Expression of p53 in colorectal cancer and dysplasia complicating ulcerative colitis

Artificial Intelligence Program to Predict p53 Mutations in Ulcerative Colitis-Associated Cancer or Dysplasia

BACKGROUND

The diagnosis of colitis-associated cancer or dysplasia is important in the treatment of ulcerative colitis. Immunohistochemistry of p53 along with hematoxylin and eosin (H&E) staining is conventionally used to accurately diagnose the pathological conditions. However, evaluation of p53 immunohistochemistry in all biopsied specimens is expensive and time-consuming for pathologists. In this study, we aimed to develop an artificial intelligence program using a deep learning algorithm to investigate and predict p53 immunohistochemical staining from H&E-stained slides.

METHODS

We cropped 25 849 patches from whole-slide images of H&E-stained slides with the corresponding p53-stained slides. These slides were prepared from samples of 12 patients with colitis-associated neoplasia who underwent total colectomy. We annotated all glands in the whole-slide images of the H&E-stained slides and grouped them into 3 classes: p53 positive, p53 negative, and p53 null. We used 80% of the patches for training a convolutional neural network (CNN), 10% for validation, and 10% for final testing.

RESULTS

The trained CNN glands were classified into 2 or 3 classes according to p53 positivity, with a mean average precision of 0.731 to 0.754. The accuracy, sensitivity (recall), specificity, positive predictive value (precision), and F-measure of the prediction of p53 immunohistochemical staining of the glands detected by the trained CNN were 0.86 to 0.91, 0.73 to 0.83, 0.91 to 0.92, 0.82 to 0.89, and 0.77 to 0.86, respectively.

CONCLUSIONS

Our trained CNN can be used as a reasonable alternative to conventional p53 immunohistochemical staining in the pathological diagnosis of colitis-associated neoplasia, which is accurate, saves time, and is cost-effective.

PHLDA1 expression in ulcerative colitis: A potential role in the management of dysplasia

Pleckstrin homology-like domain, family A, member 1 (PHLDA1) is a protein involved in cell proliferation, adhesion and migration in colon cancer. In normal large intestinal mucosa, this protein is expressed only in the crypts. By contrast, its expression in adenomas and cancers of the large intestine is spread throughout the glandular ducts, and it has been reported that PHLDA1 may be involved in the process of carcinogenesis. PHLDA1 may also be involved in the pathogenesis of ulcerative colitis (UC). The expression levels of PHLDA1 in tissues from patients with UC were analyzed using immunohistochemistry, and its relationship with the development of UC-associated colorectal cancer (UC-CRC) was examined. Overall, tissue samples from 143 lesions (90 colitis lesions, 39 dysplastic lesions and 14 UC-CRC lesions) were prepared from excised specimens of 49 patients with UC who underwent surgery in Tokyo Medical and Dental University Hospital between January 2004 and December 2017. Subsequently, immunostaining for PHLDA1 was performed. PHLDA1 expression was evaluated in UC-CRC and dysplastic tissues within the entire lesion area on the slide and in colitis over the area of the accompanying duct. The cytoplasmic staining intensity was classified into four levels, and the expression score (0-2 points) was calculated. The median PHLDA1 expression score was 0.295 for colitis, 0.607 for dysplasia and 0.865 for UC-CRC. The dysplasia expression score was significantly higher than the colitis score (P<0.001), while the UC-CRC expression score was significantly higher than the dysplasia score (P=0.003). The expression levels of PHLDA1 in UC cases were higher in colitis, followed by dysplasia and UC-CRC, which suggested that this protein may be involved in the carcinogenesis of UC-CRC. In addition, PHLDA1 immunostaining may help in the diagnosis of dysplasia, which is a type of precancerous lesion.

Sporadic adenoma or ulcerative colitis associated neoplasia? The endoscopist's information has an impact on diagnosis and patient management

BACKGROUND

Diagnosing low grade intraepithelial neoplasia (LGIN) in patients with ulcerative colitis (UC) is difficult. Distinguishing between sporadic adenoma (SA) and UC associated LGIN is even more challenging but has clinical impact. We aimed to examine, if the morphological distinction between both entities is reliably possible, how it influences patient's outcome and the role of the endoscopist in this decision with respect to current endoscopy classification schemes.

METHODS

Seven pathologists retrospectively reevaluated 425 cases of LGIN in UC patients, diagnosed between 2009 and 2017 with preceding expert consensus and follow up in two separate readings, based on published morphological differentiation criteria. In the first evaluation, the observers were blinded to any clinical data. In the second evaluation, they knew patients' age as well as endoscopic features. They also rated their subjective diagnostic certainty.

RESULTS

Diagnostic correctness improved significantly in the second assessment as did the pathologists' confidence in their diagnoses (p < 0.001 - p = 0.019). Knowledge of clinical and endoscopical data led to a higher percentage of SA (71.8% vs. 85.6%). UC associated LGIN showed significant earlier LGIN relapse as well as more high grade intraepithelial neoplasia and carcinoma during follow up (p < 0.001, p < 0.001, p = 0.005).

CONCLUSIONS

Distinction between SA and UC associated LGIN is important as it has an impact on patients' follow up and treatment. Morphological distinction remains difficult with moderate interobserver variability. Adequate clinical information significantly improves pathologists' diagnoses as well as their confidence in their diagnoses.

Expression of ATF6 as a marker of pre-cancerous atypical change in ulcerative colitis-associated colorectal cancer: a potential role in the management of dysplasia

BACKGROUND

Diagnosis of low-grade dysplasia (LGD) is important in the management of ulcerative colitis (UC), but it is often difficult to distinguish LGD from inflammatory regenerative epithelium. The unfolded protein response (UPR) is activated in inflammatory bowel disease and malignancies. We aimed to identify a UPR-related gene that is involved in the development of non-UC and UC-associated colorectal cancer (CRC), and to investigate whether the target gene is useful for the diagnosis of LGD.

