Abstract
Background: We previously found foci of p53 up-regulation in dysplasia in colorectal adenomas (CRAs). The present study aimed at exploring the frequency of this phenomenon in CRAs with and without submucosal invasive carcinoma. Materials and Methods: Sections from 568 polypectomies or surgical resections harbouring a CRA (without or with submucosal invasion) or overt colorectal carcinomas were challenged with p53 immunostaining. The largest section from single colorectal neoplasias was measured by the aid of a calibrated ocular scale in a conventional microscope. Lesions were divided into small adenomas (≤10 mm in size), large adenomas (≥11 mm in size), adenomas with submucosal invasion, and overt invasive carcinomas (without any recognizable adenoma remnant tissue). Results: CRAs with three or more dysplastic foci of p53-up-regulation gradually increased from 8% in small adenomas (size: ≤10 mm) to 48% in large adenomas (size: ≥11 mm), and to 65% in the adenomatous tissue in adenomas displaying submucosal invasion), but plummeted to 13% in the submucosal carcinomatous tissue and to 11% in overt carcinomas. In contrast, extensive p53 up-regulation predominated in the submucosal carcinomatous tissue (87%) and in overt carcinomas (89%). Conclusion: The frequency of foci of dysplastic glands with up-regulation of p53 (hotspots) gradually increased from small to larger CRAs, being highest in the adenomatous tissue of CRAs with submucosal invasive carcinoma. The foci of p53 up-regulation became confluent (appreciated as extensive up-regulation) in the submucosal carcinomatous tissue and in overt carcinomas. It is concluded that a high number of foci with p53 up-regulation in adenomatous tissue might be required before submucosal invasive carcinoma ensues.
Colorectal adenomas (CRAs) are foci of highly proliferating dysplastic cells with biochemical and molecular alterations carrying a particular propensity to progress to colorectal carcinoma (CRC) (1). The colorectal adenoma–carcinoma sequence embodies a well-known paradigm for sequential cancer development driven by the accumulation of genomic mutations. Years ago Vogelstein et al. proposed a multi-step molecular paradigm of colorectal tumorogenesis, whereby, following Adenomatous polyposis coli (APC) (5q21) inactivation, accumulation of β-catenin in the nucleus, hypomethylation and V-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations, tubular adenomas with low-grade dysplasia progressed to villous adenomas with high-grade dysplasia (2). Allelic loss of the chromosome 17p region containing the p53 gene in conjunction with mutation of the second allele would give rise to bi-allelic inactivation. This functional loss of p53 in CRC was proposed to be a late event in the transition from adenoma to carcinoma.
The p53 transcription factor (encoded by the human gene TP53) is a key tumor suppressor, as it regulates several signaling pathways involved in carcinogenesis (3). The wild-type TP53 gene hosts the vital anti-proliferative function that empowers cells to arrest their proliferation cycle in G1, aiming to repair the DNA of highly damaged cells and to prevent the expansion of potential cancer clones. Cells that fail to repair their DNA undergo apoptosis, but when the wild-type p53 is inactivated by the mutant p53, this mechanism collapses (4).
Mutations in TP53 have been found in 50% of patients with colonic cancer (5). Given the logistical difficulties and resources associated with direct sequencing of the TP53 gene, most investigations have used immunohistochemistry as a means of detecting mutant p53, with the assumption that overexpression of p53 is often associated with a mutation, while the lack of expression is generally indicative of wild-type p53 (6).
According to the prevailing theory, most CRCs arise from adenomas. Nonetheless, histological studies have demonstrated that the majority of the CRAs are indolent neoplasias in which invasive carcinoma does not occur. In fact, only 1% to 3% out of endoscopically-resected CRAs of all sizes revealed invasive carcinoma at histological examination (7). In contrast, the study of 926 CRAs demonstrated in several surveys (8-17) that p53-up-regulation occurs in up to 64% of the cases (8). Hence, despite extensive studies of the TP53 gene and of its encoded protein p53 in colorectal carcinogenesis, our understanding of why only occasional indolent CRAs are selected to progress to invasive carcinoma remains elusive.
