Trastuzumab emtansine versus capecitabine plus lapatinib in patients with previously treated HER2-positive advanced breast cancer (EMILIA): a descriptive analysis of final overall survival results from a randomised, open-label, phase 3 trial

Abstract

Background: The antibody-drug conjugate trastuzumab emtansine is indicated for the treatment of patients with HER2-positive metastatic breast cancer previously treated with trastuzumab and a taxane. Approval of this drug was based on progression-free survival and interim overall survival data from the phase 3 EMILIA study. In this report, we present a descriptive analysis of the final overall survival data from that trial.

Methods: EMILIA was a randomised, international, open-label, phase 3 study of men and women aged 18 years or older with HER2-positive unresectable, locally advanced or metastatic breast cancer previously treated with trastuzumab and a taxane. Enrolled patients were randomly assigned (1:1) via a hierarchical, dynamic randomisation scheme and an interactive voice response system to trastuzumab emtansine (3·6 mg/kg intravenously every 3 weeks) or control (capecitabine 1000 mg/m² self-administered orally twice daily on days 1-14 on each 21-day cycle, plus lapatinib 1250 mg orally once daily on days 1-21). Randomisation was stratified by world region (USA vs western Europe vs or other), number of previous chemotherapy regimens for unresectable, locally advanced, or metastatic disease (0 or 1 vs >1), and disease involvement (visceral vs non-visceral). The coprimary efficacy endpoints were progression-free survival (per independent review committee assessment) and overall survival. Efficacy was analysed in the intention-to-treat population; safety was analysed in all patients who received at least one dose of study treatment, with patients analysed according to the treatment actually received. On May 30, 2012, the study protocol was amended to allow crossover from control to trastuzumab emtansine after the second interim overall survival analysis crossed the prespecified overall survival efficacy boundary. This study is registered with ClinicalTrials.gov, number NCT00829166.

Findings: Between Feb 23, 2009, and Oct 13, 2011, 991 eligible patients were enrolled and randomly assigned to either trastuzumab emtansine (n=495) or capecitabine and lapatinib (control; n=496). In this final descriptive analysis, median overall survival was longer with trastuzumab emtansine than with control (29·9 months [95% CI 26·3-34·1] vs 25·9 months [95% CI 22·7-28·3]; hazard ratio 0·75 [95% CI 0·64-0·88]). 136 (27%) of 496 patients crossed over from control to trastuzumab emtansine after the second interim overall survival analysis (median follow-up duration 24·1 months [IQR 19·5-26·1]). Of those patients originally randomly assigned to trastuzumab emtansine, 254 (51%) of 495 received capecitabine and 241 [49%] of 495 received lapatinib (separately or in combination) after study drug discontinuation. In the safety population (488 patients treated with capecitabine plus lapatinib, 490 patients treated with trastuzumab emtansine), fewer grade 3 or worse adverse events occurred with trastuzumab emtansine (233 [48%] of 490) than with capecitabine plus lapatinib control treatment (291 [60%] of 488). In the control group, the most frequently reported grade 3 or worse adverse events were diarrhoea (103 [21%] of 488 patients) followed by palmar-plantar erythrodysaesthesia syndrome (87 [18%]), and vomiting (24 [5%]). The safety profile of trastuzumab emtansine was similar to that reported previously; the most frequently reported grade 3 or worse adverse events in the trastuzumab emtansine group were thrombocytopenia (70 [14%] of 490), increased aspartate aminotransferase levels (22 [5%]), and anaemia (19 [4%]). Nine patients died from adverse events; five of these deaths were judged to be related to treatment (two in the control group [coronary artery disease and multiorgan failure] and three in the trastuzumab emtansine group [metabolic encephalopathy, neutropenic sepsis, and acute myeloid leukaemia]).

Interpretation: This descriptive analysis of final overall survival in the EMILIA trial shows that trastuzumab emtansine improved overall survival in patients with previously treated HER2-positive metastatic breast cancer even in the presence of crossover treatment. The safety profile was similar to that reported in previous analyses, reaffirming trastuzumab emtansine as an efficacious and tolerable treatment in this patient population.

Funding: F Hoffmann-La Roche/Genentech.

Figure 1. Trial profile

ITT=intention to treat.*Patients might have had multiple reasons for ineligibility. †Eight patients in the lapatinib plus capecitabine group and five patients in the trastuzumab emtansine group did not receive study treatment; reasons for why these patients did not receive study treatment are not available. ‡Non-protocol therapy included: capecitabine (n=53), lapatinib (n=82), trastuzumab (n=235), pertuzumab (n=12), and trastuzumab emtansine (n=7); patients could have received more than one of the treatments listed because data were collected on individual drugs, not complete regimens. §Non-protocol therapy included: capecitabine (n=254), lapatinib (n=241), trastuzumab (n=177), pertuzumab (n=21), and trastuzumab emtansine (n=11); patients could have received more than one of the treatments listed.

Figure 2. Overall survival

Tick marks on the curve represent censored patients. (A) Descriptive analysis of final overall survival in the intention-to-treat population. (B) Sensitivity analysis for overall survival in which patients were censored at the time of crossover from capecitabine plus lapatinib to trastuzumab emtansine. OS=overall survival.

Figure 3. Overall survival in clinically relevant subgroups at the time of the second (confirmatory) interim overall survival analysis and at the descriptive analysis of final overall survival

Error bars represent 95% CIs. HR=hazard ratio. *Data cutoff July 31, 2012. †Data cutoff Dec 31, 2014. ‡Hazard ratios are from unstratified analyses.