Signaling network connects cancer types and collaborators
What could have prompted the scientific director of the Center for Thoracic Cancers at MGH to cross over into colorectal cancer research? The answer lies in a signaling network that links the two cancer types. “Cancers of the lung and the gastrointestinal tract use similar signaling pathways to survive and grow,” explains Jeffrey Engelman, MD, PhD (MGH), a former postdoc in the laboratory of Lewis Cantley, PhD (BIDMC), a world-renowned authority on the PI3 kinase (PI3K) pathway. Their shared scientific interest in the PI3K pathway is what drew the two investigators to collaborate on a SPORE project of the Gastrointestinal Malignancies Program.
The PI3K pathway, which drives many epithelial cancers, plays an important role in many colorectal cancers (CRC). This is underscored by the presence of activating mutations in the PIK3CA gene, which encodes PI3K, in 20-30% of cases. In their SPORE project, co-PIs Engelman and Cantley are studying how regulation of the PI3K pathway influences sensitivity of CRC to cetuximab (Erbitux), a humanized monoclonal antibody against the epidermal growth factor receptor (EGFR). Cetuximab elicits responses in a subset of patients with metastatic CRC, but to date, its efficacy has been modest.
Receptor tyrosine kinases, like EGFR, collude in the initiation and progression of several cancer types including lung cancer, breast cancer, and gastrointestinal stromal tumor (GIST), says Engelman. When targeted therapies for these tumors are effective, they typically inhibit the RTK and, at the same time, downregulate the PI3K signaling network — a dual mechanism that increases effectiveness. Cetuximab, by contrast, appears to fall short in this respect. In their recent studies of cetuximab-sensitive CRC cells, Engelman and Cantley found that while cetuximab affected the ERK pathway, it had less impact on the PI3K pathway. “We think that’s the reason why cetuximab as a single agent does not produce dramatic responses,” concludes Engelman.
Based on their work in cell lines, investigators also believe that most colorectal cancers are not as addicted to EGFR as are lung cancers with mutant EGFR. Unlike those cancers, CRCs lack an alteration in the EGF receptor that predicts for sensitivity to cetuximab. While searching for markers in CRC, scientists have discovered that cells with Kras or Braf mutations or amplification of Her2 are resistant to cetuximab, calling for an alternate therapeutic approach for patients with one of these genetic profiles.
“For the subset of CRCs that do benefit from cetuximab, we’re hoping we can transform mild responses into gangbusters,” says Engelman. Toward that end, he and colleagues have been testing combinations of cetuximab with numerous other agents to boost the potency and synergism of anti-EGFR therapy. In cell lines, investigators have experimented combining cetuximab with inhibitors of the insulin growth factor receptor, the platelet-derived growth factor receptor, as well as PI3K inhibitors that are already in clinical development. They will soon apply the same therapeutic strategies in xenografts and then complement those studies with experiments in genetically engineered models of CRC to confirm their in vitro findings.
“Based on knowing the KRAS, BRAF, and PIK3CA mutational status of CRC cells, we are now developing a very good idea of which therapies work in the lab,” notes Engelman. For example, combining a PI3K inhibitor with cetuximab works well in KRAS wild type CRC cells, but fails to induce any response in KRAS mutant cells. In those cases, investigators treat with an ERK pathway inhibitor in combination with either a PI3K inhibitor or an RTK inhibitor that prevents the activation of PI3K. The insights gained from both cell cultures and animal models will provide the data needed to initiate clinical trials, which are already planned.
“In cancers of the lung, the GI tract, and the breast, we know you have to downregulate PI3K and ERK signaling to get dramatic responses,” says Engelman. “That’s why we may eventually treat some KRAS-mutant lung cancer quite similarly to KRAS-mutant colorectal cancer — it’s all in the genetics and signaling networks.”