A personalized neoantigen vaccine reprograms the immune landscape of glioblastoma

A personalized neoantigen vaccine reprograms the immune landscape of glioblastoma Jack Y. Ghannam1,4,5,18*, Daniel Kovarsky14*, Jakob Weiss4, Benson Z. Wu1,4,5, Landon K. Oetjen1,7, L. Nicolas Gonzalez Castro4,11,13,15,16, Elena Kuehner1, Michael H. Sun9, Chloe R. Tu1,2, Cleo Forman1,2, Christine McCluskey1, Laine Marrah1,4,5, Oriol Olive1, Haley Greenslade1, McKayla Van Orden1,3, J Thomas Janes 3rd1,3, Bohoon Shim1,2, Jiaxun Li1,3, Melis Akinci1,3, Marwa Belhaj1,3, Allison Vanasse1,3, Ayşe Dila Gemalmayan1,3, Isabel Carulli1,3, Julian Bryan1,4,5,10,17, Patrick Y. Wen1,5,16, Ugonma N. Chukwueke15, Lakshmi Nayak15, Elisa Aquilanti5,15, Rameen Berouhkim1,4,5, Giacomo Oliveira1,4, Rifaquat Rahman16, E. Antonio Chiocca5,12, Shuqiang Li1,3,4, Donna S. Neuberg2, Anita Giobbie-Hurder2, Kenneth J. Livak3, Edward F. Fritsch1,4, Mario L. Suvà4,5,11,13, Keith L. Ligon4,5,9,10, Patrick A. Ott1,4,5,7,8, Derin B. Keskin1,3,4,6‡,, Itay Tirosh14‡, David A. Reardon1,5,8‡, Catherine J. Wu1,4,5,8,19‡ *These authors contributed equally; ‡Co-senior authors Affiliations 1Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA 2Department of Data Science, Dana-Farber Cancer Institute, Boston, MA, USA 3Translational Immunogenomics Laboratory, Dana-Farber Cancer Institute, Boston, MA, USA 4Broad Institute of MIT and Harvard, Cambridge, MA, USA 5Harvard Medical School, Boston, MA, USA 6Department of Computer Science, Metropolitan College, Boston University, Boston, MA, USA 7Center for Immuno-Oncology, Dana-Farber Cancer Institute, Boston, MA, USA 8Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA 9Department of Oncologic Pathology, Dana-Farber Cancer Institute, Boston, MA, USA 10Department of Pathology, Brigham and Women’s Hospital, Boston, MA, USA 11Department of Pathology, Massachusetts General Hospital, Boston, MA, USA 12Department of Neurosurgery, Mass General Birgham and Center for Tumors of the Nervous System, Mass General Brigham Cancer Institute, Boston, MA, USA 13Massachusetts General Hospital, Krantz Family Center for Cancer Research, Boston, MA, USA 14Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel. 15Center for Neuro-Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. 16Department of Neurology, Brigham and Women’s Hospital, Boston, MA, USA 17Molecular Diagnostics Laboratory, Division of Pathology and Laboratory Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, USA 18Harvard/MIT MD-PhD Program and Harvard Immunology PhD Program, Harvard Medical School, Boston, MA, USA 19Lead contact Correspondence: [e-mail]; [e-mail] Summary Personalized neoantigen vaccines combined with immune checkpoint blockade (ICB) have shown promise across solid tumors, but their efficacy in glioblastoma (GBM) remains poorly defined. We conducted a phase I trial of neoantigen-targeting vaccination with pembrolizumab in 39 patients with newly diagnosed GBM, varying by MGMT methylation status and pembrolizumab timing across four cohorts. Among patients with MGMT-unmethylated tumors, survival was improved with pre-vaccine ICB initiation, baseline mesenchymal state, and circulating ex vivo neoantigen-specific T cell responses. In patients with MGMT-methylated GBM, despite temozolomide-induced lymphopenia and diminished ex vivo responses, circulating and intratumoral putative vaccine-reactive (PVR) clonotypes were persistently detectable. Across cohorts, intratumoral PVR T cells were predominantly CD4 tissue-resident with a post-effector phenotype at peri-necrotic regions. Recurrent tumors showed selective contraction of vaccine-targeted subclones and increased HLA class I expression on residual malignant cells. Our findings link ICB timing, chemotherapy, and tumor transcriptional state to NeoVax anti-tumor activity in GBM. This repository contains supplementary data files 1-7: Data S1, List of patient mutations, related to Figure 1 Data S2, Patient clinical and tumor characteristics, related to Figure 1 Data S3, Supporting data for ELISpot analysis, related to Figure 2 Data S4, Supporting single-nucleus RNA sequencing data, related to Figure 3 Data S5, Supporting Olink circulating soluble protein data, related to Figure 3 Data S6, Supporting bulk and single-nucleus TCR data for putative vaccine-reactive T cell clonotypes, related to Figure 4 Data S7, Supporting data for Xenium analysis, related to Figures 5 and 6 Data S8, Supporting data for WES PhylogicNDT analysis, related to Figure 7