Dr. Larkin presented updated results from coBRIM, including progression-free survival, response rate, and treatment outcomes of patients who harbored additional oncogenic mutations in their pretreatment tumor samples at the 2015 ASCO Annual Meeting.
James Larkin, MD, PhD
Patients with advanced melanoma harboring mutations inBRAFV600may benefit from targeted therapy with BRAF inhibitors such as vemurafenib (Zelboraf). However, resistance to vemurafenib monotherapy often occurs via multiple adaptive pathways, including the reactivation of mitogen-activated protein kinase (MAPK) through the MEK signaling pathway.1-3The coBRIM study was designed to address this resistance mechanism via cotreatment of patientsBRAFV600-mutated melanomas undergoing vemurafenib therapy with the MEK inhibitor cobimetinib.3The study’s results were presented at the 2015 ASCO Annual Meeting by James Larkin, MD, PhD, medical oncologist at the Royal Marsden Hospital, London, England. Larkin presented updated results from coBRIM, including progression-free survival (PFS), response rate (RR), and treatment outcomes of patients who harbored additional oncogenic mutations in their pretreatment tumor samples.4
Patients in coBRIM (n = 495) hadBRAFV600-mutantpositive melanoma, as determined using a polymerase chain reaction-based test (COBAS), and were treatment naïve, with histologically confirmed, locally advanced, unresectable stage IIIC and IV melanoma.3Patients were randomized to vemurafenib and placebo, or to vemurafenib with cobimetinib at a 1:1 ratio. Treatment continued until disease progression or unacceptable toxicity, and investigator-assessed PFS was the primary endpoint.3Notably, the continuation of study treatment, or crossover following progression, was not permitted in coBRIM.
Patient characteristics were well balanced between the treatment arms and were consistent with advanced melanoma populations from other trials, with approximately 75% of patients being of Eastern Cooperative Oncology Group (ECOG) performance status 0, approximately 60% having M1c disease, and approximately 45% of having elevations in lactate dehydrogenase (LDH). The initial findings of coBRIM, published in theNew England Journal of Medicinein 2014, showed a significant benefit of the vemurafenib/cobimetinib combination in prolonging PFS after 7.3 months of follow-up (median PFS 9.9 vs 6.2 months; HR, 0.51;P<.0001), with additional benefits in overall response rate (68% vs 45%;P<.001) observed with the combination versus vemurafenib alone.3
Larkin presented the results of an updated analysis of coBRIM, with a median follow up of 14.2 months.4Kaplan-Meier (KM) analysis showed an early separation between the treatment curves, leading to a median PFS of 12.25 months in the combination arm versus 7.20 months in the monotherapy arm (HR, 0.58; 95% CI, 0.460-0.719) for the intent-to-treat (ITT) population in the updated analysis (number of PFS events, 57.9% vs 72.6% for combination vs monotherapy, respectively). These results suggested maintenance of the initially observed treatment benefit with longer follow-up (~1 year after enrollment of the last patient).4The treatment benefit of the combination versus monotherapy was consistently observed among all of the prespecified subgroups that were examined, inclusive of patients with elevated versus normal baseline LDH, and also among patients withBRAFV600KorV600E-mutationpositive disease.4
In the updated analysis, the benefit in RR was also maintained, with 15.8% and 53.8% of patients in the combination arm experiencing a complete response (CR) or partial response (PR), respectively, versus corresponding rates of 10.5% and 39.5%, respectively, in the monotherapy arm, resulting in a difference in overall response rate (ORR) of nearly 20% (69.6% vs 50.0% for the combination and monotherapy groups, respectively). In addition, there was a longer duration of response with the combination versus monotherapy (12.98 months vs 9.23 months).4
It is known that BRAF inhibitor therapy is associated with a robust antimelanoma response, but initial tumor responses are frequently incomplete, providing a source for the development of adaptive resistance and subsequent disease progression.1,2In experimental studies, treatment with BRAF inhibitors has been associated with early upregulation of other signaling mechanisms and increased levels of receptor tyrosine kinases (RTKs).2In particular, a role for the adaptive upregulation of the phosphoinositide 3 kinase (PI3K) and phosphatase and tensin homolog (PTEN)-protein kinase B (AKT) pathway has been implicated in the adaptive resistance of melanomas to targeted therapy.2Thus, an examination of treatment responses in relation to relevant biomarkers and molecular alterations continues to be an important avenue of research in targeted therapies for melanoma.
