No conflicts of interest were declared.
The biological and clinical value of p53 expression in pelvic high-grade serous carcinomas†
Version of Record online: 8 JUN 2010
Copyright © 2010 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
The Journal of Pathology
Volume 222, Issue 2, pages 191–198, October 2010
How to Cite
Köbel, M., Reuss, A., Bois, A. d., Kommoss, S., Kommoss, F., Gao, D., Kalloger, S. E., Huntsman, D. G. and Gilks, C. B. (2010), The biological and clinical value of p53 expression in pelvic high-grade serous carcinomas. J. Pathol., 222: 191–198. doi: 10.1002/path.2744
- Issue online: 1 SEP 2010
- Version of Record online: 8 JUN 2010
- Manuscript Accepted: 2 JUN 2010
- Manuscript Revised: 25 MAY 2010
- Manuscript Received: 5 APR 2010
- National Cancer Institute of Canada. Grant Number: 017051
- Michael Smith Foundation for Health Research Unit. Grant Number: INRUA006045
- Cheryl Brown Ovarian Cancer Outcomes Unit of the British Columbia Cancer Agency
Studies on the p53 expression and outcome for women with ovarian carcinoma have produced conflicting results. The observed heterogeneity may be due to the range of cut-offs used to define overexpression and the mix of histotypes of the study cohorts. We aimed to examine the association between p53 expression and biological properties of tumours as well as outcome in 502 pelvic high-grade serous carcinomas (HGSCs) derived from two population-based cohorts from British Columbia representing cases with or without residual tumour after initial surgery, respectively, and one clinical trial cohort from Germany (AGO-OVAR-3). p53 expression was assessed on tissue microarrays by immunohistochemistry using the DO-7 antibody. p53 expression was scored in three tiers as complete loss of expression, focal expression or overexpression (defined as more than 50% positive tumour cell nuclei) and correlated with survival using multivariate Cox regression models. p53 was completely absent in 30.3%, focally expressed in 12.0%, and overexpressed in 57.7% of HGSCs, which was an inverse pattern compared to clear cell and endometrioid types of ovarian carcinomas, where 76% and 69% of cases showed focal expression, respectively (p < 0.001, chi square test). Pelvic HGSCs show either complete absence of p53 expression or p53 overexpression in 88% of cases; thus, aberrant p53 expression is a ubiquitous feature of HGSCs. HGSCs with p53 overexpression were associated with a reduced risk of recurrence compared to cases with complete absence of p53 in the British Columbia cohort with residual tumour (HR = 0.71, 95% CI 0.51–0.99) and for a combination of all three cohorts (HR = 0.70, 95% CI 0.55–0.89) in multivariate analysis including age, stage, residual tumour, and stratification by cohort. The association of complete absence of p53 expression with unfavourable outcome suggests functional differences of TP53 mutations underlying overexpression, compared to those underlying complete absence of expression. Copyright © 2010 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.