Premium
Additive effect of the AZGP1, PIP, S100A8 and UBE2C molecular biomarkers improves outcome prediction in breast carcinoma
Author(s) -
Parris Toshima Z.,
Kovács Anikó,
Aziz Luaay,
Hajizadeh Shahin,
Nemes Szilárd,
Semaan May,
ForssellAronsson Eva,
Karlsson Per,
Helou Khalil
Publication year - 2013
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.28497
Subject(s) - gene signature , biology , oncology , medicine , microarray analysis techniques , microarray , tissue microarray , breast cancer , proportional hazards model , cancer research , cancer , gene , gene expression , genetics
The deregulation of key cellular pathways is fundamental for the survival and expansion of neoplastic cells, which in turn can have a detrimental effect on patient outcome. To develop effective individualized cancer therapies, we need to have a better understanding of which cellular pathways are perturbed in a genetically defined subgroup of patients. Here, we validate the prognostic value of a 13‐marker signature in independent gene expression microarray datasets ( n = 1,141) and immunohistochemistry with full‐faced FFPE samples ( n = 71). The predictive performance of individual markers and panels containing multiple markers was assessed using Cox regression analysis. In the external gene expression dataset, six of the 13 genes ( AZGP1, NME5, S100A8, SCUBE2, STC2 and UBE2C ) retained their prognostic potential and were significantly associated with disease‐free survival ( p < 0.001). Protein analyses refined the signature to a four‐marker panel [AZGP1, Prolactin‐inducible protein (PIP), S100A8 and UBE2C] significantly correlated with cycling, high grade tumors and lower disease‐specific survival rates. AZGP1 and PIP were found in significantly lower levels in invasive breast tissue as compared with adjacent normal tissue, whereas elevated levels of S100A8 and UBE2C were observed. A predictive model containing the four‐marker panel in conjunction with established clinical variables outperformed a model containing the clinical variables alone. Our findings suggest that deregulated AZGP1, PIP, S100A8 and UBE2C are critical for the aggressive breast cancer phenotype, which may be useful as novel therapeutic targets for drug development to complement established clinical variables.