Novel tumor proteomics strategy discriminates patient sub-groups in HPV related gynecological cancer
[PRESS RELEASE 2012-05-10] A novel strategy to analyze tumor proteomics data was employed by researchers at the Karolinska Institutet and Science for Life Laboratory to increase the understanding of molecular pathways in the gynecological cancer vulvar squamous cell carcinoma and how patient sub-groups that do or do not relapse regardless of infection with human papilloma virus (HPV) can be discriminated.
Vulvar squamous cell carcinoma (VSCC) is the forth most common gynecological cancer and is divided into two sub-groups, one related to infection with human papilloma virus (HPV) and one that is HPV negative. The aim of the present study, published in Molecular and Cellular Proteomics (Epub ahead of print), was to identify patient sub-groups with low risk of cancer relapse. “This study has increased the molecular understanding of VSCC and a number of proteins and pathways have been found that could potentially result in a less invasive treatment of patient sub-groups,” says Janne Lehtiö, group leader at SciLifeLab Stockholm and principal investigator of the study.
A novel mass spectrometry (MS) proteomics strategy based on the analysis of pathways in clinical samples, analyzed as a group or on an individual tumor level basis, was employed to extract as much data as possible. Four proteins that were expressed differently from normal cells and associated with cancer relapse, independent of HPV status, were identified and further validated with immunohistochemistry (IHC). Using both MS and IHC, a patient sub-group that were HPV negative but did relapse in VSCC could be discriminated from other sub-groups based on the levels of these four selected proteins.
In this study, tumor proteomics using high-resolution isoelectric focusing and LC-MS/MS was used, followed by combining biological pathway mapping on individual tumor level with multivariate analysis to compare HPV and relapse status groups. After analysis and filtering in several steps on average 546 proteins from each individual tumor sample, were estimated to have a quantitative change relative to the sample mean. These 1566 quantified proteins in total can be found in a downloadable and searchable database (http://tools.scilifelab.seVSCCtpd).
AnnSofi Sandberg, Gunnel Lindell, Brita Nordström Källström, Rui Mamede Branca, Kristina Gemzell Danielsson, Mats Dahlberg, Barbro Larson, Jenny Forshed, Janne Lehtiö. “Tumor proteomics by multivariate analysis on individual pathway data for characterization of vulvar cancer phenotypes”. Molecular and Cellular Proteomics, First published on April 12, 2012, doi: 10.1074/mcp.M112.016998