889P - Immunohistochemistry (IHC) evaluation of a novel 4-protein prognostic and predictive biomarker panel in endometrial cancer (EC)

Date 08 October 2016
Event ESMO 2016 Congress
Session Poster Display
Topics Endometrial Cancer
Pathology/Molecular Biology
Personalised/Precision Medicine
Basic Scientific Principles
Basic Principles in the Management and Treatment (of cancer)
Presenter Bihani Kularatne
Citation Annals of Oncology (2016) 27 (6): 296-312. 10.1093/annonc/mdw374
Authors B. Kularatne1, R. Arora2, G. Elshstein3, N. Guppy3, A. Kirkwood4, T. Meyer1, R.S. Kristeleit5
  • 1Medical Oncology, University College London Cancer Institute, WC1E 6DD - London/GB
  • 2Histopathology, University College London Hospital, WC1E 6JJ - London/GB
  • 3Ucl Advanced Diagnostics, University College London Hospital UCLH NHS Foundation Trust, wc1e 6jj - London/GB
  • 4Cancer Research Uk & Ucl Cancer Trials Centre, UCL - University College London, W1T 4TJ - London/GB
  • 5Oncology, University College London, Cancer Institute, London/GB



EC is common and incidence has increased by 65% in 40 years. There are no validated biomarkers or approved targeted therapies in clinical use. This study evaluates a novel biomarker panel in EC by correlating clinico-pathological features and tumour tissue expression levels of p53, PTEN, phospho-P70S6K (pS6), phospho-Stathmin (pSTMN). pS6 and pSTMN activity is influenced by PI3K/Akt pathway activation which frequently occurs in EC.


The 144 EC patients who had primary surgery from January 2006 to December 2010 at University College London Hospital were retrospectively identified and included in this analysis. Patient characteristics are shown in Table 1. Antibodies for p53, PTEN, pS6 and pSTMN were optimised for use in this study. IHC was performed on surgical resection specimens. Standard scoring methods incorporating percentage of cells stained and staining intensity were applied.


Univariate analysis for disease specific survival (DSS) showed, as expected, non-endometrioid histology, grade 3 tumour, presence of lymphovascular invasion (LVI) or myometrial invasion (MI) and advanced International Federation of Gynaecology and Obstetrics (FIGO) stage conferred poor DSS. The overexpression of p53 and pSTMN was also associated with poor DSS. In multivariate analysis grade 3 histology, MI, p53 and pSTMN overexpression were the only factors that remained significantly associated with poor DSS.

Cox proportional hazard regression model comparing clinico-pathological features and biomarker expression with DSS. NA-Not applicable. HR-Hazard ratio. CI- Confidence interval.

Variable Categories Univariate Analysis Multivariate Analysis
n Deaths HR (95% CI) p value HR(95% CI) p value
Age >65 ≥65 59 85 7 17 1.91 (0.79-4.62) 0.15 3.43 (0.33-35.14) 0.3
Histological sub-type Endometrioid Non-endometrioid 128 16 16 8 4.90 (2.09-11.48)


We demonstrate for the first time that p53 and pSTMN overexpression are independent predictors of DSS in EC and may be useful prognostic biomarkers. Overexpression of pSTMN may predict sensitivity to PI3K pathway inhibitors in EC. Prospective evaluation is warranted in clinical studies.

Clinical trial identification

Legal entity responsible for the study

UCL Cancer Institute


UCL Cancer Institute


All authors have declared no conflicts of interest.