genomic profiling of 40 invasive breast cancers using matrix-comparative genomic hybridization with an average resolution of 0.5 Mb was conducted on bacterial artificial chromosome microarrays. The data were subjected to classification and unsupervised hierarchical cluster analyses. Expression of candidate genes was analyzed in tumor samples.

RESULTS

The highest discriminating power was achieved when combining the aberration patterns of chromosome arms 1q and 16p, which were significantly more often gained in ILC. These regions were further narrowed down to subregions 1q24.2-25.1, 1q25.3-q31.3, and 16p11.2. Located within the candidate gains on 1q are two genes, FMO2 and PTGS2, known to be overexpressed in ILC relative to invasive ductal carcinoma. Assessment of four candidate genes on 16p11.2 by real-time quantitative PCR revealed significant overexpression of FUS and ITGAX in ILC with 16p copy number gain. Unsupervised hierarchical cluster analysis identified three molecular subgroups that are characterized by different aberration patterns, in particular concerning gain of MYC (8q24) and the identified candidate regions on 1q24.2-25.1, 1q25.3-q31.3, and 16p11.2. These genetic subgroups differed with regard to histology, tumor grading, frequency of alterations, and estrogen receptor expression.

CONCLUSIONS

Molecular profiling using bacterial artificial chromosome arrays identified DNA copy number imbalances on 1q and 16p as significant classifiers of histologic and molecular subgroups.

Comparative expressed sequence hybridization reveals differential gene expression in morphological breast cancer subtypes

In this study, comparative expressed sequence hybridization (CESH) has been used to compare gene expression patterns in three morphologically different breast cancer subtypes: classic-type invasive lobular carcinoma (ILC), poorly differentiated ERBB2-negative invasive ductal carcinoma-not otherwise specified (IDC-NOS), and poorly differentiated ERBB2-positive IDC-NOS. CESH allows global detection of chromosomal regions with differential gene expression in a way similar to that of comparative genomic hybridization (CGH). Eight cases of each breast cancer subtype were included in the study. For each subtype, two pools of four cases each were constructed. CESH was used to compare both pools within the same morphological subtype, followed by a comparison of pools belonging to different subtypes. This revealed three chromosomal regions that were differentially expressed in ductal and lobular carcinomas, including relative overexpression at 8q13-q23 and 16q22, and relative underexpression at 8p21-p22. In addition, an expression signature characterized by relative overexpression at 3q24-q26.3, 14q23-31, 17q12, and 20q12-13 was identified for ERBB2-positive IDC-NOS. In summary, CESH analysis highlights chromosomal regions of differential gene expression that are associated with morphologically defined breast cancer subtypes and suggests that regions on chromosome 8 are of interest in the discrimination between ductal and lobular carcinomas. In addition, using CESH, it was possible to identify an ERBB2 expression signature, comprising four chromosomal regions with potential significance in the aggressive behaviour of ERBB2-positive IDC-NOS.

Lobular breast cancer: excess familiality observed in the Utah Population Database

Family history of breast cancer (BC) is a strong predictor for developing female BC. Whether this excess familiality differs within morphological BC subgroups remains unclear. We assessed the risk of lobular breast cancer (LOB) and any BC among relatives of probands with LOB. We used the Utah Population Database (UPDB) to estimate familial relative risks (FRR) as well as average relatedness, using the genealogical index of familiality (GIF) statistic. The UPDB, a population-based resource, links genealogical data from over 2 million individuals to the Utah Cancer Registry. Consistent with other studies, analysis of all BC cases showed significantly increased risk of BC to relatives (first-degree relative [FDR]: FRR = 1.83, 95% confidence interval [CI] = 1.75-1.90). Morphology-specific risks showed that relatives of LOB probands had an increased risk of LOB (FDR: FRR = 4.51, 95% CI = 2.79-6.89) and an increased risk of any BC (FDR: FRR = 2.47, 95% CI = 2.12-2.85); both measures were significantly greater than the all BC FRR estimates, and surpassed even generalized early-onset BC risk. GIF analyses corroborated the FRR results and indicated that the excess relatedness of LOB cases extended to third-degree relatives. Our findings suggest that LOB has a heritable component and may represent a genetically homogeneous form of BC. Pedigrees with excess LOB may be useful in isolating additional BC predisposition genes. Relatives of women with LOB are at higher risk for BC than relatives of other BC subtypes; a more rigorous BC screening regime may be warranted for these individuals.

