Gene-expression profiles in hereditary breast cancer

Integrating pharmacogenomics into drug development

The first observations of inherited differences in drug effects in the 1950s led to the recognition of a genetic basis for drug response. With the development of genetics and molecular biology, it became clear that certain drug responses could be associated with specific genetic variations or polymorphisms. There are now examples of polymorphisms that affect response to drugs ranging from common analgesics to chemotherapeutics. The goal of pharmacogenetics is to identify polymorphisms that can serve as predictive markers of drug response. This review summarizes how existing pharmacogenomic technologies can be applied advantageously throughout drug development to bring drugs successfully to market along with diagnostic tests that ensure their appropriate use.

Simulation of cDNA microarrays via a parameterized random signal model

cDNA microarrays provide simultaneous expression measurements for thousands of genes that are the result of processing images to recover the average signal intensity from a spot composed of pixels covering the area upon which the cDNA detector has been put down. The accuracy of the signal measurement depends on using an appropriate algorithm to process the images. This includes determining spot locations and processing the data in such a way as to take into account spot geometry, background noise, and various kinds of noise that degrade the signal. This paper presents a stochastic model for microarray images. There are over 20 model parameters, each governed by a probability distribution, that control the signal intensity, spot geometry, spot drift, background effects, and the many kinds of noise that affect microarray images owing to the manner in which they are formed. The model can be used to analyze the performance of image algorithms designed to measure the true signal intensity because the ground truth (signal intensity) for each spot is known. The levels of foreground noise, background noise, and spot distortion can be set, and algorithms can be evaluated under varying conditions.

Strong feature sets from small samples

For small samples, classifier design algorithms typically suffer from overfitting. Given a set of features, a classifier must be designed and its error estimated. For small samples, an error estimator may be unbiased but, owing to a large variance, often give very optimistic estimates. This paper proposes mitigating the small-sample problem by designing classifiers from a probability distribution resulting from spreading the mass of the sample points to make classification more difficult, while maintaining sample geometry. The algorithm is parameterized by the variance of the spreading distribution. By increasing the spread, the algorithm finds gene sets whose classification accuracy remains strong relative to greater spreading of the sample. The error gives a measure of the strength of the feature set as a function of the spread. The algorithm yields feature sets that can distinguish the two classes, not only for the sample data, but for distributions spread beyond the sample data. For linear classifiers, the topic of the present paper, the classifiers are derived analytically from the model, thereby providing an enormous savings in computation time. The algorithm is applied to cancer classification via cDNA microarrays. In particular, the genes BRCA1 and BRCA2 are associated with a hereditary disposition to breast cancer, and the algorithm is used to find gene sets whose expressions can be used to classify BRCA1 and BRCA2 tumors.

Recent developments in DNA microarrays

DNA microarrays are used to quantify tens of thousands of DNA or RNA sequences in a single assay. Upon their introduction approximately six years ago, DNA microarrays were viewed as a disruptive technology that would fundamentally alter the scientific landscape. Supporting this view, the number of applications of DNA microarray technology has since expanded exponentially. Here, we review recent advances in microarray technology and selected new applications of the technology.

Microarray technology and gene expression analysis for the study of angiogenesis

Angiogenesis plays a major role in multiple disease processes including cancer, and new agents that modulate angiogenesis are rapidly entering clinical trials. The understanding of the biological mechanisms and downstream effects for many of these agents is poorly understood. It is therefore important that methods evolve to understand how an agent regulates angiogenesis, in order to promote a higher percentage of successful drug candidates. With the emergence of microarray technology for the evaluation of gene expression, researchers have a powerful tool for dissecting the biological mechanisms of angiogenesis. However, huge data sets and complex statistics pose a hurdle for the investigator to obtain useful and meaningful data. To eliminate problems in data analysis, proper design and planning prior to performing a microarray experiment is crucial to making valid conclusions. This review will discuss the critical factors in designing, performing and analysing microarray experiments, and the utility of various models of angiogenesis for microarray analysis.

Role of gene expression microarray analysis in finding complex disease genes

The promise of gene expression studies using microarray technology has inspired much new hope for finding complex diseases genes. It has become clear that complex diseases result from collective actions of many genetic and nongenetic factors. Therefore, genetic dissection of complex diseases should be carried out in a global context. The technology of gene expression microarray analysis (GEMA) can provide such global information on transcription activities of essentially all genes simultaneously. It is hoped that this promising technology can be applied to samples drawn from large-scale, well-defined genetic epidemiological studies and help us untangle the web of pathways leading to complex diseases. However, extremely noisy GEMA data pose serious challenges in terms of the statistical methodologies needed. Extensive work is needed in order to respond to the challenges before one can fully utilize the potential power provided by GEMA. We begin in this paper by identifying several statistical problems related to the application of GEMA to genetic epidemiological analysis, and consider study designs that might benefit from this promising new technology. While it is still too early to tell how much of the enormous potential of GEMA will be realized ultimately, its success will probably depend most critically on the ability of statistical genetics to rise to the challenge of mining information from a sea of noise.