In a preliminary study, we found haphazardly distributed foci of p53 up-regulation in CRAs (18). It was speculated that these foci could highlight autonomous groups of dysplastic glands supposedly committed to cancer progression.
The purpose of the present work was to explore whether the frequency of foci of p53 up-regulation correlated with the presence of submucosal invasive carcinoma in a large cohort of CRAs.
Materials and Methods
The study comprised of 568 non-consecutive polypectomies or surgical resections harbouring a CRA with or without submucosal invasion, as well as overt CRCs.
The Institutional Review Board approved this retrospective study. No individual patient identification was involved and no study-driven clinical intervention was performed.
CRAs with invasive carcinoma were regarded as those exhibiting neoplastic cells or glands invading through the muscularis mucosae into the submucosal tissues, without reaching the confines of the margin of resection. In CRAs with invasive carcinoma, the term adenomatous tissue is used to define the adenomatous area; the area with invasive carcinoma was separately analysed, appearing as such in results and Tables.
Overt carcinomas were regarded those invasive carcinomas without any recognizable adenoma remnant tissue (19).
Sections were challenged with the primary mouse monoclonal antibody (IgG1, kappa) directed against human p53 protein (anti-p53, DO-7; Ventana Medical System, Inc, Roche, Switzerland).
Histological evaluation. CRAs were classified according to their structural configuration into tubular adenomas (TA), villous adenomas (VA), and traditional serrated adenomas (TSA), and according to their degree of cellular dysplasia into: low-grade dysplasia (LGD), high-grade dysplasia (HGD) (20-22).
Size assessment. Previous investigations revealed that polyp size assessment by endoscopists, radiologists and pathologists was unreliable when compared to the gold standard size (23-25). To by-pass this problem, the largest section from single colorectal neoplasias was measured by the aid of a calibrated ocular scale in a conventional microscope. Lesions were divided into small adenomas (≤10 mm in size), large adenomas (≥11 mm in size), adenomas with submucosal invasion, and overt invasive carcinomas (without any recognizable adenoma remnant tissue).
To illustrate this work, the largest section exhibiting p53-up-regulation was scanned in an EPSON Perfection 4990 (Epson America, Inc., Long Beach, CA, USA), together with a sidewise placed conventional ruler. Hence, the illustrations presented in Figures 1, 2, 3, 4, 5, 6, and 7, provide the reader with the actual size of the lesion scanned.
Scoring p53 expression. Dysplastic cells or tumour cells were regarded as overexpressing the p53 protein when strongly immunostained.
p53-up-regulation was found either focally or extensively distributed. Focal p53 was present as a single assemblage or focus of dysplastic glands, or as several foci (assemblages) of dysplastic glands. Extensive p53 up-regulation was noted when present in the entire adenomatous tissue, in the entire submucosal invasive carcinoma, or in the entire cancerous tissue in overt carcinomas.
Statistical analysis. Kruskal–Wallis one-way analysis of variance was used to compare difference between groups. Statistical significance was defined as p<0.05.
Results
Size of CRAs. Out of the 568 neoplastic lesions investigated, 38% (n=215) were small adenomas (mean= 8 mm, range=4-10 mm), 38% (n=218) large adenomas (mean= 18 mm, range=11-28 mm), 12% (n=68) were adenomas displaying submucosal invasion (mean=16 mm, range=8-30 mm), the remaining 12% (n=67), being overt carcinomas. Out of the 68 adenomas with submucosal invasion, 88.2% (n=60) measured ≥11 mm (mean=17 mm, range=12-27 mm).
Histological phenotype of CRAs. Out of the 568 neoplastic lesions reported in Table I, 88% (n=501) were CRAs. Histology revealed that the majority of small adenomas (58%) were TA with LGD, whereas the majority of large adenomas (51%) and adenomas with submucosal invasion (57%) were VA with HGD.
Frequency of p53 up-regulation in different colorectal neoplastic lesions. From Table II, it may be deduced that the p53 protein was up-regulated in 28% (141/501) of the adenomas/adenomatous tissue, and in 51% (69/135) of the invasive carcinomas, either in the submucosa or in overt carcinomas; (p53 up-regulation in CRAs vs. carcinomas, p<0.05).
