Phase I study of vismodegib in children with recurrent or refractory medulloblastoma: a pediatric brain tumor consortium study

Abstract

Purpose: To investigate the safety, dose-limiting toxicities, and pharmacokinetics of the smoothened inhibitor vismodegib in children with refractory or relapsed medulloblastoma.

Experimental design: Initially, vismodegib was administered daily at 85 mg/m(2) and escalated to 170 mg/m(2). The study was then revised to investigate a flat-dosing schedule of 150 mg for patients with small body surface area (BSA, 0.67-1.32 m(2)) or 300 mg for those who were larger (BSA, 1.33-2.20 m(2)). Pharmacokinetics were performed during the first course of therapy, and the right knees of all patients were imaged to monitor bone toxicity. Immunohistochemical analysis was done to identify patients with Sonic Hedgehog (SHH)-subtype medulloblastoma.

Results: Thirteen eligible patients were enrolled in the initial study: 6 received 85 mg/m(2) vismodegib, and 7 received 170 mg/m(2). Twenty eligible patients were enrolled in the flat-dosing part of the study: 10 at each dosage level. Three dose-limiting toxicities were observed, but no drug-related bone toxicity was documented. The median (range) vismodegib penetration in the cerebrospinal fluid (CSF) was 0.53 (0.26-0.78), when expressed as a ratio of the concentration of vismodegib in the CSF to that of the unbound drug in plasma. Antitumor activity was seen in 1 of 3 patients with SHH-subtype disease whose tumors were evaluable, and in none of the patients in the other subgroups.

Conclusions: Vismodegib was well tolerated in children with recurrent or refractory medulloblastoma; only two dose-limiting toxicities were observed with flat dosing. The recommended phase II study dose is 150 or 300 mg, depending on the patient's BSA. Clin Cancer Res; 19(22); 6305-12. ©2013 AACR.

Fig 1

Variations in target doses of vismodegib reflect interpatient differences in body surface area (BSA). Deliverable doses vary from targeted doses due to BSA-based dosing and limited drug formulation. Orange symbols (+) represent the initial cohort who received 25-mg capsules, and the black symbols (○) represent the flat-dosing cohort who received 150-mg capsules. Dosing strategy was based on minimizing the difference between the deliverable and targeted doses. Variation from the targeted doses was substantially less in the initial study than in the flat-dosing trial due to the initial availability of the 25-mg capsules. With a targeted dose of 170 mg/m² in the flat-dosing cohort, BSA ranges were chosen to minimize the difference between the targeted and the deliverable doses. Individual patient BSAs are plotted just above the X-axis.

Fig. 2

Pharmacokinetics of vismodegib after a 28-day administration of 85 or 170 mg/m². Plasma concentrations of total vismodegib after 85 mg/m² (A) or 170 mg/m² (C) and unbound vismodegib after 85 mg/m² (B) or 170 mg/m² (D) are shown for each patient. (E) The medians and scatter of steady-state plasma concentration (C_ss) were also measured on Day 21. The total vismodegib C_ss (E) and unbound vismodegib C_ss (F) significantly differed between the 2 groups (Mann-Whitney; p=0.026 and p=0.022, respectively).

Fig. 3

Diagram of medulloblastoma (MB) charateristics, as determined by immunohistochemistry and histologic analyses. Immunohistochemical analysis grouped samples as SHH, WNT, or non-SSH/WNT tumors. Histologic categories included anaplastic (AN), classic (CL), desmoplastic nodular (D/N), large-cell (LC), or nonspecified (ns) tumors.