The publication data currently available has been vetted by Vanderbilt faculty, staff, administrators and trainees. The data itself is retrieved directly from NCBI's PubMed and is automatically updated on a weekly basis to ensure accuracy and completeness.
If you have any questions or comments, please contact us.
BACKGROUND - The TNM staging system originated as a response to the need for an accurate, consistent, universal cancer outcome prediction system. Since the TNM staging system was introduced in the 1950s, new prognostic factors have been identified and new methods for integrating prognostic factors have been developed. This study compares the prediction accuracy of the TNM staging system with that of artificial neural network statistical models.
METHODS - For 5-year survival of patients with breast or colorectal carcinoma, the authors compared the TNM staging system's predictive accuracy with that of artificial neural networks (ANN). The area under the receiver operating characteristic curve, as applied to an independent validation data set, was the measure of accuracy.
RESULTS - For the American College of Surgeons' Patient Care Evaluation (PCE) data set, using only the TNM variables (tumor size, number of positive regional lymph nodes, and distant metastasis), the artificial neural network's predictions of the 5-year survival of patients with breast carcinoma were significantly more accurate than those of the TNM staging system (TNM, 0.720; ANN, 0.770; P < 0.001). For the National Cancer Institute's Surveillance, Epidemiology, and End Results breast carcinoma data set, using only the TNM variables, the artificial neural network's predictions of 10-year survival were significantly more accurate than those of the TNM staging system (TNM, 0.692; ANN, 0.730; P < 0.01). For the PCE colorectal data set, using only the TNM variables, the artificial neural network's predictions of the 5-year survival of patients with colorectal carcinoma were significantly more accurate than those of the TNM staging system (TNM, 0.737; ANN, 0.815; P < 0.001). Adding commonly collected demographic and anatomic variables to the TNM variables further increased the accuracy of the artificial neural network's predictions of breast carcinoma survival (0.784) and colorectal carcinoma survival (0.869).
CONCLUSIONS - Artificial neural networks are significantly more accurate than the TNM staging system when both use the TNM prognostic factors alone. New prognostic factors can be added to artificial neural networks to increase prognostic accuracy further. These results are robust across different data sets and cancer sites.
BACKGROUND & AIMS - Two case-control studies have shown that folate may protect against neoplasia in ulcerative colitis. This historical cohort study was performed to better define this association.
METHODS - The records of 98 patients with ulcerative colitis who had disease proximal to the splenic flexure for at least 8 years were reviewed. Documented folate use of at least 6 months was deemed a positive exposure.
RESULTS - Of the patients, 29.6% developed neoplasia and 40.2% took folate supplements. The adjusted relative risk (RR) of neoplasia for patients taking folate was 0.72 (95% confidence interval [CI], 0.28-1.83). The dose of folate varied with the risk of neoplasia (RR, 0.54 for 1.0 mg folate; RR, 0.76 for 0.4 mg folate in a multivitamin compared with patients taking no folate). Folate use also varied with the degree of dysplasia (RR for cancer, 0.45; RR for high-grade dysplasia, 0.52; RR for low-grade dysplasia, 0.75 compared with patients with no dysplasia) (P = 0.08).
CONCLUSIONS - Although not statistically significant, the RR for folate supplementation on the risk of neoplasia is < 1 and shows a dose-response effect, consistent with previous studies. Daily folate supplementation may protect against the development of neoplasia in ulcerative colitis.
Immunohistochemistry (IHC) for intranuclear p53 gene product is a simple, routine alternative to molecular genetic analysis of the p53 gene. Several methods for antigen enhancement are currently in use for IHC. This study evaluates the effect of extreme antigen enhancement for p53, using a monoclonal antibody (DO7) and a polyclonal antibody (CM1). The cases studied were five colorectal carcinomas, two specimens of normal colorectal mucosa, and four colorectal carcinomas with genetic alterations which are expected to preclude p53 gene product expression, namely mutation to a STOP codon in the p53 gene detected by denaturing gradient gel electrophoresis with subsequent sequencing and allelic loss of 17p in the region where p53 is located, detected by restriction fragment length polymorphism analysis. The findings suggest that extreme antigen enhancement may cause false-positive results with a distinct nuclear staining pattern when MAb DO7 is used as a primary antibody. It is concluded that all antigen enhancement methods should be thoroughly tested to evaluate their validity and that there may be a limit to the extent to which antigen enhancement can be applied in IHC for p53 protein.
