A phase I study of the oral gamma secretase inhibitor R04929097 in combination with gemcitabine in patients with advanced solid tumors (PHL-078/CTEP 8575).

PURPOSE: To establish the recommended phase II dose of the oral γ-secretase inhibitor RO4929097 (RO) in combination with gemcitabine; secondary objectives include the evaluation of safety, tolerability, pharmacokinetics, biomarkers of Notch signaling and preliminary anti-tumor activity. METHODS: Patients with advanced solid tumors were enrolled in cohorts of escalating RO dose levels (DLs). Tested RO DLs were 20 mg, 30 mg, 45 mg and 90 mg. RO was administered orally, once daily on days 1-3, 8-10, 15-17, 22-24. Gemcitabine was administered at 1,000 mg/m(2) on d1, 8, and 15 in 28 d cycles. Dose limiting toxicities (DLTs) were assessed by CTCAE v4. Serial plasma was collected for RO (total and unbound) and gemcitabine pharmacokinetic analysis. Biomarkers of Notch signaling were assessed by immunohistochemistry in archival tissue. Antitumor activity was evaluated (RECIST 1.1). RESULTS: A total of 18 patients were enrolled to establish the recommended phase II dose. Of these, 3 patients received 20 mg RO, 7 patients received 30 mg RO, 6 patients received 45 mg RO and 2 patients received 90 mg RO. DLTs were grade 3 transaminitis (30 mg RO), grade 3 transaminitis and maculopapular rash (45 mg RO), and grade 3 transaminitis and failure to receive 75 % of planned RO doses secondary to prolonged neutropenia (90 mg); all were reversible. The maximum tolerated dose was exceeded at 90 mg RO. Pharmacokinetic analysis of both total and free RO confirmed the presence of autoinduction at 45 and 90 mg. Median levels of Notch3 staining were higher in individuals who received fewer than 4 cycles (p = 0.029). Circulating angiogenic factor levels did not correlate with time to progression or ≥ grade 3 adverse events. Best response (RECIST 1.1) was partial response (nasopharyngeal cancer) and stable disease > 4 months was observed in 3 patients (pancreas, tracheal, and breast primary cancers). CONCLUSIONS: RO and gemcitabine can be safely combined. The recommended phase II dose of RO was 30 mg in combination with gemcitabine 1,000 mg/m(2). Although RO exposure was limited by the presence of autoinduction, RO levels achieved exceeded the area under the concentration-time curve for 0-24 h (AUC(0-24)) predicted for efficacy in preclinical models using daily dosing. Evidence of clinical antitumor activity and prolonged stable disease were identified.