A streamlined platform for high-content functional proteomics of primary human specimens

Achieving information content of satisfactory breadth and depth remains a formidable challenge for proteomics. This problem is particularly relevant to the study of primary human specimens, such as tumor biopsies, which are heterogeneous and of finite quantity. Here we present a functional proteomics strategy that unites the activity-based protein profiling and multidimensional protein identification technologies (ABPP-MudPIT) for the streamlined analysis of human samples. This convergent platform involves a rapid initial phase, in which enzyme activity signatures are generated for functional classification of samples, followed by in-depth analysis of representative members from each class. Using this two-tiered approach, we identified more than 50 enzyme activities in human breast tumors, nearly a third of which represent previously uncharacterized proteins. Comparison with cDNA microarrays revealed enzymes whose activity, but not mRNA expression, depicted tumor class, underscoring the power of ABPP-MudPIT for the discovery of new markers of human disease that may evade detection by other molecular profiling methods.

Reproductive factors and subtypes of breast cancer defined by hormone receptor and histology

Reproductive factors are associated with reduced risk of breast cancer, but less is known about whether there is differential protection against subtypes of breast cancer. Assuming reproductive factors act through hormonal mechanisms they should protect predominantly against cancers expressing oestrogen (ER) and progesterone (PR) receptors. We examined the effect of reproductive factors on subgroups of tumours defined by hormone receptor status as well as histology using data from the NIHCD Women's Contraceptive and Reproductive Experiences (CARE) Study, a multicenter case–control study of breast cancer. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) as measures of relative risk using multivariate unconditional logistic regression methods. Multiparity and early age at first birth were associated with reduced relative risk of ER + PR + tumours (P for trend=0.0001 and 0.01, respectively), but not of ER − PR − tumours (P for trend=0.27 and 0.85), whereas duration of breastfeeding was associated with lower relative risk of both receptor-positive (P for trend=0.0002) and receptor-negative tumours (P=0.0004). Our results were consistent across subgroups of women based on age and ethnicity. We found few significant differences by histologic subtype, although the strongest protective effect of multiparity was seen for mixed ductolobular tumours. Our results indicate that parity and age at first birth are associated with reduced risk of receptor-positive tumours only, while lactation is associated with reduced risk of both receptor-positive and -negative tumours. This suggests that parity and lactation act through different mechanisms. This study also suggests that reproductive factors have similar protective effects on breast tumours of lobular and ductal origin.

Comparison of cryopreservation and standard needle biopsy for gene expression profiling of human breast cancer specimens

Given the growing importance of molecular profiling of breast cancer, we initiated a small study to test whether human breast biopsies obtained via cryopreservation large core needle biopsy (C-LCNB) provided similar gene expression profiles compared with 'optimally handled' standard large core needle biopsy (S-LCNB) specimens. Five matched pairs of C-LCNB versus S-LCNB were obtained at the same visit, and subjected to gene array expression analysis using the Affymetrix system with U133A chips. No significant changes in gene expression were identified comparing the C-LCNB versus the matched S-LCNB from individual patients. This was corroborated by a paired t-test analysis, which supported the hypothesis that the S/C biopsies measured equivalent samples. A small number of genes (17) showed decreased expression when second biopsies were compared with first biopsies, suggesting a slight patient response to the first biopsy. A scatter plot analysis comparing first biopsy versus second biopsy values disclosed a slope of 0.859, further indicating that the first biopsy affects the second biopsy measurement. It therefore appears that conventional biopsies, when handled appropriately, provide RNA which is equivalent to RNA from biopsies which are frozen immediately, but that multiple biopsy protocols may introduce additional complexities.

Basal cytokeratins and their relationship to the cellular origin and functional classification of breast cancer

Recent publications have classified breast cancers on the basis of expression of cytokeratin-5 and -17 at the RNA and protein levels, and demonstrated the importance of these markers in defining sporadic tumours with bad prognosis and an association with BRCA1-related breast cancers. These important observations using different technology platforms produce a new functional classification of breast carcinoma. However, it is important in developing hypotheses about the pathogenesis of this tumour type to review the nomenclature that is being used to emphasize potential confusion between terminology that defines clinical subgroups and markers of cell lineage. This article reviews the lineages in the normal breast in relation to what have become known as the 'basal-like' carcinomas.

Effects of anastrozole on the intratumoral gene expression in locally advanced breast cancer