Inactivation of the p53 tumour-suppressor gene is common in a wide variety of human neoplasms. In the majority of cases, single point mutations in the protein-encoding sequence of p53 lead to positive immunohistochemistry (IHC) for the p53 protein, and are accompanied by loss of the wild-type allele. Recently, the WAF1/Cip1 gene was identified as one of the genes induced by wild-type p53, and increased expression of p21WAF1/Cip1 has been found to reflect the status of the p53 tumour-suppressor pathway. We investigated the inactivation of p53 in a relatively small, but well-characterised, group of 46 colorectal carcinomas that were previously studied for allelic alterations, ras oncogene mutations and DNA aneuploidy. Alterations in p53 were identified by IHC, loss of 17p and DNA sequence analysis of exons 5-8, whereas p21WAF1/Cip1 protein expression was determined by IHC. p53 mutations were identified in 19 of the 46 tumours (41%), whereas positive IHC for p53 was found in 21 of the 46 tumours (46%). Positive IHC for p21WAF1/Cip1 was detected in 16 of 42 cases (38%). We found no relationship between p21WAF1/Cip1 staining and p53 protein expression or p53 mutational status. Inactivating mutations in the p53 gene correlated with LOH at 17p but not with LOH at 5q or 18q, Dukes' stage, tumour grade or DNA ploidy. There was a higher survival rate independent of Dukes' stage in the group with no alterations in p53 compared with those with evidence of dysfunction of p53, but the difference was not statistically significant. We conclude that inactivation of p53 and altered expression of p21WAF1/Cip1 are common in colorectal carcinoma but do not correlate with each other or with the clinical or pathological parameters investigated.
Colorectal cancer is the second leading cause of death from cancer in the United States. Continuous use of aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) has been shown to reduce the risk of colorectal cancer in humans by 40-50%. Patients with familial adenomatous polyposis who take NSAIDs, such as sulindac, undergo a regression of intestinal adenomas. Rodents exposed to carcinogens that cause colon cancer have a 50-60% reduction in the size and number of colonic tumors when treated continuously with NSAIDs. One common target for these drugs is prostaglandin endoperoxide synthase, also referred to as cyclooxygenase (COX). We and others have shown recently that COX-2 levels are increased dramatically in 85-90% of human colorectal adenocarcinomas and in 40-50% of colonic adenomas. We prepared intestinal epithelial cells that express the COX-2 gene permanently and found that they have altered adhesion properties and resist undergoing apoptosis. We report here that these cells also have a 3-fold increase in the duration of G1, lower levels of cyclin D1 protein, and a marked decrease in retinoblastoma kinase activity associated with cyclin-dependent kinase 4. The delay in G1 transit may relate to the resistance of these cells to undergo programmed cell death, which could affect their tumorigenic potential.
BACKGROUND/AIMS - Several clinical, epidemiological, and animal studies indicate that nonsteroidal anti-inflammatory drugs (NSAIDs) may alter the incidence of colorectal cancer. A likely target for NSAIDs is cyclooxygenase, a key enzyme in arachidonic acid metabolism. Two isoforms of this enzyme have been identified; cyclooxygenase (COX) 1 and COX-2. The present study was undertaken to determine if there is differential expression of these isoforms in colorectal neoplasia, and, if so, at what stage in malignant transformation this occurs.
METHODS - COX-1 and COX-2 messenger RNA (mRNA) levels were determined by Northern blot analysis of poly(A)+ RNA isolated from human colorectal cancers, adenomas, and accompanying normal mucosa.
RESULTS - There was a marked increase in COX-2 mRNA levels in 12 of 14 carcinomas (86%) compared with paired normal mucosa. In contrast, there was equivalent intensity of the COX-1 mRNA transcript between the normal mucosa and cancer in all 14 cases. In six pairs of colorectal adenomas and normal mucosa, three showed up-regulation of COX-2 in the adenoma compared with the normal mucosa. Because COX-2 expression is low to undetectable in normal colorectal mucosa, 14 unpaired adenomas were examined for COX-2 expression; a clearly detectable transcript was identified in six (43%).
CONCLUSIONS - COX-2, but not COX-1, gene expression is markedly elevated in most human colorectal cancers compared with accompanying normal mucosa. Furthermore, COX-2 expression seems to be increased in a subset of adenomas. COX-2 may provide an attractive therapeutic target in colorectal neoplasia.
The utility of ordinal logistic regression in the prediction of colorectal neoplasia was demonstrated in a group of 461 consecutive patients undergoing colonoscopy in a community practice. One hundred twenty-nine patients had adenomatous polyps and 34 had colorectal adenocarcinoma. An ordinal logistic regression model developed in a random subset (292 patients) identified five predictors of colorectal neoplasia. Colorectal neoplasia risk could be predicted using the patient's age, sex, hematocrit, fecal occult blood test result and indication for colonoscopy. The risk of colorectal neoplasia in the remaining subset of patients (169) could be reliably estimated from the model. Ordinal logistic regression analysis in this select group of patients can accurately estimate the likelihood of colorectal neoplasia. Because the generalizability of our findings are unknown, the model should not be applied to other patients. However, application of this technique to an unselected group of patients not already referred for colonoscopy could provide unbiased estimates of colorectal neoplasia risk in individual patients.