In:
Journal of Clinical Oncology, American Society of Clinical Oncology (ASCO), Vol. 30, No. 15_suppl ( 2012-05-20), p. 4561-4561
Abstract:
4561 Background: Current clinical parameters are not accurate in the prediction of overall survival of prostate cancer patients. Methods: Driven by cancer stem cell hypothesis and by using a simple bioinformatics analysis, we identified 641 genes with either consistently high or low level of expression in human embryonic stem cells. We showed that these 641 genes had strong power to classify cancers into different biological subtypes and named them as embryonic stem cell gene predictors (ESCGPs). Nineteen selected ESCGPs and five control genes were analyzed by multiplex quantitative PCR in prostate fine needle aspiration (FNA) samples taken at diagnosis. The cohort included 189 patients diagnosed during 1986-2001. Of these patients, 97.9% had overall and cancer specific survival data and 77.9% were treated by medical or surgical castration as the primary treatment. The cohort was divided into a discovery and two validation subsets. The univariate and multivariate Cox proportional hazards and Kaplan-Meier plots were used in the survival analysis. A published dataset was used for an external validation. Results: Ten gene markers F3, WNT5B, VGLL3, CTGF, IGFBP3, c-MAF-a, c-MAF-b, AMACR, MUC1 and EZH2, showed a significant correlation to overall or cancer specific survival. Four of these genes, F3, IGFBP3, CTGF and AMACR, were independent of all clinical parameters. An expression signature of F3, VGLL3 and IGFBP3 characterized patients into three subtypes. The median overall survival was 2.60 years in the high risk, 3.85 years in the intermediate risk and 7.98 years in the low risk subtype, corresponding to a HR of 5.86 (95% CI 2.91-11.78, p 〈 0.001) for the high risk and 3.45 (95% CI 1.79-6.66, p 〈 0.001) for the intermediate risk over the low risk subtype. Two gene markers, F3 and EZH2, were further verified by the external validation. Conclusions: The new ESCGP gene markers and the expression signatures can be used to predict the overall and cancer specific survival of prostate cancer patients.
Type of Medium:
Online Resource
ISSN:
0732-183X
,
1527-7755
DOI:
10.1200/jco.2012.30.15_suppl.4561
Language:
English
Publisher:
American Society of Clinical Oncology (ASCO)
Publication Date:
2012
detail.hit.zdb_id:
2005181-5
Permalink
