Signaling Pathways in Colorectal Cancer


Through close collaborations between laboratory and clinic, we are working to better understand signaling pathways in colorectal cancer. Our goal is to use this understanding to develop mechanism-based therapies. Activation of the MAPK (mitogen-activated protein kinase) pathway, which relays signals that promote cell proliferation and survival to the nucleus, is common in the disease.

Mutational activation of KRAS, BRAF, and NRAS — key signaling proteins in this pathway — occurs respectively in about 40 percent, 8 percent, and 2 percent of metastatic colorectal cancer cases, and activation of the upstream epidermal growth factor receptor (EGFR) due to increased ligand levels can be seen in approximately 10 percent of cases. New agents targeting proteins in this pathway are now being assessed in clinical trials. A deeper understanding of these signaling pathways will enable us to apply anti-EGFR and chemotherapeutic agents more effectively to treat the disease.

Working with medical oncologists, surgical oncologists, pathologists, and scientists, we recently found that mutations in KRAS develop with acquired resistance to cetuximab and panitumumab, which are anti-EGFR targeting antibodies. Experiments in preclinical models surprisingly revealed that KRAS wild-type cells that are sensitive to cetuximab develop hotspots (areas of DNA likely to mutate), activating mutations in KRAS with acquired resistance to cetuximab. These new mutations are causally associated with resistance.

A review of tumors collected before treatment regimens with anti-EGFR antibodies that develop resistance to these agents revealed that many cases (six out of ten) had new hotspot KRAS mutations after treatment. Deep sequencing of patient tumors following cytotoxic chemotherapy confirmed that mutations in KRAS do not develop with chemotherapy. This finding concurs with our understanding of KRAS mutation as an early event. We hypothesize that treatment with anti-EGFR agents creates a selective pressure for MAPK activation, and that KRAS mutations activate this pathway, bypassing the anti-EGFR agents. We hope these findings will spur larger clinical trials and guide the development of new treatments.

In the clinic, we are now testing a new strategy to treat BRAF-mutant metastatic colorectal cancer based on an improved understanding of BRAF signaling and adaptation to selective inhibitors. This is largely based on work from the laboratory of Memorial Sloan Kettering’s Neal Rosen. This clinical trial will test the combination of the RAF inhibitor vemurafenib and panitumumab.

Vemurafenib has minimal single-agent activity in this population. Work from Dr. Rosen’s laboratory suggests a mechanistic explanation. Vemurafenib inhibits mutated BRAF only, and mutated BRAF signals as a monomer. Paradoxically, vemurafenib activates MAPK signaling in tumors with wild-type BRAF, where RAF signals as a dimer that forms in response to RAS activation. Negative feedback loops maintain low activated RAS levels in BRAF mutant tumors. Because of this, RAF dimers are low.

Treatment with a RAF inhibitor releases this feedback inhibition and leads to activation of upstream receptors and RAS, resulting in the formation of RAF dimers and insensitivity to vemurafenib. Several groups have recently reported that this adaptive resistance occurs rapidly in the colon because of relatively high levels of EGFR and its ligands. Based on this data, we hope that combined RAF and EGFR inhibition will lead to a clinical benefit for our patients.