METHODS

Using our microarray gene expression database of 152 CRCs, we identified activating transcription factor 6 (ATF6) as a target gene. Immunohistochemistry (IHC) of ATF6 were analyzed in 137 surgically resected CRCs, 95 endoscopically resected adenomas and pTis cancers, and 136 samples from 51 UC patients (93 colitis without neoplasia, 31 dysplasia, and 12 UC-associated CRC). The diagnostic accuracy of ATF6 and p53 as markers of LGD was assessed.

RESULTS

ATF6 expression was detectable in all CRCs but not in normal colonic mucosa, was elevated with increase in cellular atypia (adenoma with moderate atypia < severe atypia < pTis CRC, p < 0.001), and higher in dysplasia and CRC than in non-neoplastic colitis (p < 0.001). Notably, the difference between colitis and LGD was significant. Compared to p53-IHC, ATF6-IHC had better diagnostic accuracy for distinguishing LGD from background inflammatory mucosa (sensitivity 70.8 vs. 16.7%, specificity 78.5 vs.71.0%, respectively).

CONCLUSIONS

ATF6 was expressed in lesions undergoing pre-cancerous atypical change in both non-UC and UC-associated CRC and may be used to distinguish LGD from inflammatory regenerative epithelium in UC patients.

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

Background

Tumor suppressor gene p53 expression has been reported in patients with ulcerative colitis (UC). However, the correlation between p53 expression and UC remains controversial. The aim of this meta-analysis was to investigate the association between p53 expression and different pathological types of UC.

Methods

Publications were searched in the PubMed, Embase, EBSCO, Wangfang, and CNKI databases. The overall odds ratios (ORs) and their corresponding 95% confidence intervals (95% CIs) were summarized in this study.

Results

Final 19 papers were identified in this meta-analysis, including 1068 patients with UC and 130 normal tissue samples. Immunohistochemical p53 expression was significantly higher in UC without dysplasia and carcinoma (UC group) compared to normal tissue samples (OR = 3.14, P = 0.001), higher in UC with dysplasia than in UC group (OR = 10.76, P < 0.001), and higher in UC with colorectal cancer (CRC) than in UC with dysplasia (OR = 1.69, P = 0.035). Subgroup analysis of ethnicity (UC group vs. normal tissues) showed that p53 expression was correlated with UC in Asians, but not in Caucasians. When UC with dysplasia was compared to UC group, p53 expression was linked to UC with dysplasia among both Asians and Caucasians. When UC-CRC was compared to UC with dysplasia, p53 expression was not associated with UC-CRC in both Caucasians and Asians.

Conclusions

p53 expression was closely associated with UC-CRC development. p53 expression showed different ethnic characteristics among different pathological types of UC.

Electronic supplementary material

The online version of this article (10.1186/s12876-017-0665-y) contains supplementary material, which is available to authorized users.

KRAS and TP53 mutations in inflammatory bowel disease-associated colorectal cancer: a meta-analysis

Although KRAS and TP53 mutations are common in both inflammatory bowel disease-associated colorectal cancer (IBD-CRC) and sporadic colorectal cancer (S-CRC), molecular events leading to carcinogenesis may be different. Previous studies comparing the frequency of KRAS and TP53 mutations in IBD-CRC and S-CRC were inconsistent. We performed a meta-analysis to compare the presence of KRAS and TP53 mutations among patients with IBD-CRC, S-CRC, and IBD without dysplasia. A total of 19 publications (482 patients with IBD-CRC, 4,222 with S-CRC, 281 with IBD without dysplasia) met the study inclusion criteria. KRAS mutation was less frequent (RR=0.71, 95%CI 0.56-0.90; P=0.004) while TP53 mutation was more common (RR=1.24, 95%CI 1.10-1.39; P<0.001) in patients with IBD-CRC compared to S-CRC. Both KRAS (RR=3.09, 95%CI 1.47-6.51; P=0.003) and TP53 (RR=2.15, 95%CI 1.07-4.31 P=0.03) mutations were more prevalent in patients with IBD-CRC compared to IBD without dysplasia. In conclusion, IBD-CRC and S-CRC appear to have biologically different molecular pathways. TP53 appears to be more important than KRAS in IBD-CRC compared to S-CRC. Our findings suggest possible roles of TP53 and KRAS as biomarkers for cancer and dysplasia screening among patients with IBD and may also provide targeted therapy in patients with IBD-CRC.

Immunohistochemical assessment of a unique basal pattern of p53 expression in ulcerative-colitis-associated neoplasia using computer-assisted cytometry

Background

The basal pattern of p53 expression, defined as its immunoreactivity confined to the basal half of the glands, is associated with early neoplastic lesions in ulcerative colitis (UC). However, their clinical utility of this finding is limited by the use of “visual estimation” (approximate immunoreactivity on the basis of scanning the stained slide, without formal counting). This study was designed to analyze the basal pattern of p53 using computer-assisted cytometry and to identify the optimal cutoff value for discriminating between UC-associated early-stage neoplasia and regenerative atypia.

Methods

The specimens were obtained from eight UC patients undergoing colectomy and were classified according to the criteria by the Research Committee of Inflammatory Bowel Disease of the Ministry of Health and Welfare in Japan. Patients with classes UC-IIa (indefinite for dysplasia, probably regenerative), UC-IIb (indefinite for dysplasia, probably dysplastic), and UC-III (definitive dysplasia) were enrolled in the study. Based on the percentage of immunoreactive cells in the basal half of the crypt with visual estimation, basal positivity of p53 was classified into three categories: grade 1 (1 - 9%), grade 2 (10 - 19%), and grade 3 (≥20%). Next, crypts classified as grade 3 by visual estimation were analyzed by computer-assisted image analysis.

Results

Using visual estimation, grade-3 p53 basal positivity was observed in 46.0% of UC-IIa crypts (128 of 278), 61.9% of UC-IIb crypts (39 of 63), and 94.2% of UC-III crypts (81 of 86). Using image analysis, the median p53 basal positivities were 30.3% in UC-IIa, 52.3% in UC-IIb, and 65.4% in UC-III (P ≤0.002). A receiver operating characteristics curve was generated to determine the method’s diagnostic utility in differentiating UC-IIa from UC-III. In this cohort, the sensitivity was 0.78; the specificity was 0.98; the negative predictive value was 87.4%; the positive predictive value was 95.5%, and the accuracy was 90.2% with a cutoff value for p53 basal positivity of 46.1%.

Conclusions

Our findings indicate that assessing p53 basal positivity by image analysis with an optimal threshold represents an alternative to visual estimation for the accurate diagnosis of UC-associated early-stage neoplasia.

Virtual Slides

The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/3588120501252608

Stool DNA testing for the detection of colorectal neoplasia in patients with inflammatory bowel disease

BACKGROUND

Current approaches to the detection of colorectal neoplasia associated with inflammatory bowel disease (IBD-CRN) are suboptimal.

AIM

To test the feasibility of using stool assay of exfoliated DNA markers to detect IBD-CRN.

METHODS

This investigation comprised tissue and stool studies. In the tissue study, gene sequencing and methylation assays were performed on candidate genes using tissue DNA from 25 IBD-CRNs and from 25 IBD mucosae without CRN. Mutations on p53, APC, KRAS, BRAF or PIK3CA genes were insufficiently informative, but several aberrantly methylated genes were highly discriminant. In the stool study, we evaluated candidate methylated genes (vimentin, EYA4, BMP3, NDRG4) in a prospective blinded study on buffered stools from 19 cases with known IBD-CRN and 35 age- and sex-matched IBD controls without CRN. From stool-extracted DNA, target genes were assayed using quantitative allele-specific real-time target and signal amplification method.

RESULTS

IBD-CRN cases included 17 with ulcerative colitis (UC) and two with Crohn's disease (CD); nine had cancer and 10 had dysplasia. Controls included 25 with UC and 10 with CD. Individually, BMP3, vimentin, EYA4 and NDRG4 markers showed high discrimination in stools with respective areas under the ROC curve of 0.91, 0.91, 0.85 and 0.84 for total IBD-CRN and of 0.97, 0.97, 0.95 and 0.85 for cancer. At 89% specificity, the combination of BMP3 and mNDRG4 detected 9/9 (100%) of CRC and 80% of dysplasia, 4/4 (100%) of high grade and 4/6 (67%) of low grade.

CONCLUSION

These findings demonstrate the feasibility of stool DNA testing for non-invasive detection of colorectal neoplasia associated with inflammatory bowel disease.

Expression of activation-induced cytidine deaminase in ulcerative colitis-associated carcinogenesis

AIMS

Activation-induced cytidine deaminase (AID) is a DNA/RNA-editing enzyme that is essential for hypermutation and class-switch recombination in immunoglobulin genes. The aim of this study was to investigate the expression of AID and its association with p53 mutation in ulcerative colitis (UC)-associated carcinogenesis.

METHODS AND RESULTS

The expression of AID was examined in 25 patients with UC-associated neoplasia, 20 UC patients without neoplasia, 18 patients with non-inflamed colorectal mucosa unaffected by UC, and 19 patients with sporadic colorectal cancer, by immunohistochemistry and quantitative reverse transcription polymerase chain reaction analysis. Mutational analysis and immunohistochemistry for p53 were also performed. The degree of AID expression was not different between UC-associated neoplasia and sporadic colorectal cancer. However, AID was expressed in both UC-associated neoplasia and UC without neoplasia. Whereas AID expression in UC-associated neoplasia was not correlated with the grade of dysplasia, expression in non-neoplastic mucosa of UC was correlated with the histological grade of inflammation. In UC-associated neoplasia, there was no significant correlation between AID expression and p53 mutation.

CONCLUSIONS

AID is associated with inflammation in UC, whereas it may not specifically contribute to carcinogenesis in UC.

Outcome of sporadic adenomas and adenoma-like dysplasia in patients with ulcerative colitis undergoing polypectomy

BACKGROUND

Ulcerative colitis (UC) patients are at increased risk of colorectal dysplasia and cancer. Few studies have examined the clinical outcomes of dysplastic polyps resembling sporadic adenomas that are removed with endoscopic polypectomy.

METHODS

A centralized diagnostic index identified patients evaluated between 1994 and 2004 with UC and polypoid dysplasia who were followed from the time of polypectomy until the most recent colonoscopy. They were stratified into two groups by polyp occurrence, either within (adenoma-like dysplasia) or outside (sporadic adenoma) the most proximal endoscopic or histologic extent of colitis. The endpoints of interest were the development of subsequent colorectal neoplasia, flat dysplasia, or cancer. The cumulative probabilities of these endpoints were estimated using the Kaplan-Meier method, and the association with clinical factors assessed using Cox proportional hazards regression.

RESULTS

Ninety-five patients were found to have polypoid dysplasia; of these, 77 underwent polypectomy. The cumulative probability of subsequent colorectal neoplasia in polypectomy patients was 18% at 1 year and 69% at 5 years. After polypectomy, cumulative incidence of cancer or flat dysplasia was 2% at 1 year and 13% at 5 years. The proportional hazards models indicated that these outcomes were not significantly associated with polyp type, primary sclerosing cholangitis, family history of colorectal cancer, 5-aminosalicylate use, extent of colitis, or duration of disease.

CONCLUSIONS

While polypectomy may be safe for the management of adenomas occurring in most UC patients, the 5-year cumulative incidence of a combined endpoint (cancer or flat dysplasia) was 13%. Such patients should be followed closely.

The value of histological changes and immunohistochemical markers Ki67 and p53 in the assessment of ulcerative colitis related dysplasia

The risk of carcinoma increases in patients with a 10-year or longer duration of ulcerative colitis (UC). To search for a more objective parameter to assess epithelial dysplasia. The study comprised 25 cases of longstanding UC: 7 cases with regenerative atypia, 7 with low grade dysplasia, 7 with high grade dysplasia, and 4 cases indefinite for dysplasia. The colonic biopsies obtained during endoscopy were stained with H&E to identify the aforementioned categories. Seventy-five sections from biopsy specimens were stained immunohistochemically to detect differences in the frequency and pattern of nuclei positive for the proliferation marker Ki67 and p53. In high grade dysplasia, the distribution of Ki67 positive cells was diffuse throughout the full length of the crypt, whereas low grade dysplasia and epithelium indefinite for dysplasia, as well as regenerative epithelium, showed an expanded basal zone. None of the regenerative atypia cases showed strong intensity p53 staining compared to dysplasia cases. None of the high grade dysplasia cases showed restricted p53 staining to the lower two thirds of the crypt. All the cases of HGD showed extension of Ki67 and p53 staining above the basal two thirds of the crypt. Ki67 and p53 immunostained cell assessment combined with routine histological evaluation of colorectal mucosa can improve the diagnostic accuracy, as well as the assessment of malignant transformation risk.

Ulcerative colitis and cancer

Patients with ulcerative colitis (UC) are at an increased risk for the development of colorectal cancer (CRC). Unlike sporadic CRC, the cancer in UC patients arises from a focal or multifocal dysplastic mucosa in areas of inflammation. The clinical features of UC-associated cancer are similar to those found in patients with hereditary non-polyposis colorectal cancer. As with other varieties of CRC, UC-associated cancer exhibits a variety of genetic and molecular changes/abnormalities. These abnormalities are however clustered in areas of mucosae with histological abnormalities. The magnitude and timing of these changes are however significantly different. Surveillance and identification of patients at risk for cancer are a challenging problem.

p53 mutations are associated with dysplasia and progression of dysplasia in patients with Crohn's disease

BACKGROUND

Mutations in the tumor suppressor gene p53 are associated with neoplasia in ulcerative colitis, but little is understood of their significance in Crohn's disease (CD).

PURPOSE

To explore p53 expression as a marker of neoplasia in CD patients.

METHODS

This is a retrospective review of CD patients who underwent p53 IHC staining in our center between 1995 and 2003. The p53 status was correlated to the presence and grade of neoplasia at the time of staining and in subsequent follow-up.

RESULTS

Fourteen CD patients had p53 assessment: eight were p53 positive and six were p53 negative. Seven of eight p53+ had dysplasia (six LGD, one HGD); one of six p53-had dysplasia (LGD) (P = 0.03). Four p53+ patients with follow-up had persistent dysplasia and two had progression to a higher grade. Three p53- patients with follow-up remained free of dysplasia.

CONCLUSIONS

This limited study shows that p53 over expression in CD patients is associated with dysplasia that may progress to a higher grade of neoplasia over time.

Long-term cancer risk of the anorectal cuff following restorative proctocolectomy assessed by p53 expression and cuff dysplasia

OBJECTIVE

Restorative proctocolectomy (RP) for ulcerative colitis (UC) retains a 'cuff' of columnar rectal epithelium that has unknown risk of malignant change. Markers of malignant potential in UC include aberrant p53 expression and dysplasia. We undertook a prospective study comprising serial surveillance biopsy and assessed the occurrence of aberrant p53 expression, epithelial dysplasia and carcinoma in the retained anorectal cuff following stapled RP.

METHOD

A total of 110 patients who underwent stapled RP for UC between 1988 and 1998 were followed up by cuff surveillance biopsies under general anaesthesia. Histological samples were analysed by a specialist colorectal pathologist for the presence of rectal mucosa, dysplasia and carcinoma. Immunohistochemistry for p53 expression was performed for each most recent cuff biopsy. Median follow-up was 56 months (12-145) and median time since diagnosis of UC was 8.8 years (2-32).

RESULTS

Rectal mucosa was obtained from the cuff in 65% of biopsies. No overt carcinomas developed during the follow-up period and there was no dysplasia or carcinoma in any cuff biopsy. The p53 overexpression was identified in 38 specimens (50.6%), but was also identified in controls (3/3 colitis, 3/3 ileal pouch and 6/6 stapled haemorrhoidectomy donuts).

CONCLUSION

The lack of carcinoma and dysplasia in the columnar cuff epithelium specimens is reassuring. The lack of stabilized p53 and absence of a relationship between p53 stabilization and dysplasia up to 12 years after pouch formation suggests neoplastic transformation is a rare event. Furthermore, p53 expression was not useful in surveillance of cuff biopsies from patients who have undergone RP for UC and the search should continue for alternative predysplastic markers. These data suggest that in patients who do not have high-grade dysplasia or colorectal cancer at the time of RP, cuff surveillance in the first decade after pouch formation is unnecessary. However, we consider regular cuff surveillance biopsies should continue for patients with high-grade dysplasia, whether or not there was a carcinoma in the original colectomy specimen.

Abnormal p53 immunohistochemistry is associated with an increased colorectal cancer-related mortality in patients with ulcerative colitis

OBJECTIVE

Immunohistochemical evaluation of p53 staining patterns has been considered as a complementary test for dysplasia in ulcerative colitis-related colorectal cancer surveillance, but usefulness would be particularly important if it were a marker associated with a poor prognosis. The aim of this study was to determine whether abnormal p53 staining of the tumor correlated with cancer-related mortality in ulcerative colitis patients.

METHODS

An historical cohort study was designed to examine all ulcerative colitis patients who developed colorectal cancer between 1978 and 1997 and who had tumor tissue blocks available for staining. Tissue was recut and stained for abnormal p53 immunohistochemistry using the DO-7 antibody. Tumors were considered to be p53 positive if at least 5% of nuclei in a high power field were positive for staining.

RESULTS

Among 75 patients entered in the study, 38 (50.7%) had p53 positive tumors. Fourteen patients (36.8%) with p53 positive tumors died from colorectal cancer, compared to five (13.5%) with p53 negative tumors (p < 0.04, log-rank test). The adjusted relative risk of cancer-related death among patients with p53 positive tumors was 3.03 (95% CI = 1.05-8.73).

CONCLUSIONS

Abnormal p53 immunohistochemistry of tumors is associated with a poor prognosis among ulcerative colitis patients who develop colorectal cancer. As such, p53 immunohistochemical staining could be a useful histological marker to complement routine histology in cancer surveillance programs in ulcerative colitis patients.

Screening for p53 and K-ras mutations in whole-gut lavage in chronic inflammatory bowel disease

OBJECTIVES

Molecular screening for frequently mutated genes may increase the likelihood of identifying cancer risk groups, such as patients with longstanding inflammatory bowel disease. This study investigated the prevalence and time course of p53 and K-ras mutations in colonic lavage fluid of patients with inflammatory bowel disease.

METHODS

Colonic lavage fluid from 190 patients with ulcerative colitis (73), Crohn's disease (58) or controls (49 non-tumour, 10 colorectal cancer) was studied by oligomer-specific hybridization for K-ras mutations and single-strand conformation polymorphism (SSCP) for p53 mutations. Follow-up investigations were carried out after 1-3 years.

RESULTS

Mutations were most frequent in carcinomas (5/10, 50%) and rare in non-tumour controls (1/49, 2.0%). They were found in Crohn's colitis in 15.4%, in extensive ulcerative colitis in 18.6%, in left-sided ulcerative colitis in 13.3%, and in distal ulcerative colitis in 6.7% (P > 0.05). There was a positive association with disease duration (> or =11 years, P < 0.05). Follow-up investigations detected the same mutation in four patients and revealed new mutations in three patients.

CONCLUSIONS

In our large series of patients with inflammatory bowel disease, K-ras and p53 mutations could be detected with reasonable frequency and confirmed at follow-up in at least some patients. Our data encourage the use of molecular screening for the detection of malignant precursor lesions in at-risk patients.

A unique basal pattern of p53 expression in ulcerative colitis is associated with mutation in the p53 gene

AIMS

The p53 protein is implicated in the control of cell proliferation, differentiation, and death. As part of a study characterizing p53 alterations in colonic mucosa of patients with ulcerative colitis, we identified a unique pattern of basal p53 immunoreactivity.

METHODS AND RESULTS

Tissue samples (n=180) from 42 ulcerative colitis patients were evaluated for p53 alterations by immunohistochemistry, loss of heterozygosity analysis, polymerase chain reaction-single-strand conformation polymorphism and direct sequencing. In addition, the expression of the p53- associated proteins p21waf1/cip1 and MDM2 was evaluated immunohistochemically. Three basic patterns of p53 immunoreactivity were observed: (i) isolated immunoreactive cells in the crypt bases; (ii) strongly positive cells confined to the basal half of the glands; and (iii) diffusely staining cells. The basal staining pattern was observed in both non-neoplastic tissues and in some areas of dysplasia, and was associated with normal expression of p21waf1/cip1 in all cases, and with p53 mutation in seven of 11 cases.

CONCLUSIONS

The basal pattern of p53 expression is associated with mutation in the p53 gene, and appears to be an early change in a subgroup of ulcerative colitis patients. The significance of this pattern of immunoreactivity and the mechanism by which it develops are discussed.

Colorectal neoplasia in ulcerative colitis-recent advances

It is recognized that ulcerative colitis (UC) predisposes to the development of colorectal adenocarcinoma (CRC), and the molecular pathway for this process differs from that for sporadic CRCs. However, several important details regarding the risk factors for and the molecular changes underlying UC-related colorectal carcinogenesis have only come to light lately. First, recent data suggest environmental factors related to long-standing inflammation contribute more to this increased cancer risk than an inherited susceptibility. Second, molecular changes that may represent the first steps in the development of neoplasia are being increasingly identified in non-dysplastic, colitic mucosa. Third, there is now good evidence suggesting that UC-related CRC may develop along more than one molecular pathway. These emerging data will hopefully contribute to attempts to prevent the development of UC-related CRC, e.g. through refining surveillance programmes. Details of the molecular heterogeneity of UC-related dysplasia and CRC may also help develop reliable tools for diagnosing the former and for predicting the behaviour of the latter. Finally, there is increasing awareness of non-epithelial colorectal malignancies which are associated with UC and may potentially increase in incidence with changes in the medical management of this inflammatory disease.

Microsatellite instability in inflamed and neoplastic epithelium in ulcerative colitis

BACKGROUND

Several genetic alterations have been documented in dysplasia and cancer developing in ulcerative colitis (UC). However, the microsatellite instability (MSI) status has rarely been described, especially in the inflamed epithelium of UC.

AIMS

To study MSI status during neoplastic and inflammatory changes in UC.

METHODS

Seventy five surgically resected samples of colorectal mucosa, taken from 16 colectomy specimens of patients with UC were examined: five patients had a long duration with dysplasia or cancer (UC-LD with neoplasm), seven patients had a long duration without neoplastic changes (UC-LD without neoplasm), and four patients had a short duration without neoplastic changes (UC-SD). In addition to MSI status examined by six microsatellite markers, p53 expression was compared among the three groups.

RESULTS

With regard to non-neoplastic inflamed epithelium, MSI in two or more loci (MSI> or =2) was seen more frequently in the UC-LD without neoplasm group than in the UC-SD group (six of 14 v one of 12; p = 0.060), and significantly more often than in the UC-LD with neoplasm group (six of 14 v two of 23; p = 0.016). In the UC-LD without neoplasm group, MSI> or =2 was detected significantly more frequently in patients with severe inflammation than in those with mild inflammation (six of nine v none of five; p = 0.028). With regard to neoplastic epithelium in the UC-LD with neoplasm group, MSI in two or more loci was found in three of 17, and p53 overexpression was seen in 11 of 17 of the neoplastic lesions.

