Kathryn L. DeJulius

BRAF mutations in colorectal cancer are associated with distinct clinical characteristics and worse prognosis.

BACKGROUND: Colorectal cancer is a heterogeneous disease with multiple underlying genetic mutations causing different clinical phenotypes. Mutation in the BRAF oncogene is a key step in malignant transformation within the methylator pathway to colorectal cancer. However, there is a paucity of information about BRAF mutant colorectal tumors. OBJECTIVE: This study defines the clinical characteristics and oncologic outcome associated with colorectal cancer BRAF mutations. DESIGN: Colorectal adenocarcinomas from a single-institution frozen-tumor biobank were studied. Genomic DNA was isolated and analyzed for mutations in the BRAF oncogene by polymerase chain reaction amplification followed by direct sequencing. A sample was classified as mutant if any of the tested loci were mutated. Patient and tumor characteristics were recorded including patient age, sex, tumor location, tumor differentiation, and microsatellite instability. MAIN OUTCOME MEASURES: Statistical associations with BRAF mutant tumors were determined by the Fisher exact probability test, &khgr;2 test, or Wilcoxon analysis. Kaplan-Meier estimates and multivariate Cox regression analysis were performed for overall survival. RESULTS: Four hundred seventy-five colorectal adenocarcinomas were included in the study population; 56 samples harbored a BRAF mutation (12%). There were significant differences between BRAF wild-type and mutant tumors in age (66 vs 75 years, p = 0.004), female sex (44% vs 71%, p < 0.001), proximal tumor location (44% vs 95%, p < 0.001), and frequency of microsatellite instability (16% vs 76%, p < 0.001). There was no difference in cancer stage between BRAF mutant and wild-type populations. Survival data were analyzed for 322 patients with stage I to III disease, and patients with a BRAF mutation had decreased overall survival than those without a mutation (p = 0.018). With the use of Cox regression analysis, BRAF mutation conferred a worse overall survival (HR 1.79, CI 1.05–3.05, p = 0.03) independent of microsatellite instability status. CONCLUSIONS: BRAF mutations in colorectal cancers are associated with distinct clinical characteristics and worse prognosis.

Genetic and epigenetic classifications define clinical phenotypes and determine patient outcomes in colorectal cancer

A molecular classification of colorectal cancer has been proposed based on microsatellite instability (MSI), CpG island methylator phenotype (CIMP), and mutations in the KRAS and BRAF oncogenes. This study examined the prevalence of these molecular classes, and differences in clinical presentation and outcome.

Relative role of methylator and tumor suppressor pathways in ulcerative colitis-associated colon cancer

Background: Chronic ulcerative colitis (UC) is associated with an increased colorectal cancer risk which may be secondary to repetitive mucosal injury. Both epigenetic methylation and the classic adenoma‐to‐carcinoma sequence have been implicated in this malignant transformation, but the underlying molecular mechanisms remain poorly defined. This study compares the molecular characteristics of colitis‐associated and common colorectal cancers. Methods: Nineteen patients with colorectal adenocarcinomas arising within UC were matched for age and cancer site with 54 patients with sporadic adenocarcinomas. Tumor tissue was examined for BRAF mutations, CpG island methylator phenotype (CIMP), and MLH1 promoter methylation. Mutations of KRAS and p53 were assessed by sequencing. Results: Patient demographics were similar for the two groups. CIMP was observed in 22% of sporadic colorectal cancers and in 5% of UC cancers (P = 0.162). Rates of BRAF mutation (4% vs 5%, P = 1.0), MLH1 methylation (9% versus 5%, P = 0.682), and KRAS mutations (24% versus 32%, P = 0.552) were similar between the groups. However, colitis‐associated colorectal cancers were more likely to have a p53 mutation compared to sporadic adenocarcinomas (95% versus 53%, P = 0.001). The dominant mutation for colitis‐associated cancers was a mutation in codon 4, representing half of the mutations. Furthermore, colitis‐associated cancers had a higher rate of mutation in codon 8 (48% versus 6%, P < 0.001) than sporadic counterparts. Conclusions: Unlike other inflammatory gastrointestinal cancers, colitis‐associated colorectal cancers do not preferentially arise via a methylator pathway when compared to sporadic colorectal cancers. Chromosomal instability remains an important etiology, but with a unique p53 frequency and mutation pattern. (Inflamm Bowel Dis 2010;)

