Full speed ahead for tumor screening

Bayesian statistical learning for big data biology

Bayesian statistical learning provides a coherent probabilistic framework for modelling uncertainty in systems. This review describes the theoretical foundations underlying Bayesian statistics and outlines the computational frameworks for implementing Bayesian inference in practice. We then describe the use of Bayesian learning in single-cell biology for the analysis of high-dimensional, large data sets.

Tissue microarrays in clinical urology--technical considerations

The tissue microarray (TMA) technique is a 'high-throughput' tool which enables a rapid and concurrent analysis of molecular targets in large numbers of specimens at the DNA, RNA and protein levels under standardized conditions. In addition, TMA is a cost- and time-efficient method. This review focuses on its technical side, outlining its advantages and weaknesses. It also gives an overview of some major studies of bladder and prostate cancer in which the TMA technique was used.

Molecular profiling techniques and bioinformatics in cancer research

AIMS

Our aim was to describe the commonly used molecular profiling techniques in cancer research, to examine their limitations and to discuss the challenges of bioinformatics.

METHODS

A literature search was performed using the PubMed database to identify publications relevant to this review. Citations from these articles were also examined to yield further relevant publications.

RESULTS

We describe the use of DNA microarrays, comparative genomic hybridisation, tissue microarrays and digital differential display. The limitations of these technologies, their contribution to cancer research and the challenges of bioinformatics are also discussed.

CONCLUSIONS

Although these high throughput technologies each have their own limitations they are rapidly developing and contributing significantly to our understanding of cancer genetics. They have also led to the emergence of bioinformatics as a rapidly developing and vital field.

Technology insight: Identification of biomarkers with tissue microarray technology

High-throughput technologies have been developed in the hope of increasing the pace of biomedical research, and accelerating the rate of translation from bench to bedside. Using such technology in target discovery has resulted in the need for systematic validation of the targets in an equally rapid manner. For example, gene expression microarrays have highlighted many potential targets in cancer, and tissue microarrays have emerged as a powerful tool to validate these targets by measuring tumor-specific protein expression and linking it to clinical outcome. Automated quantitative analysis of the tissue microarray 'spots' is beginning to take the technology a step further, removing observer bias, and providing standards for quality control and the potential for high-throughput analysis. The validation required for translation of tissue biomarkers from the research lab to the clinical lab will probably rely heavily on the combination of tissue microarray technology with automated quantitative analysis.

Tissue microarrays – automated analysis and future directions

Tissue microarrays have rapidly become a vital tool in the analysis of protein biomarkers on large patient cohorts. Composed of hundreds of miniature histologic tissue cores taken from patient tumors and arrayed in a grid pattern, tissue microarrays have great potential for automation. However, unlike DNA/RNA expression arrays, they contain spatial information, such that successful automation requires both the ability to quantitateandlocalize biomarkers within the tissue. This review traces tissue microarray analysis from the simplest of systems (visual inspection) to dedicated, automated quantitative analyzers capable of localizing biomarkers to sub-cellular compartments. It focuses on the importance of automating the process of tissue microarray analysis, while commenting on the inherent problems associated with both manual and automated analysis. Finally, it discusses the future of automated tissue microarray analysis, emphasizing the many areas of development and potential improvement.

Protein arrays and their role in proteomics

Arraying technologies have shown the way to smaller sample volumes, more efficient analyses and higher throughput. Proteomics is a field, which has grown in significance in the last five years. This review outlines recent developments in protein arrays and their applications in proteomics, and discusses the requirements, current limitations and the potential and future perspectives of the technology.

Molecular techniques for studying gene expression in carcinogenesis

Many genes and signaling pathways controlling cell proliferation, death, differentiation, and genomic integrity are involved in cancer development. Various methods are available for detection and quantification of messenger RNA. Older methods such as Northern blots, nuclease protection, plaque hybridization, and slot blots suffer from being inherently serial, measure a single mRNA at a time, or being difficult to automate. New techniques for analysis of gene expression include: (a) comprehensive open systems such as serial analysis of gene expression (SAGE), differential display (DD) analysis, RNA arbitrarily primer (RAP)-PCR, restriction endonucleolytic analysis of differentially expressed sequences (READS), amplified restriction fragment-length polymorphism (AFLP), total gene expression analysis (TOGA), and use of internal standard competitive template primers (CTs) in a quantitative multiplex RT-PCR method [StaRT-(PCR)], and (b) focused closed systems such as: high density cDNA filter hybridization (HDFCA) analysis, suppression subtractive hybridization (SSH), differential screening (DS), several forms of high-density cDNA arrays, or oligonucleotide chips, and tissue microarrays. Sometimes, a combination of these systems is used to enhance the sensitivity and specificity of the assays. While closed systems are excellent for the initial screening of large number of sequences, the value of the information generated is generally limited to an often arbitrarily chosen known sequence. On the other hand, only the open system platform has the potential to evaluate the expression patterns of tens of thousands of genes that have not yet been cloned or partially sequenced in a quantitative manner. A cost analysis of the most commonly used expression technologies is provided. A method for purifying tumors from surrounding stroma and normal tissue employing laser microdissection, and subsequent RNA isolation/amplification from few cells employing sensitive kits are also discussed.