Design of studies using DNA microarrays

DNA microarrays are assays that simultaneously provide information about expression levels of thousands of genes and are consequently finding wide use in biomedical research. In order to control the many sources of variation and the many opportunities for misanalysis, DNA microarray studies require careful planning. Different studies have different objectives, and important aspects of design and analysis strategy differ for different types of studies. We review several types of objectives of studies using DNA microarrays and address issues such as selection of samples, levels of replication needed, allocation of samples to dyes and arrays, sample size considerations, and analysis strategies.

Familial breast carcinoma risks by morphology: a nationwide epidemiologic study from Sweden

BACKGROUND

Familial risks in patients with breast carcinoma have not been assessed by morphologic types of medically verified cancers. Reliable data on familial risks would help to establish prevention programs and guide clinical decisions.

METHODS

We used the nationwide Swedish Family-Cancer Database to calculate standardized incidence ratios (SIRs) and 95% confidence intervals (CIs) for invasive and in situ breast carcinomas in women with mothers and sisters. This database has information on 10.2 million individuals and on more than 13,000 morphology-specific breast carcinomas.

RESULTS

SIRs for all invasive breast carcinomas were 1.82 (95% CI 1.71-1.93) for breast carcinoma in the mother and 1.89 (1.70-2.01) for breast carcinoma in a sister. The respective risks were 1.81 and 1.85 for a mother and sisters with ductal breast carcinoma. The SIRs were equally for lobular, tubuloductal, comedo, and mucinous breast carcinomas. However, the SIRs for lobular carcinoma were lower than those for the ductal type, whereas the opposite trend was noted for the comedo and mucinous type; none of the differences were significant. The risks for all morphologic types were highest when both a mother and a sister were affected, SIR 3.19 (2.36-4.22). The risks for in situ breast carcinomas were 2.09 (1.78-2.44) for an affected mother, 2.24 (1.88-2.85) for an affected sister, and 5.23 (2.59-9.39) when both a mother and a sister were affected.

CONCLUSIONS

The data suggest that the familial risk of breast carcinoma is independent of the morphologic type. The higher risks in in situ cancer may be due to medical surveillance. The risks were identical from a mother or sister proband, suggesting that recessive effects are unlikely as a heritable cause of breast carcinoma.

Empirical bayes methods and false discovery rates for microarrays

In a classic two-sample problem, one might use Wilcoxon's statistic to test for a difference between treatment and control subjects. The analogous microarray experiment yields thousands of Wilcoxon statistics, one for each gene on the array, and confronts the statistician with a difficult simultaneous inference situation. We will discuss two inferential approaches to this problem: an empirical Bayes method that requires very little a priori Bayesian modeling, and the frequentist method of "false discovery rates" proposed by Benjamini and Hochberg in 1995. It turns out that the two methods are closely related and can be used together to produce sensible simultaneous inferences.

Computational selection of distinct class- and subclass-specific gene expression signatures

In this investigation we used statistical methods to select genes with expression profiles that partition classes and subclasses of biological samples. Gene expression data corresponding to liver samples from rats treated for 24 h with an enzyme inducer (phenobarbital) or a peroxisome proliferator (clofibrate, gemfibrozil or Wyeth 14,643) were subjected to a modified Z-score test to identify gene outliers and a binomial distribution to reduce the probability of detecting genes as differentially expressed by chance. Hierarchical clustering of 238 statistically valid differentially expressed genes partitioned class-specific gene expression signatures into groups that clustered samples exposed to the enzyme inducer or to peroxisome proliferators. Using analysis of variance (ANOVA) and linear discriminant analysis methods we identified single genes as well as coupled gene expression profiles that separated the phenobarbital from the peroxisome proliferator treated samples and discerned the fibrate (gemfibrozil and clofibrate) subclass of peroxisome proliferators. A comparison of genes ranked by ANOVA with genes assessed as significant by mixedlinear models analysis [J. Comput. Biol. 8 (2001) 625] or ranked by information gain revealed good congruence with the top 10 genes from each statistical method in the contrast between phenobarbital and peroxisome proliferators expression profiles. We propose building upon a classification regimen comprised of analysis of replicate data, outlier diagnostics and gene selection procedures to utilize cDNA microarray data to categorize subclasses of samples exposed to pharmacologic agents.

The fundamental role of epigenetic events in cancer

Patterns of DNA methylation and chromatin structure are profoundly altered in neoplasia and include genome-wide losses of, and regional gains in, DNA methylation. The recent explosion in our knowledge of how chromatin organization modulates gene transcription has further highlighted the importance of epigenetic mechanisms in the initiation and progression of human cancer. These epigenetic changes -- in particular, aberrant promoter hypermethylation that is associated with inappropriate gene silencing -- affect virtually every step in tumour progression. In this review, we discuss these epigenetic events and the molecular alterations that might cause them and/or underlie altered gene expression in cancer.

