Safety of trastuzumab emtansine (T-DM1) in patients (pts) with HER2-positive locally advanced or metastatic breast cancer (mBC): Final results from KAMILLA Cohorts 1 (global) and 2 (Asia).

1039 Background: KAMILLA is an open-label, single-arm, phase 3b safety study of T-DM1 in pts with HER2-positive advanced BC (NCT01702571). The treated (safety) population of KAMILLA comprises 2 cohorts: a larger global Cohort 1 (n=2002) and a smaller Asia Cohort 2 (n=181 [China, n=154; Thailand, n=15; Indonesia, n=12]). Here we report results from Cohort 2 in the context of those previously reported for Cohort 1. Methods: Pts had HER2-positive, locally advanced or mBC with progression after chemotherapy and anti-HER2 therapy or ≤6 months (mo) of completing adjuvant therapy. T-DM1 3.6 mg/kg was given intravenously every 3 weeks until disease progression, consent withdrawal, or unacceptable toxicity. Primary endpoints were grade ≥3 (G≥3) adverse events of primary interest (AEPIs), specifically hepatic events, allergic reactions, thrombocytopenia (TCP), and hemorrhage events; all other G≥3 treatment-related AEs (TRAEs); and all-grade pneumonitis. Results: As of 31 July 2019,KAMILLA enrolled 2185 pts (Cohort 1, n=2003; Cohort 2, n=182), of which 2002 and 181 in each cohort, respectively, received ≥1 study dose and were included in the safety population. Baseline characteristics were generally similar between cohorts. Median (range) T-DM1 exposure was 5.6 mo (0–46) for Cohort 1 and 5.0 mo (0–31) for Cohort 2. The overall G≥3 AEPI rate was higher in Cohort 2 vs Cohort 1 (Table), mostly driven by a higher G≥3 TCP rate in Cohort 2. In Cohort 2, G≥3 TCP (the most frequently reported G≥3 AEPI) did not appear to be associated with G≥3 hemorrhagic events — the majority of G≥3 TCP events (128/138) fully resolved, with a duration of ≤15 days for 98/138 of these events. G≥3 TRAE rates were 18.4% in Cohort 1 and 48.6% in Cohort 2, the latter mainly due to TCP and platelet count decreased; any-grade pneumonitis rates were 1.0% and 2.2%, respectively. No other safety signals were identified. Median progression-free survival and overall survival were similar for both cohorts (Table). Conclusions: These data confirm prior observations in an Asian subgroup of the phase 3 EMILIA trial of T-DM1 in pts with previously treated mBC. Although the G≥3 TCP rate was higher in the Asia cohort, the majority of these events resolved fully. OS and PFS were similar in both cohorts. These data reinforce the favorable T-DM1 benefit-risk profile in mBC. Clinical trial information: NCT01702571 .[Table: see text]