Modified RANO Affords a More Accurate Assessment of Outcomes in Recurrent Glioblastoma

Standard response assessment in neuro-oncology (RANO)- and immunotherapy RANO (iRANO)-defined progression-free (PFS) did not correlate with overall survival (OS) vs modified RANO (mRANO)-defined PFS, which strongly correlated with OS in patients with recurrent glioblastoma treated with MDNA55 in a phase 2 trial (NCT02858895), suggesting that mRANO may be the optimal means of therapeutic response assessment in recurrent glioblastoma.

In patients who died and had confirmed progression on standard RANO, there was no correlation between PFS and OS (local, P = .47; central, P = .34). Using iRANO, there was minimal association between confirmed PFS and OS via local site measurements (P = .017), but not central measurements (P = .18). A strong correlation was observed between mRANO PFS and OS for local (R² = .66; P <.0001) and centrally determined measurements (R² = .57; P = .0007).

“This phase 2b trial has generated compelling data, and we are very pleased to have them published in such a prestigious peer-reviewed journal. In particular, we are highly encouraged by results showing that early determination of PFS using mRANO may be a strong surrogate for OS in recurrent glioblastoma,” said Fahar Merchant, PhD, president and chief executive officer of Medicenna, in a press release.

MDNA55 is an engineered circularly permutated IL4 fused to a modified sequence of the Pseudomonas aeruginosa exotoxin A specifically designed for glioblastoma.

Although response assessment using the standard RANO criteria is accepted criteria by the FDA, it fails to adequately characterize responses to immunotherapy. As such, investigators performed response assessment based on mRANO criteria and iRANO criteria.

Forty-seven patients with recurrent glioblastoma were enrolled in the single-arm, open-label, multicenter phase 2 study, which evaluated convention-enhanced delivery of the IL4R-targeted immunotoxin.

Eligible patients included those at least 18 years of age with histologically proven primary de novo glioblastoma without an IDH1/IDH2 mutation that had recurred or progressed per standard RANO criteria and not indicated for resection at relapse.

The study was conducted from April 11, 2017, through October 31, 2019. All patients must have received surgery and radiotherapy with or without chemotherapy and discontinued any previous standard or investigational therapy. Patients must have had a life expectancy of more than 12 weeks and a Karnofsky performance status (KPS) of at least 70. Patients had to have a minimum tumor diameter of at least 1 cm by at least 1 cm up to 4 cm in any direction by preinterventional MRI within 14 days of planned treatment and had no features that made the tumor a poor target for convection enhanced delivery.

MDNA55 was administered intraperitumorally and peritumorally via 1 to 4 surgically placed catheters. Patients received a dose concentration range of 1.5 to 9.0 mg/mL and dose volume range of 0* to 66 mL. Of the 46 patients treated, the volume range was 12 mL to 66 mL.

The maximum anticipated duration of study participation for each patient was 12.5 months, including up to 14 days of screening, up to a 3-day planning period, and a 12-month follow-up period following the day of catheter placement and start of infusion.

Bidirectional tumor measurements were made by local sites and centrally by an independent radiologic faculty. Standard RANO, iRANO, and mRANO criteria were applied.

Of 47 patients, 41 were evaluable for response. Patients had a median age of 56 years (range, 34-77); the majority were male (n = 25; 61%) and had a KPS of 90 and 100 (n = 22; 54%). Patients had a median maximum tumor dimension of 29.7 mm (range, 12-8.5).

Additional findings demonstrated that local site–determined PFS was significantly shorter using standard RANO vs iRANO (HR, 0.28; median PFS = 34 days vs 351 days; log-rank, P <.0001) and mRANO (HR, 0.29; median PFS = 34 days vs 122 days; P <.0001). Centrally determined PFS was also significantly shorter using standard RANO criteria vs iRANO (HR, 0.27; median PFS = 35 days vs 155 days; P <.0001) and mRANO (HR, 0.28; median PFS = 35 vs 100 days; P <.0001).

Of 41 patients, 24 (59%) were censored using iRANO because they did not have confirmed progression 3 months after initial progression.

Moreover, the incorporation of clinical status into standard RANO assessment did not result in a significant correlation between PFS and OS for local measurements (R² = 0.0534; P = .2463) or central measurements (R² = 0.03243; P = .3410). However, a statistically significant correlation was observed between PFS and OS when clinical status was incorporated into iRANO assessment (local, R² = 0.6344; P =.0001; central, R² = 0.5079, P = .0029).

Similar to radiographic evaluations alone, PFS and OS were highly correlated using the combination of mRANO criteria and clinical status (local, R² = 0.5958, P <.0001; central, R² = 0.6435, P <.0001).

“We believe this finding and the positive mRANO PFS and overall survival [OS] data from our proposed patient population bode well for the outcome of the planned phase 3 trial, which utilizes an innovative open label hybrid design with OS as the primary end point. Medicenna is pursuing a partnership strategy to continue the phase 3 development,” concluded Merchant.

References

1. Ellingson BM, Sampson J, Achrol AS, et al. Modified RANO, immunotherapy RANO, and standard RANO response to convection-enhanced delivery of IL4R-targeted immunotoxin MDNA55 in recurrent glioblastoma. Clin Cancer Res. Published April 16, 2021. Accessed May 21, 2021. doi:10.1158/1078-0432.CCR-21-0446
2. Medicenna announces the peer-reviewed publication of clinical data from phase 2b trial evaluating MDNA55 in recurrent glioblastoma. News release. Medicenna. Published May 7, 2021. Accessed May 21, 2021. https://bit.ly/34aHJaY