Phase I/II trial testing safety and immunogenicity of the multipeptide IMA950/poly-ICLC vaccine in newly diagnosed adult malignant astrocytoma patients

Abstract

Background: Peptide vaccines offer the opportunity to elicit glioma-specific T cells with tumor killing ability. Using antigens eluted from the surface of glioblastoma samples, we designed a phase I/II study to test safety and immunogenicity of the IMA950 multipeptide vaccine adjuvanted with poly-ICLC (polyinosinic-polycytidylic acid stabilized with polylysine and carboxymethylcellulose) in human leukocyte antigen A2+ glioma patients.

Methods: Adult patients with newly diagnosed glioblastoma (n = 16) and grade III astrocytoma (n = 3) were treated with radiochemotherapy followed by IMA950/poly-ICLC vaccination. The first 6 patients received IMA950 (9 major histocompatibility complex [MHC] class I and 2 MHC class II peptides) intradermally and poly-ICLC intramuscularly (i.m.). After protocol amendment, IMA950 and poly-ICLC were mixed and injected subcutaneously (n = 7) or i.m. (n = 6). Primary endpoints were safety and immunogenicity. Secondary endpoints were overall survival, progression-free survival at 6 and 9 months, and vaccine-specific peripheral cluster of differentiation (CD)4 and CD8 T-cell responses.

Results: The IMA950/poly-ICLC vaccine was safe and well tolerated. Four patients presented cerebral edema with rapid recovery. For the first 6 patients, vaccine-induced CD8 T-cell responses were restricted to a single peptide and CD4 responses were absent. After optimization of vaccine formulation, we observed multipeptide CD8 and sustained T helper 1 CD4 T-cell responses. For the entire cohort, CD8 T-cell responses to a single or multiple peptides were observed in 63.2% and 36.8% of patients, respectively. Median overall survival was 19 months for glioblastoma patients.

Conclusion: We provide, in a clinical trial, using cell surface-presented antigens, insights into optimization of vaccines generating effector T cells for glioma patients.

Trial registration: Clinicaltrials.gov NCT01920191.

Keywords: IMA950; glioma; immune response; peptide vaccine; poly-ICLC.

© The Author(s) 2019. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Figures

Fig. 1

Study protocol. Patients were screened between surgery and end of radiochemotherapy. Treatment weeks are defined starting from the initiation of radiochemotherapy. Vaccinations were administered over a period of 24 weeks at least 1 week after the end of radiochemotherapy. V: vaccinations in the initial study protocol; V*: vaccinations in the modified study protocol. T1 to T8 represent blood samples collected for immunomonitoring. A punch biopsy was taken 48 h after delayed type hypersensitivity. RT, radiotherapy.

Fig. 2

Illustrative local injection site reactions and MR response patterns. (A) Axial plane post gadolinium T1-weighted imaging MRI shows impressive radiological pseudoprogression after vaccination number 4 in Patient 8, with eventual complete local response. a: Jun 2014: left parieto-occipital lesion before initial gross total resection, b: immediate post-surgical MRI shows hemorrhagic changes with methemoglobin rim surrounding the resection cavity, c: Oct 2014: local relapse after V4 toward the left ventricle trigone, d: Dec 2014: stable disease, e: Feb 2015: partial response, f: Jun 2017: long term follow-up imaging shows relapse toward left insula and internal capsule but local complete response at the original surgical bed. (B) Illustrative example of peritumoral edema after 4 immunizations in Patient 2. Jun 2013: coronal T1 post-contrast images show right temporal lesion before (a) and after (b) surgical removal. Axial fast spin echo (FSE) T2 (c1–c2, 2 different anatomical levels) and T1 (c3) images post-surgery. Oct. 2013: d1–d2 FSE T2 images denote prominent edema after vaccination number 4, with minimal tumor progression on T1 post-contrast image (d3). Nov 2013: e1–e3 show the same pattern of vaccine-induced edema after vaccination number 6. Jan 2014: f1–f3: right temporal heterogeneously enhancing multifocal tumor relapse.

Fig. 3

Vaccine-induced CD8 and CD4 T-cell responses before and after protocol amendment. (A) Percentage of patients with an HBV-, tumor antigen-specific CD8 or CD4 T-cell response in the initial (n = 6)/modified protocol (n = 13) and for all patients. (B) Percentage of patients with a tumor antigen-specific CD8 T-cell response to one or multiple peptides in the initial (n = 6)/modified protocol (n = 13) and for all patients. (C) Mean ± SD of the percentage of HLA-A2/peptide multimer+ CD8+ T cells specific for the BCAN478-486, NLGN4X131-139, and PTPRZ11347-1355 antigens in the pre-vaccination (T1 + T2), post-vaccination 1 (T3 + T4), 2 (T5 + T6), and 3 (T7 + T8) timepoints. (D) Percentage of patients with a tumor antigen-specific CD4 T-cell response to one or multiple peptides in the initial (n = 6)/modified protocol (n = 13) and for all patients. (E) Mean ± SD of the percentage of TNF-α-secreting CD4+ T cells specific for the MET651-667 and BIRC597-111 antigens in the pre-vaccination, post-vaccination 1, 2, and 3 timepoints.

Fig. 4

Patient survival. (A) Percentage surviving patients in the whole (n = 19) and GBM-only cohort (n = 16). (B) Percentage of patients with PFS in the whole (n = 19) and GBM-only cohort (n = 16). (C) Percentage surviving patients vaccinated with the initial (n = 6) or modified (n = 13) protocol. (D) Percentage surviving patients vaccinated in the s.c. (n = 7) or i.m. (n = 6) arm of the modified protocol. Ticks denote censored patients (5 living patients).