Metabolomics,Unknown,Transcriptomics,Genomics,Proteomics

Dataset Information

Molecular Profiling provides evidence of the existence of two functionally distinct classes of ovarian cancer stroma

ABSTRACT: RNA microarray profiling of 45 tissue samples was carried out using the Affymetrix (U133) gene expression platform. Laser capture microdissection (LCM) was employed to isolate cancer cells from the tumors of 18 serous ovarian cancer patients (Cepi). For 7 of these patients, a matched set of surrounding cancer stroma (CS) was also collected. For controls, surface ovarian epithelial cells (OSE) were isolated from the normal (non-cancerous) ovaries of 12 individuals including matched sets of samples of OSE and normal stroma (NS) from 8 of these patients. Unsupervised hierarchical clustering of the microarray data resulted in the expected separation between the OSE and Cepi samples. Consistent with models of stromal activation, we also observed significant separation between the NS and CS samples. Unexpectedly, the CS samples sub-divided into two distinct groups. Analysis of expression patterns of genes encoding signaling molecules and compatible receptors in the CS and Cepi samples are consistent with the hypothesis that the two CS sub-groups differ significantly in their relative propensities to support tumor growth.The results indicate the existence of distinct categories of ovarian cancer stroma and suggest that functionally significant variability exists among ovarian cancer patients in the ability of the microenvironment to modulate cancer development. Gene expression analysis of 8 normal stroma (NS) and 8 matched normal ovarian surface epithelium (OSE) from 12 individuals, along with 7 cancer stroma and 7 matched cancer epithelium from 18 additional ovarian cancer patients

ORGANISM(S): Homo sapiens

SUBMITTER: Loukia Lili

PROVIDER: E-GEOD-38666 | biostudies-arrayexpress |

REPOSITORIES: biostudies-arrayexpress

ACCESS DATA

Similar Datasets

Project description:A data set of normal epithelium, serous ovarian surface epithelial-stromal tumors (benign and type II malignancies), stroma distal to tumor, and stroma adjacent to tumor (50 samples total). Additional cel files are included which represent replicate sampling from patients, and cel files that failed quality control but may be bioinformatically interesting. Additional replicate or failed cel files were not included in the final analysis (and so these samples were not included in the matrix). Background: Ovarian cancer is the most lethal gynecologic cancer in the United States. If caught in early stages, patient survival rate can reach 94%, when diagnosed at late stages survival rates drop to 28%. Correct diagnosis depends on the presence of definite symptoms: while ~90% of diagnosed ovarian cancers have symptoms, they tend to be unfocused and subacute. A definitive and early molecular signature of disease is thus desired. To further progress towards this goal, we present an Affymetrix™ human exon array data set measuring ovarian tumor expression, assembled using best practices. Method: Samples were collected from patients with benign and malignant (type II) serous ovarian surface epithelial-stromal tumors. Normal epithelium, tumor, stroma adjacent to tumor, and distal stroma were selected based on histopathology, and isolated using laser capture microdissection. Nugen products were used to perform random-primed mRNA amplification procedures (for full transcript capture) before hybridization to Affymetrix exon chips. Tumor expression and paracrine signaling was assessed using GC-RMA and a two-way Model I ANOVA. Single enrichment ontological analysis and gene network construction were performed to guide inferences about biological context. Results: In total, across 50 microarrays, ~270 million measurements were obtained. Based on comparisons to known ovarian cancer properties as established in molecular genetics literature, the initial analysis presented emphasizes data quality. Major trends between sample classes included: apical surface and tight junction activity, mitotic activity, benign tumor suppression, epithelial-mesenchymal transitioning, tumor oncogene activity, and paracrine signaling. A list of differentially expressed transcripts has been produced to enable rapid comparisons with published biomarker lists, but it is expected that detailed alternative transcript analysis will refine these predictions. Conclusions: A data set of 50 arrays, from carefully dissected serous ovarian surface epithelial-stromal tumors, has been produced, from which high quality measurements were obtained. While relatively small in number, this represents an important addition to the community pool of ovarian tumor samples, and the chosen platform enables bridging between 3' expression and exome sequencing data sets. This represents a significant contribution to the ovarian cancer genetics community. A total of 50 human ovary samples were used in analysis: 23 tissue samples laser capture microdissected from an ovary with a benign serous tumor (specifically 4 normal epithelium samples, 5 tumor samples, 6 stroma samples adjacent to the tumor, and 8 stroma samples distal of the tumor), and 27 tissue samples laser capture microdissected from an ovary with a malignant serous tumor (specifically 5 normal epithelium samples, 8 tumor samples, 7 stroma samples adjacent to the tumor, and 7 stroma samples distal of the tumor). Additional cel files were provided which, although were used in the quality control of the data set, were not used in the final experimental analysis. 8 replicates are included as cel files (1 from each cohort previously listed). 14 cel files were also included which failed quality control. Although the replicate and failed cel files were not used in the final analysis, they may still be interesting in other research.

Dataset Information

Molecular Profiling provides evidence of the existence of two functionally distinct classes of ovarian cancer stroma

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure