Inhibiting both BRAF and EGFR within the MAPK pathway proves promising for staunching tumor growth in select cases of metastatic colorectal cancer (mCRC). Interim results of the phase II SWOG S1406 trial reveal that the addition of vemurafenib to cetuximab and irinotecan significantly prolonged progression-free survival (PFS) and the disease control rate in patients with BRAFV600E-mutated mCRC (Abstract 520).
Dr. Scott Kopetz
Previous work indicates that vemurafenib can elicit a tumor response in mCRC harboring the BRAF V600E mutation, although transiently. When vemurafenib blocks BRAF activity and cuts off signaling within the MAPK pathway, this kicks on a feedback mechanism leading to upregulation of upstream EGFR, once again driving signaling through the MAPK pathway upon which these tumors are so dependent. To prevent this molecular escape route, the investigators hypothesized that adding the EGFR inhibitor cetuximab along with irinotecan—already a standard second-line regimen for patients with KRAS/NRAS wild-type disease—may provide sufficient inhibition to stop mCRC growth.
SWOG S1406 tested this approach. A total of 99 patients with BRAFV600E-mutated and RAS wild-type mCRC enrolled within the study and underwent random assignment to irinotecan and cetuximab either with or without vemurafenib. Patients in the control arm who experienced disease progression could switch over to the experimental arm and receive add-on vemurafenib.
Individuals with either measurable or nonmeasurable metastatic disease were eligible for the study to permit patients with peritoneal lesions who are typically excluded from clinical trials. In addition, participants could have received one or two prior regimens, including those containing irinotecan, but not any EGFR, BRAF, or MEK inhibitors. Roughly 40% of patients had received prior irinotecan, and approximately 36% had received two prior regimens.
The addition of vemurafenib to irinotecan and cetuximab more than doubled the primary endpoint of median PFS, extending the duration from 2.0 months to 4.4 months (HR 0.42, 95% CI [0.26, 0.66]; p = 0.0002).
The objective response rate was higher with the addition of vemurafenib to cetuximab/irinotecan (16% vs. 4%). However, the triple-drug combination excelled with regard to the disease control rate, which reached 67% with the addition of vemurafenib and 22% without.
Some severe toxicities did occur significantly more often with the addition of vemurafenib to irinotecan and cetuximab, including grade 3/4 neutropenia (28% vs. 7%), anemia (13% vs. 0%), and nausea (15% vs. 0%). “These rates may be attributed to an increased duration of exposure and are similar to prior second-line studies of cetuximab and irinotecan,” Dr. Kopetz said. No increases in skin toxicity were observed—as can occur with vemurafenib—nor were any new safety signals.
Dr. Kopetz outlined several future analyses of the SWOG S1406 data that are planned. Overall survival will be analyzed when the data are mature. Of note, 48% of patients in the irinotecan/cetuximab control arm crossed over to the experimental arm at progression and began to receive vemurafenib. Other planned work includes subgroup analysis according to microsatellite instability status, as well as an examination of the emergence and mechanisms of treatment resistance. Patient-derived xenografts have also been generated from a subset of patients at baseline for pharmacodynamic studies, marking the first time this has been accomplished within a cooperative group study.
Dr. Kopetz was asked whether the SWOG findings warrant routine use of vemurafenib in combination with cetuximab and irinotecan in patients with BRAFV600E-mutated disease in the clinic. “It’s a great question. This is a rare subset of [patients with] CRC. I think that … replicating this original study may not be worth it,” he said. Dr. Kopetz then alluded to additional efforts underway to potentially improve on the vemurafenib/cetuximab/irinotecan outcomes by adding an MEK inhibitor to the combination. In the interim, however, Dr. Kopetz said he welcomed the input of the community regarding how the SWOG data should be applied to clinical practice.
—Kara Nyberg, PhD