In small adenomas, p53 up-regulation was recorded in 6%, in 45% of large adenomas, and in 46% in the adenomatous tissue (Table II) (p53 up-regulation small adenomas vs. large adenomas/adenomatous tissue, p<0.05). p53 was found up-regulated in 34% of the submucosal invasive carcinomas, and in 69% of the overt carcinomas (p<0.05).
Distribution of p53 up-regulation in colorectal neoplastic lesions. i) Focal p53 up-regulation (Figures 1, 2, 3, 4, 5): Out of the 210 lesions exhibiting p53 up-regulation reported in Table III, focal up-regulation was found in 57% (n=119): in 83% of small adenomas, in 80% of large adenomas, and in 74% in the adenomatous tissue, but only in 13% in submucosal carcinomas and 11% in overt carcinomas (focal p53-up-regulation in adenomas/adenomatous tissue vs. submucosal carcinoma/overt carcinomas, p≤0.05).
Colonic adenoma with one focus of p53 up-regulation (arrow). Endoscopic polypectomy, p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Rectal adenoma with two foci of p53 up-regulation (arrows). Endoscopic polypectomy, p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Multiple foci with p53 up-regulation: From Table IV it may be deduced that ≥3 up-regulated p53 foci were found in 8% (1/12) of the small adenomas, in 48% (47/98) of the large adenomas, in 65% (20/31) of the adenomatous tissue, but only in 13% (3/23) of the submucosal-invasive carcinomas and in 11% (5/46) of the overt carcinomas, (≥3 p53 up-regulated foci in large adenomas/ adenomatous tissue vs. small adenomas/submucosal invasion/ overt carcinomas, p≤0.05).
Colonic adenoma with three foci of p53 up-regulation (arrows). Endoscopic polypectomy, p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Colonic adenoma with four foci of p53 up-regulation (arrows). Endoscopic polypectomy. p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
ii) Extensive p53 up-regulation (Figures 6 and 7). Out of the 210 lesions exhibiting p53 up-regulation in Table V, extensive p53 up-regulation was found in 43% (n=91): in 21% (30/141) of small and large adenomas/adenomatous tissue, and in 88% (59/67) of submucosal invasion/overt carcinomas (p53 up-regulation in CRAs vs. invasive carcinomas, p<0.05). This difference was retained when the various groups were separately analyzed in Table VI; extensive p53 up-regulation was recorded in 17% of small adenomas, in 20% of large adenomas, in 26% of the adenomatous tissue, but in 86% in submucosal invasion and in 89% in overt carcinomas (p<0.05).
Colonic adenoma with multiple foci of p53 up-regulation with submucosal carcinoma (arrow). Endoscopic polypectomy. p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Colonic adenoma with confluent foci of p53 up-regulation with carcinoma invading into the muscularis propria (arrow). Surgical specimen. p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Overt carcinoma of the colon with extensive p53-up-regulation. Surgical specimen. p53 immunostain. Section was scanned with EPSON Perfection 4990; scale in mm.
Histology in 568 colorectal neoplastic lesions: 501 colorectal adenomas (CRAs) and 67 overt invasive carcinomas.
Frequency of p53 up-regulation in 568 colorectal neoplastic lesions.
Focal p53 up-regulation in 210 colorectal neoplastic lesions exhibiting p53 up-regulation.
Discussion
The present investigation showed that the frequency of dysplastic glands with foci of p53 up-regulation significantly increased from 6% in small adenomas, to 36% in large adenomas and to 69% in the adenomatous tissue. Pari passu, the frequency of neoplastic lesions with three or more dysplastic foci of p53 up-regulation (hotspot mutations) gradually increased from 8% in small adenomas, to 48% in large adenomas, and to 65% in the adenomatous tissue. Thus, the highest proportion of dysplastic cells with p53 up-regulation was recorded in the adenomatous tissue synchronously exhibiting submucosal invasion. In light of these findings, we are prone to speculate that p53 up-regulation might had reached “ the point of no-return” in the adenomatous tissue, an event that might have sent signals aimed to trigger tumour penetration into the submucosal tissues. In contrast, extensive p53 up-regulation predominated in tissues with invasion: in 87% with submucosal invasion and in 89% of overt carcinomas, and focal p53 up-regulation plummeted to 13% and to 11%, respectively. The rational explanation for this phenomenon might be that the foci of p53 up-regulation became confluent (extensive) in invasive carcinomas.
