Mass spectrometric and linear discriminant analysis of N-glycans of human serum alpha-1-acid glycoprotein in cancer patients and healthy individuals

N-glycan oligosaccharides of human serum α 1-acid glycoprotein (AGP) samples isolated from 43 individuals (healthy individuals and patients with lymphoma and with ovarian tumor) were analyzed by MALDI-TOF mass spectrometry and a multivariate statistical method (linear discriminant analysis, LDA). 34...

Full description

Saved in:
Bibliographic Details
Published in:Journal of proteomics 2008-07, Vol.71 (2), p.186-197
Main Authors: Imre, Tímea, Kremmer, Tibor, Héberger, Károly, Molnár-Szöllősi, Éva, Ludányi, Krisztina, Pócsfalvi, Gabriella, Malorni, Antonio, Drahos, László, Vékey, Károly
Format: Article
Language:English
Subjects:
Alpha-1-acid glycoprotein (AGP)
Biomarker
Biomarkers, Tumor
Cancer
Discriminant Analysis
Female
Glycosylation
Humans
Linear discriminant analysis (LDA)
Lymphoma
Lymphoma - blood
MALDI-TOF
Mass Spectrometry
Orosomucoid - chemistry
Orosomucoid - metabolism
Ovarian Neoplasms - blood
Ovarian tumor
Polysaccharides - chemistry
Proteomics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:N-glycan oligosaccharides of human serum α 1-acid glycoprotein (AGP) samples isolated from 43 individuals (healthy individuals and patients with lymphoma and with ovarian tumor) were analyzed by MALDI-TOF mass spectrometry and a multivariate statistical method (linear discriminant analysis, LDA). 34 different glycan structures have been identified. From the glycosylation pattern determined by mass spectrometry fucosylation and branching indices have been calculated. These parameters show only small differences between the patient groups studied, but these differences are not sufficiently large to use as a potential biomarker. LDA analysis, on the other hand shows a very good separation between the three groups (with a classification of 88%). Cross-validation indicates that the method has predictive power: Identifying cancerous vs. healthy individuals shows 96% selectivity and 93% specificity; identification of lymphoma vs. the mixed group of healthy and ovarian tumor cases is also promising (72% selectivity and 84% specificity). The pilot study presented here demonstrates that mass spectrometry combined with linear discriminant analysis (LDA) may provide valuable data for identifying and studying the pathophysiology of malignant diseases.
ISSN:1874-3919
DOI:10.1016/j.jprot.2008.04.005