investigator_user investigator user funding collaborators pending menu bell message arrow_up arrow_down filter layers globe marker add arrow close download edit facebook info linkedin minus plus save share search sort twitter remove user-plus user-minus
  • Project leads
  • Collaborators

Prediction of Therapy Response in Colorectal Cancer

Thomas Ried

1 Collaborator(s)

Funding source

National Cancer Institute (NIH)
The clinical study CAO-/ARO-/AIO-94 of the German Rectal Cancer Study Group led clinicians around the world to embrace neoadjuvant chemoradiotherapy as the treatment of choice for patients with locally advanced rectal cancer because it reduces the rate of local recurrence. A follow-up prospective randomized phase III clinical trial is now in place (CAO-/ARO-/AIO-04). The goal of this trial is to address whether tumor response and prognosis can be further improved by the addition of oxaliplatin to the existing combined modality therapy (5-FU and radiation). The gene expression analyses, aCGH experiments, and measurement of miRNA profiles have been performed on patients enrolled in this trial and are presently being summarized for publication. Our collaborations with Goettingen are integrated with a Clinical Research Unit with the title "The biological basis of individual tumor response in patients with rectal cancer" ( The Clinical Research Unit consists of eight subprojects conducive to interdisciplinary research. We are actively involved in two subprojects, "Gene expression signature and genetic polymorphisms for response prediction of rectal carcinomas to preoperative chemoradiotherapy" and "Functional validation of genes involved in resistance of rectal carcinomas to preoperative chemoradiation". In our laboratory at the NCI, we have complemented these analyses by using aCGH for profiling genomic copy number aberrations, and by genotyping germline DNA using the Illumina 1 Million platform. Together, we were able to pursue a holistic approach to the problem of treatment failure and disease prognostication in patients with rectal cancer. There is a wide spectrum of tumor responsiveness of rectal adenocarcinomas to preoperative chemoradiotherapy ranging from complete response to complete resistance. In a previous study, we collected pre-therapeutic biopsies from 30 locally advanced rectal carcinomas (determined by rectal ultrasound as uT3 and uT4) and analyzed these samples for gene expression signatures using microarrays. All patients were participants of the above-mentioned phase III clinical trial (CAO/ARO/AIO-94, German Rectal Cancer Trial) and were randomized to receive a preoperative combined modality therapy including 5-fluorouracil and radiation. We identified 54 genes (p0.001) that were differentially expressed between responders and non-responders. The ability to predict response to therapy using gene expression profiles was rigorously evaluated using leave-one-out cross-validation (LOOCV). Tumor behavior was correctly predicted in 83% of patients (p=0.02). Sensitivity (correct prediction of response) was 78% and specificity (correct prediction of non-response) was 86% with a positive predictive value of 78% and negative predictive value of 86%. These results suggest that pre-therapeutic gene expression profiling may assist in response prediction of rectal adenocarcinomas to preoperative chemoradiotherapy. We hypothesized that transcriptional silencing of genes overexpressed in resistant tumors could render these tumors more sensitive to chemoradiotherapy. Of note, we found the transcription factor Transcription factor 7-like 2 (TCF7L2, also known as TCF4), which is major mediator of Wnt signaling, to be expressed at higher levels in tumors that were resistant to chemoradiotherapy. TCF7L2 represents the key transcription factor that mediates canonical Wnt/beta-catenin signaling, which plays a central role in embryonic development and in the maintenance of tissue homeostasis. Aberrant pathway activity promotes colorectal cancer development. Binding of Wnt ligands to cell surface receptors of the Frizzled family inhibits glycogen synthase kinase-3beta (GSK-3beta)-mediated phosphorylation of the co-transcription factor beta-catenin, leading to its stabilization and subsequent accumulation in the nucleus. This results in binding to members of the TCF/LEF (T cell factor and lymphoid enhancer-binding factor) family of transcription factors, which in turn induces or represses transcription of a plethora of target genes ( Because Wnt signaling has not yet been associated with treatment response, we aimed to investigate whether TCF4 mediates chemoradioresistance. RNA-interference mediated silencing of TCF4 was employed in three colorectal cancer cell lines, and sensitivity to (chemo-) radiotherapy was assessed using a standard colony formation assay. Silencing of TCF4 caused a significant sensitization of colorectal cancer cells to clinically relevant doses of X-rays. This effect was restricted to tumor cells with high TCF reporter activity, but, presumably, in a beta-catenin independent manner. Radiosensitization was the consequence of (i) a transcriptional deregulation of Wnt/TCF4 target genes, (ii) a silencing-induced G2/M phase arrest, (iii) an impaired ability to adequately halt cell cycle progression after radiation, and (iv) a compromised DNA double strand break repair as assessed by gamma-H2AX staining. Taken together, our results indicate a novel mechanism through which the Wnt transcription factor TCF4 mediates chemoradioresistance. Moreover, they suggest that TCF4 is a promising molecular target to sensitize resistant tumor cells to (chemo-) radiotherapy. We analyzed 57 rectal cancer samples and matched normal mucosa for differentially expressed miRNAs. Our comprehensive analysis uncovered novel miRNAs and pathways associated with rectal cancer. A subset of deregulated miRNAs was specific to rectal cancer. Of clinical impact, miR-135b expression correlated significantly with disease-free and cancer-specific survival in an independent multicenter cohort of 116 patients. This study was supported through a CRADA with Exiqon. Sequence analysis of more than 90 rectal cancers samples (the patients were part of a phase-III clinical trial and treated with neoadjuvant chemoradiotherapy) revealed KRAS mutations in 48%, and notably a high number of mutations in exons 61 and 146. BRAF mutations were not present in rectal cancers, which is different from tumors located in the colon. The presence of activating KRAS mutations is significantly correlated with an upregulation of DUSP4, a MAP-kinase phosphatase, and SMYD3, a histone methyltransferase. These findings suggest a potential approach to overcome resistance to treatment with antibodies against the epithelial growth factor receptor in patients with KRAS-mutant rectal carcinomas. Like normal colorectal epithelium, colorectal carcinomas (CRC) are organized hierarchically and potentially include populations of cells with stem-like properties. We were therefore curious whether stem-like positive and negative fraction in primary CRC, derived cell lines, and mouse models were different with respect to their genome and transcriptome profiles. The genomic aberration profiles of putative CRC stem cells (defined by positivity for CD133) was different from the bulk primary tumors in 7/12 tumors, and accompanied by specific gene expression profiles. However, CD133 positive and negative fraction of CRC cell lines did not show differences in the genomic aberration pattern, nor were they different as to their tumorigenicity.

Related projects