Phase I safety and bioimaging trial of ifabotuzumab in patients with glioblastoma

Objectives: Glioblastoma multiforme (GBM) is the most frequent and lethal primary brain neoplasm, with only 10% of patients surviving 5 years. EphA3 is a tumor restricted antigen expressed in various solid tumors and the tumor vasculature of 100% of GBM. Ifabotuzumab is a non-fucosylated IgG1κ humaneered antibody targeting the EphA3 receptor, and a Phase I study of ifabotuzumab in haematological malignancies showed it was well tolerated and clinically active. Here we report on a Phase I dose escalation and biodistribution study of ifabotuzumab in recurrent GBM. Methods The primary objective is to determine the safety and recommended Phase II dose of ifabotuzumab in GBM patients (pts). Secondary objectives are to determine the biodistribution and pharmacokinetics (PK) of ⁸⁹Zr- ifabotuzumab, the frequency of EphA3 positive GBM and response rates. On day 1, eligible pts with measurable tumors received a trace (5mg) dose of zirconium labelled ifabotuzumab (⁸⁹Zr-ifab) followed by sequential PET imaging over 1 week to determine its biodistribution, frequency of in situ EphA3 expression and quantitative tumor uptake. Safety assessments and PK sampling were also undertaken. On day 8, pts commenced weekly ifabotuzumab infusions over 2 hours in one of two cohorts (3.5mg/kg, 5.25 mg/kg). On day 36, pts received both ⁸⁹Zr-ifab and ifabotuzumab, allowing assessment of receptor occupancy. Response rate (RANO) and survival data were collected. Pts then continued on ifabotuzumab until progression. Results In total, 12 pts have been enrolled, including 6 in the 3.5mg/kg and 6 in the 5.25 mg/kg dose cohorts. Mean age was 51.6 years (±14.24) and 7/12 pts were male. Treatment emergent adverse events included infusion reactions in 4 pts, seizures in 3 pts, cerebral oedema in 1, rash in 1, headaches in 8, eye disorder in 1. Most were considered related to study drug except seizure in 2 pts, headaches and eye disorder. Seizures and infusion reactions were readily managed with increased premedications after the first occurrence. The best response was stable disease for 23 weeks. ⁸⁹Zr-ifab-PET scans showed rapid, tumor-specific targeting at all known tumor sites and in all pts, but with no normal tissue uptake. MRI scans showed predominant T2/FLAIR changes, occasionally marked, which were consistent with treatment effect of ifabotuzumab on tumor vasculature. The mean ± SD (n=12) PK parameters for first infusion ⁸⁹Zr-ifab were T½α= 9.03 ± 4.45 hr, T½β = 92.50 ± 65.65 hr, V1 = 3.75 ± 0.67 L, CL= 132.11 ± 70.16 mL/hr.

Conclusions: Ifabotuzumab demonstrates highly sensitive, specific and reproducible targeting of the tumor and tumor microenvironment in all patients in this study. The imaging changes suggest direct modulation of the tumor vasculature. Additional studies are planned to evaluate ifabotuzumab as part of an antibody-drug conjugate in various solid tumor types.