Before this updated analysis of coBRIM, results from BRIM-2, a phase II study, assessed changes in the activity of MAPK, and cell cycle progression, using serially collected biopsy samples in patients receiving daily doses of vemurafenib (960 mg twice daily).1This study allowed for a molecular analysis of tumor samples (formalin-fixed and paraffin-embedded biopsies) that were obtained before and during vemurafenib therapy, as well as at the time of progression. Results from this analysis showed that patients withBRAFV600mutant tumors had a high activation of MAPK signaling, as assessed by phosphorylated ERK (pERK), and showed high levels of the proliferative markers Ki-67 and cyclin D1, and low levels of the cell cycle inhibitor p27.1
Following vemurafenib therapy, there was a strong decrease in pERK by day 15 of therapy, while there was a less pronounced decrease in phosphorylated MEK (pMEK), and PTEN and phosphorylated AKT (pAKT) were relatively unchanged. These molecular changes were associated with decreased proliferation as assessed by reduced Ki-67 and cyclin D1 levels, and increased p27 levels.1An important finding from this earlier study was the analysis of acquired resistance to vemurafenib, which showed a strong upregulation of MAPK activity and pERK at progression, with no notable changes in PTEN or pAKT, along with corresponding upregulation of Ki-67 and cyclin D1.1These and other previous findings have suggested that the escape from vemurafenib inhibition upon progression is driven, at least in part, by MAPK reactivation. A second objective of the updated analysis of coBRIM was thus to report on treatment outcomes according to several relevant biomarkers in this advanced melanoma population.
DNA for the biomarker analysis in coBRIM could have been from archival or fresh baseline pretreatment tissues, and was examined using Ion Torrent next-generation sequencing (NGS) technology, which entailed a targeted sequencing of 528 hotspots in 17 oncogenes.4The number of samples analyzed was 423 and median sequencing depth was 3600×, and a sample was defined as mutation positive if the mutant allele was detected in 3% or more of the reads. Mutations detected included genes of the RAS/RAF pathway (HRAS,KRAS,NRAS,BRAF), RTKs such as epidermal growth factor receptor (EGFR), fibroblast growth factor receptors (FGFR1, FGFR3), FLT3, KIT, MET, RET, and platelet-derived growth factor receptor alpha (PDGFRα), and a number of other genes of interest, such asABL1,AKT1,AKT2,JAK2, andPIK3CA.4
Of note, 55 of the 423 patients in the study (13%) were found to have coexistent mutations in the genes that were evaluated (inclusive of those patients with one or more coexistent mutations), and 46 (11%) were found to have a mutation inRAS,RAF, or one of the RTKs. Interestingly, an analysis of pERK in the available samples using reverse phase protein assay showed that, compared to patients with wild-type disease (n = 44 samples), those with a comutation inRAS,RAF, orRTK(n = 18 samples) had a detectably higher activation of the MAPK pathway, as assessed by pERK (P=.001), presumably as a result of these coexistent mutations. Despite this, there was no apparent difference in the efficacy of therapy, with an RR of 60% and 61% in patients who were wild-type or who hadRAS/RAF/RTKcomutations, respectively. Similarly, KM analysis for PFS also showed no relevant difference in treatment effect between these two populations of patients (HR, 0.92; 95% CI, 0.63-1.32).4
Taken together, the updated survival results from coBRIM continue to favor use of combination therapy with vemurafenib and cobimetinib over vemurafenib alone; results of the final analysis are expected by the end of 2015. Given the multiple mechanisms accounting for the development of adaptive resistance in patients undergoing vemurafenib therapy, which are driven at least in part by adaptive upregulation of MAPK activity, it is also notable that, based on these updated findings, the presence of relevant comutations in this population does not appear to impact the efficacy of the combination.