Gene signature is associated with early stage rectal cancer recurrence

BACKGROUND Despite expected excellent outcomes of surgical resection for early stage rectal cancers, 20% of stage I and II rectal cancers recur. Identifying biologic factors that predict the subset prone to recur could allow more directed therapy. This study identifies a tumor gene expression profile that accurately predicts disease recurrence. STUDY DESIGN Stage I/II rectal cancer patients treated by surgery alone at a single institution were included and classified as having recurrent or nonrecurrent cancer. Tumor mRNA was isolated from frozen tissue and evaluated for total genome gene expression by microarray analysis. Background-corrected and normalized microarray data were analyzed using BAMarray software. Selected genes were further analyzed using unsupervised clustering and nearest-centroid classification. A balanced K-fold scoring-pair algorithm using 1,000 independent replications was used for gene signature development. RESULTS Sixty-nine patients with disease-free survival and 31 patients with recurrent disease were included at a median follow-up of 105 months (interquartile range 114 months) and 32 months (interquartile range 25 months), respectively. Demographics and tumor characteristics between groups were similar. Fifty-two genes from 43,148 probes were differentially expressed, and a 36-gene signature was found to be statistically associated with recurrence using a scoring-pair algorithm. Accuracy to identify recurrence as measured by area under the receiver operating characteristic curve was 0.803. CONCLUSIONS Differential gene expression within rectal cancers is associated with recurrence of early stage disease. A 36-gene signature correlates with an increased risk of more or less aggressive tumor behavior. This information obtainable at biopsy may assist in determining treatment decisions.

Gene expression profile is associated with chemoradiation resistance in rectal cancer

Patients with rectal cancer who achieve a complete pathological response after preoperative chemoradiation (CRT) have an improved oncological outcome. Identifying factors associated with a lack of response could help our understanding of the underlying biology of treatment resistance. This study aimed to develop a gene expression signature for CRT‐resistant rectal cancer using high‐throughput nucleotide microarrays.

Genetic and molecular diversity of colon cancer hepatic metastases

BACKGROUND Colon cancer arises through distinct molecular pathways resulting in diverse tumor populations demonstrated by differences in microsatellite instability (MSI), CpG island methylator phenotype (CIMP), and mutations in oncogenes KRAS and BRAF. Although these molecular differences are well-described for primary neoplasms, the molecular nature of hepatic metastases is not well-characterized. This study seeks to describe molecular characteristics of colon cancer hepatic metastases in terms of oncogenic pathway. METHODS Tumor DNA was isolated from fresh frozen hepatic metastases from colon cancer and analyzed for MSI by polymerase chain reaction (PCR)-based microsatellite analysis and for CIMP using MethyLight quantitative PCR. KRAS and BRAF oncogenes were analyzed for DNA mutations. Metastases were classified by their molecular and genetic features. Unfortunately, tissue from the primary neoplasms from these patients were not available RESULTS Thirty patients with liver metastases from colon cancer were studied. Molecular analysis revealed 10% (3/30) were MSI-H, 10% (3/30) were CIMP positive, 33% (10/30) had KRAS mutations, and none had BRAF mutations. Literature describing primary colon cancers reports an incidence of approximately 20% MSI-H, 20% CIMP-positive, 35% KRAS mutants, and 17% BRAF mutants. CONCLUSION Hepatic metastases from colon cancer, like primary colon adenocarcinomas, show genetic and molecular diversity. Furthermore, hepatic metastases may have a different incidence of MSI and methylation compared with primary neoplasms. These differences could impact treatment decisions.

CEACAM-7: A predictive marker for rectal cancer recurrence

BACKGROUND The identification of rectal cancer patients predisposed to developing recurrent disease could allow directed adjuvant therapy to improve outcomes while decreasing unnecessary morbidity. This study evaluates carcinoembryonic antigen cellular adhesion molecule-7 (CEACAM-7) expression in rectal cancer as a predictive recurrence factor. METHODS A single-institution colorectal cancer database and a frozen tissue biobank were queried for rectal cancer patients. CEACAM-7 messenger RNA (mRNA) expression from normal rectal mucosa and rectal cancers was analyzed using quantitative real-time polymerase chain reaction (PCR). Expression-level differences among normal tissue, disease-free survivors, and those that developed recurrence were analyzed. RESULTS Eighty-four patients were included in the study, which consisted of 37 patients with nonrecurrent disease (median follow-up, 170 months), 29 patients with recurrent disease, and 18 patients with stage IV disease. CEACAM-7 expression was decreased 21-fold in rectal cancers compared with normal mucosa (P = .002). The expression levels of CEACAM-7 were relatively decreased in tumors that developed recurrence compared with nonrecurrence, significantly for stage II patients (14-fold relative decrease, P = .002). For stages I-III, disease-free survival segregates were based on relative CEACAM-7 expression values (P = .036), specifically for stage II (P = .018). CONCLUSION CEACAM-7 expression is significantly decreased in rectal cancer. Expression differences between long-term survivors and those with recurrent disease introduce a potential tumor marker to define a subset of patients who benefit most from adjuvant therapy.

High-throughput arrays identify distinct genetic profiles associated with lymph node involvement in rectal cancer.

