Background Trastuzumab emtansine (T-DM1) was studied in the EMILIA trial as a second line of treatment for HER-2 positive metastatic breast cancer (MBC), following the trastuzumab-taxane first-line scheme. However, the demonstration of the superiority of pertuzumab-trastuzumab-taxane as the first line of MBC after the CLEOPATRA trial, means that T-DM1 is currently used in a different scenario from the one studied in the EMILIA.

Purpose Our objective is to provide more real-world data of the efficacy of T-DM1 in specific subgroups of patients, in order to know which patients will benefit more from T-DM1 therapy.

Material and methods A retrospective, longitudinal, observational study was conducted between December 2016 and September 2017. Patients who started T-DM1 for MBC between October 2014 and September 2017 were included. Patients who had received T-DM1 in clinical trials were not included.

Data collected were demographic data, previous treatments for MBC, Eastern Cooperative Oncology Group (ECOG) status at baseline, hormone receptor status, dates of therapy start, progression and discontinuation, and adverse events. Subsequently, data were analysed with Stata14^®.

Results Thirty-one patients were included. 32.3% of them had a hormone-sensitive tumour. The median of previous treatments for the MBC was 2 (range 0–6). Median progression-free survival (PFS) for all patients was 4.14 months (9.6 months in EMILIA). 38.7% of patients had a serious haematological adverse event.

19.3% had previously received trastuzumab-pertuzumab-taxane. They achieved a median PFS of 2.86 months, compared to 4.47 months for non-treated patients.

35.5% of patients were previously treated with more than one previous scheme. They had a median PFS of 3.88 months, compared to 6.37 for more pretreated patients.

Conclusion There may be a profile of patients who respond in an excellent manner to T-DM1, as others appear not to have a good response.

This could be related to the number of previous regimens received and number of different regimens. It seems not to be related, in our study, with ECOG or age at the beginning of treatment.

Characterising patients prior to initiating therapy may be complex but advisable to obtain optimal results with the therapy chosen.

No conflict of interest