PD1 inhibition and GITR agonism in combination with fractionated stereotactic radiotherapy for the treatment of recurrent glioblastoma: A phase 2, multi-arm study.

Abstract

2004 Background: Despite preclinical evidence that fractionated stereotactic radiotherapy (FSRT) is immunostimulatory, there remains a paucity of data regarding immune effects of FSRT in patients treated with immune checkpoint blockade (ICB). To evaluate the immune impact of FSRT in glioblastoma (GBM), which is refractory to single-agent ICB, and to evaluate the efficacy of the addition of FSRT to ICB for this disease, we conducted a phase 2 trial of the combination of retifanlimab (anti-PD1), INCAGN01876 (GITR agonist), and FSRT (8 Gy x 3 fractions) in patients with recurrent GBM (rGBM). Methods: This single-center trial enrolled patients with no prior bevacizumab and dexamethasone dose ≤ 2mg/day. The study included a single-arm, primary efficacy cohort (Cohort A) treated with retifanlimab, INCAGN01876, and FSRT, and a two-arm neoadjuvant window-of-opportunity cohort (Cohort B) for patients needing tumor resection. In Cohort B, patients on Arm B1 received single doses of retifanlimab/INCAGN01876 plus FSRT prior to surgical resection, and patients on Arm B2 received single doses of retifanlimab/INCAGN01876 without FSRT prior to surgical resection. We used flow cytometry on tumor tissue and peripheral blood mononuclear cells and plasma proteomics to explore the immune consequences of neoadjuvant ICB, with vs. without neoadjuvant FSRT. Primary endpoint: ORR in Cohort A (null hypothesis, ORR 5%; alternative hypothesis, ORR 25%; n = 16 for 80% power, α = 0.05). Secondary endpoints: progression-free survival (PFS) and overall survival (OS), separately in Cohorts A, B1, B2. Data cut-off: January 18, 2023. Results: 32 evaluable patients: Cohort A, n = 16; Cohort B, n = 16 (Arm B1, n = 8; Arm B2, n = 8). Median follow-up time: 20 months. Most common grade 3/4 treatment-related adverse events: cerebral edema (34%), fatigue (16%). Efficacy in Cohort A: no objective responses, best response of stable disease in 9/16 patients (56%), median PFS 3.9 months (95% CI 2.1 – 6.2 months), median OS 9.4 months (95% CI 8.2 – 10.6 months). Median PFS and OS were significantly longer in Cohort B1 (neoadjuvant immunotherapy + FSRT) vs. Cohort B2 (neoadjuvant immunotherapy, no FSRT): PFS 11.7 vs. 2.0 months (p = 0.0002), OS 20.1 vs. 9.4 months (p = 0.001). Inflammatory cytokine responses were stronger and more sustained in Cohort B1 vs. Cohort B2, with concomitant increases in proliferative T cell responses. Conclusions: The combination of retifanlimab, INCAGN01876, and FSRT did not demonstrate efficacy in patients with rGBM when administered without surgical resection. However, among patients receiving neoadjuvant immunotherapy prior to surgical resection, the addition of neoadjuvant FSRT was associated with a significant survival advantage as well as increased inflammatory and cellular immune responses. Neoadjuvant FSRT + ICB warrants further evaluation in rGBM. Clinical trial information: NCT04225039 .