Towards an integrated classification of adult acute lymphoblastic leukemia

Acute lymphoblastic leukemia (ALL) represents a biologically and clinically heterogeneous group of diseases characterized by the abnormal proliferation and accumulation of immature lymphoid cells within the bone marrow and lymphoid tissues. Following a diagnostic work-up, prognostic data are routinely achieved through physical examination, serum biochemical profiles, peripheral blood count and bone marrow morphology. Over the years, information obtained through karyotype, molecular genetics, extensive immunophenotype, multidrug resistance and, more recently, genomic profiling is progressively contributing to a better understanding of the biology of this complex disease, to the identification of subgroups of patients with a different clinical outcome, to the more precise monitoring of minimal residual disease, to the use of different therapeutic protocols based on prognostic indicators and, recently, also to the design of innovative and specific treatment strategies. In the present review, we will discuss how an integrated approach is now mandatory for the optimal management of adult ALL.

Gene expression signature of benign prostatic hyperplasia revealed by cDNA microarray analysis

BACKGROUND

Despite the high prevalence of benign prostatic hyperplasia (BPH) in the aging male, little is known regarding the etiology of this disease. A better understanding of the molecular etiology of BPH would be facilitated by a comprehensive analysis of gene expression patterns that are characteristic of benign growth in the prostate gland. Since genes differentially expressed between BPH and normal prostate tissues are likely to reflect underlying pathogenic mechanisms involved in the development of BPH, we performed comparative gene expression analysis using cDNA microarray technology to identify candidate genes associated with BPH.

METHODS

Total RNA was extracted from a set of 9 BPH specimens from men with extensive hyperplasia and a set of 12 histologically normal prostate tissues excised from radical prostatectomy specimens. Each of these 21 RNA samples was labeled with Cy3 in a reverse transcription reaction and cohybridized with a Cy5 labeled common reference sample to a cDNA microarray containing 6,500 human genes. Normalized fluorescent intensity ratios from each hybridization experiment were extracted to represent the relative mRNA abundance for each gene in each sample. Weighted gene and random permutation analyses were performed to generate a subset of genes with statistically significant differences in expression between BPH and normal prostate tissues. Semi-quantitative PCR analysis was performed to validate differential expression.

RESULTS

A subset of 76 genes involved in a wide range of cellular functions was identified to be differentially expressed between BPH and normal prostate tissues. Semi-quantitative PCR was performed on 10 genes and 8 were validated. Genes consistently upregulated in BPH when compared to normal prostate tissues included: a restricted set of growth factors and their binding proteins (e.g. IGF-1 and -2, TGF-beta3, BMP5, latent TGF-beta binding protein 1 and -2); hydrolases, proteases, and protease inhibitors (e.g. neuropathy target esterase, MMP2, alpha-2-macroglobulin); stress response enzymes (e.g. COX2, GSTM5); and extracellular matrix molecules (e.g. laminin alpha 4 and beta 1, chondroitin sulfate proteoglycan 2, lumican). Genes consistently expressing less mRNA in BPH than in normal prostate tissues were less commonly observed and included the transcription factor KLF4, thrombospondin 4, nitric oxide synthase 2A, transglutaminase 3, and gastrin releasing peptide.

CONCLUSIONS

We identified a diverse set of genes that are potentially related to benign prostatic hyperplasia, including genes both previously implicated in BPH pathogenesis as well as others not previously linked to this disease. Further targeted validation and investigations of these genes at the DNA, mRNA, and protein levels are warranted to determine the clinical relevance and possible therapeutic utility of these genes.

Initiating oncogenic event determines gene-expression patterns of human breast cancer models

Molecular expression profiling of tumors initiated by transgenic overexpression of c-myc, c-neu, c-ha-ras, polyoma middle T antigen (PyMT) or simian virus 40 T/t antigen (T-ag) targeted to the mouse mammary gland have identified both common and oncogene-specific events associated with tumor formation and progression. The tumors shared great similarities in their gene-expression profiles as compared with the normal mammary gland with an induction of cell-cycle regulators, metabolic regulators, zinc finger proteins, and protein tyrosine phosphatases, along with the suppression of some protein tyrosine kinases. Selection and hierarchical clustering of the most variant genes, however, resulted in separating the mouse models into three groups with distinct oncogene-specific patterns of gene expression. Such an identification of targets specified by particular oncogenes may facilitate development of lesion-specific therapeutics and preclinical testing. Moreover, similarities in gene expression between human breast cancers and the mouse models have been identified, thus providing an important component for the validation of transgenic mammary cancer models.

Diagnosis of multiple cancer types by shrunken centroids of gene expression

We have devised an approach to cancer class prediction from gene expression profiling, based on an enhancement of the simple nearest prototype (centroid) classifier. We shrink the prototypes and hence obtain a classifier that is often more accurate than competing methods. Our method of "nearest shrunken centroids" identifies subsets of genes that best characterize each class. The technique is general and can be used in many other classification problems. To demonstrate its effectiveness, we show that the method was highly efficient in finding genes for classifying small round blue cell tumors and leukemias.

The pathology of familial breast cancer: predictive value of immunohistochemical markers estrogen receptor, progesterone receptor, HER-2, and p53 in patients with mutations in BRCA1 and BRCA2

PURPOSE

The morphologic and molecular phenotype of breast cancers may help identify patients who are likely to carry germline mutations in BRCA1 and BRCA2. This study evaluates the immunohistochemical profiles of tumors arising in patients with mutations in these genes.

