Development and validation of a prognostic gene-expression signature for lung adenocarcinoma

Although several prognostic signatures have been developed in lung cancer, their application in clinical practice has been limited because they have not been validated in multiple independent data sets. Moreover, the lack of common genes between the signatures makes it difficult to know what biological process may be reflected or measured by the signature. By using classical data exploration approach with gene expression data from patients with lung adenocarcinoma (n = 186), we uncovered two distinct subgroups of lung adenocarcinoma and identified prognostic 193-gene gene expression signature associated with two subgroups. The signature was validated in 4 independent lung adenocarcinoma cohorts, including 556 patients. In multivariate analysis, the signature was an independent predictor of overall survival (hazard ratio, 2.4; 95% confidence interval, 1.2 to 4.8; p = 0.01). An integrated analysis of the signature revealed that E2F1 plays key roles in regulating genes in the signature. Subset analysis demonstrated that the gene signature could identify high-risk patients in early stage (stage I disease), and patients who would have benefit of adjuvant chemotherapy. Thus, our study provided evidence for molecular basis of clinically relevant two distinct two subtypes of lung adenocarcinoma.     

Glycoprotein profiles of human breast cells demonstrate a clear clustering of normal/benign versus malignant cell lines and basal versus luminal cell lines

Gene expression profiling has defined molecular subtypes of breast cancer including those identified as luminal and basal. To determine if glycoproteins distinguish various subtypes of breast cancer, we obtained glycoprotein profiles from 14 breast cell lines. Unsupervised hierarchical cluster analysis demonstrated that the glycoprotein profiles obtained can serve as molecular signatures to classify subtypes of breast cancer, as well as to distinguish normal and benign breast cells from breast cancer cells. Statistical analyses were used to identify glycoproteins that are overexpressed in normal versus cancer breast cells, and those that are overexpressed in luminal versus basal breast cancer. Among the glycoproteins distinguishing normal breast cells from cancer cells are several proteins known to be involved in cell adhesion, including proteins previously identified as being altered in breast cancer. Basal breast cancer cell lines overexpressed a number of CD antigens, including several integrin subunits, relative to luminal breast cancer cell lines, whereas luminal breast cancer cells overexpressed carbonic anhydrase 12, clusterin, and cell adhesion molecule 1. The differential expression of glycoproteins in these breast cancer cell lines readily allows the classification of the lines into normal, benign, malignant, basal, and luminal groups.     

SWATH‐based proteomics reveals processes associated with immune evasion and metastasis in poor prognosis colorectal tumours

Newly emerged proteomic methodologies, particularly data‐independent acquisition (DIA) analysis–related approaches, would improve current gene expression–based classifications of colorectal cancer (CRC). Therefore, this study was aimed to identify protein expression signatures using SWATH‐MS DIA and targeted data extraction, to aid in the classification of molecular subtypes of CRC and advance in the diagnosis and development of new drugs. For this purpose, 40 human CRC samples and 7 samples of healthy tissue were subjected to proteomic and bioinformatic analysis. The proteomic analysis identified three different molecular CRC subtypes: P1, P2 and P3. Significantly, P3 subtype showed high agreement with the mesenchymal/stem‐like subtype defined by gene expression signatures and characterized by poor prognosis and survival. The P3 subtype was characterized by decreased expression of ribosomal proteins, the spliceosome, and histone deacetylase 2, as well as increased expression of osteopontin, SERPINA 1 and SERPINA 3, and proteins involved in wound healing, acute inflammation and complement pathway. This was also confirmed by immunodetection and gene expression analyses. Our results show that these tumours are characterized by altered expression of proteins involved in biological processes associated with immune evasion and metastasis, suggesting new therapeutic options in the treatment of this aggressive type of CRC.     

Class discovery analysis of the lung cancer gene expression data

Traditional histological classification of lung cancer subtypes is informative, but incomplete. Recent studies of gene expression suggest that molecular classification can be used for effective diagnostic and prediction of the treatment outcome. We attempt to build a molecular classification based on the public data available from a few independent sources. The data is reanalyzed with a new cluster analysis algorithm. This algorithm allows us to preserve the high dimensionality of data and produce the cluster structure without preliminary selection of significant genes or any other presumption about the relation between different cancer and normal tissue samples. The resulting clusters are generally consistent with the histological classification. However, our analysis reveals many additional details and subtypes of previously defined types of lung cancer. Large histological cancer types can be further divided into subclasses with different patterns of gene expression. These subtypes should be taken into account in diagnostics, drug testing, and treatment development for lung cancer patients.     

