Phase 1b/2 safety and efficacy of TAK-228 (MLN0128), plus exemestane (E) or fulvestrant (F) in postmenopausal women with ER + /HER2- metastatic breast cancer (MBC)

Abstract

Background: TAK-228, an investigational, oral, highly selective, ATP-competitive inhibitor of TORC1/2, may mitigate feedback activation within the PI3K/AKT/mTOR pathway. This represents a potential cause of resistance to TORC1 selective inhibitors. Thus TORC1/2 inhibition may restore sensitivity to endocrine therapies in patients ( pts) who have progressed on such agents plus everolimus. We report the recommended phase 2 dose (RP2D), safety, pharmacokinetics (PK), and preliminary efficacy of TAK-228 + E or F. Methods: Postmenopausal (≥1 y) women ≥18 y with advanced or MBC following everolimus + E or F were eligible. Phase 1b: part 1, unmilled TAK-228 5 mg continuously once-daily (QD) + E 25 mg QD or F 500 mg monthly; part 2, milled TAK-228 3 or 4 mg QD + E or F. Modified 3 + 3 dose escalation rules were applied, and pts were treated until disease progression. Results: N = 24 were enrolled into phase 1b (median age 58.5 y [33-75]; median prior lines of therapy/chemotherapy 5 [1-10]/1 [0-2]). In part 1 (n = 12), pts received 5 mg unmilled TAK-228 QD + E or F (each n = 6). No dose-limiting toxicities (DLT) were seen, with no apparent differences in tolerability between the two groups. In part 2 (n = 12), 6 pts received 3 mg milled TAK-228 QD and no DLTs were seen. Another 6 pts received 4 mg milled TAK-228 QD; 1 pt treated with TAK-228 + E had DLTs of grade 3 nausea and diarrhea. The most common grade ≥3 drug-related adverse events (≥2 pts) were: diarrhea (13%) and increased ALT, fatigue, hyperglycemia, nausea, rash and stomatitis (each 8%). PK data suggest there was a dose related increase in total systemic exposure from 3 to 5 mg TAK-228, with no readily apparent difference between milled and unmilled TAK-228. Five pts (21%) had an overall response (1 complete response [CR], 4 partial responses [PRs]), and 12 (50%) had stable disease (SD). The overall response rate was 21% (7% E vs 44% F) and the disease control rate (CR + PR + SD) was 71% (60% E vs 89% F). The median duration of clinical benefit was 6 mos for TAK-228 + E vs 10 mos for TAK-228 + F. Conclusions: Safety was established for milled TAK-228 at the RP2D of 4 mg QD + E or F. The encouraging preliminary efficacy warrants further evaluation in the phase 2 portion.