CONCLUSIONS

A high incidence of MSI in long standing UC with severe inflammation probably reflects genomic instability caused by repeated inflammatory stress. Thus, the influence of inflammation should be considered when estimating MSI in UC. It is possible that changes in p53 expression are important in the development of cancer in UC.

Mutations of p53 in morphologically non-neoplastic mucosa of long-standing ulcerative colitis

Two cases of ulcerative colitis (UC)-associated carcinoma or dysplasia and morphologically non-neoplastic mucosa with p53 protein overexpression (MNNM-p53OE) were selected. DNA was extracted from the paraffin blocks of these lesions and exons 5 - 8 of the p53 gene were analyzed by PCR and direct sequencing. In addition, mutations in K-ras codon 12 were analyzed by PCR-RFLP methods. MNNM-p53OE was located surrounding and adjoining a coexisting carcinoma and / or dysplasia. A p53 mutation was detected in 12 / 22 (54.5%) MNNM-p53OE samples, 4 / 8 (50%) dysplasia samples and 8 / 8 (100%) carcinoma samples. The p53 mutations detected in MNNM-p53OE were identical to those demonstrated in the adjoining carcinoma and / or dysplasia. No K-ras codon 12 mutation was detected in any of the samples. These results indicate that MNNM-p53OE may share an identical clonal linkage with a coexisting carcinoma and / or dysplasia, and may be an initial and submorphological form of UC-associated neoplasia. Recognition of MNNM-p53OE in biopsy specimens may help to identify patients with UC at risk of developing colorectal carcinoma.

Immunohistochemical assessment of Ki67 and p53 expression assists the diagnosis and grading of ulcerative colitis-related dysplasia

AIMS

To assess whether Ki67 and p53 immunostaining may assist the diagnosis and grading of ulcerative colitis-related dysplasia.

METHODS AND RESULTS

Location of Ki67 staining and location and intensity of p53 staining were assessed in ulcerative colitis (UC) cases showing the features of high-grade dysplasia (HGD, n = 14), low-grade dysplasia (LGD, n = 22), 'indefinite for dysplasia' (n = 12), or regenerative atypia (RA, n = 22). Good intra- and inter-observer reproducibilities were demonstrated in the performance of these assessments. All the dysplasia cases showed extension of Ki67 staining above the basal third of the crypt. Moderate intensity p53 staining was seen in 10/22 RA cases, but strong intensity p53 staining was seen only in cases of dysplasia. All the cases of HGD showed extension of Ki67 and p53 staining above the basal two thirds of the crypt.

CONCLUSIONS

Restriction of Ki67 staining to the basal third of the crypt appears to exclude a diagnosis of dysplasia whereas strong intensity p53 staining suggests a diagnosis of dysplasia. Restriction of Ki67 or p53 staining to the basal two-thirds of the crypt appears to exclude a diagnosis of HGD.

Dysplasia and cancer in the dextran sulfate sodium mouse colitis model. Relevance to colitis-associated neoplasia in the human: a study of histopathology, B-catenin and p53 expression and the role of inflammation

Animal models of colitis, which develop dysplasia and cancer similar to human ulcerative colitis are needed to further investigate the dysplasia cancer sequence. This study describes the expression of B-catenin and p53 along with the histopathology and inflammation scores as they relate to dysplasia and cancer in the dextran sulfate sodium (DSS) colitis model. Swiss Webster mice were fed with 5% DSS as follows: group A, four cycles of DSS, 84 days total (1 cycle = 7 days DSS + 14 days H(2)O); group B, four cycles DSS followed by 120 days H(2)O, 204 days total; group C, 7 days DSS followed by 180 days H(2)O, 187 days total; group D, 7 days DSS followed by 90 days H(2)O, 97 days total. The incidences of dysplasia and/or cancer were 15.8, 37.5, 18.1 and 0% in groups A-D, respectively. Dysplasia and/or cancer occurred as flat lesions or as dysplasia-associated lesion or mass (DALM) as observed in the human. Thirty-three percent of cancers had associated dysplasia. Within group A, inflammation scores were significantly higher in animals with dysplasia and/or cancer compared with those without dysplasia and/or cancer (P < 0. 05-P < 0.0001). Inflammation scores were significantly higher in animals with cancers versus those with dysplasia (P < 0.015) and in flat dysplasia and/or cancer versus DALM (P < 0.0042). B-catenin showed translocation from the cell membrane to the cytoplasm and/or nucleus in 100% of DALM and 5.8% of flat dysplasia and/or cancer. A total of 94.2% of flat dysplasia and/or cancer had exclusive cell membrane expression compared with 0% DALM (P < 0.0001). Only 7.4% of dysplasia and/or cancer showed nuclear expression of p53. In colitis-associated dysplasia and/or cancer in the DSS model: (i) histology resembles that in the human; (ii) inflammation plays a significant role in the dysplasia cancer sequence and whether dysplasia and/or cancer grows as a flat lesion or a DALM; (iii) the early molecular pathways are different for flat dysplasia and/or cancer versus DALM, with nuclear/cytoplasmic translocation of B-catenin as an early event in DALM but not flat dysplasia and/or cancer; and (iv) p53 has little or no role in dysplasia and/or cancer. This well characterized model provides an excellent vehicle for studying the roles of inflammation, the molecular events and the role of chemopreventive agents in colitis-associated neoplasia.

