Phase I Dose-Escalation Study of 5-Day Intermittent Oral Lapatinib Therapy in Patients With Human Epidermal Growth Factor Receptor 2–Overexpressing Breast Cancer

Abstract

Purpose The highly effective treatment of human epidermal growth factor receptor (HER) 2–amplified breast cancer has proven challenging because of a signal buffering capacity inherent in the functionally relevant HER2-HER3 target. HER2-HER3 signaling can be inactivated by doses of lapatinib that fully inactivate the HER2 kinase. In mouse models, such doses are not tolerable in continuous administration, but they are tolerable and highly effective in intermittent dosing. We pursued the clinical translation of this treatment hypothesis. Patients and Methods We conducted a phase I dose-escalation study in women with advanced HER2-overexpressing breast cancer. Lapatinib was administered on days 1 through 5 of repeating 14-day cycles. Dose escalation was conducted using a 3+3 design with plasma lapatinib level monitoring. Results Forty patients were evaluable for toxicity, and 34 patients were evaluable for dose-limiting toxicity (DLT). Lapatinib dose was escalated to 7,000 mg per day in twice-daily dosing with no DLTs; however, plasma lapatinib concentrations plateaued in this dose range. Additional cohorts evaluated strategies to increase lapatinib exposure, including the food effect, CYP3A4 inhibition, and dose fractionation. Of these, only ketoconazole was able to increase lapatinib exposure, despite highly variable lapatinib bioavailability. Intolerable exposure levels were not encountered. Eight patients (20%) experienced grade 3 diarrhea. Six patients achieved a response, and dramatic responses were seen in three patients with lapatinib concentrations approaching 10,000 ng/mL. Conclusion Lapatinib exposure can be safely and significantly increased through intermittent dosing but reaches a ceiling that currently impedes clinical translation of the treatment hypothesis. Preliminary efficacy data suggest that exposures approaching those seen in mouse models can result in highly significant tumor responses.