ONCAlert | 2017 San Antonio Breast Cancer Symposium

Dabrafenib, Trametinib Pooled Analysis Confirms Safety Profile for Advanced Melanoma

Darcy Lewis
Published Online: 3:56 PM, Mon June 6, 2016
The combination of dabrafenib (Tafinlar) and trametinib (Mekinist) shows promise in the treatment of advanced melanoma. Safety data and adverse event (AE) data have been available for individual registration trials involving the combined regimen; however, no pooled analysis had previously been performed.
The pooled analysis, presented at the 2016 ASCO Annual Meeting, examined the phase II BRF113220 trial and the phase III COMBI-d and COMBI-v trials. It shows that the safety profile of dabrafenib and trametinib is consistent with reports from the three individual studies.
The pooled analysis also revealed that the highest AE rates occur during the first 6 months of treatment. Analysis showed that pyrexia, the most common AE, does not predict patient response or disease progression.
“This is important information because it means that oncologists should treat pyrexia in the most effective way they can, without the concern that doing so will compromise patients’ treatments,” said Jean-Jacques Grob, MD, professor at Aix-Marseille University and APHM Hospital CHU Timone, in Marseille, France.
A total of 614 patients were randomized to receive dabrafenib (150 mg BID) plus trametinib (2 mg once daily) in all three trials: BRF113220 (n = 55), COMBI-d (n = 209), and COMBI-v (n = 350). Patient safety assessments included physical examinations and monitoring for vital signs, AEs, and laboratory data (which included hematological monitoring, urinalysis, electrocardiograms and echocardiograms).
The researchers considered a number of baseline factors, including age, sex, disease stage, number of organs, BRAF mutation status (V-600E/K), and whether or not patients developed pyrexia. They also used patient treatment status as a baseline factor, specifically, whether patients had received prior non-ipilimumab treatment or prior ipilimumab adjuvant treatment.
Of the 614 patients, 603 (98%) had at least one adverse event. The most common were: pyrexia (n = 350, 57%), nausea (n = 224, 36%), fatigue (n = 223, 36%), chills (n = 213, 35%), and diarrhea (n = 210, 34%). In all, 20 types of AEs had frequency rates of 10% of patients or more. The median follow-up was 20 months.
Other AEs of interest occurred, too, including hyperkeratosis (n = 41, 7%), palmar-plantar erythrodysesthesia (n = 23, 4%), photosensitivity reaction (n = 22, 4%), skin papilloma (n = 14, 2%), and squamous cell carcinoma/keratoacanthoma (n = 13, 2%).
A significant number of patients (n = 165, 27%) received the combined regimen for 24 months or longer.
Grob and his colleagues found that AE incidence generally peaked within the first 6 months of treatment, then declined. Consider the pyrexia rate: from 0-6 months, more than half the patients developed pyrexia (n = 350, 57%), but that rate fell dramatically in months 6 to 12, to 26%. At that point the pyrexia rate remained unchanged from months 12 to 18 at 26%, then dropped slightly in months 18 to 24 to 22%).
Alternatively, the AE incidence rate for the other common AEs declined to less than 10% by months 18 to 24.
The study also included multivariate Cox proportional hazards model analysis of outcomes by baseline factors and pyrexia during treatment. For each finding, the researchers calculated the hazard ratio, 95% confidence interval, and the P value for progression free survival (PFS), complete response (CR) versus non-CR, and CR/partial response (PR) versus non-CR/PR. For PFS for pyrexia events during treatment, the HR was .949, the confidence interval was .773 to 1.166, and the P value was .6194.
The researchers analyzed the patient demographics of age and sex. They also examined baseline disease characteristics of stage, BRAF mutation status, ECOG performance status, lactate dehydrogenase (LDH), number of organ sites, and lesion diameter. Additionally, the analysis included prior treatment status.
Grob noted that as far as limitations, one of the studies had no control arm and that the control arms of the other two trials differed from each other.
“As soon as you mix trials, you risk losing some information due to the differing designs,” he said. “But by increasing the number of patients [in a pooled analysis], you may see data that are more relevant than in each trial separately, like our finding that pyrexia is not prognostic in advanced melanoma.”
Grob JJ, Flaherty K, Long GV, et al. Pooled analysis of safety over time and link between adverse events and efficacy across combination dabrafenib and trametinib (D+T) registration trials. J Clin Oncol 34, 2016 (suppl; abstr 9534).

Copyright © TargetedOnc 2017 Intellisphere, LLC. All Rights Reserved.