Date of Graduation


Document Type

Thesis (MS)

Program Affiliation

Clinical and Translational Sciences

Degree Name

Masters of Science (MS)

Advisor/Committee Chair

Scott Kopetz

Committee Member

Gary Gallick

Committee Member

David Menter

Committee Member

Michael Overman

Committee Member

Imad Shureiqi


Introduction: BRAF V600E mutations are associated with poor clinical outcomes for patients with metastatic colorectal cancer (mCRC). Unlike other tumors with the same mutation, BRAF inhibitors are ineffective as monotherapy. CRC tumors with BRAF V600E mutations are associated with global hypermethylation, which may turn off tumor suppressor gene expression. We studied demethylation in BRAF V600E mCRC to restore sensitivity to BRAF inhibitors.

Methods: Tumor databanks were investigated for genes differentially expressed according to BRAF mutation status to identify genes which may be particularly susceptible to epigenetic influence. Mouse xenograft models of BRAFV600E mCRC were treated with vemurafenib or azacitidine, alone or in combination, to assess for changes in tumor size. Tumors and cell lines exposed to azacitidine were analyzed for methylation status and for gene expression differences, with particular emphasis on genes identified from the bioinformatics analysis.

Results: The addition of azacitidine did not restore sensitivity to vemurafenib in two xenograft models of BRAF V600E mCRC. Genes critical to negative regulation of Wnt/β-catenin signaling like RNF43 and Axin2 were significantly underexpressed in BRAF V600E mutant tumors when compared to their wild-type counterparts. These genes were hypermethylated in the xenograft models, which could be reversed with a demethylating agent.

Conclusions: The combination of a BRAF inhibitor and a demethylating agent does not appear to have promising anti-tumor activity in preclinical models of BRAF V600E mCRC. Negative regulators of Wnt/β-catenin signaling are influenced by hypermethylation. Future clinical trials incorporating these genes as integral biomarkers should consider gene expression given the relevant non-genomic alterations.


Colorectal cancer, BRAF, MAPK, metastasis, epigenetics, signaling, beta-catenin