Dataset Information

Expression profiling of 275 lung cancer specimens

ABSTRACT: From ~1,700 non-small cell lung cancer specimens collected at the MD Anderson Cancer Center over the years 1997 to 2005, we selected 914 tumors with detailed clinical and pathological information. We extracted RNA and DNA from frozen tissues using histology quality control from 700 cases. RNA was examined for quality using Agilent Bioanalyzer and RNA integrity number (RIN) was obtained for all specimens. A final selection of 275 tumor specimens was based on the following criteria: 1) Select mainly adenocarcinomas (n = 183) and squamous carcinomas (n = 80); 2) Tissue material contains at least 70% tumor by section examination; 3) within the tumor section, there should be at least 30% tumor cells (as opposed to stromal cells); 4) RIN should be at least 4.0 (range 0 - 10). Various profiling experiments were then performed including mRNA expression (this study), miRNA profiling, aCGH (Agilent), Reverse-phase protein array (RPPA), mutation analysis of 20 genes (Sequenom) and MS-MALDI-TOFF analysis. 275 lung cancer specimens collected at the MD Anderson Cancer Center were profiled on Illumina WG6-V3 expression arrays. Detailed clinico-pathological information were also collected.

ORGANISM(S): Homo sapiens

SUBMITTER: Sato M

PROVIDER: S-ECPF-GEOD-41271 | biostudies-other | 2013 Jun

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:BACKGROUND:Identification of predictive molecular alterations in lung adenocarcinoma is essential for accurate therapeutic decisions. Although several molecular approaches are available, a number of issues, including tumor heterogeneity, frequent material scarcity, and the large number of loci to be investigated, must be taken into account in selecting the most appropriate technique. MALDI-TOF mass spectrometry (MS), which allows multiplexed genotyping, has been adopted in routine diagnostics as a sensitive, reliable, fast, and cost-effective method. Our aim was to test the reliability of this approach in detecting targetable mutations in non-small cell lung cancer (NSCLC). In addition, we also analyzed low-quality samples, such as cytologic specimens, that often, are the unique source of starting material in lung cancer cases, to test the sensitivity of the system. METHODS:We designed a MS-based assay for testing 158 mutations in the EGFR, KRAS, BRAF, ALK, PIK3CA, ERBB2, DDR2, AKT, and MEK1 genes and applied it to 92 NSCLC specimens and 13 liquid biopsies from another subset of NSCLC patients. We also tested the sensitivity of the method to distinguish low represented mutations using serial dilutions of mutated DNA. RESULTS:Our panel is able to detect the most common NSCLC mutations and the frequency of the mutations observed in our cohort was comparable to literature data. The assay identifies mutated alleles at frequencies of 2.5-10%. In addition, we found that the amount of DNA template was irrelevant to efficiently uncover mutated alleles present at high frequency. However, when using less than 10 ng of DNA, the assay can detect mutations present in at least 10% of the alleles. Finally, using MS and a commercial kit for RT-PCR we tested liquid biopsy from 13 patients with identified mutations in cancers and detected the mutations in 4 (MS) and in 5 samples (RT-PCR). CONCLUSIONS:MS is a powerful method for the routine predictive tests of lung cancer also using low quality and scant tissues. Finally, after appropriate validation and improvement, MS could represent a promising and cost-effective strategy for monitoring the presence and percentage of the mutations also in non-invasive sampling.

Dataset Information

Expression profiling of 275 lung cancer specimens

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure