Abstract P5-18-04: Tolerability and efficacy of targeting both mTOR and HER2 signaling in trastuzumab-refractory HER2+ metastatic breast cancer

Background: Single agent approaches to targeting the HER2-PI3K-AKT-mTOR pathway in HER2+ breast cancer have been met with limitations. The efficacy of direct inhibition of HER2 may be limited by mutational activation of the downstream PI3K pathway. Pharmacologic inhibition of the downstream mTOR kinase results in feedback upregulation of upstream receptor tyrosine kinases such as HER3 that may diminish the benefit of this class of agents. We hypothesize that dual inhibition of HER2 and mTOR can overcome feedback signaling and provide clinical benefit. We therefore conducted a phase I/II trial examining the tolerability and efficacy of temsirolimus (T), an mTOR inhibitor, and neratinib (N), a HER1/2 kinase inhibitor, in patients with trastuzumab-refractory HER2+ metastatic breast cancer (MBC). Tumor biopsies were obtained to identify specific genotypes and biomarkers that might predict response to the T-N combination. Methods: The phase I study utilized a 3+3 dose escalation design to determine the maximum tolerated dose (MTD) of T (IV weekly) with N (fixed dose 240 mg oral daily) in patients with HER2+ trastuzumab-refractory MBC. Cycle length was 4 weeks; response was evaluated radiographically every 8 weeks and toxicity assessed every 2 weeks. The phase II study is ongoing and utilizes a Simon two-stage design to assess the overall response rate by RECIST to this combination at the established MTD. Patients must have HER2+ trastuzumab-refractory MBC with no more than four lines of prior therapy for advanced disease. All patients undergo biopsy of metastatic disease for biomarker assessment prior to therapy. Activating mutations in PIK3CA are assayed using the Sequenom MassARRAY system and expression of PTEN assessed by immunohistochemistry utilizing a published scoring system. Next generation sequencing on ∼250 candidate cancer related genes is being performed on a subset of tumors. Results: Eight patients enrolled in the phase I trial and received a median of 5 (range 1–24) cycles of therapy. All patients had received trastuzumab and a median of 5 (range 2–12) prior lines of therapy. Grade 3 diarrhea was the dose-limiting toxicity. The MTD was determined to be T at 8 mg IV weekly with neratinib at 240 mg daily. As of June 1, 2012, nineteen patients (target accrual thirty-four patients) have enrolled in the phase II trial and received a median of 2 (range 1–8) cycles of therapy. Among the 25 patients treated at the MTD (phase I n=6; phase II n=19), the most frequent treatment-related grade 2/3 events were diarrhea (gr 2 44%/gr 3 24%), mucositis (20%/12%), hyperglycemia (24%8%), and rash (20%/0%). Twenty patients treated at the MTD are evaluable for response; 11 patients had PR (8 confirmed PRs) and 1 had SD≥6 months for a RR of 40%. Molecular analyses revealed a set of patients with more than 1 mutation in the HER2-PI3K-AKT-mTOR pathway had prolonged responses to T-N therapy. Conclusions: Temsirolimus and neratinib is a tolerable regimen that has clinical activity in trastuzumab-refractory HER2+ MBC patients. The phase II study is ongoing and additional efficacy, safety, and biomarker data will be presented. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-18-04.