A trial of targeted treatment with improved drug delivery to the brain met its primary end point for overall survival in patients with recurrent glioblastoma.
Convection-enhanced delivery (CED) administration of bizaxofusp (formerly MDNA55) showed favorable overall survival (OS) and tumor control as well as acceptable safety in the phase 2b MDNA55-05 trial (NCT02858895) of patients with recurrent glioblastoma (rGBM), according to results published in Neuro-Oncology.1
The open-label, single-arm phase 2b trial of bizaxofusp at escalating doses resulted in a 46% 12-month OS rate and a median OS of 11.64 months in treated patients, with improved efficacy in patients with high interleukin-4 receptor (IL-4R) expression and patients with low IL4R expression who had a higher dose. The rate of adverse events (AEs) did not appear to be dose dependent.
Bizaxofusp is an immunotoxin that targets IL-4R in tumor cells. High IR4R expression is associated with poor outcomes in GBM and remains expressed in recurrent disease.
Investigators in the MDNA55-05 trial administered bizaxofusp intratumorally with CED using up to 4 catheters and real-time image guidance to improve distribution of the agent. CED is a minimally invasive procedure that bypasses the blood-brain barrier. A total of 47 patients were enrolled from March 23, 2017 to September 12, 2019 at 8 sites in the United States and 1 in Poland. The primary end point was OS with a null hypothesis of 8.0 months median OS versus an alternative of 11.5 months, based on previous studies of approved therapies. Overall response rate (ORR) and progression-free survival (PFS) were secondary end points. Modified RANO (Response Assessment in Neuro-Oncology) criteria were used to account for the high rate of pseudo-progression in this population.
Patients enrolled in the trial had 1 or 2 prior recurrences and had received treatment including surgery and radiotherapy with or without chemotherapy. Patients received a single treatment of bizaxofusp whose concentration and volume were equivalent to a pre-determined total dose ranging from 18 to 240 µg, less than or equal to the established maximum tolerated dose.
The median age of the patient population in the intent-to-treat (ITT) population was 56; 64% were male and 36% were female. Approximately half of patients had Karnofsky performance score of 70% to 80%; the rest had a score between 81% and 100%. Twenty-three patients (49%) had high IL-4R expression, 19 (40%) had a low expression, and in 5 (11%) it was unknown. Thirty-nine (83%) had a total resection of their tumor at diagnosis, and 10 (21%) had 2 prior relapses.
Four different drug concentrations were administered: 1.5, 3.0, 6.0, and 9.0 µg/mL, with a median infusion volume of 30 mL for a median duration of 26.57 hours. The median total dose infused was 180 µg, with nearly all patients receiving the planned total dose and infusion volume.
In the ITT population, the median OS was 10.2 months (1-sided 80% CI, 8.39-12.75); in patients who were treated per protocol, it was 11.56 months (1-sided 80% CI, 8.62-15.02), meeting the primary end point. Twelve-month OS rate was 43% (95% CI, 29%-57%) in the ITT population and 46% (95% CI, 31%-60%) in the per-protocol population.
Patients with high IL-4R expression had a median OS of 15.02 months, whereas those with low expression had a median OS of 8.4 months; this difference was not statistically significant (P = .215). However, patients who had low IL-4R expression but received at least 180 µg of bizaxofusp had a 12-month OS rate of 53% with median OS not reached, whereas other patients with low IL4R expression had a median OS of 8.0 months. When the high IL-4R and low IL-4R with high dose groups were combined, the median OS was 15.0 months, and 12-month OS rate was 55%.
There was no significant difference in OS depending on dose received, and patients with unmethylated MGMT promoters did not show significant difference in OS.
Only 1 patient had an objective response, for an ORR of 2.4%, but tumor shrinkage or stabilization was observed in 31 of 41 evaluable patients. One patient exhibited radiographic changes consistent with pseudo-progression followed by evidence of a durable complete response. The total tumor control rate (TCR) was 75.6%. In patients who had high IL-4R or had low IL-4R but received a high dose, the TCR was 81%. The median PFS was 3.61 months (95% CI: 2.62, 7.70).
The most common AEs patients experienced during the treatment period and follow-up included 20 patients with seizure (42.6%), 19 with fatigue (40.4%), 18 with headache (38.3%), and 15 with muscular weakness (31.9%). Thirty-two patients (68.1%) had AEs that were at least possibly considered to be related to treatment. The most frequent AEs were seizure in 10 patients (21.3%), fatigue in 9 (19.1%), headache in 8 (17.0%), and pyramidal tract syndrome in 8 (17.0%). Eighteen patients had a history of seizures prior to enrollment. The overall incidence of AEs was similar across dose groups based on both infusion volume and total dose.
There were 8 grade 5 AEs, 6 of which were deemed unrelated to treatment. One patient experienced a cerebral hemorrhage after completing infusion and died 14 days later; this was possibly related to the infusion procedure.
A propensity-matched external control arm of patients with similar baseline characteristics was employed, concluding that the benefit in the trial was attributable to the treatment. The investigators acknowledged that this was not part of the prespecified statistical analysis plan for this trial, though they have received approval from the FDA to use this study design in a registrational trial.
“Combining targeted treatment and advanced drug delivery techniques employed in this study provides opportunities to potentially deliver substantive benefit in patients with rGBM and to explore the efficacy of MDNA55 in a pivotal trial,” the investigators concluded in their report.
1. Sampson JH, Singh Achrol A, Aghi MK, et al. Targeting the IL4 receptor with MDNA55 in patients with recurrent glioblastoma: Results of a phase IIb trial. Neuro Oncol. 2023;25(6):1085-1097. doi:10.1093/neuonc/noac285