The present invention relates to biomarkers allowing predicting
breast tumor and
solid tumor outcome using hypoxia related genes. More specifically, the present invention relates to a method for predicting the survival of a patient suffering from
cancer, said method comprising the steps of (a) measuring the expression of at least five genes selected from the group consisting of GLUT1, PGK1, LDHA, ENO1, CAIX, NHERF1, TPI, AMF / GPI, VEGFA, TGFB3, ENG, LEP, EDN1, MDR1, AK3, MXR1, TGM2, CDH1, MMP2, CK1 9, VIM, CXCR4, UPAR, CATHD,
CTGF, C0X2, MET, IGF-2, CCND1, EPO, NDRG1, BNIP3, NIX, ETS1, PHD2, TWIST1, DEC1, SNAH, CEBPA, CITED2, F0X03A, NUR77, BRCA1,
PTEN, VHL and
ERBB2 in a biological sample of said patient, and (b) analyzing the expression values to generate a risk
score of relapse, wherein a risk
score superior or equal to three is indicative of high risk of relapse and a risk
score inferior or equal to two is indicative of a low risk of relapse. In particular the following genes: EPO, ETS1, ENO1, PGK1, LDHA, TPI and optionally VEGFA were significantly over-expressed in patients with relapse.