MATERIALS AND METHODS

Samples of breast cancers obtained from the International Breast Cancer Linkage Consortium were characterized morphologically and immunohistochemically using antibodies to estrogen receptor, progesterone receptor, HER-2 (c-erbB-2 oncogene), and p53 protein.

RESULTS

Breast cancers in patients with BRCA1 germline mutations are more often negative for estrogen receptor, progesterone receptor, and HER-2, and are more likely to be positive for p53 protein compared with controls. In contrast, BRCA2 tumors do not show a significant difference in the expression of any of these proteins compared with controls.

CONCLUSION

BRCA1 has a distinctive morphology and immunohistochemical phenotype. The combined morphologic and immunohistochemical data can be used to predict the risk of a young patient harboring a germline mutation in BRCA1. The BRCA2 phenotype is currently not well defined.

Analyzing array data using supervised methods

Pharmacogenomics is the application of genomic technologies to drug discovery and development, as well as for the elucidation of the mechanisms of drug action on cells and organisms. DNA microarrays measure genome-wide gene expression patterns and are an important tool for pharmacogenomic applications, such as the identification of molecular targets for drugs, toxicological studies and molecular diagnostics. Genome-wide investigations generate vast amounts of data and there is a need for computational methods to manage and analyze this information. Recently, several supervised methods, in which other information is utilized together with gene expression data, have been used to characterize genes and samples. The choice of analysis methods will influence the results and their interpretation, therefore it is important to be familiar with each method, its scope and limitations. Here, methods with special reference to applications for pharmacogenomics are reviewed.

Altered methylation patterns in cancer cell genomes: cause or consequence?

CpG islands are associated with at least half of all cellular genes and are normally methylation-free. Dense methylation of cytosine residues within islands causes strong and heritable transcriptional silencing. Such silencing normally occurs almost solely at genes subject to genomic imprinting or to X chromosome inactivation. Aberrant methylation of CpG islands associated with tumor suppressor genes has been proposed to contribute to carcinogenesis. However, questions of mechanisms underlying the cancer changes and the precise consequences for tumorigenesis exist in the field, and must continue to be addressed before the importance of abnormalities in genomic methylation patterns in carcinogenesis can be fully understood. In this article, two workers in DNA methylation, one concentrating on cancer biology and the other on developmental biology, address recurrent questions about cancer epigenetics from different perspectives. The goal is to highlight important controversies in the field which can be productive targets of ongoing and future research.

Gene expression analysis reveals matrilysin as a key regulator of pulmonary fibrosis in mice and humans

Pulmonary fibrosis is a progressive and largely untreatable group of disorders that affects up to 100,000 people on any given day in the United States. To elucidate the molecular mechanisms that lead to end-stage human pulmonary fibrosis we analyzed samples from patients with histologically proven pulmonary fibrosis (usual interstitial pneumonia) by using oligonucleotide microarrays. Gene expression patterns clearly distinguished normal from fibrotic lungs. Many of the genes that were significantly increased in fibrotic lungs encoded proteins associated with extracellular matrix formation and degradation and proteins expressed in smooth muscle. Using a combined set of scoring systems we determined that matrilysin (matrix metalloproteinase 7), a metalloprotease not previously associated with pulmonary fibrosis, was the most informative increased gene in our data set. Immunohistochemisry demonstrated increased expression of matrilysin protein in fibrotic lungs. Furthermore, matrilysin knockout mice were dramatically protected from pulmonary fibrosis in response to intratracheal bleomycin. Our results identify matrilysin as a mediator of pulmonary fibrosis and a potential therapeutic target. They also illustrate the power of global gene expression analysis of human tissue samples to identify molecular pathways involved in clinical disease.

Multiplex standardized RT-PCR for expression analysis of many genes in small samples

Standardized RT-PCR (StaRT-PCR) enables numerical quantification as well as intra- and inter-laboratory comparison of gene expression. Multiplex StaRT-PCR, using two rounds of amplification, was conducted on Stratagene Universal Reference RNA. In the first round, cDNA, competitive template (CT) mix, and primers for up to 96 genes were amplified for varying numbers of cycles. Next, products from round one were diluted, combined with primers for one gene, and amplified for an additional 35 cycles. No additional cDNA or CT mix was added. Expression values obtained by uniplex and multiplex StaRT-PCRs were highly correlated (R=0.993, p<0.001). Products from round one could be diluted as much as 100,000-fold and still be quantified following round two amplification. Thus, using multiplex StaRT-PCR, 96 genes were measured in the same amount of cDNA typically used to measure one gene with uniplex StaRT-PCR. Multiplex StaRT-PCR was also used to measure 18 genes in the fine needle biopsy of a primary lung carcinoma.