BACKGROUND: Preoperative clinical diagnosis of lymph node involvement guides treatment decisions for rectal cancer. Unfortunately, clinical staging still suffers from a lack of accuracy. OBJECTIVE: The aim of this study was to evaluate objective genetic differences in primary rectal cancers with and without associated lymph node metastasis. DESIGN: cDNA microarrays were generated from fresh-frozen tumors. Normalized data underwent global unsupervised hierarchical clustering analysis, and discriminating genes were mapped. Top discriminating genes were compared between stage II and III rectal cancers by use of an empirical Bayes 2 group t test with the Statistical Analysis of Microarrays and the Reproducibility-Optimized Test Statistic software separately to guide data reduction and deal with the difficulties of simultaneous statistical inference. Ingenuity Pathways Analysis software was used to analyze discriminating genes in terms of function and biological processes. PATIENTS: Fifty-five patients with stage II and 22 patients with stage III rectal adenocarcinomas not treated with chemoradiation were included. RESULTS: Two major unsupervised clusters emerged representing stage II and III cancers. In 1 cluster, 11 of 12 patients (92%) had stage III cancer; in the other cluster, 54 of 65 patients (83%) had stage II (p < 0.001). Five significantly differentially expressed genes characterized the stage III cluster: interleukin-8, 3-hydroxy-3-methylglutaryl coenzyme A synthase, carbonic anhydrase, ubiquitin, and cystatin (all p < 0.05). Of the 12 patients with differential expression of the 5 marker genes, only one had stage II cancer. Fifty-four of 55 stage II patients clustered with alternative expression patterns of the predictor genes. Differentially expressed genes are related to cancer-associated processes, pathways, and networks. LIMITATIONS: The identified gene signatures have not yet been validated in independent patient populations. CONCLUSIONS: Distinct gene expression signatures from primary rectal adenocarcinomas can help differentiate the presence or absence of lymph node metastases. These data are informative, and validation of this gene signature may provide a novel approach for more appropriate individualized treatment selection.

Transcriptional profiles underpin microsatellite status and associated features in colon cancer

INTRODUCTION While microsatellite instability is associated with prognosis and distinct clinical phenotypes in colon cancer, the basis for this remains incompletely defined. Novel bioinformatic techniques enable a detailed interrogation of the relationship between gene expression profiles and tumor characteristics. AIM We aimed to determine if microsatellite instability high (MSI-H) and microsatellite stable (MSS) tumors could be differentiated by gene expression profiles. We investigated the basis of this using a system and network based algorithmic approach. METHODS Microsatellite status was established using a polymerase chain reaction (PCR) panel and fragment length analysis. Gene expression was determined using Illumina© microarrays comprising 48,701 transcripts, and scaling normalization was conducted using Limma in R. Following filtration for non-significant changes a meta-gene was established and subjected to unsupervised hierarchical clustering using Chipster©. A supervised learning algorithm (PAM) was used to generate a gene-expression based clinical-outcome predictor that was further tested using an independent validation group. A network based linkage analysis was conducted using Ingenuity© focusing on canonical, functional pathways, and associated therapeutic modalities. RESULTS MSI-H and MSS tumors clustered separately following an unsupervised hierarchical clustering analysis. A transcriptomic classifier (with 19 component genes) was generated that reliably and reproducibly predicted microsatellite status. MSI-H associated canonical pathways were predominantly immune or inflammation related converging on increased IL-1B and thymidylate synthase expression. The network linkage analysis identified canakinumab, IL-trap and MDX-1100 as the strongest therapeutic candidates that remain to be assessed in the colon cancer setting. CONCLUSIONS Microsatellite status is underpinned by transcriptional events and can be accurately and reliably defined by differential gene expression. A specific transcriptomic profile is pathognomonic and provides insight into the differences in biology between MSS and MSI-H colon cancers.

Transcriptomic Profiles Differentiate Normal Rectal Epithelium and Adenocarcinoma

Adenocarcinoma is a histologic diagnosis based on subjective findings. Transcriptional profiles have been used to differentiate normal tissue from disease and could provide a means of identifying malignancy. The goal of this study was to generate and test transcriptomic profiles that differentiate normal from adenocarcinomatous rectum. Comparisons were made between cDNA microarrays derived from normal epithelium and rectal adenocarcinoma. Results were filtered according to standard deviation to retain only highly dysregulated genes. Genes differentially expressed between cancer and normal tissue on two-groups t test (P < 0.05, Bonferroni P value adjustment) were further analyzed. Genes were rank ordered in terms of descending fold change. For each comparison (tumor versus normal epithelium), those 5 genes with the greatest positive fold change were grouped in a classifier. Five separate tests were applied to evaluate the discriminatory capacity of each classifier. Genetic classifiers derived comparing normal epithelium with malignant rectal epithelium from pooled stages had a mean sensitivity and specificity of 99.6% and 98.2%, respectively. The classifiers derived from comparing normal and stage I cancer had comparable mean sensitivities and specificities (97% and 98%, respectively). Areas under the summary receiver-operator characteristic curves for each classifier were 0.981 and 0.972, respectively. One gene was common to both classifiers. Classifiers were tested in an independent Gene Expression Omnibus-derived dataset. Both classifiers retained their predictive properties. Transcriptomic profiles comprising as few as 5 genes are highly accurate in differentiating normal from adenocarcinomatous rectal epithelium, including early-stage disease.