Molecular signatures of lung cancer: defining new diagnostic and therapeutic paradigms

Molecular profiling holds great promise for improving our ability to diagnose, prognosticate, and select individualized treatments for lung cancer patients. However, using multidimensional data and novel technologies to derive these profiles is limited by our ability to employ the assay in a clinical scenario where it can impact the course of disease. Although many molecular signatures have been reported in lung cancer, as of yet, few have been sufficiently validated for widespread clinical use. Recently, several novel signatures have been reported, which address critical aspects of patient care and/or demonstrate improved efforts for appropriate clinical validation. Here, we present our opinion on the current state of the field of molecular signatures in lung cancer.     

Messina: A Novel Analysis Tool to Identify Biologically Relevant Molecules in Disease

Background

Morphologically similar cancers display heterogeneous patterns of molecular aberrations and follow substantially different clinical courses. This diversity has become the basis for the definition of molecular phenotypes, with significant implications for therapy. Microarray or proteomic expression profiling is conventionally employed to identify disease-associated genes, however, traditional approaches for the analysis of profiling experiments may miss molecular aberrations which define biologically relevant subtypes.

Methodology/Principal Findings

Here we present Messina, a method that can identify those genes that only sometimes show aberrant expression in cancer. We demonstrate with simulated data that Messina is highly sensitive and specific when used to identify genes which are aberrantly expressed in only a proportion of cancers, and compare Messina to contemporary analysis techniques. We illustrate Messina by using it to detect the aberrant expression of a gene that may play an important role in pancreatic cancer.

Conclusions/Significance

Messina allows the detection of genes with profiles typical of markers of molecular subtype, and complements existing methods to assist the identification of such markers. Messina is applicable to any global expression profiling data, and to allow its easy application has been packaged into a freely-available stand-alone software package.     

CAFET algorithm reveals Wnt/PCP signature in lung squamous cell carcinoma

We analyzed the gene expression patterns of 138 Non-Small Cell Lung Cancer (NSCLC) samples and developed a new algorithm called Coverage Analysis with Fisher's Exact Test (CAFET) to identify molecular pathways that are differentially activated in squamous cell carcinoma (SCC) and adenocarcinoma (AC) subtypes. Analysis of the lung cancer samples demonstrated hierarchical clustering according to the histological subtype and revealed a strong enrichment for the Wnt signaling pathway components in the cluster consisting predominantly of SCC samples. The specific gene expression pattern observed correlated with enhanced activation of the Wnt Planar Cell Polarity (PCP) pathway and inhibition of the canonical Wnt signaling branch. Further real time RT-PCR follow-up with additional primary tumor samples and lung cancer cell lines confirmed enrichment of Wnt/PCP pathway associated genes in the SCC subtype. Dysregulation of the canonical Wnt pathway, characterized by increased levels of β-catenin and epigenetic silencing of negative regulators, has been reported in adenocarcinoma of the lung. Our results suggest that SCC and AC utilize different branches of the Wnt pathway during oncogenesis.     

A Novel Five Gene Signature Derived from Stem-Like Side Population Cells Predicts Overall and Recurrence-Free Survival in NSCLC

Gene expression profiling has been used to characterize prognosis in various cancers. Earlier studies had shown that side population cells isolated from Non-Small Cell Lung Cancer (NSCLC) cell lines exhibit cancer stem cell properties. In this study we apply a systems biology approach to gene expression profiling data from cancer stem like cells isolated from lung cancer cell lines to identify novel gene signatures that could predict prognosis. Microarray data from side population (SP) and main population (MP) cells isolated from 4 NSCLC lines (A549, H1650, H460, H1975) were used to examine gene expression profiles associated with stem like properties. Differentially expressed genes that were over or under-expressed at least two fold commonly in all 4 cell lines were identified. We found 354 were upregulated and 126 were downregulated in SP cells compared to MP cells; of these, 89 up and 62 downregulated genes (average 2 fold changes) were used for Principle Component Analysis (PCA) and MetaCore™ pathway analysis. The pathway analysis demonstrated representation of 4 up regulated genes (TOP2A, AURKB, BRRN1, CDK1) in chromosome condensation pathway and 1 down regulated gene FUS in chromosomal translocation. Microarray data was validated using qRT-PCR on the 5 selected genes and all showed robust correlation between microarray and qRT-PCR. Further, we analyzed two independent gene expression datasets that included 360 lung adenocarcinoma patients from NCI Director''s Challenge Set for overall survival and 63 samples from Sungkyunkwan University (SKKU) for recurrence free survival. Kaplan-Meier and log-rank test analysis predicted poor survival of patients in both data sets. Our results suggest that genes involved in chromosome condensation are likely related with stem-like properties and might predict survival in lung adenocarcinoma. Our findings highlight a gene signature for effective identification of lung adenocarcinoma patients with poor prognosis and designing more aggressive therapies for such patients.     

