Dataset Information

Characterization of Primary Melanomas Using High-Resolution Array Based CGH Technology

ABSTRACT: Malignant melanoma is an aggressive heterogeneous disease for which new biomarkers for diagnosis and clinical outcome are needed. We investigated by array-CGH the presence of DNA gains and losses to provide better genomic profile of primary malignant melanoma and to explore the possibility to distinguish metastatic from non metastatic melanomas using this technology. High resolution array-CGH (Agilent Technologies, Palo Alto, CA), with more than 40.000 probes, has been used to analyze 20 frozen tissues of vertical growth phase primary melanoma with a minimum follow-up time of 36 months. Eight patients developed nodal metastatic disease and twelve did not. For results validation, 83 additional melanoma samples with similar clinical characteristics were analysed by FISH. DNA copy number aberrations (CNA) were observed in 19 out of 20 cases. The most frequent changes were complete or partial losses in chromosomes 9 (12 cases, 60%) and 10 (9 cases, 45%), partial gains or trisomies of chromosome 7 (10 cases, 50%); and monosomy of chromosome 19 (7 cases, 35%). Sixty-four recurrent aberrant regions (SORIs) were precisely delimited and used as variables for clustering. Unsupervised Cluster analysis allowed the segregation of samples into two genomic groups that naturally fitted with the metastatic condition of the cases. Four of these aberrant regions were chosen for their biological interest and were confirmed as aberrant using FISH technique on fixed paraffin embedded tissues. Conclusion: Supervised classification allowed obtaining aberrations useful to separate samples with different clinical outcome. Obtained results are useful to improve the knowledge about melanoma tumorigenesis but unfortunately cannot be used as a marker for metastatic progression. Keywords: Comparative Genomic Hybridization - array; Genomic Instability

ORGANISM(S): Homo sapiens

PROVIDER: GSE11109 | GEO | 2009/01/01

SECONDARY ACCESSION(S): PRJNA106997

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Low grade flat ductal intraepithelial neoplasia (DIN1a, flat epithelial atypia) is one of the earliest morphologically recognizable neoplastic lesions of the breast. Frequently, it occurs in association with lobular intraepithelial neoplasia (LIN). The aim of this study was to elucidate chromosomal aberrations in these early neoplastic breast lesions using array comparative genomic hybridization (CGH) analysis. Laser capture microdissection of 12 archival formalin-fixed, paraffin-embedded specimens harbouring both foci of DIN1a as well as LIN was performed. All analyzed cases of DIN1a and LIN showed chromosomal gains and losses. The aberration encountered most often was loss on 16q in 7 DIN1a (70%) and 10 LIN (91%) cases. Regarding changes in chromosome 1, four DIN1a (40%) and 7 LIN (64%) cases showed a gain on 1q. The results of our study show concurrent chromosomal aberrations of 1q gains and 16q losses in several cases with coexisting LIN and low grade flat DIN. These aberrations are known to be common in low grade invasive ductal carcinomas as well as more advanced (conventional) types of low grade DIN (low grade ductal carcinoma in-situ). Our results raise the possibility of similar molecular-genetic pathways in most of the cases with coexisting LIN and low grade flat DIN. In this study, 11 cases of classic lobular intraepithelial neoplasias and 10 cases with several areas of low grade flat DIN (DIN1a or flat epithelial atypia) were analyzed by array CGH. Sections of formalin-fixed paraffin-embedded (FFPE) specimens cut at 7-8µm were mounted on special foil-coated slides (Molecular Devices, San Diego, USA). The sections were then deparaffinized with Xylene, processed in decreasing concentrations of ethanol and stained with haematoxylin. Lasercapture microdissection for both lesions was performed at multiple sites using Veritas Arcturus. For reference-DNA, female mammary tissue without histomorphological changes obtained from reduction mammoplasty specimens was procecessed and laser-microdissected as explained above. Cells were digested in 10µl TE, pH 9, and 0.5µl proteinase K (20mg/ml) for 48h at 55°C. After inactivation of proteinase K at 99°C for 10min, the digest was stored at -20°C. Without any further purification, the complete digest was used for whole genome amplification by means of the WGA (Whole Genome Amplification) kit from Sigma following the manufacturer’s recommendations. Array CGH was performed as described previously. In brief, two µg of amplified tumor and reference DNA were labelled by random priming (BioPrime® Total Genomic Labeling System, Invitrogen, Carlsbad, CA) with Alexa Fluor® 3 and Alexa Fluor® 5, respectively, and hybridized onto a tiling path BAC array, consisting of the human 32k BAC Re-Array Set (BACPAC Resources Center; http://bacpac.chori.org/pHumanMinSet.htm; DNA kindly provided by Pieter de Jong) and a 1Mb Resolution BAC set (clones kindly provided by Nigel Carter, Wellcome Trust Sanger Centre). All protocols are provided in detail on our website (http://www.molgen.mpg.de/~abt_rop/molecular_cytogenetics/) and more information concerning this platform have been submitted to the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/; GPL5114). For the analysis and visualization of array CGH data, our software-package CGH-PRO was employed. No background subtraction was applied. Raw data were normalized by “Subgrid LOWESS”. For the assessment of copy number gains and losses, we used circular binary segmentation in combination with log2 ratio thresholds of 0.15 and -0.15, respectively.

Dataset Information

Characterization of Primary Melanomas Using High-Resolution Array Based CGH Technology

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure