Adam M. Brufsky, MD, PhD
Neratinib (Nerlynx) combined with capecitabine (Xeloda) reduced the risk of disease progression or death by 24% compared with lapatinib (Tykerb) plus capecitabine in patients with HER2-positive metastatic breast cancer who received at least 2 prior lines of HER2-targeted therapy.
The results from the phase III NALA trial, which were presented at the 2019 ASCO Annual Meeting, showed that the HR for PFS favoring neratinib was 0.76 (95% CI, 0.63-0.93; log-rank P
value = .0059). Landmark analysis showed that the PFS curves began to separate after 6 months with 6-month PFS rates of 47% versus 38%, 1-year rates of 29% versus 15%, and 18-month rates of 16% versus 7% for the neratinib arm versus the lapatinib arm, respectively.
A prespecified restricted means analysis for PFS was conducted that was limited to a follow-up of 24 months. The analysis showed a mean PFS of 8.8 months for the neratinib arm compared with 6.6 months for the lapatinib arm (P
For overall survival (OS), there was a numerical but not a statistically significant benefit with neratinib. A prespecified restricted means analysis showed a mean OS of 24.0 months for the neratinib arm compared with 22.2 months for the lapatinib arm (HR, 0.88; 95% CI, 0.72-1.07; log-rank P
= .2086). The analysis was restricted to a follow-up of 48 months.
“The trial results suggest that neratinib and capecitabine is an effective option for treating patients with progressive HER2-positive metastatic breast cancer,” said NALA study investigator Adam M. Brufsky, MD, PhD, University of Pittsburgh Medical Center.
The international, open-label, phase III NALA trial (NCT01808573) enrolled 621 patients with metastatic breast cancer and centrally confirmed HER2-positive disease. Patient characteristics were well balanced at baseline. Nearly 80% of patients in each arm were aged <65 years and about 80% of patients in each arm had visceral disease.
Sixty-nine percent of patients had 2 prior HER2-targeted treatments for metastatic breast cancer, with the remaining 31% having received 3 or more. Prior HER2-targeted regimens included single-agent trastuzumab (Herceptin; 38%), trastuzumab plus pertuzumab (Perjeta; 7.5%), trastuzumab plus T-DM1 (Kadcyla; 20%), and trastuzumab plus pertuzumab plus T-DM1 (35%).
Patients were randomized in a 1:1 ratio to 21-day cycles of either neratinib (n = 307; 240 mg daily) plus capecitabine (1500 mg/m2
on days 1-14) or lapatinib (n = 314; 1250 mg daily) plus capecitabine (2000 mg/m2
on days 1-14). Patients on the neratinib arm also received antidiarrheal prophylaxis with loperamide.
The coprimary endpoints were PFS and OS. The study was conducted under an FDA Special Protocol Assessment stipulating that the study would be considered positive if either of these coprimary endpoint measures were met: P
<.01 for PFS or P
<.04 for OS.
Response data were available for 256 patients in the neratinib arm and 270 patients in the lapatinib arm. The objective response rate was 33% in the neratinib arm versus 27% in the lapatinib group. The clinical benefit rate was 45% versus 36%, respectively. The median duration of response was 8.5 months versus 5.6 months in the neratinib versus lapatinib arms, respectively.
Brufsky also noted, “Fewer patients required intervention for CNS metastases with neratinib plus capecitabine [P
= .043], suggestive of a delay in CNS progression.”
The median duration of treatment was 5.7 months for neratinib compared with 4.4 months for lapatinib. Dose reductions and holds occurred in 24% versus 20% and 48% versus 43% with neratinib versus lapatinib, respectively. The median duration of treatment, dose reductions, and dose holds, were similar for capecitabine between the 2 arms.
All-grade diarrhea occurred in 83% of the neratinib group compared to 66% of the lapatinib arm. Grade 3 diarrhea occurred in 24% versus 13% of the 2 arms, respectively. There were no cases of grade 4 diarrhea reported in either arm. The median time to first onset of grade 2/3 diarrhea was 9 days in the neratinib arm compared with 18 days in the lapatinib arm. The median time to first onset of grade 3 diarrhea was 11 days versus 38 days, respectively.
The median cumulative duration of diarrhea was 7 days for grade 2/3 diarrhea in the neratinib arm versus 9 days in the lapatinib arm. The median cumulative duration of grade 3 diarrhea was 4 days for both arms. Discontinuation due to diarrhea occurred in 2.6% of the neratinib arm compared with 2.3% of the lapatinib arm.
Discontinuations due to treatment-emergent adverse events (TEAEs) occurred in 10.9% of the neratinib arm compared with 14.5% of the lapatinib arm. Beyond diarrhea, the other most common all-grade TEAEs for the neratinib arm were nausea (53% versus 42% in the lapatinib arm), vomiting (46% vs 31%, respectively), hand-foot syndrome (46% vs 56%), decreased appetite (35% vs 22%), and fatigue (34% vs 31%).