Identification of genes associated with head and neck carcinogenesis by cDNA microarray comparison between matched primary normal epithelial and squamous carcinoma cells

In order to identify genes involved in head and neck carcinogenesis, we compared the gene expression profile in matched primary normal epithelial cells and primary head and neck cancer cells from the same patients. A cDNA microarray analysis consisting of 12 530 human genes revealed significant changes in the expression of 213 genes, with 91 genes being up-regulated and 122 being down-regulated. This comprehensive list of genes includes those associated with signal transduction (growth factors), cell structure, cell cycle, transcription, apoptosis, and cell-cell adhesion. Further analysis of nine genes involved in cell-cell interaction, using Western blot and/or reverse transcription (RT)-PCR of four paired cell lines supported the reliability of our microarray analysis. More specifically, our study provides the first evidence that claudin-7 and connexin 31.1 are down-regulated in head and neck squamous cell carcinomas (HNSCC) compared to normal cells. These findings provide a large body of information regarding gene expression profiles associated with head and neck carcinogenesis, and also represent a source of potential targets for HNSCC prevention and/or therapeutics.

Identifying pre-post chemotherapy differences in gene expression in breast tumours: a statistical method appropriate for this aim

Although widely used for the analysis of gene expression microarray data, cluster analysis may not be the most appropriate statistical technique for some study aims. We demonstrate this by considering a previous analysis of microarray data obtained on breast tumour specimens, many of which were paired specimens from the same patient before and after chemotherapy. Reanalysing the data using statistical methods that appropriately utilise the paired differences for identification of differentially expressed genes, we find 17 genes that we can confidently identify as more expressed after chemotherapy than before. These findings were not reported by the original investigators who analysed the data using cluster analysis techniques.

Large-scale gene expression analysis in molecular target discovery

The evolution of simple arrays consisting of a few genes to ones composed of thousands of genes and/or ESTs has allowed investigators unprecedented views of the molecular mechanisms within cells. Due to the enormous quantities of information derived from microarray analysis, new types of problems have surfaced, such as where to store all of the data. The ability to solve database or statistical problems has required the bench biologist to collaborate with database developers, software designers and statisticians to determine solutions for storage, analysis and interpretation of microarray data. The collaborative effort between these extremely diverse disciplines has led to the development of creative database query and gene expression analysis tools, producing significant reductions in the time required by researchers to filter through the datasets and discover the key processes perturbed by the diseases of interest. Both unsupervised and supervised analysis methods have been applied to gene expression data leading to the discovery of novel therapeutic targets and diagnostic markers. Furthermore, tumor classification based on their respective molecular fingerprints has led to the classification of cancer subtypes and the discovery of novel molecular taxonomies that may eventually lead to improved patient stratification and superior therapeutic strategies.

DNA microarrays in clinical oncology

Aberrant gene expression is critical for tumor initiation and progression. However, we lack a comprehensive understanding of all genes that are aberrantly expressed in human cancer. Recently, DNA microarrays have been used to obtain global views of human cancer gene expression and to identify genetic markers that might be important for diagnosis and therapy. We review clinical applications of these novel tools, discuss some important recent studies, identify promising avenues of research in this emerging field of study, and discuss the likely impact that expression profiling will have on clinical oncology.

Gene expression profile of long-lived snell dwarf mice

To gain further insight into the basis for the extended longevity and delayed aging of Snell dwarf (dw/dw) mice, we have measured levels of expression of 2352 genes in liver of mice at 6 months of age. We find 60 genes for the which the Student's t statistic meets the arbitrary criterion of p <.001, and among these 17 meet the Bonferroni-adjusted significance criterion at p <.05, which corresponds to a nominal value of p <.00002. Using the Bonferroni criterion, we find that dwarf mice show increases in liver mRNA for two mannose-binding lectins, two DNA binding proteins, serum amyloid P component, corticosteroid-binding globulin, and insulin-like growth factor-binding protein 2, as well as decreases in a two phosphodiesterases, a pheromone-binding urinary protein, insulin-like growth factor-I (IGF-I), a calcium-binding protein calgranulin B, a deubiquitinating enzyme, a hydroxysteroid dehydrogenase, a DNA methyltransferase, a glycine transporter, and a placental lactogen. We also use this data set to compare the results of different suggested criteria for evaluating intergroup differences in gene expression. Of the 2352 genes examined, 524 (22%) showed a twofold difference between dwarf and normal mice, but most of these fail to meet the conventional significance criterion of p <.05, let alone criteria that have been adjusted to compensate for multiple comparison artifacts. The list of genes that show reliable differences between dwarf and control animals provides new insights into the range of changes induced by deficiencies in growth hormone, thyroid-stimulating hormone, and prolactin, and it will help to guide further studies of the pathways by which these hormone deficiencies contribute to delayed aging in these mutant mice.