Chromosomal alterations in ulcerative colitis-related and sporadic colorectal cancers by comparative genomic hybridization

Both ulcerative colitis (UC)-related and sporadic colorectal cancers are thought to evolve through a multistep process of genomic instability, accumulation of genomic alterations, and clonal expansion. This process may involve different genomic changes in UC-related cancers than in sporadic cancers because of the origin of UC-related cancers in an inflammatory field. This study was designed to define the specific genomic events occurring in UC-related cancers. Comparative genomic hybridization (CGH) was performed on 32 UC-related and 42 stage-matched sporadic colorectal cancers. The mean number of chromosomal alterations per case was similar in the UC-related and sporadic tumor groups (8.6 in UC, 8.1 in sporadic). The 2 tumor groups shared many chromosomal alterations: losses on 18q (78% UC v69% sporadic), 8p (53% v50%), 17p (44% v57%), and gains on 8q (63% v45%), 20q (44% UC v67%), and 13q (44% UC v38%). However, differences in the frequency and timing of specific alterations were observed. Chromosome 5q was lost in 56% of UC-related but in only 26% of sporadic cancers. Alterations of chromosome 8 were associated with stage progression in UC-related, but not in sporadic cancers. In contrast, 18q loss was associated with stage progression in sporadic cancers only. Thus, differences in the frequency and timing of individual chromosomal alterations suggest that genetic progression in these 2 tumor groups may follow multiple pathways.

p53 immunohistochemistry of ulcerative colitis-associated with dysplasia and carcinoma

In order to evaluate the usefulness of p53 immunohistochemistry (IHC) in the diagnosis of ulcerative colitis-associated colorectal carcinoma (UCACRC), ordinary paraffin sections were examined in 61 cases with ulcerative colitis (UC) and 29 control cases without UC. Among the 61 cases with UC, 11 were complicated by carcinoma coexisting with dysplasia, three with dysplasia, and two cases with adenoma. There were a total of 38 dysplasias, including 33 low grade dysplasias (LGD) and five mixed low and high grade dysplasias (LGD + HGD). The results of p53 IHC were divided into diffuse, nested, scattered and sporadic patterns for 29 control cases. Diffuse and nested patterns were presumed to reflect mutant forms of p53 protein and were defined as overexpression of p53 protein. In non-neoplastic mucosa of UC, the frequency of p53 positive tubules was significantly higher in active phase (13.5-17.9%) than in resolving phase (3.9-6.5%) and in remission (0.7-2.4%), regardless of association with neoplasia. Eight of the 37 lesions of dysplasia (21. 6%) showed p53 overexpression: 12.5% in LGD and 80.0% in LGD + HGD. The rate of p53 overexpression was significantly higher in UCACRC (90.9%) than in non-neoplastic mucosa of UC (0%), LGD and sporadic colorectal carcinoma (54.5%), but it did not differ between UCACRC and LGD + HGD. Interestingly, the mucosa without dysplasia showed p53 overexpression in one case of UCACRC. The biopsy specimen taken 4 years before the diagnosis of carcinoma revealed p53 overexpression in another case with UCACRC. These results suggest that p53 abnormalities play an important role in UC-associated tumorigenesis in its relatively early phase. For the diagnosis of dysplasia and carcinoma in UC, p53 IHC seems to be useful.

Evaluation of the usefulness of testing for p53 mutations in colorectal cancer surveillance for ulcerative colitis

OBJECTIVE

Immunohistochemical staining for p53 suppressor gene mutations is sensitive and, therefore, has potential for use as a complementary test for dysplasia to improve ulcerative colitis (UC) cancer surveillance program performance.

METHODS

A cohort of 95 patients with long standing pan-UC enrolled in a surveillance program was studied. Archival colonic biopsy specimens were stained for p53 mutations and clinical information was obtained from medical records.

RESULTS

The 37 patients who developed p53 mutations were significantly more likely to develop dysplasia or cancer (relative risk [RR] 4.53, 95% confidence interval [CI] 2.16-9.48). The p53 mutations developed approximately 8 months before low grade dysplasia, 26 months before high grade dysplasia, and 38 months before cancer. Three of seven cancer patients with p53 mutations had Dukes' stage C or D, whereas only one of five cancer patients without p53 mutations had Dukes' C or D; all three patients who died from metastatic cancer had p53 mutations (three of 37 vs 0 of 58, p < 0.03). Folic acid supplementation had a small, significant protective effect for p53 mutations (RR 0.97, CI 0.94-1.00).

CONCLUSION

p53 Mutations 1) are associated with, and likely precede, dysplasia and cancer, 2) are associated with cancer-related mortality, and 3) may possibly be prevented by folic acid supplementation.

Ulcerative colitis with overexpression of p53 preceding overt histological abnormalities of the epithelium

A 53-year-old man with a 22-year history of ulcerative colitis(UC) (pancolitis), had an ulcerating rectal tumor. Resection of the rectum and sigmoid colon was performed. Pathology showed an expansive ulcerating adenocarcinoma tumor (type 2) invading the adventitia against a background of UC in a resolving phase. Dysplasia was also found in granular and flat mucosa adjacent to the invading carcinoma. Immunostaining for p53 showed diffuse positivity in both the carcinoma and dysplasia, and also in the mucosa with indefinite dysplasia or no dysplasia neighboring the dysplasia and carcinoma. Mapping of neoplasms and the area with p53 protein overexpression showed that the grade of dysplasia increased as the lesion approached the invasive carcinoma and that the mucosa with dysplasia was surrounded by mucosa without dysplasia or indefinite for dysplasia, but with p53 protein overexpression. In some areas without dysplasia showing p53 overexpression, there was significant morphometric enlargement of the area and diameter of the nucleus p53 Immunostaining is a good marker for assessing the genetic alterations that precede histological abnormalities and for diagnosing carcinoma in UC. Objective findings such as p53-protein overexpression and morphometric values should be used to evaluate cytological abnormalities in UC, as well as in common colorectal cancer.

p53 expression in ulcerative colitis: a longitudinal study

p53 Mutation is a late event in the development of sporadic colorectal carcinoma (CRC). The timing of p53 mutations in the development of ulcerative colitis associated colorectal carcinoma (UCACRC) is, however, less certain and some reports suggest that it may be a relatively early event. This study sought to establish the timing of p53 mutations in neoplasia arising in ulcerative colitis. Blocks of 10 resected colorectal specimens from patients who had had a biopsy positive for dysplasia at least one year prior to resection were retrieved from the archives of St Mark's Hospital. Immunochemistry using the monoclonal antibody DO-7, specific for both mutant and wild type p53, was performed on sections from both the resection specimens and dysplastic biopsy specimens. Seven of the 10 resection specimens were positive for p53. Two of these seven specimens showed p53 overexpression in specimens taken two years before the development of carcinoma or high grade dysplasia. Five of the seven specimens were negative for p53 overexpression between one and four years prior to resection. These results suggest that p53 overexpression is usually a late event in the development of UCACRCs.

