Purpose: Recent transcriptomic analyses have identified four distinct molecular subtypes of colorectal cancer with evident clinical relevance. However, the requirement for sufficient quantities of bulk tumor and difficulties in obtaining high-quality genome-wide transcriptome data from formalin-fixed paraffinembedded tissue are obstacles toward widespread adoption of this taxonomy. Here, we develop an immunohistochemistrybased classifier to validate the prognostic and predictive value of molecular colorectal cancer subtyping in a multicenter study.Experimental Design: Tissue microarrays from 1,076 patients with colorectal cancer from four different cohorts were stained for five markers (CDX2, FRMD6, HTR2B, ZEB1, and KER) by immunohistochemistry and assessed for microsatellite instability. An automated classification system was trained on one cohort using quantitative image analysis or semiquantitative pathologist scoring of the cores as input and applied to three independent clinical cohorts.Results: This classifier demonstrated 87% concordance with the gold-standard transcriptome-based classification. Application to three validation datasets confirmed the poor prognosis of the mesenchymal-like molecular colorectal cancer subtype. In addition, retrospective analysis demonstrated the benefit of adding cetuximab to bevacizumab and chemotherapy in patients with RAS wild-type metastatic cancers of the canonical epithelial-like subtypes.Conclusions: This study shows that a practical and robust immunohistochemical assay can be employed to identify molecular colorectal cancer subtypes and uncover subtypespecific therapeutic benefit. Finally, the described tool is available online for rapid classification of colorectal cancer samples, both in the format of an automated image analysis pipeline to score tumor core staining, and as a classifier based on semiquantitative pathology scoring.

Purpose: Chemotherapy treatment of metastatic colon cancer ultimately fails due to development of drug resistance. Identification of chemotherapy-induced changes in tumor biology may provide insight into drug resistance mechanisms.Experimental Design: We studied gene expression differences between groups of liver metastases that were exposed to preoperative chemotherapy or not. Multiple patient-derived colonosphere cultures were used to assess how chemotherapy alters energy metabolism by measuring mitochondrial biomass, oxygen consumption, and lactate production. Genetically manipulated colonosphere-initiated tumors were used to assess how altered energy metabolism affects chemotherapy efficacy.Results: Gene ontology and pathway enrichment analysis revealed significant upregulation of genes involved in oxidative phosphorylation (OXPHOS) and mitochondrial biogenesis in metastases that were exposed to chemotherapy. This suggested chemotherapy induces a shift in tumor metabolism from glycolysis towards OXPHOS. Indeed, chemotreatment of patientderived colonosphere cultures resulted in an increase of mitochondrial biomass, increased expression of respiratory chain enzymes, and higher rates of oxygen consumption. This was mediated by the histone deacetylase sirtuin-1 (SIRT1) and its substrate, the transcriptional coactivator PGC1a. Knockdown of SIRT1 or PGC1a prevented chemotherapy-induced OXPHOS and significantly sensitized patient-derived colonospheres as well as tumor xenografts to chemotherapy.Conclusions: Chemotherapy of colorectal tumors induces a SIRT1/PGC1a-dependent increase in OXPHOS that promotes tumor survival during treatment. This phenomenon is also observed in chemotherapy-exposed resected liver metastases, strongly suggesting that chemotherapy induces long-lasting changes in tumor metabolism that potentially interfere with drug efficacy. In conclusion, we propose a novel mechanism of chemotherapy resistance that may be clinically relevant and therapeutically exploitable.

Identifying druggable targets in the Wnt-signaling pathway can optimize colorectal cancer treatment. Recent studies have identified a member of the RNA helicase family DDX3 (DDX3X) as a multilevel activator of Wnt signaling in cells without activating mutations in the Wnt-signaling pathway. In this study, we evaluated whether DDX3 plays a role in the constitutively active Wnt pathway that drives colorectal cancer.We determined DDX3 expression levels in 303 colorectal cancers by immunohistochemistry. 39% of tumors overexpressed DDX3. High cytoplasmic DDX3 expression correlated with nuclear β-catenin expression, a marker of activated Wnt signaling. Functionally, we validated this finding in vitro and found that inhibition of DDX3 with siRNA resulted in reduced TCF4-reporter activity and lowered the mRNA expression levels of downstream TCF4-regulated genes. In addition, DDX3 knockdown in colorectal cancer cell lines reduced proliferation and caused a G1 arrest, supporting a potential oncogenic role of DDX3 in colorectal cancer.RK-33 is a small molecule inhibitor designed to bind to the ATP-binding site of DDX3. Treatment of colorectal cancer cell lines and patient-derived 3D cultures with RK-33 inhibited growth and promoted cell death with IC50 values ranging from 2.5 to 8 μM. The highest RK-33 sensitivity was observed in tumors with wild-type APC-status and a mutation in CTNNB1.Based on these results, we conclude that DDX3 has an oncogenic role in colorectal cancer. Inhibition of DDX3 with the small molecule inhibitor RK-33 causes inhibition of Wnt signaling and may therefore be a promising future treatment strategy for a subset of colorectal cancers.

The recent discovery of ‘molecular subtypes’ in human primary colorectal cancer has revealed correlations between subtype, propensity to metastasize and response to therapy. It is currently not known whether the molecular tumor subtype is maintained after distant spread. If this is the case, molecular subtyping of the primary tumor could guide subtype-targeted therapy of metastatic disease. In this study, we classified paired samples of primary colorectal carcinomas and their corresponding liver metastases (n=129) as epithelial-like or mesenchymal-like, using a recently developed immunohistochemistry-based classification tool. We observed considerable discordance (45%) in the classification of primary tumors and their liver metastases. Discordant classification was significantly associated with the use of neoadjuvant chemotherapy. Furthermore, gene expression analysis of chemotherapy-exposed versus chemotherapy naive liver metastases revealed expression of a mesenchymal program in pre-treated tumors. To explore whether chemotherapy could cause gene expression changes influencing molecular subtyping, we exposed patient-derived colonospheres to six short cycles of 5-fluorouracil. Gene expression profiling and signature enrichment analysis subsequently revealed that the expression of signatures identifying mesenchymal-like tumors was strongly increased in chemotherapy-exposed tumor cultures. Unsupervised clustering of large cohorts of human colon tumors with the chemotherapy-induced gene expression program identified a poor prognosis mesenchymal-like subgroup. We conclude that neoadjuvant chemotherapy induces a mesenchymal phenotype in residual tumor cells and that this may influence the molecular classification of colorectal tumors.