Application of genome-wide gene expression profiling by high-density DNA arrays to the treatment and study of inflammatory bowel disease

Identification of factors involved in the initiation, amplification, and perpetuation of the chronic immune response and the identification of markers for the characterization of patient subgroups remain critical objectives for ongoing research in inflammatory bowel disease (IBD). The Human Genome Project and the development of the expressed sequence tag (EST) clone collection and database have made possible a new revolution in gene expression analysis. Instead of measuring one or a few genes, parallel DNA microarrays are capable of simultaneously measuring expression of thousands of genes, providing a glimpse into the logic and functional grouping of gene programs encoded by our genome. Applied to clinical specimens from affected and normal individuals, this methodology has the potential to provide a new level of information about disease pathogenesis not previously possible. Two dominant platforms for the construction of high-density microarrays have emerged: cDNA arrays and GeneChips. The first involves robotic spotting of DNA molecules, often derived from EST clone collections, onto a suitable solid phase matrix such as a glass slide. The second involves direct in situ synthesis of sets of gene-specific oligonucleotides on a silicon wafer by an eloquent derivative of the photolithography process. Both cDNA and oligonucleotide arrays are interrogated by hybridization with a fluorescent-labeled cDNA or cRNA representation of the original tissue mRNA. This enables measurement of the expression levels for thousands of mucosal genes in a single experiment. These technologies have recently become less expensive and more widely accessible to all researchers. This review details the principles and methods behind DNA array technology, data analysis and mining, and potential application to research and treatment of IBD.

Breast cancer activism: past lessons, future directions

Breast cancer activism has become a fixture in the United States, where fundraising events are ubiquitous and government financing of research into the disease has skyrocketed. Activists in other countries are now reporting similar accomplishments. Here, predominantly using the United States as a case study, I analyse the recent successes of breast cancer activism. I also raise a series of questions about the future goals of activism.

Multivariate approach for selecting sets of differentially expressed genes

An important problem addressed using cDNA microarray data is the detection of genes differentially expressed in two tissues of interest. Currently used approaches ignore the multidimensional structure of the data. However it is well known that correlation among covariates can enhance the ability to detect less pronounced differences. We use the Mahalanobis distance between vectors of gene expressions as a criterion for simultaneously comparing a set of genes and develop an algorithm for maximizing it. To overcome the problem of instability of covariance matrices we propose a new method of combining data from small-scale random search experiments. We show that by utilizing the correlation structure the multivariate method, in addition to the genes found by the one-dimensional criteria, finds genes whose differential expression is not detectable marginally.

The evolving role of information technology in the drug discovery process

Information technologies for chemical structure prediction, heterogeneous database access, pattern discovery, and systems and molecular modeling have evolved to become core components of the modern drug discovery process. As this evolution continues, the balance between in silico modeling and 'wet' chemistry will continue to shift and it might eventually be possible to step through the discovery pipeline without the aid of traditional laboratory techniques. Rapid advances in the industrialization of gene sequencing combined with databases of protein sequence and structure have created a target-rich but lead-poor environment. During the next decade, newer information technologies that facilitate the molecular modeling of drug-target interactions are likely to shift this balance towards molecular-based personalized medicine -- the ultimate goal of the drug discovery process.

A novel sensitive microarray approach for differential screening using probes labelled with two different radioelements

We have developed a novel microarray approach for differential screening using probes labelled with two different radioelements. The complementary DNAs from the reverse transcription of mRNAs from two different biological samples were labelled with radioelements of significantly different energies (3H and 35S or 33P). Radioactive images corresponding to the expressed genes were acquired with a MicroImager, a real time, high resolution digital autoradiography system. An algorithm was used to process the data such that the initially acquired radioactive image was filtered into two subimages, each representative of the hybridisation result specific for one probe. The simultaneous screening of gene expression in two different biological samples requires <100 ng mRNA without any amplification. In such conditions, the technique is sensitive enough to directly quantify the amount of mRNA even when present in small amounts: 10(7) molecules in the probe as assessed with an added control sequence and 2 x 10(5) molecules with an endogenous tyrosine hydroxylase mRNA. This novel technique of double radioactive labelling on a microarray is thus suitable for the comparison of gene expression in two different biological samples available in only small quantities. Consequently, it has great potential for various biological fields, such as neuroscience.

Relation between resistance of Philadelphia-chromosome-positive acute lymphoblastic leukaemia to the tyrosine kinase inhibitor STI571 and gene-expression profiles: a gene-expression study

BACKGROUND

The ABL tyrosine kinase inhibitor STI571 is a promising agent for treatment of advanced Philadelphia-chromosome-positive (Ph+) acute lymphoblastic leukaemia. However, resistance to this drug develops within a few months in most patients. We aimed to predict resistance to STI571.

METHODS

Total RNA was extracted from 25 bone-marrow samples from 19 patients with Ph+ acute lymphoblastic leukaemia who were enrolled into a phase II study. 17 samples were obtained before STI571 treatment was started: ten from individuals who were classified as good responders to STI571 (sensitive), and seven from individuals who did not to respond to STI571 (primary resistance). Eight samples were obtained from patients during treatment with STI571. We analysed six matched samples, which were obtained before and during treatment with STI571. Oligonucleotide microarray analysis of samples was done with high-density microarrays.

FINDINGS

We identified 95 genes whose expression could be used to predict sensitivity of leukaemic cells to STI571. On this basis, all the STI571-sensitive samples could clearly be distinguished from the resistant cases. 56 highly differentially expressed genes were identified in leukaemic cells that were secondarily resistant to STI571. Resistant leukaemic cells expressed high levels of Bruton's tyrosine kinase and two ATP synthetases (ATP5A1 and ATP5C1), and showed significantly reduced expression of the proapoptotic gene BAK1 and the cell-cycle control gene p15 INK4b.

