Project description:In patients with metastatic melanoma, the identification and validation of accurate prognostic biomarkers will assist rational treatment planning. Studies based on "-omics" technologies have focussed on a single high-throughput data type such as gene or microRNA transcripts. Occasionally, these features were evaluated in conjunction with limited clinico-pathologic data. With the increased availability of multiple data types, there is a pressing need to tease apart which of these sources contain the most valuable prognostic information. We evaluated and integrated several data types derived from the same tumor specimens in AJCC stage III melanoma patients - gene, protein, and microRNA expression as well as clinical, pathologic and mutation information - to determine their relative impact on prognosis. We used classification frameworks based on pre-validation and bootstrap multiple imputation classification to compare the prognostic power of each data source, both individually as well as integratively. We found that the prognostic utility of clinico-pathologic information was not out-performed by various "-omics" platforms. Rather, a combination of clinico-pathologic variables and mRNA expression data performed best. Furthermore, a patient-based classification analysis revealed that the prognostic accuracy of various data types was not the same for different patients, providing useful insights for ongoing developments in the individualized treatment of melanomas patients. SPECIAL NOTE: In this study, survival data were re-extracted from the MIA research database for all patients and brought up to date, revealing discrepancies affecting survival class in the case of four patients compared with the previous dataset (GSE53118: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE53118). The current survival data are considered to be more accurate although the expression information has not changed. In addition, there were 5 samples (122, 144, 195, 264, and 358) for which gene expression information were not available at the time of analysis. However, the associated clinical information for these samples is provided since it was analysed elsewhere in the accompanying publication. Samples eligible for this study (n=84) were obtained from lymph node specimens (Melanoma Institute Australia (MIA) Biospecimen Bank) in which macroscopic tumor was observed, obtained from patients believed to be without distant metastases at the time of tumor banking based on clinical examination and computerised axial tomographic scanning of the brain, chest, abdomen and pelvis. Specimens were macro-dissected at time of banking and subsequently reviewed to meet minimum criteria for tumor cell content (>80%) and amount of necrosis (<30%). Linked clinical and pathologic data were obtained from the MIA research database. We previously analyzed the distribution of survival times in these samples and identified more favorable and less favorable groups as patients having time from surgery to death from melanoma greater than 4 years with no sign of relapse (n=25) or less than 1 year (n=22), respectively (Mann et al. 2013, PMID: 22931913). Since that publication, survival data have been re-extracted from the MIA research database for all patients and brought up to date, revealing discrepancies affecting survival class in the case of four patients compared with the previous dataset. The current survival data are considered to be more accurate. MRNA expression profiling and somatic mutation profiling, were performed as previously described in Mann et al. 2013 (PMID: 22931913).

Project description:In patients with metastatic melanoma, the identification and validation of accurate prognostic biomarkers will assist rational treatment planning. Studies based on "-omics" technologies have focussed on a single high-throughput data type such as gene or microRNA transcripts. Occasionally, these features were evaluated in conjunction with limited clinico-pathologic data. With the increased availability of multiple data types, there is a pressing need to tease apart which of these sources contain the most valuable prognostic information. We evaluated and integrated several data types derived from the same tumor specimens in AJCC stage III melanoma patients - gene, protein, and microRNA expression as well as clinical, pathologic and mutation information - to determine their relative impact on prognosis. We used classification frameworks based on pre-validation and bootstrap multiple imputation classification to compare the prognostic power of each data source, both individually as well as integratively. We found that the prognostic utility of clinico-pathologic information was not out-performed by various "-omics" platforms. Rather, a combination of clinico-pathologic variables and mRNA expression data performed best. Furthermore, a patient-based classification analysis revealed that the prognostic accuracy of various data types was not the same for different patients, providing useful insights for ongoing developments in the individualized treatment of melanomas patients. SPECIAL NOTE: In this study, survival data were re-extracted from the MIA research database for all patients and brought up to date, revealing discrepancies affecting survival class in the case of four patients compared with the previous dataset (GSE53118: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE53118). The current survival data are considered to be more accurate although the expression information has not changed. In addition, there were 5 samples (122, 144, 195, 264, and 358) for which gene expression information were not available at the time of analysis. However, the associated clinical information for these samples is provided since it was analysed elsewhere in the accompanying publication.

Dataset Information

Somatic mutation detection and classification through probabilistic integration of clonal population information.

Publications

Somatic mutation detection and classification through probabilistic integration of clonal population information.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure