906P - A prostate cancer expression signature to predict survival benefit of primary hormone therapy

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Prostate Cancer
Translational Research
Presenter Chunde Li
Authors C. Li1, Z. Peng2, L. Skoog2, M. Hjelm-Eriksson1, U. Harmenberg1, L. Ährlund-Richter1, G. Cohn Cedermark1, Y. Pawitan3, M. Nistér2, S. Nilsson1
  • 1Clinical Oncology, Karolinska University Hosiptal, 171 76 - Stockholm/SE
  • 2Oncology-pathology, Karolinska Instiutet, SE 171 76 - Stockholm/SE
  • 3Medical Epidemiology And Biostatistics, Karolinska Institutet, SE 171 76 - Stockholm/SE



At the ASCO annual meeting 2012, we have presented the identification of an embryonic stem cell gene predictor (ESCGP) signature in a cohort of 189 patients with 7-21 years complete survival follow up. The expression signature is composed of F3, IGFBP3 and VGLL3 and can significantly improve the accuracy of survival prediction at the time of diagnosis. We further hypothesize that measuring stem cell gene expression signature in prostate cancer biopsy samples taken at the time of diagnosis can predict the survival benefit of hormone therapy.


In the previous cohort, 139 of the 189 patients were primarily treated by hormone therapy. Of these 139 patients 65 had data of the ESCGP signature. Kaplan-Meier plots were used in the survival analysis of this subset of patients.


Of the 65 patients, 22 were classified as having high risk, 25 as having intermediate risk and 18 as having low risk subtype cancers. The obvious overall and cancer specific survival difference between the three subtypes was still maintained (Kaplan-Meier survival curve, p < 0.001 by log rank test) in these 65 patients. At 8 years after the diagnosis, the overall survival rate was 0% in patients with the high risk subtype cancer, 16% in patients with the intermediate risk subtype cancer and 55.6% in patients with the low risk subtype cancer. This obvious survival difference by the ESCGP classification was independent of age, PSA level, tumor grade and clinical stage.


The expression signature of F3, IGFBP3 and VGLL3 can be accurately measured in prostate fine needle aspiration samples. If the finding is further validated in a large and independent cohort, the ESCGP expression signature can be used to predict the survival benefit of primary hormone treatment for patients with newly diagnosed prostate cancer.


C. Li: Chunde Li is the major stock holder of Chundesell Medcals AB.

All other authors have declared no conflicts of interest.