Cell cycle regulators (p21, p53, pRb) in oligodendrocytic tumors: a study by novel tumor microarray technique

Using the novel tissue microarray technique, we studied immunohistochemical expression of cell cycle regulators p53, p21, pRb in 42 grade II oligodendrogliomas, 16 grade III anaplastic oligodendrogliomas, 10 primary and 4 recidive grade II oligoastrocytomas, 10 grade III oligoastrocytomas and 2 other grade II mixed gliomas. The p53 immunopositivity associated with malignant histology of the tumor (p = 0.01, Mann-Whitney test) and high pRb expression (p = 0.015). The p21 score associated strongly with histological grade (p < 0.001). The immunopositive tumors had a significantly higher rate of proliferation (p = 0.021). The p21 immunopositivity correlated positively with p53 immunopositivity: among the 33 p21 immunopositive tumors 30 (91%) were p53 immunopositive and only 3 were p53 immunonegative (p = 0.017). Patients with p21 immunonegative primary tumors had significantly better prognosis: among them 42 of the 46 (91%) survived, whereas only 18 of the 30 patients (60%) with p21 immunopositive primary tumors survived until the follow-up date (p = 0.0017). Statistical significance was reached in multivariate analysis as well (p = 0.01, exp(B) = 5.5). The pRb immunopositive tumors had higher proliferation rate than immunonegative tumors (p = 0.002). In multivariate variance analysis, comparing the effects of different regulatory proteins on cell proliferation, only the amount of pRb expression reached statistical significance (p = 0.004). In conclusion, the expression of p21 in oligodendrocytic tumors seems to be upregulated by p53 expression which rises with cell proliferation and malignancy as in attempt to halt cell cycle but seems to be overrun by other factors. The amount of p21 expression has independent prognostic significance and could be used in diagnosis to help the difficult evaluation of the malignancy potential of oligodendrogliomas and oligoastrocytomas.

Feature (gene) selection in gene expression-based tumor classification

There is increasing interest in changing the emphasis of tumor classification from morphologic to molecular. Gene expression profiles may offer more information than morphology and provide an alternative to morphology-based tumor classification systems. Gene selection involves a search for gene subsets that are able to discriminate tumor tissue from normal tissue, and may have either clear biological interpretation or some implication in the molecular mechanism of the tumorigenesis. Gene selection is a fundamental issue in gene expression-based tumor classification. In the formation of a discriminant rule, the number of genes is large relative to the number of tissue samples. Too many genes can harm the performance of the tumor classification system and increase the cost as well. In this report, we discuss criteria and illustrate techniques for reducing the number of genes and selecting an optimal (or near optimal) subset of genes from an initial set of genes for tumor classification. The practical advantages of gene selection over other methods of reducing the dimensionality (e.g., principal components), include its simplicity, future cost savings, and higher likelihood of being adopted in a clinical setting. We analyze the expression profiles of 2000 genes in 22 normal and 40 colon tumor tissues, 5776 sequences in 14 human mammary epithelial cells and 13 breast tumors, and 6817 genes in 47 acute lymphoblastic leukemia and 25 acute myeloid leukemia samples. Through these three examples, we show that using 2 or 3 genes can achieve more than 90% accuracy of classification. This result implies that after initial investigation of tumor classification using microarrays, a small number of selected genes may be used as biomarkers for tumor classification, or may have some relevance in tumor development and serve as a potential drug target. In this report we also show that stepwise Fisher's linear discriminant function is a practicable method for gene expression-based tumor classification.

Recursive partitioning for tumor classification with gene expression microarray data

Precise classification of tumors is critically important for cancer diagnosis and treatment. It is also a scientifically challenging task. Recently, efforts have been made to use gene expression profiles to improve the precision of classification, with limited success. Using a published data set for purposes of comparison, we introduce a methodology based on classification trees and demonstrate that it is significantly more accurate for discriminating among distinct colon cancer tissues than other statistical approaches used heretofore. In addition, competing classification trees are displayed, which suggest that different genes may coregulate colon cancers.

Protein and antibody arrays and their medical applications

Many new gene products are being discovered by large-scale genomics and proteomics strategies, the challenge is now to develop high throughput approaches to systematically analyse these proteins and to assign a biological function to them. Having access to these gene products as recombinantly expressed proteins, would allow them to be robotically arrayed to generate protein chips. Other applications include using these proteins for the generation of specific antibodies, which can also be arrayed to produce antibody chips. The availability of such protein and antibody arrays would facilitate the simultaneous analysis of thousands of interactions within a single experiment. This chapter will focus on current strategies used to generate protein and antibody arrays and their current applications in biological research, medicine and diagnostics. The shortcomings of these approaches, the developments required, as well as the potential applications of protein and antibody arrays will be discussed.

