Unknown,Transcriptomics,Genomics,Proteomics

Dataset Information

Carcinoma-associated fibroblasts’ transcriptomic program predicts clinical outcome in stage II/III colorectal cancer

ABSTRACT: We obtained fibroblast cultures from fresh surgical specimen ressected from patients with primary colorectal carcinoma: normal colonic fibroblasts (NCF=9) from the normal colonic mucosa at least 5-10cm from the surgical margin, carcinoma-associated fibroblasts from the primary tumor (CAF-PT=14) and carcinoma-associated fibroblasts (CAF-LM=11) from fresh surgical specimens of liver metastases. We identified 277 probes, in common between the three types of fibroblasts, whose expression level is sequentially deregulated according to cancer progression (NCF→CAF-PT→CAF-LM; fold change Log2 normalized expression>1.5 in each step). Prediction Analysis of Microarrays was applied to obtain a 25-gene signature that better characterizes each fibroblast class. The signature is able to classify patients carrying primary tumors according to prognosis. This fact was exploited to obtain a 19-gene signature (from the 277 deregulated probes) predicting recurrence with high accuracy in stage II/III colorectal cancer patients. Signature validation has been carried out in two independent datasets and in a meta-cohort of 336 stage II/III patients. Since the 25-gene signature was obtained regardless of gene expression data of tumor specimens or patient’s clinical data, the prognostic power of this signature provides strong evidence of the link between the tumor stroma and cancer progression. Furthermore, the 19-gene signature was able to identify low-risk patients with very high accuracy, especially relevant for those high-risk stage-II patients. We hybridised fibroblast RNA in Affymetrix GeneChip 1.0 st arrays

ORGANISM(S): Homo sapiens

SUBMITTER: david mollevi

PROVIDER: E-GEOD-46824 | biostudies-arrayexpress |

REPOSITORIES: biostudies-arrayexpress

ACCESS DATA

Similar Datasets

Project description:Introduction: Malignant pleural mesothelioma (MPM) is an aggressive malignancy with poor prognosis. Unlike many other cancers, MPM is mostly characterized by inactivation of tumor suppressor genes. Its highly malignant nature in absence of tumor driving oncogene mutations indicates an extrinsic supply of stimulating signals by cells of the tumor microenvironment (TME). Cancer-associated fibroblasts (CAFs) are an abundant cell type of the TME and have been shown to drive the progression of several cancer types. The aim of the current study was to isolate and characterize patient-derived mesothelioma-associated fibroblasts (Meso-CAFs), and evaluate their impact on MPM cells. Methods: Meso-CAFs were isolated from surgical specimens of MPM patients and analyzed by comparative genomic hybridization, transcriptomics and proteomics. Human MPM cell lines were retrovirally transduced with GFP. The impact of Meso-CAFs on tumor cell growth, migration, and response to pathway inhibitors and cisplatin was investigated in 2D and 3D co-culture models by videomicroscopy and automated image analysis. Results: Meso-CAFs possess normal genomes without gene copy number aberrations typical for MPM cells. They express CAF markers and lack MPM marker expression. Their proteome and secretome profiles clearly differ from normal lung fibroblasts with particularly strong differences in the fraction of actively secreted proteins. The presence of Meso-CAFs in co-culture resulted in increased proliferation and migration of MPM cells. A similar effect on cell growth was induced by conditioned medium. Met/PI3K or WNT signaling inhibition abolished the Meso-CAF-mediated growth stimulation. While Meso-CAFs reduced the efficacy of EGFR inhibition in co-cultures, they did not provide protection of tumor cells against cisplatin. Conclusion: Our study provides the first characterization of human patient-derived Meso-CAFs and demonstrates a strong impact of Meso-CAFs on tumor cell growth and migration, two key aspects of MPM aggressiveness, indicating a substantial role of Meso-CAFs in driving MPM progression. Moreover, we show that Meso-CAFs significantly influence drug response and identify signaling pathways required for Meso-CAF-mediated growth stimulation. These data could be relevant for improving therapeutic strategies in MPM.

Dataset Information

Carcinoma-associated fibroblasts’ transcriptomic program predicts clinical outcome in stage II/III colorectal cancer

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets