p53 and Ki-ras as prognostic factors for Dukes' stage B colorectal cancer

Abstract

Mutations of the TP53 and Ki-ras genes have been reported to be of prognostic importance in colorectal carcinomas. An increased intracellular concentration of the p53 protein, although not identical to, is sometimes seen in tumours with TP53 mutation and has been correlated with poor prognosis in some tumour types. Previous colorectal cancer studies, addressing the prognostic importance of Ki-ras mutation and TP53 aberrations, yielded contradictory results. The aim of this study was to determine in a clinically and therapeutically homogeneous group of 122 sporadic Dukes' B colorectal carcinomas with a median follow-up of 67 months (3–144 months) whether or not p53 protein expression, TP53 mutation and K-ras mutation correlated with prognosis. p53 staining was performed by immunohistochemistry, using the monoclonal antibody DO7 on paraffin-embedded tissue. Mutations in exons 5–8 of the TP53 gene and in codons 12 and 13 of the K-ras gene were assayed in paraffin-embedded tissue by the single-strand conformation polymorphism (SSCP) assay. Nuclear p53 staining was found in 57 (47%) tumours. Aberrant migration patterns indicating mutation of the TP53 gene were found in 39 (32%) tumours. Forty-six carcinomas (38%) showed a mutation of the Ki-ras codons 12 or 13. In a univariate analysis, patients with wild-type TP53 status showed a trend towards better survival, compared with those with mutated TP53 (log-rank test, P=0.051). Likewise, tumours immunohistochemically positive for p53 showed a worse prognosis than p53-negative tumours (P=0.010). The presence or absence of mutations in Ki-ras did not correlate with prognosis (P=0.703). In multivariate analysis, only p53 immunoreactivity emerged as an independent marker for prognosis hazard ratio (HR)=2.16, 95% confidence interval (CI) 1.12–4.11, P=0.02). Assessment of p53 protein expression is more discriminative than TP53 mutation to predict the outcome of Dukes' stage B tumours and could be a useful tool to identify patients who might benefit from adjuvant therapy.

Introduction

The clinical behaviour of colorectal cancer is highly variable. Amongst various clinicopathological parameters such as histological degree of differentiation, vascular invasion, mucin production, size and location of the tumour, tumour stage is the most significant prognostic indicator [1]. However, there is important heterogeneity within a single stage category since approximately 1/3 of patients with Dukes' stage B and 2/3 of patients with Dukes' stage C will die of recurrent disease [2]. Approximately one-third of all colorectal tumours are Dukes' stage B and as yet, there are no parameters available allowing the subdivision of this group of patients into more accurate prognostic strata and no marker to reliably predict which patients are likely to benefit from adjuvant therapy.

It is now widely accepted that colonic carcinogenesis is a multistep process in which multiple mutational events, including activation of Ki-ras, inactivation of adenomatous polyposis coli (APC), deleted in colorectal cancer (DCC), TP53 and inactivation of DNA mismatch repair genes are implicated.

Ki-ras mutations occur early in colorectal carcinogenesis, mostly before adenoma formation, whereas TP53 mutations are implicated in the progression from adenoma to carcinoma. In colorectal cancers, the TP53 gene often shows missense mutations in one allele, accompanied by loss of the wild-type allele [3]. These missense mutations result in mutant p53 proteins with a prolonged half-life, which renders them detectable by immunohistochemical analysis 3, 4, 5. For as much as mutational events are the driving force in the process of colonic carcinogenesis, the hypothesis that the profile of mutations in a particular cancer would predict its behaviour seems justified. However, since the reported results of such studies are very contradictory, the impact of cancer associated gene mutations on the prognosis of colorectal carcinoma is still unresolved. Indeed, some studies showed that TP53 aberrations and Ki-ras mutations were not correlated with prognosis [6] whereas other studies indicated that mutations of these genes correlated with poor prognosis 7, 8, 9, 10. In contrast, Dix and colleagues observed that patients whose tumours showed p53 protein expression by immunohistochemistry had a significantly better prognosis than those whose tumours had no p53 immunoreactivity [4]. Part of this controversy might stem from the fact that in most reported series rather heterogeneous patient groups were studied. We, therefore, examined a large retrospective but clinically and therapeutically homogeneous, series of primary Dukes' stage B colorectal carcinomas, addressing the question of whether Ki-ras mutations and/or TP53 aberrations, as detected by molecular genetic analysis and immunohistochemistry, have any prognostic value in this group of patients.