The role of tumor rejection antigens in host antitumor defense mechanisms

BACKGROUND

Normal cells undergo contact inhibition of growth when their surface molecules interact. Tumor cells, however, have undergone a mutation that prevents this arrest of growth upon contact inhibition and allows constant growth. Thus, growth inhibition fails to occur despite the interaction of surface molecules. In recent years a subgroup of these surface molecules has been of interest to cancer investigators. This subgroup has been termed the tumor rejection antigens (TRAs). As the name implies, these are specific to the tumor of origin and may direct the immune system of the host to target the tumor cells and kill them.

METHODS

A literature search was carried out on TRAs to ascertain the current thinking on the subject.

RESULTS

Initial studies of TRAs have revealed that some of them may be heat shock proteins (HSPs). In particular, grp96, a number of the HSP90 family, has been implicated. More recent studies, however, have shown that HSPs alone may not be immunogenic but may act as carrier proteins for tumor specific peptides.

CONCLUSION

Such findings have led to speculation that HSPs or their associated peptides may have a role in the diagnosis and/or treatment of specific cancers. Immunotherapy and bispecific antibodies in particular are areas in which HSPs may prove to be useful.

p53 protein, PCNA staining, and DNA content in follicular neoplasms of the thyroid gland

Forty-nine follicular adenomas and 11 follicular carcinomas of the thyroid were investigated by immunohistochemistry for the expression of p53 protein and proliferating cell nuclear antigen (PCNA). The DNA ploidy and the S-phase fraction (SPF) of the neoplasms were analysed by flow cytometry. Twelve adenomas (24 per cent) and six carcinomas (55 per cent) were DNA non-diploid (P = 0.07). The carcinomas had a higher proliferation rate than the adenomas when assessed either by SPF size (median 9.9 per cent vs. 2.9 per cent, P = 0.0003) or by PCNA staining intensity (P < 0.0001). Some scattered nuclei in two (4 per cent) adenomas and in three (27 per cent) carcinomas stained positively for p53 (P = 0.04). The two adenomas with positive staining for p53 were subserially sectioned, but no signs of invasion were found; both patients are alive and well 6 and 7 years after surgery. One of the two adenomas showing positive p53 nuclear staining was DNA aneuploid, and both were positive in PCNA staining, but their SPFs were low (2.1 and 3.3 per cent). We conclude that p53 protein expression is not confined to follicular carcinomas; scattered p53-positive cells may also be present in histologically and clinically benign follicular adenomas. Because both follicular adenomas and carcinomas may be DNA aneuploid and their SPF and PCNA staining distributions overlap, the distinction between follicular adenoma and carcinoma should still be based on histological criteria.

Cancer in inflammatory bowel disease

Patients with inflammatory bowel disease, including both ulcerative colitis (UC) and Crohn's disease, are at increased risk for the development of gastrointestinal carcinoma, particularly colorectal adenocarcinoma. The current options to reduce this cancer risk include prophylactic colectomy, periodic endoscopic screening with colectomy performed in those patients found to have dysplastic colonic mucosa, or expectant management with no routine surveillance regimen. Despite the lack of data demonstrating effectiveness of surveillance colonoscopy, this approach has become the standard of care in most communities in the United States. Although it has fallen out of fashion in recent years, prophylactic colectomy remains a good option for reducing cancer risk for select patients with UC.

p53 protein expression in ulcerative colitis-associated colorectal dysplasia and carcinoma

The frequency and timing of p53 inactivation in ulcerative colitis (UC)-associated tumorigenesis were investigated using immunohistochemistry (IHC) to detect p53 protein overexpression in 56 carcinomas and 40 dysplastic epithelia derived from 58 patients with UC undergoing colectomy for neoplasia. p53 DNA in 25 of the carcinomas also was evaluated by single-strand conformation polymorphism analysis (SSCP) to detect point mutations in exons 5-8 and by loss of heterozygosity analysis to detect allelic deletions. Point mutations were detected in 20 of the 25 carcinomas (80.0%) undergoing both IHC and DNA analysis. One carcinoma contained an allelic deletion but no mutations of the corresponding allele within the region tested. p53 overexpression occurred in 16 (76.2%) of the 21 carcinomas with point mutations and/or allelic deletions but not in any of those with wild type DNA. Of the 56 carcinomas evaluated by IHC, p53 overexpression occurred in 34 carcinomas (60.7%). The proportion of positive tumors was independent of stage, anatomic location, differentiation, and histological subtype. Overexpression was observed in nine of 20 dysplastic masses devoid of and situated remote from carcinoma (45.0%) and correlated positively with increasing grade of dysplasia (P < .025). In contrast, overexpression occurred in 16 of 20 dysplastic epithelia situated adjacent to carcinoma (80.0%) and correlated with overexpression by the corresponding carcinomas but not with the grade of dysplasia present (P = .013). It is concluded that p53 overexpression can be detected by IHC in most, although not all, UC-associated carcinomas with p53 mutations and/or allelic deletions. Based on this method, p53 overexpression occurs frequently in UC-associated carcinomas regardless of stage and pathological characteristics, in noncancerous dysplastic masses with high grade dysplasia, and in dysplasias of all grades situated adjacent to carcinomas. These findings implicate p53 inactivation in the progression from dysplasia to carcinoma in UC and suggest that its occurrence in dysplastic epithelium may be an independent marker of malignant potential.