INTERPRETATION

We have shown the clinical relevance of gene expression data for the pretreatment assessment of acute lymphoblastic leukaemia. Our results have implications for future clinical studies of tyrosine kinase inhibitors.

Tumor immunology--towards a paradigm of reciprocal research

Recent advances in the field of tumor immunology highlight the difficulties involved in generating and maintaining a tumor-specific immune response. The tendency of T cells to be tolerized in vivo, and the tendency of tumors to escape immune recognition represent significant barriers to successful immunotherapy. The results of early clinical trials illustrate these points and underscore the critical importance of an interactive dialog between laboratory and clinical research efforts.

Cancer genomics

The draft human genome sequence and the dissemination of high throughput technology provides opportunities for systematic analysis of cancer cells. Genome-wide mutation screens, high resolution analysis of chromosomal abberations and expression profiling all give comprehensive views of genetic alterations in cancer cells. From these analyses will come a complete list of the genetic changes that drive malignant transformation and of the therapeutic targets that may be exploited for clinical benefit.

Clinical and functional target validation using tissue and cell microarrays

Expression levels of thousands of genes or proteins can be readily determined using microarray techniques. However, this represents only the first step in understanding the biological and medical significance of these molecules. New high-throughput techniques, such as tissue and cell microarrays, will facilitate clinical and functional analysis of molecular targets.

Core biopsies can be used to distinguish differences in expression profiling by cDNA microarrays

The primary focus of this work was to determine the feasibility of obtaining representative expression array profiles from clinical core biopsies. For this purpose we performed six 16-gauge needle core biopsies and an excision biopsy on each of two different human xenografts, one from an Ewing's sarcoma cell line and the second from neuroblastoma cell line grown in Beige-Scid mice. Three of the six core biopsies were processed separately and the remaining three were pooled and processed together. As the initial RNA material isolated from the core biopsies was not sufficient for microarray analysis, an amplification procedure using a modified Eberwine protocol was performed, and the amplified products applied onto a 6000-feature human cDNA microarray. Comparisons of the array results from core biopsies (amplified RNA) and surgical specimens (non-amplified RNA) showed maintenance of the expression profile as assessed by hierarchical clustering. Gene expression profiles obtained from microarray analysis clearly differentiated the Ewing's sarcoma from the neuroblastoma with both core and excisional biopsies as starting material. Pooling the core biopsies did not improve the concordance with excisional biopsies. In conclusion, our results suggest that core biopsies can be used as a suitable and reliable material for the determination of tumor genetic profiles.

Gene expression profiling predicts clinical outcome of breast cancer

Breast cancer patients with the same stage of disease can have markedly different treatment responses and overall outcome. The strongest predictors for metastases (for example, lymph node status and histological grade) fail to classify accurately breast tumours according to their clinical behaviour. Chemotherapy or hormonal therapy reduces the risk of distant metastases by approximately one-third; however, 70-80% of patients receiving this treatment would have survived without it. None of the signatures of breast cancer gene expression reported to date allow for patient-tailored therapy strategies. Here we used DNA microarray analysis on primary breast tumours of 117 young patients, and applied supervised classification to identify a gene expression signature strongly predictive of a short interval to distant metastases ('poor prognosis' signature) in patients without tumour cells in local lymph nodes at diagnosis (lymph node negative). In addition, we established a signature that identifies tumours of BRCA1 carriers. The poor prognosis signature consists of genes regulating cell cycle, invasion, metastasis and angiogenesis. This gene expression profile will outperform all currently used clinical parameters in predicting disease outcome. Our findings provide a strategy to select patients who would benefit from adjuvant therapy.

Cancer susceptibility and the functions of BRCA1 and BRCA2

Inherited mutations in BRCA1 or BRCA2 predispose to breast, ovarian, and other cancers. Their ubiquitously expressed protein products are implicated in processes fundamental to all cells, including DNA repair and recombination, checkpoint control of cell cycle, and transcription. Here, I examine what is known about the biological functions of the BRCA proteins and ask how their disruption can induce susceptibility to specific types of cancer.

Identification of high risk breast-cancer patients by gene expression profiling

We previously used DNA array analyses in the molecular profiling of breast cancers. By cluster analysis of 55 patients, we identified a subpopulation of breast cancers-designated class A-that contained a high number of nodal-positive tumours and that had frequently developed distant metastases at the time of diagnosis. We have now analysed follow-up data from these patients. We found that, despite a median of only 23.5 months of follow-up, 11 of 22 patients in class A progressed to metastatic disease, and three of five patients classified as having a nodal status of N0 in this subpopulation developed distant metastases. Our analysis identifies breast-cancer patients with a high risk of disease recurrence, and could act as a first step towards improved patient-adapted therapy.