Gene expression profiling of endobronchial ultrasound (EBUS)‐derived cytological fine needle aspirates from hilar and mediastinal lymph nodes in non‐small cell lung cancer

R. Lee, D. J. Cousins, E. Ortiz‐Zapater, R. Breen, E. McLean and G. Santis
Gene expression profiling of endobronchial ultrasound (EBUS)‐derived cytological fine needle aspirates from hilar and mediastinal lymph nodes in non‐small cell lung cancer Objective: Endobronchial ultrasound (EBUS) allows minimally invasive sampling of hilar and mediastinal lymph nodes and has an established role in non‐small cell lung cancer (NSCLC) diagnosis and staging. Molecular biomarkers are being explored increasingly in lung cancer research. Gene expression profiling (GEP) is a microarray‐based technology that comprehensively assesses genome‐wide changes in gene expression that can provide tumour‐specific molecular signatures with the potential to predict prognosis and treatment responsiveness. We assessed the feasibility of using EBUS‐derived aspirates from benign and tumour‐infiltrated lymph nodes for GEP. Methods: RNA was extracted from EBUS‐directed transbronchial fine needle aspiration samples in routine clinical practice. GEP was subsequently performed in six patients with NSCLC, three of whom had tumour‐infiltrated nodes and three who had benign lymph nodes; the differences in gene expression were then compared. Results: RNA was successfully extracted in 29 of 32 patients, 12 of whom were diagnosed with NSCLC. RNA yield (median, 12.1 μg) and RNA integrity (median, 6.3) were sufficient after amplification for GEP. Benign and malignant nodes in adenocarcinoma were discriminated by principal component analysis and hierarchical clustering with different expression patterns between malignant and benign nodes. Conclusion: We have demonstrated the feasibility of RNA extraction and GEP on EBUS‐derived transbronchial fine needle aspirates from benign and tumour‐infiltrated lymph nodes in patients with known NSCLC in routine clinical practice. Further studies on larger patient cohorts are required to identify expression profiles that robustly differentiate benign from malignant lymph nodes in NSCLC.     

Iterative local Gaussian clustering for expressed genes identification linked to malignancy of human colorectal carcinoma

Gene expression profiling plays an important role in the identification of biological and clinical properties of human solid tumors such as colorectal carcinoma. Profiling is required to reveal underlying molecular features for diagnostic and therapeutic purposes. A non-parametric density-estimation-based approach called iterative local Gaussian clustering (ILGC), was used to identify clusters of expressed genes. We used experimental data from a previous study by Muro and others consisting of 1,536 genes in 100 colorectal cancer and 11 normal tissues. In this dataset, the ILGC finds three clusters, two large and one small gene clusters, similar to their results which used Gaussian mixture clustering. The correlation of each cluster of genes and clinical properties of malignancy of human colorectal cancer was analysed for the existence of tumor or normal, the existence of distant metastasis and the existence of lymph node metastasis.     

A comparison of cluster analysis methods using DNA methylation data

MOTIVATION: Aberrant DNA methylation is common in cancer. DNA methylation profiles differ between tumor types and subtypes and provide a powerful diagnostic tool for identifying clusters of samples and/or genes. DNA methylation data obtained with the quantitative, highly sensitive MethyLight technology is not normally distributed; it frequently contains an excess of zeros. Established tools to analyze this type of data do not exist. Here, we evaluate a variety of methods for cluster analysis to determine which is most reliable. RESULTS: We introduce a Bernoulli-lognormal mixture model for clustering DNA methylation data obtained using MethyLight. We model the outcomes using a two-part distribution having discrete and continuous components. It is compared with standard cluster analysis approaches for continuous data and for discrete data. In a simulation study, we find that the two-part model has the lowest classification error rate for mixture outcome data compared with other approaches. The methods are illustrated using DNA methylation data from a study of lung cancer cell lines. Compared with competing hierarchical clustering methods, the mixture model approaches have the lowest cross-validation error for detecting lung cancer subtype (non-small versus small cell). The Bernoulli-lognormal mixture assigns observations to subgroups with the lowest uncertainty. AVAILABILITY: Software is available upon request from the authors. SUPPLEMENTARY INFORMATION: http://www-rcf.usc.edu/~kims/SupplementaryInfo.html     

Complementary hierarchical clustering

When applying hierarchical clustering algorithms to cluster patient samples from microarray data, the clustering patterns generated by most algorithms tend to be dominated by groups of highly differentially expressed genes that have closely related expression patterns. Sometimes, these genes may not be relevant to the biological process under study or their functions may already be known. The problem is that these genes can potentially drown out the effects of other genes that are relevant or have novel functions. We propose a procedure called complementary hierarchical clustering that is designed to uncover the structures arising from these novel genes that are not as highly expressed. Simulation studies show that the procedure is effective when applied to a variety of examples. We also define a concept called relative gene importance that can be used to identify the influential genes in a given clustering. Finally, we analyze a microarray data set from 295 breast cancer patients, using clustering with the correlation-based distance measure. The complementary clustering reveals a grouping of the patients which is uncorrelated with a number of known prognostic signatures and significantly differing distant metastasis-free probabilities.