236P - Possible use of urinary prostate proteins glycosylation profile as a diagnostic biomarker for prostate cancer

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Prostate Cancer
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Tijl Vermassen
Citation Annals of Oncology (2014) 25 (suppl_4): iv58-iv84. 10.1093/annonc/mdu326
Authors T. Vermassen1, N. Lumen2, C. van Praet2, D. Vanderschaeghe3, N. Callewaert3, P. Hoebeke2, S. Van Belle4, J. Delanghe5, S. Rottey4
  • 1Department Of Medical Oncology, Ghent University Hospital, 9000 - Ghent/BE
  • 2Department Of Urology, University Hospital Ghent, 9000 - Ghent/BE
  • 3Unit For Medical Biotechnology, Inflammation Research Center, Flemisch Institute for Biotechnology, 9052 - Zwijnaarde/BE
  • 4Department Of Medical Oncology, Ghent University Hospital, Ghent/BE
  • 5Department Of Clinical Chemistry, Ghent University Hospital, Ghent/BE



Serum Prostate Specific Antigen (sPSA) is widely used for screening and early diagnosis of prostate cancer (PCa). This analysis is associated with considerable sensitivity and specificity problems, especially in the diagnostic grey zone (sPSA between 4 and 10 ng/mL). Other biomarkers have emerged but only few have shown clinical significance. Because of aberrant glycosylation changes in tumorogenesis, we explored the use of urinary prostate proteins and its glycosylation profile as a new biomarker for PCa.


We determined sPSA and urinary biomarkers in healthy volunteers (HV; n = 54), patients with benign prostate hyperplasia (BPH; n = 99) and PCa patients (n = 74). Urinary prostate protein N-glycans were determined by fluorophore-assisted carbohydrate electrophoresis.


N-glycan profile analyses have pointed out differences between the subject groups: a decrease in overall fucosylation was noticed in PCa patients compared to patients with BPH and HV (p = 0.001 and p < 0.0001, respectively) and a decrease in total amount of triantennary structures (p = 0.005 and p < 0.0001, respectively). These differences, combined into a urinary glycosylation marker (UGM) and divided by the prostate volume, were not statistically better than sPSA for BPH compared to PCa (AUC after overall ROC curve analysis: 0.75 [0.68–0.83] and 0.71 [0.63–0.79] for sPSA screening and UGM respectively). However, multivariate logistic regression showed that UGM gives an added value to sPSA screening (Table 1). Combining sPSA and UGM resulted in an overall AUC of 0.84 [0.78–0.90] which was significantly higher than for sPSA alone (p = 0.009). In the diagnostic grey zone, UGM was significantly better than sPSA (p = 0.004) with AUC of 0.55 [0.42–0.69] and 0.81 [0.71–0.91] for sPSA and UGM respectively.

Multivariate logistic regression
Variables OR (95% CI) P-value
Age 0.99 (0.94–1.04) n.s.
SPSA 1.19 (1.09–1.30) 0.0001
UGM 1.50 (1.26–1.79) < 0.0001


Our urinary marker was able to differentiate between BPH and PCa. These changes in N-glycosylation could lead to the discovery of a new biomarker for PCa, which seems particularly useful in the diagnostic gray zone of sPSA concentration between 4 and 10 ng/mL.


All authors have declared no conflicts of interest.