p53 expression in patients with ulcerative colitis - associated with dysplasia and carcinoma: a systematic meta-analysis

As one common pathological type of inflammatory bowel diseases (IBD), ulcerative colitis (UC) is characterized by a relapsing, idiopathic, and chronic inflammatory disease which usually affects the entire colon and rectum [1, 2]. The prevalence of this disease has been rising in the world, patients with UC are correlated with an increased risk of developing colorectal cancer (CRC) [3, 4].

Although the exact mechanism of UC is not fully understood, increasing evidence reports that genetic and environmental factors are important in the pathogenesis of UC [5‐7]. Given the increased risk of CRC in UC, dysplasia is precancerous lesions, it is essential to clinical screening of UC patients with dysplasia [8]. The p53, a key tumor-suppressor gene (TSG), is mapped to the short arm of chromosome 17 (17p13) [9]. p53 gene encodes the p53 protein and is responsible for the regulation of cell cycle, DNA repair, and apoptosis [10, 11]. The p53 mutation is the most common event in human carcinomas, due to the accumulation of mutant p53, p53 protein often shows nuclear staining, which makes it easily detected using immunohistochemistry (IHC) method [12, 13]. The expression of p53 protein can be measured by IHC in different histological types of UC [14, 15].

Although Du et al. reported the association between KRAS and TP53 mutations with IBD-associated colorectal cancer [16], the correlation between p53 expression and the risk of dysplasia and cancer in patients with UC remained to be elucidated. Because the sample size of an individual study was small in UC, moreover, there were some inconsistent results with respect to p53 expression in UC. For example, Li 2004 et al. reported that no immunoreactivity for p53 expression was found in UC and normal tissue samples [17]. Klump 1997 et al. reported that p53 expression was noted in UC without dysplasia/carcinoma and UC with dysplasia [15]. Thus, we first conducted this meta-analysis involving more eligible articles to determine the correlation between p53 expression and different histological types of UC, including normal tissue samples, UC without dysplasia/carcinoma, UC with dysplasia, and UC with CRC.

TSG p53 expression has been frequently reported in human malignancies [38, 39]. Some studies suggest that positive expression of the p53 gene is linked to the development and progression of tumors [40‐42]. Immunohistochemical analysis of p53 protein expression is also found in patients with UC [32, 34]. However, the results regarding p53 expression in different pathological types of UC are still inconsistent and conflicting. The present study was carried out to analyze whether p53 expression was associated with the increased risk of UC with dysplasia and carcinoma.

p53 gene showed different expression levels in UC-associated different pathological types, ranging from 0% [14, 17, 23, 33] to 40% [37] in normal tissue samples, and ranging from 0% [17, 24, 29, 36] to 88.5% [37] in UC without dysplasia and carcinoma. The expression of the p53 gene had different frequencies in UC with dysplasia or carcinoma, with a range from 20% [34] to 90% [27] in UC with dysplasia, and with a range from 21.4% [35] to 100% [15] in UC with CRC. This study integrated all available studies involving a large population showed that immunohistochemical detection of p53 protein expression was significantly higher in UC with CRC than in UC with dysplasia, higher in UC with dysplasia than in UC without dysplasia and carcinoma, and higher in UC without dysplasia and carcinoma than in normal tissue samples, which suggested that p53 expression was closely linked to UC-CRC carcinogenesis and the progression of UC with dysplasia. In addition, a slight heterogeneity was detected in UC with dysplasia versus UC without dysplasia and carcinoma (P = 0.027 < 0.1). Thus, we conducted a sensitivity analysis to assess the change of the pooled OR and heterogeneity based on the omission of one study. We removed one study [15], and a re-calculated OR was not significantly changed, with no evidence of heterogeneity (P = 0.181), suggesting the stability of the current result. The reasons of heterogeneity from this study and other studies were not clear, perhaps due to the use of inappropriate or different conditions in IHC method, which may cause an observed bias.

Subgroup analysis of ethnicity (the Asian and Caucasian populations) was performed to find different correlation among different subgroups. When UC without dysplasia and carcinoma was compared to normal tissue samples, the result of subgroup analysis revealed that only Asians were susceptible to p53 expression. When UC with dysplasia was compared to UC without dysplasia and carcinoma, the result showed that the Asian and Caucasian populations were susceptible to p53 expression. When UC with carcinoma was compared to UC with dysplasia, the result demonstrated that Asians and Caucasians were not susceptible to p53 expression.

Some limitations should be noted in the present meta-analysis. First, although a range of online electronic databases was systematically searched to identify available publications. Egger’s test demonstrated a slight publication bias in UC with dysplasia vs. UC without dysplasia and carcinoma. Only papers published in English or Chinese were included in this meta-analysis. The unpublished articles or conference abstracts were excluded because of insufficient information. Articles with positive conclusions were usually published than these articles with negative conclusions. Second, the main ethnic groups consisted of Asians and Caucasians in this study, but other study populations, such as African population, were lacking. Finally, three eligible studies reported the data between p53 expression and dysplasia and cancer in patients with Crohn’s disease (CD) [37, 43, 44], more studies with large sample sizes are essential to further perform a meta-analysis to assess the correlation of p53 expression in CD in the future.

This study suggests that p53 gene has a notably higher expression frequency in UC-CRC than in UC with dysplasia, higher expression level in UC with dysplasia than in UC without dysplasia and carcinoma, and higher expression rate in UC without dysplasia and carcinoma than in normal tissue samples. Further large-scale clinical trials with large sample sizes are needed to validate our conclusions in the future.