A gene expression classifier of node-positive colorectal cancer.

We used digital long serial analysis of gene expression to discover gene expression differences between node-negative and node-positive colorectal tumors and developed a multigene classifier able to discriminate between these two tumor types. We prepared and sequenced long serial analysis of gene expression libraries from one node-negative and one node-positive colorectal tumor, sequenced to a depth of 26,060 unique tags, and identified 262 tags significantly differentially expressed between these two tumors (P < 2 x 10(-6)). We confirmed the tag-to-gene assignments and differential expression of 31 genes by quantitative real-time polymerase chain reaction, 12 of which were elevated in the node-positive tumor. We analyzed the expression levels of these 12 upregulated genes in a validation panel of 23 additional tumors and developed an optimized seven-gene logistic regression classifier. The classifier discriminated between node-negative and node-positive tumors with 86% sensitivity and 80% specificity. Receiver operating characteristic analysis of the classifier revealed an area under the curve of 0.86. Experimental manipulation of the function of one classification gene, Fibronectin, caused profound effects on invasion and migration of colorectal cancer cells in vitro. These results suggest that the development of node-positive colorectal cancer occurs in part through elevated epithelial FN1 expression and suggest novel strategies for the diagnosis and treatment of advanced disease.

Somatic mutations to CSMD1 in colorectal adenocarcinomas.

The short arm of chromosome 8 is frequently deleted in advanced human colorectal cancers, suggesting the presence of one or more tumor suppressor genes having a major role in tumor progression and metastasis. Comprehensive sequencing of over 18,000 genes in colon and breast cancers identified somatic mutations in CUB and Sushi Multiple Domains 1 Gene (CSMD1)which is located on the p arm of chromosome 8. In this report, we describe a novel, robust, high-throughput gene mutation profiling strategy based on massively parallel picotiter plate pyrosequencing and have used this approach to identify additional somatic mutations to CSMD1 in early and late stage colorectal cancers. Using this strategy, we identified five nonsynonymous somatic mutations in CSMD1 among 26 colorectal cancers. Interestingly, these mutations occurred predominantly in advanced colorectal tumors,suggesting a role for CSMD1 in the development of late-stage metastatic disease.

The consensus coding sequences of human breast and colorectal cancers.

The elucidation of the human genome sequence has made it possible to identify genetic alterations in cancers in unprecedented detail. To begin a systematic analysis of such alterations, we determined the sequence of well-annotated human protein-coding genes in two common tumor types. Analysis of 13,023 genes in 11 breast and 11 colorectal cancers revealed that individual tumors accumulate an average of approximately 90 mutant genes but that only a subset of these contribute to the neoplastic process. Using stringent criteria to delineate this subset, we identified 189 genes (average of 11 per tumor) that were mutated at significant frequency. The vast majority of these genes were not known to be genetically altered in tumors and are predicted to affect a wide range of cellular functions, including transcription, adhesion, and invasion. These data define the genetic landscape of two human cancer types, provide new targets for diagnostic and therapeutic intervention, and open fertile avenues for basic research in tumor biology.

Characterization of the gammadelta T cell response to acute leukemia.

BACKGROUND: Previous work from our center has suggested a correlation between increased donor-derived Vdelta1+ gammadelta T cells and long-term relapse-free survival following bone marrow transplantation for leukemia. Questions remain, however, as to whether this observation can be explained by a gammadelta T cell-based immune response against primary leukemia. METHODS: We examined gammadelta T cell receptor (TCR) phenotype, cell proliferation, and cytolytic activity following culture with irradiated primary leukemia blasts from a haploidentical first-degree relative. Subsequently, we also studied the gammadelta TCR phenotype and complimentarity determining region 3 (CDR3) cDNA sequences from 17 newly diagnosed leukemia patients. RESULTS: In 17/28 (61%) of in vitro cultures, gammadelta T cells proliferated in culture with primary blasts. Vdelta1+ T cells were proportionally increased in all cultures and were the predominant cell population in 6/17. In the 7 cultures where cytotoxicity could be assessed, 6 (86%) showed some degree of cytotoxicity to the primary leukemia. Vdelta1+ T cells were also the predominant gammadelta T cell subtype in pre-treatment leukemia patients principally due to loss of Vdelta2+ T cells rather than expansion of Vdelta1+ cells. The Vdelta1 CDR3-region cDNA sequence from these patients revealed exclusive use of the Jdelta1 constant region and sequence conservation in 4/11 patients. CONCLUSIONS: gammadelta T cells exhibit an in vitro response to primary leukemia blasts that is manifested by proliferation, an increased proportion of Vdelta1+ T cells, and cytotoxicity to the primary leukemia blasts. The Vdelta1+ T cell population is also predominant in newly diagnosed leukemia patients likely due to a loss of circulating Vdelta2+ T cells. A small proportion of newly diagnosed patients showed Vdelta1 CDR3 region similarity. These findings suggest a role for gammadelta T cells in the immune response to leukemia.