The patient population will be molecularly defined and include adult patients with newly diagnosed K27M-mutant histone-3.1 (H3.1K27M) or histone-3.3 (H3.3K27M) diffuse midline gliomas (DMG).
Within this trial, a long peptide vaccine containing a K27M-mutated histone-3 sequence, will be administered subcutaneously in addition to standard radiotherapy and thereafter in combination with the human anti-PD-L1 antibody Atezolizumab.
Fifteen patients (pts. 1-15) will receive 11 doses of H3K27M peptide vaccine in total starting with standard radiotherapy (RT) and 14 doses of the human anti-PD-L1 antibody Atezolizumab (every three weeks, q3w) starting four weeks after completion of RT. The first 3 vaccines will be given bi-weekly (q2w) in combination with RT. One dose of vaccination will be given at the beginning of recovery (RE) period following RT. Vaccines 5-11 (q6w) will be initiated with Atezolizumab after completion of RE. In a safety lead-in, the first three patients (pts. 1-3) will be enrolled sequentially.
Following the last IMP administration, a safety / immunogenicity follow-up is planned for 24 weeks until end of study (EOS). To be able to assess safety, tolerability and immunogenicity of the peptide vaccine in combination with Atezolizumab 15 evaluable patients will be enrolled.
Diffuse gliomas of the thalamus, brain stem, spinal cord or other midline structures represent 3-4% of high-grade glioma and harbor H3.1K27M or H3.3K27M mutations as a characteristic founder mutation in \> 70% of cases. H3K27M-mutant gliomas typically occur in children and adolescents but also in adult patients. After biopsy or resection, the standard of care consists of involved-field radiotherapy. Adding alkylating chemotherapy to radiotherapy does not offer additional benefit in retrospective case series and prospective clinical trials in children with pontine gliomas, probably as hypermethylation of the MGMT promoter in DMG is typically lacking. After radiotherapy gliomas frequently recur with a median 12-month progression-free survival of 20 %. Importantly, at recurrence particularly in the adult patient population, there is frequent distant progression and leptomeningeal dissemination, arguing for the necessity of systemic therapy upfront.
From an immunological point of view H3K27M represents an attractive tumor antigen specifically expressed in tumor but not normal cells. Patients with H3K27M-mutant gliomas may harbor mutation-specific T cells, indicating that H3K27M is specifically presented to and recognized by the immune system in a mutation-specific manner. Vaccination of humanized mice with a long H3K27M vaccine results in an anti-tumor immune response effective in controlling H3K27M-expressing tumors in a preventive and a therapeutic manner without causing toxicity. In addition, eight adult patients with H3.3K27M-mutated gliomas were treated with an H3K27M 27 amino acid long peptide vaccine on a compassionate-use basis. None of the patients treated either with the peptide vaccine alone or in combination with an anti-human PD(L)-1 antibody showed any clinical or laboratory sign of treatment-related toxicity except for grade 1 injection site reactions. Importantly, all patients developed H3K27M-specific T cell responses with one patient of the combination treatment group showing long-term response with no sign of tumor progression for \>24 months after vaccination and one patient experiencing complete response after pseudoprogression (unpublished observations). Based on these observations we hypothesize that checkpoint inhibition targeting PD-L1 is required for optimal amplification of a vaccine-induced H3K27M-specific T cell response.
The aim of this phase I trial is to evaluate the safety and immune response to the H3K27M peptide vaccine in combination with Atezolizumab in patients with H3K27M-mutant diffuse midline gliomas.
