LAUSR

OBJECTIVES Common therapies for HER2-positive advanced breast cancer (ABC) are associated with heterogeneity in prognosis and treatment benefit. Prognostic models of survival outcomes with ado-trastuzumab-emtansine (T-DM1) have not been evaluated. MATERIAL AND METHODS A pre-treatment prognostic model for overall survival (OS) and progression-free survival (PFS) based on clinicopathological factors was developed for HER2-positive ABC patients initiating second-line and later T-DM1 using data from the randomised clinical trials EMILIA and TH3RESA (n = 893). Pre-treatment prognostic groups were identified via recursive partitioning analysis. RESULTS The most significant OS/PFS pre-treatment risk predictors were metastatic sites count (≤2 versus > 2) and ECOG performance-status (0 versus ≥ 1) (P < 0.05). Based on these two factors, patients can be characterised as one of three prognostic groups (good = 0 factors; intermediate = 1 factor; poor = 2 factors). The prognostic groups were identified as significantly associated with OS (P < 0.001) and PFS (P < 0.001). Median OS for the good, intermediate and poor prognostic groups were 40 (95%CI: 36-48), 25 (23-30) and 16 (14-19) months, respectively, and median PFS was 12 (10-15), 8 (7-9) and 6 (4-7) months. CONCLUSION Pre-treatment prognostic groups with significant differences in OS and PFS for HER2-positive ABC patients initiating second-line and later T-DM1 were identified. For HER2-positive ABC patients considering initiating second-line and later T-DM1, the prognostic groups enable more personalized expectations of disease control, survival and absolute treatment benefit.