Tuma, Rabiya S. PhD
WASHINGTON—Three types of cancer are commonly driven by mutations in the BRAF gene: melanoma, colon, and thyroid. Responses to the BRAF inhibitor vemurafenib, though, are very different in melanoma and colon cancer, which are the diseases with the most data thus far. In melanoma, vemurafenib produces dramatic responses, whereas patients with BRAF-mutant colorectal cancer rarely respond to the drug.
Now, with an ongoing trial testing vemurafenib in BRAF-mutant thyroid cancer, non-trial data already call the approach into question. Preclinical data, though, suggest that combining vemurafenib with lapatinib, a small molecule inhibitor of HER2/HER3 signaling, may reduce the intrinsic resistance, researchers reported here at the American Association for Cancer Research Annual Meeting (Abstract 3402).
“Based on our experience, thyroid cancer patients perhaps lie in between [the other] two,” said James Fagin, MD, Chief of the Endocrinology Service at Memorial Sloan-Kettering Cancer Center, during a presentation at the meeting. “They are certainly not responding as well as patients with melanoma, but we are probably seeing slightly better responses than in colorectal cancer patients.”
The primary investigator for the company-sponsored Phase II trial, Maria E. Cabanillas, MD, Assistant Professor in the Department of Endocrine Neoplasia and Hormonal Disorders at the University of Texas MD Anderson Cancer Center, said in an email message that the trial completed enrollment last year and the hope is that the results will be presented sometime in the next year or so.
However, based on other data—including the Phase I trial data, currently in press, her “off-label” experience with vemurafenib, and her ongoing neoadjuvant trial—she said she concurs with Fagin's assessment: “Compared with the results for melanoma, the responses to vemurafenib are not as impressive. These patients can have partial responses, but we have not seen complete responses—as in melanoma patients—so far at MD Anderson.”
Vemurafenib Triggers HER3 Signaling
To learn why thyroid cancer patients lack the robust responses seen in melanoma patients, Cristina Montero-Conde, PhD, a postdoctoral fellow in Fagin's laboratory, examined what happens in thyroid cancer cells after treatment with vemurafenib. The results showed that unlike melanoma cells, which show sustained inhibition of BRAF signaling, including a lasting drop in phosphorylated ERK (pERK), inhibition of BRAF signaling is only transient in the thyroid cancer cells. Within 48 hours of adding vemurafenib to the cell culture, pERK is back to pre-treatment levels.
She subsequently found that the rebound in signaling is induced by activation of signaling through HER2/HER3 heterodimers. It appears that BRAF inhibition allows transcription of HER3, which is blocked when BRAF signaling is active.
The newly produced HER3 receptor protein binds to neuregulin, which is constitutively secreted by thyroid cancer cells. The activated HER3 protein then dimerizes with HER2 and initiates signaling through the PI3 kinase/AKT pathway, effectively bypassing the BRAF blockade.
In contrast, colon cancer cells or melanoma don't secrete neuregulin and, therefore, are not susceptible to this form of innate resistance to vemurafenib.
Benefits for the Combination
Based on those observations, the investigators hypothesized that combining vemurafenib and lapatinib would prevent the resistance by blocking BRAF signaling and the HER3/HER2 bypass. As predicted, the combination prevented reactivation of pERK and activation of signaling through AKT.
Additionally, in the presence of lapatinib, the concentration of vemurafenib required to kill the cancer cells dropped substantially, suggesting that the combination is synergistic.
The team saw similar results in a genetic mouse model of BRAF-mutant thyroid cancer. Tumors grew in the presence of either single agent, but when the team treated animals with vemurafenib plus lapatinib, there was significant tumor shrinkage in four out of five animals and stable disease in the fifth.
The findings are “very exciting,” Cabanillas said. They do need, though, of course, to be replicated in humans before a combination trial is initiated in patients. Her team is in the process of doing that in their ongoing neoadjuvant study.
“We are studying the mechanisms of resistance in patients before and after treatment with vemurafenib,” she said, and subsequently presented some of the findings during the “Trials in Progress” session at the American Society of Clinical Oncology Annual Meeting (see next page). “Most likely there will be more than one mechanism of resistance—in melanoma there are at least six such mechanisms of resistance—and therefore this needs further study before proceeding with a clinical trial,” she said.
Asked for a comment for this article, Robert I. Haddad, MD, Chief of the Head and Neck Oncology Program at Dana-Farber Cancer Institute, called the data from Fagin's group exciting: “This gives us a framework when we start thinking about which agents to combine. Here we have preclinical data to support this combination.”
From a practical point of view, though, he cautioned against anyone trying the combination off-study. “These drugs have side effects; they are not benign. So if you are going to combine them, you really have to be on a clinical trial.”