Abstract 2262: The prognostic value of p53 abnormalities in colorectal cancers varies based on detection methods

Abstract

Abstract Background: Although abnormalities in p53 (gene mutations or nuclear accumulation) are considered, by most, but not all, investigators, as prognostic markers for colorectal cancers (CRCs), their clinical utility remains controversial. The conflicting findings may be attributed, in part, to technical variations in detecting p53 abnormalities. Therefore, this study examined the effect of detection methods on the prognostic value of p53 abnormalities in CRCs. Methods: We evaluated 107 CRCs and their corresponding normal snap-frozen and formalin-fixed paraffin-embedded (FFPE) tissues for p53 abnormalities. To detect p53 mutations, DNA extracted from frozen and FFPE tissues were analyzed by single-strand conformation polymorphisms (SSCP) and by direct sequencing using exon-specific primers, respectively. Also, the FFPE tissue sections were evaluated for p53 nuclear accumulation (p53nac) by immunohistochemistry (IHC), using an anti-human p53 monoclonal antibody (Clone BP53.12.1, BioGenex, San Ramon, CA), with and without antigen retrieval (AR). A correlation coefficient test was performed to determine the concordance between the molecular and IHC methods with different cutoff-values (>0%, 10%, 20%, and 50% positive cells) to categorize tumors as positive for p53nac. Data derived by the molecular and IHC methods were correlated independently with patient survival by Kaplan-Meier analyses. Subsequently, with an expanded CRC cohort of 339 tissues, we validated the findings obtained from the exon-wise direct DNA sequencing and IHC without AR. Results The highest concordance (79.5%, Kappa co-efficient 0.612; p<0.0001) was between p53nac by IHC without AR, at a cut-off value of ≥10% positive cells and mutations detected by molecular methods. Also, this IHC method detected 95% of the p53 point mutations. Although statistically not significant, Kaplan-Meier analyses demonstrated short CRC-specific survival for patients whose tumors exhibited p53nac, detected by IHC at ≥ 10% cut-off value without AR, (log-rank p value=0.0590); for patients with mutated p53, survival rates were similar to those derived by SSCP (log-rank p value=0.0592) and by direct DNA sequencing (log-rank p value=0.096). The validation studies demonstrated that abnormal p53, detected by IHC (log-rank p value=0.042) and direct sequencing methods (log-rank p value<0.0001), was associated with shorter patient survival. Conclusions: These findings demonstrate that IHC without AR is an effective approach to detect clinically useful p53 status and can be routinely used as a prognostic marker for colorectal cancer. These studies were supported by grants 2U54-CA118948-03, R01-CA98932, and R03-CA139629 from the NCI/NIH. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2262. doi:10.1158/1538-7445.AM2011-2262