The number of foci with p53 up-regulation in 210 colorectal neoplastic lesions exhibiting p53 up-regulation.
Extensive p53 up-regulation in 210 colorectal neoplastic lesions exhibiting p53 up-regulation
Several authors studied the frequency of p53 up-regulation in CRAs (cfr. Table VI). The Table shows that that the percentage of CRAs with p53 up-regulation varied from 2% in initial adenomas (15) to 64% (8). For some authors, p53 staining was recorded as positive irrespective of the percentage of positive glands (15), whereas for others, p53 staining above 10% was considered positive (9). It is apparent that the various criteria used to define p53 up-regulation in CRAs might have influenced the disparate results in the literature.
Publications on p53 up-regulation in 926 colorectal neoplastic lesions.
According to Rivlin et al., mutations of the TP53 gene occurring at the early stages of tumourigenesis fuel uncontrolled proliferation, whereas TP53 mutations at later stages, synergizing with additional oncogenic events, trigger invasion (3). Our findings suggest that accumulation of p53 up-regulated foci of dysplastic cells in CRAs should reach a substantial level before invasion ensues.
Puzzlingly, two cases having three or more foci of p53 up-regulation in the adenomatous tissue showed absence of p53 immunoreactivity in the submucosal invasive tissue. Although the cause of this conundrum remains unclear, the possibility that p53 protein expression might have been down-regulated in the invasive tissue by one or more molecular signals cannot be totally rejected. One possible candidate for this down-regulation might be the double minute 2 homolog (MDM2) also known as E3 ubiquitin-protein ligase Mdm2 that plays a key role in regulating p53 protein (26). MDM2 binds to p53 within the p53 transactivation domain, interfering with the enrolment of transcription components. MDM2 binding can actively repress transcription when bound to p53 and through proteolytic degradation may completely eliminate p53 (27, 28). Another candidate is the Transcriptional repressor CTCF also known as 11-zinc finger protein or CCCTC-binding factor, a zinc finger protein that hampers the TP53 promoter from accumulating repressive histone marks (29). When the TP53 promoter stores such histone marks, however, the CTCF protein binds the p53, resulting in the decrease of p53 expression (29). More recently, Suzuki H et al. found in CRCs that the epigenetic inactivation by methylation of insulin-like growth factor-binding protein 7 (IGFBP7) correlated with the serine/threonine-protein kinase B-Raf mutations, the presence of the CpG island methylator phenotype (CIMP) and the absence of TP53 mutations (4). Accordingly, one or more of these candidate signals might account for the absence of p53 up-regulation in the invasion tissue in the two CRAs exhibiting p53 up-regulation in the adenomatous tissue.
In the present survey, some CRAs exhibiting foci of p53 up-regulation in dysplastic glands with HGD occasionally exhibited p53 up-regulation in glands with LGD. The latter finding suggests that p53 up-regulation might not be exclusively linked to the most severe degree of dysplastic aberration, but possibly to particular putative cellular receptor(s) present in autonomous groups of dysplastic cells.
In sum, foci of p53 up-regulation (hot spots) in dysplastic glands gradually increased, from small to large adenomas, being highest in the adenomatous tissue in adenomas with submucosal invasive carcinoma. In carcinomas (submucosal and overt), the foci of p53 up-regulation became confluent (appreciated as extensive p53-up-regulation).
It is submitted that a high number of foci with p53 up-regulation might be necessary in CRAs before submucosal invasive carcinoma ensues. Quantification of the number of such hotspots might be a useful biological marker for assessing cancer risk in CRA.
The review of the literature indicates that this is the first study showing that multiplicity of foci with p53 up-regulation in CRAs might be required before submucosal invasive carcinoma develops.
Acknowledgements
The Authors would like to thank Dr. Lorand Kis for valuable comments and constructive criticisms of the manuscript.
Footnotes
-
Conflicts of Interest
The Authors state that they have no conflict of interest.
- Received July 31, 2014.
- Revision received September 9, 2014.
- Accepted September 15, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
