Although fewer deaths were reported with neratinib in patients with HER2-positive breast cancer treated on the phase 3 ExteNET trial, the agent was not found to result in a significant improvement in overall survival after 8 years of follow-up.
Although fewer deaths were reported with neratinib (Nerlynx) in patients with HER2-positive breast cancer treated on the phase 3 ExteNET trial (NCT00878709), the agent was not found to result in a significant improvement in overall survival (OS) after 8 years of follow-up. However, neratinib did demonstrate a trend toward improved central nervous system (CNS) outcomes.1
The long-term efficacy results, which were presented during the ESMO Breast Virtual Congress 2021 showed that in the intent-to-treat (ITT) population, the estimated OS rates at 8 years with neratinib and placebo were 90.1% and 90.2%, respectively (stratified HR, 0.95; 95% CI, 0.75-1.21; P = .6916).
Subgroup analyses indicated that neratinib resulted in a numerical improvement in terms of OS in patients with hormone receptor–positive disease (n = 1631). The estimated 8-year OS rate in the neratinib arm was 91.6% in this population vs 90.1% with placebo (HR, 0.80; 95% CI, 0.58-1.12). In the subgroup of patients with hormone receptor–negative disease (n = 1209), the agent irreversible pan-HER TKI did not appear to improve OS. The estimated OS rates at 8 years in this subgroup with neratinib and placebo were 88.1% and 90.3%, respectively (HR, 1.18; 95% CI, 0.83-1.69).
Additionally, in the population of patients with centrally confirmed HER2-positive disease (ccHER2+), the 8-year OS rate was 91.2% in those who received neratinib vs 90.8% in those given placebo (stratified HR, 0.86; 95% CI, 0.63-1.18; P = .3585).
“The data suggest an association between neratinib and improved OS in patients with hormone receptor–positive disease when compared with patients with hormone receptor–negative tumors, which is consistent with the primary 2-year and 5-year analyses of invasive disease-free survival [iDFS] and distant DFS [DDFS],” lead study author Arlene Chan, of Breast Cancer Research Centre-WA and Curin University, and colleagues, wrote in a poster on the data.
Previously, neratinib was found to improve iDFS vs placebo, when administered as an extended adjuvant treatment for 1 year, in patients with HER2-positive early breast cancer following trastuzumab (Herceptin)-based adjuvant treatment.2,3
Earlier data from ExteNET showed that the agent had more marked benefits in those with hormone receptor positivity who had started treatment within 1 year of completing trastuzumab (HR+/≤1-year; n = 1334), and in those who were high risk and had residual disease following neoadjuvant therapy (HR+/≤1-year no pCR; n = 295).3,4
In the ITT population, the HR for iDFS was 0.73 (95% CI, 0.57-0.92) with neratinib and the 5-year iDFS absolute benefit rate was 2.5%; in this population, the HR for DDFS was 0.78 (95% CI, 0.60-1.01) with a 5-year DDFS absolute benefit rate of 1.7%. In the HR+/≤1-year subgroup, the HR for iDFS was 0.58 (95% CI, 0.41-0.82) with a 5-year iDFS absolute benefit rate of 5.1%, while the HR for DDFS was 0.57 (95% CI, 0.39-0.83) and the 5-year DDFS absolute benefit rate was 4.7%). In the HR+/≤1-year no pCR subgroup, the HRs for iDFS and DDFS were 0.60 (95% CI, 0.33-1.07) and 0.61 (95% CI, 0.32-1.11), respectively, while the 5-year iDFS and DDFS absolute benefit rates were 7.4% and 7.0%, respectively.
The agent has been shown to have CNS efficacy in the metastatic setting, according to data from 3 trials, but to date, no HER2-directed agent has been found to prevent CNS metastases. As such, options that prevent CNS recurrence as the first iDFS event continue to represent an important unmet need in HER2-positive disease.
In the final protocol-defined analysis of OS from the trial, investigators shared data from the ITT population and the ccHER2-positive population. They also presented findings from descriptive analyses of OS and CNS outcomes in patient subgroups of clinical interest, such as the HR+/≤1-year and HR+/≤1-year no pCR populations.
The primary end points of the trial were OS, which was defined as time from randomization to date of death of any cause, and CNS-specific end points like cumulative incidence of CNS recurrences and CNS-DFS.
The preplanned OS analyses were done in the ITT population and the ccHER2-positive population and were powered for 248 events and a HR of 0.70 with 80% power and a one-sided 0.025 type I error rate. The descriptive analyses for OS and 5-year CNS outcomes were done in the HR+/≤1-year population and high-risk patient subsets. Investigators used the Kaplan-Meier method to estimate survival rates for OS and CNS-DFS. Using Cox proportional hazard models, the HR and 95% confidence intervals were estimated; they were assessed with a log-rank test. Moreover, investigators evaluated the cumulative incidence of CNS recurrences through the use of competing risks analysis and tested it with Gray’s method.
A total of 2840 patients were randomized to receive study treatment, with 1420 patients in each group. Of these patients, 1631 had hormone receptor–positive disease; 82% of these patients (n = 1334) began treatment within 1 year of previous trastuzumab and thus represented the HR+/≤1-year population. Among those in this subset, 354 (27%) had received neoadjuvant therapy; 295 of these patients had residual invasive disease at the time of study entry, thus constituting the HR+/≤1-year no pCR population.
The median follow-up for OS was 8.1 years (range, 0.0-9.9) and 54.3% (n = 1542) completed at least 8 years of follow-up.
Additional data showed that within the HR+/≤1-year population, 7.9% (n = 53/670) and 10.2% (n = 68/664) of patients in the neratinib and placebo arms, respectively, died. The estimated OS rates at 8 years in the neratinib group was 91.5%, while it was 89.4% in the placebo group; this translated to an absolute benefit rate of 2.1% (HR, 0.79; 95% CI, 0.55-1.13).
Moreover, in the HR+/≤1-year population with no pCR (n = 295), the OS rates at 8 years with neratinib and placebo were 91.3% and 82.2%, respectively, which translated to a 9.1% absolute benefit rate (HR, 0.47; 95% CI, 0.23-0.92). Among patients in the HR+/≤1-year population who had a pCR (n = 38), the 8-year OS rates in the neratinib and placebo arms were 93.3% and 73.7%, respectively; this corresponded to a 19.6% absolute benefit (HR, 0.40; 95% CI, 0.06-1.88).
Regarding CNS outcomes, patients who received neratinib experienced a lower number of CNS events in all populations: ITT, HR+/≤1-year, and those who received neoadjuvant treatment (pCR and no pCR).
The cumulative incidence of CNS recurrences in the HR+/≤1-year population (n = 1334) was 0.7 with neratinib vs 2.1 with placebo. Among the 980 patients who did not receive previous neoadjuvant therapy, the cumulative incidence of CNS recurrences was 0.7 with neratinib vs 1.5 with placebo. In the 354 patients who did receive prior neoadjuvant treatment, these incidences were 0.7 vs 3.7, respectively. Among those without pCR (n = 295), these incidences were 0.8 vs 3.6 in the neratinib and placebo groups, respectively; in those with pCR (n = 38), these incidences were 0 and 5.0, respectively.
“Neratinib is the first HER2-directed agent to show a trend toward improved CNS outcomes in early-stage HER2-positive breast cancer,” the investigators concluded. “In all groups (ITT, HR+/≤1-year, and no pCR), consistently fewer CNS events were observed in the neratinib arm compared with placebo.”