High topoisomerase IIalpha expression associates with high proliferation rate and and poor prognosis in oligodendrogliomas

The role of molecular markers predicting the prognosis and the selection of patients for further adjuvant therapies is not well established in oligodendroglioma patients. A potential prognostic as well as a therapeutically predictive factor, topoisomerase IIalpha (topoIIalpha), is a molecular target for certain cytotoxic drugs. Its expression has been shown to correlate with the prognosis in a number of different cancers and with the chemosensitivity of cancer cells in vitro. The expression of topoIIalpha was evaluated immunohistochemically in 59 oligodendrogliomas and in 29 mixed gliomas with a predominating oligodendroglioma component by the use of a tissue microarray technique. In the gliomas, the percentage of topoIIalpha immunopositive cells protein expression varied from 0.0 to 49.1% (5.2 +/- 8.3%, mean+/- SD). In oligoastrocytomas, the mean topoIIalpha score was significantly higher in the oligodendroglioma than in the astrocytoma component of the tumour (5.37 +/- 5.58% vs. 1.89 +/- 2.49%, P = 0.018). A significant association was found between the high proportion of topoIIalpha positive cells and high grade of the tumour (P < 0.0001), high tumour proliferation rate (P < 0.0001), p53 overexpression (P = 0.01) and high expression of tumour suppressing retinoblastoma protein (P = 0.023). TopoIIalpha expression was not associated with the age or sex of patient, and the rate of apoptosis. TopoIIalpha expression associated highly significantly with patient prognosis; a significantly higher proportion of patients with low rather than with high topoIIalpha score was alive at the end of the 5-year follow-up (P = 0.03). Cox analysis was used to demonstrate that topoIIalpha had an independent prognostic value for survival (P = 0.034). In conclusion, high topoIIalpha expression characterizes oligodendrogliomas and oligoastrocytomas which are poorly differentiated, have high proliferation rate, and has prognostic value for overall survival of these patients. Therefore, topoIIalpha may be a useful marker for better targeted selection of poor prognosis oligodendroglioma patients for adjuvant therapy.

Protein arrays for gene expression and molecular interaction screening

The array format has revolutionised biomedical experimentation and diagnostics, enabling ordered high-throughput analysis. During the past decade, classic solid phase substrates, such as microtitre plates, membrane filters and microscopic slides, were turned into high-density, chip-like structures. The concept of the arrayed library was central to this development which now extends from DNA to protein. The new and versatile protein array technology allows high-throughput screening for gene expression and molecular interactions. As a major platform for functional genomics, it is already on its way into medical diagnostics.

Differential expression assay of chromosome arm 8p genes identifies Frizzled-related (FRP1/FRZB) and Fibroblast Growth Factor Receptor 1 (FGFR1) as candidate breast cancer genes

Deletions and amplifications are frequent alterations of the short arm of chromosome 8 associated with various types of cancers, including breast cancers. This indicates the likely presence of tumor suppressor genes and oncogenes. In the present study, we have used the expressed sequence tag (EST) map of 8p11-21 to assemble a set of available cDNAs representing genes from this region. DNA arrays were prepared for expression analysis and search for genes potentially involved in breast cancer. Underexpresion in tumoral breast cells (versus normal breast) was observed for 15 transcripts. Among these, the Frizzled-related gene FRP1/FRZB, was turned off in 78% of breast carcinomas, suggesting that the lack of its product may be associated with malignant transformation. Overexpression in tumoral breast cells was observed for 13 genes. The FGFR1 gene, that encodes a tyrosine kinase receptor for members of the fibroblast growth factor family, was identified as a good candidate for one amplification unit. Taken together, our results demonstrate that such a strategy can rapidly identify genes with an altered pattern of expression and provide candidate genes for malignancies.

Kidney cancer

In the USA, the incidence of kidney cancer has increased 43% since 1973. The risk of the disorder is higher in men than in women and increases with age. The von Hippel-Lindau tumour-suppressor gene is inactivated in over 75% of sporadic cases. Metastatic disease is present in 20-30% of patients at diagnosis. Early-stage kidney cancer is treated with a radical nephrectomy, but under certain circumstances a partial nephrectomy may be done. Tumour thrombus into the vena cava or right atrium requires thoracotomy and hypothermic circulatory arrest for successful removal of the tumour, but should not be done if extensive nodal or frank metastatic disease is present. Interleukin-2 is the systemic therapy of choice for metastatic disease at present, with long-term relapse-free survival of 5-8%. Several treatments including anti-angiogenesis drugs, cyclin-dependent kinase inhibitors, and differentiating agents are being actively investigated. Fluorouracil has a 10-15% response rate, and surgical excision of isolated metastases should always be considered. Therapy for metastatic renal cancer remains inadequate, but recent developments in basic and clinical research suggest future improvement.