Intratumoral levels of E1 (oestrone), E1S (oestrone sulphate) and E2 (oestradiol) are significantly reduced by treatment with the aromatase inhibitor anastrozole regardless of treatment response. The purpose of the present pilot study was to look for additional markers of biochemical response to aromatase inhibitors on mRNA expression level. Whole genome expression was studied using microarray analysis of breast cancer tissue from 12 patients with locally advanced tumors, both before and following 15 weeks of treatment with the aromatase inhibitor anastrozole (Arimidex). Intratumoral mRNA levels for a subset of genes coding for steroid metabolizing enzymes, hormone receptors and some growth mediators involved in cell cycle control were analysed by quantitative RT-PCR. There was a correlation between the two methods for some but not all genes. The mRNA expression levels of the different genes were correlated to each other and to the intratumoral levels of E1, E2 and E1S, before and after the treatment. Notably, a correlation of the E1/E2 metabolic ratio to the mRNA levels of CYP19A1 was observed before treatment (r=0.745, p<0.005). Whole genome expression analysis of these 12 breast cancer patients revealed similar tumor classification to previously published larger studies. Tumors with no or low expression of ESR1 (oestrogen receptor) clustered together and were characterized by a strong basal-like signature highly expressing keratins 5/17, cadherin 3, frizzled and apolipoprotein D, among others. The luminal epithelial tumor cluster, on the other hand, highly expressed ESR1, GATA binding protein 3 and N-acetyl transferase. An evident ERBB2 cluster was observed due to the marked over-expression of the ERBB2 gene and GRB7 and PPARBP in this patient material). Using significance analysis of microarrays (SAM), we identified 298 genes significantly differently expressed between the partial response and progressive disease groups.

Gene Expression Profiling of Breast Cancer in Relation to Estrogen Receptor Status and Estrogen-Metabolizing Enzymes: Clinical Implications

Interactions between luminal epithelial cells and their surrounding microenvironment govern the normal development and function of the mammary gland. Estradiol plays a key role in abnormal intracellular signaling, which contributes to the development and progression of breast tumors. The present article summarizes the results from a microarray whole genome gene expression analysis as well as a quantitative analysis of the mRNA expression of members of the estradiol metabolic and signaling pathways in the tumors of postmenopausal breast cancer patients. The analysis of the variation in whole genome gene expression resulted in a tumor classification comprising several distinct groups with distinct expression of the estrogen receptor (ER). The parallel study on the expression of only nine mRNA transcripts of members of the estradiol pathways resulted in two main clusters, representing ER− and ER tumors. The mRNA expression of the estradiol-metabolizing enzymes did not follow the expression of the ER in all cases, leading to the recognition of several further subclasses of tumors. When the tumor classes obtained by whole genome gene expression analysis were compared with those obtained by independent quantitation of the estradiol-metabolizing enzymes, a statistically significant association between both classification groups was observed. These findings point to a possible association between development of a tumor with a particular expression profile and its capacity to synthesize estradiol as measured by the expression of the transcripts for the necessary key enzymes. Further, whole genome expression patterns were studied in 12 patients treated with anastrozole. Using significance analysis of microarrays, we identified 298 genes significantly differently expressed between partial response and progressive disease groups.

Genomics in breast cancer—therapeutic implications

The introduction of DNA microarray techniques has had dramatic implications on cancer research, allowing researchers to analyze expression of multiple genes in concert and relate the findings to clinical parameters. The main discoveries in breast cancer, as well as in other malignancies, have so far been with respect to two key issues. First, individual tumors arising from the same organ may be grouped into distinct classes based on their gene expression profiles, independent of stage and grade. Second, the biologic relevance of such classification is corroborated by significant prognostic impact. We review how the use of microarray technologies can provide unique possibilities to explore the mechanisms of tumor behavior in vivo that will allow evaluation of prognosis and, potentially, drug resistance. However, in spite of recent advances, we are not yet at a stage where the use of these techniques should be implemented for routine clinical use, whether to define prognostic factors or to predict sensitivity to therapy.

Sample size for detecting differentially expressed genes in microarray experiments

Background

Microarray experiments are often performed with a small number of biological replicates, resulting in low statistical power for detecting differentially expressed genes and concomitant high false positive rates. While increasing sample size can increase statistical power and decrease error rates, with too many samples, valuable resources are not used efficiently. The issue of how many replicates are required in a typical experimental system needs to be addressed. Of particular interest is the difference in required sample sizes for similar experiments in inbred vs. outbred populations (e.g. mouse and rat vs. human).

Results

We hypothesize that if all other factors (assay protocol, microarray platform, data pre-processing) were equal, fewer individuals would be needed for the same statistical power using inbred animals as opposed to unrelated human subjects, as genetic effects on gene expression will be removed in the inbred populations. We apply the same normalization algorithm and estimate the variance of gene expression for a variety of cDNA data sets (humans, inbred mice and rats) comparing two conditions. Using one sample, paired sample or two independent sample t-tests, we calculate the sample sizes required to detect a 1.5-, 2-, and 4-fold changes in expression level as a function of false positive rate, power and percentage of genes that have a standard deviation below a given percentile.

Conclusions

Factors that affect power and sample size calculations include variability of the population, the desired detectable differences, the power to detect the differences, and an acceptable error rate. In addition, experimental design, technical variability and data pre-processing play a role in the power of the statistical tests in microarrays. We show that the number of samples required for detecting a 2-fold change with 90% probability and a p-value of 0.01 in humans is much larger than the number of samples commonly used in present day studies, and that far fewer individuals are needed for the same statistical power when using inbred animals rather than unrelated human subjects.