A phase I/II trial of GW572016 (lapatinib) in recurrent glioblastoma multiforme: clinical outcomes, pharmacokinetics and molecular correlation

We undertook a phase I/II study of the EGFR/erbB2 inhibitor lapatinib in patients with recurrent glioblastoma multiforme (GBM) to determine response rate, pharmacokinetics (PK) and recommended dose in patients taking enzyme-inducing anti-epileptic drugs (EIAEDs) and to explore relationships of molecular genetics to outcome. Recurrent GBM patients taking EIAEDs were enrolled on the phase I portion (starting dose of lapatinib 1,000 mg po bid). In the absence of dose-limiting toxicity (DLT), escalation continued in cohorts of three patients. Patients not on EIAEDs enrolled in the phase II arm (lapatinib 750 mg bid po). Immunohistochemical and quantitative RT PCR studies were performed on tumor to determine PTEN and EGFRvIII status, respectively. Lapatinib PK was analyzed using HPLC with tandem mass spectrometry. Phase II: Of 17 patients, 4 had stable disease and 13 progressed. Accrual ceased because of no responses. Phase I: Four patients received 1,000 mg bid and three, 1,500 mg bid. No DLT occurred, but escalation stopped because of lack of phase II efficacy. Lapatinib apparent oral clearance in patients taking EIAEDs was 106.9 L h(-1) m(-2) in comparison to 12.1 L h(-1) m(-2) in those not on EIAEDs. In 16 phase II patients, PTEN loss was seen in 6 and EGFRvIII expression in 4. No correlation was seen with outcome and molecular results. Lapatinib apparent oral clearance increased by approximately tenfold when given with EIAEDs. In this small sample, EGFRvIII expression and PTEN loss did not predict a favorable subtype. Overall, lapatinib did not show significant activity in GBM patients.