Quantitation of molecular endpoints for the dose-response component of cancer risk assessment

Cancer risk assessment involves the steps of hazard identification, dose-response assessment, exposure assessment, and risk characterization. The rapid advances in the use of molecular biology approaches has had an impact on all 4 components, but the greatest overall current and future impact will be on the dose-response assessment because this requires an understanding of the mechanisms of carcinogenesis, both background and induced by environmental agents. In this regard, hazard identification is a qualitative assessment and dose-response is a quantitative estimate. Thus, the latter will ultimately require a quantitative assessment of molecular endpoints that are used to describe the dose-response for cancer. It has been possible for many years to quantitate alterations at the level of the single gene. For example, analysis of mutation frequency by phenotypic selection, analysis of transcription (mRNA) by Northern blot, analysis of translation (proteins) by Western blot, and analysis of kinetics of metabolism from metabolite levels. However, it is becoming clear that it is necessary when considering risk for adverse health outcomes to develop quantitative approaches for whole cell phenotypes or organ effects. For example, cancer is a whole tissue phenotype, not a feature of single gene mutations, in spite of the multistep (multimutation) mode of formation of a tumor. Thus, there is the need to quantitate the circuitry of a cell: the metabolic/biochemical pathways, genetic regulation pathways, and signaling pathways in normal and stressed conditions. The hypothesis presented by Hanahan and Weinberg of the requirement for 6 acquired characteristics for tumor development, independent of tissue type and species or inducer, seems to provide a viable approach. This hypothesis can be addressed through whole cell molecular assessment using microarrays and quantitative PCR together with the emerging proteomic approaches. This is the world of the new computational cell biology.

BioArray Software Environment (BASE): a platform for comprehensive management and analysis of microarray data

The microarray technique requires the organization and analysis of vast amounts of data. These data include information about the samples hybridized, the hybridization images and their extracted data matrices, and information about the physical array, the features and reporter molecules. We present a web-based customizable bioinformatics solution called BioArray Software Environment (BASE) for the management and analysis of all areas of microarray experimentation. All software necessary to run a local server is freely available.

BRCA1 and BRCA2 in hereditary breast cancer

The hereditary breast and ovarian cancer susceptibility genes, BRCA1 and BRCA2, have established roles in genome integrity maintenance and in the control of homologous recombination. Recent work has produced valuable insights into the mechanisms of action of the gene products. This review summarizes some of these advances, and attempts to place them in the context of known functions of the genes.

Protein therapeutics: promises and challenges for the 21st century

Recent advances in massively parallel experimental and computational technologies are leading to radically new approaches to the early phases of the drug production pipeline. The revolution in DNA microarray technologies and the imminent emergence of its analogue for proteins, along with machine learning algorithms, promise rapid acceleration in the identification of potential drug targets, and in high-throughput screens for subpopulation-specific toxicity. Similarly, advances in structural genomics in conjunction with in vitro and in silico evolutionary methods will rapidly accelerate the number of lead drug candidates and substantially augment their target specificity. Taken collectively, these advances will usher in an era of predictive medicine, which will move medical practice from reactive therapy after disease onset, to proactive prevention.

Nuclear matrix proteins as biomarkers for breast cancer

Breast cancer remains a leading cause of cancer death in women despite widespread screening, in part because screening mammography has high rates of false-negative results and because many women decline to have routine mammograms. The development of sensitive and specific assays for breast tumor markers would improve detection and facilitate screening, diagnosis, therapeutic monitoring and surveillance for recurrence. Nuclear matrix proteins (NMPs) are promising candidates for tumor markers because they are involved in malignant transformation. Therefore, they may be useful for screening and early diagnosis of small tumors. Proteomic analysis was used to demonstrate that a 28.3 kD serum protein, designated NMP66, can distinguish malignant disease from benign conditions and normal controls. NMP66 is now being evaluated as a potential biomarker for early breast cancer detection in large-scale clinical trials.

Microarrays and molecular markers for tumor classification

Human cancers have traditionally been classified according to their tissue of origin, histological characteristics and, to some extent, molecular markers. Clinical studies have associated different tumor classes with differences in prognosis and in response to therapy. Measurement of the expression of thousands of genes in hundreds of cancer specimens has begun to reveal novel molecularly defined subclasses of tumor; some of these classes appear to predict clinical behavior, while others may define tumor types that are ripe for directed development of therapeutics. Unfortunately, at present, differences between studies of similar tumor types can be as striking as their similarities.

Breast cancer susceptibility genes: Options for those carrying BRCA1 mutations

The discovery of the association between breast and ovarian cancer and the BRCA genes and the development of methods for genetic testing made it possible to screen women for genetic predisposition to develop hereditary breast cancer (HBC). Parallelly, prevention strategies, including clinical surgical and medical interventions become available in order to reduce cancer risk. In a meantime, we became aware of limitations of genetic testing from the aspect of BRCA gene penetrance, negative result interpretation etc. All of these, together with data that invasive prevention strategies such as prophylactic surgery demonstrate better results in risk reduction than regimens including self and clinical-examination, face BRCA mutation carriers with difficult choice for risk reduction options. Therefore, the patients at high risk of HBC can best make informed decisions when guided by a multidisciplinary genetic counseling team.