KS03.4.A SPATIAL MULTI-OMIC PROFILING REVEALS DETERMINANTS OF TIL EXPANDABILITY IN HIGH-GRADE GLIOMAS
Abstract BACKGROUND Tumor-infiltrating lymphocyte (TIL) therapy has demonstrated efficacy in melanoma and is emerging as a promising modality for cell-based immunotherapy. However, its application in immunologically “cold” tumors like glioblastoma is limited due to sparse T cell infiltration, antigenic heterogeneity, and a suppressive tumor microenvironment. To identify genomic and spatial determinants of TIL expandability, we performed integrated, multimodal profiling of high-grade gliomas to identify molecular and spatial features associated with successful TIL expansion. MATERIALS AND METHODS We conducted comprehensive profiling of 18 TIL products using spectral flow cytometry and T Cell Receptor sequencing (TCRseq). Additionally, 4 tumors yielding ≥10⁸ viable TILs per 0.5 g tumor within two weeks of IL2 Rapid Expansion Protocol cultures, classified as TIL⁺, were analyzed and compared to 4 non-yielding tumors, classified as TIL⁻. Expanded TILs were subjected to spectral flow cytometry and TCRseq. These TIL⁺ and TIL⁻ tumors underwent single-cell RNA-seq (scRNAseq), in situ Xenium spatial transcriptomics using a customized 480-gene panel, and CODEX spatial proteomics. RESULTS The TIL⁺ tumors yielded 0.5-3.2 × 10⁹ cells with 100-1,000-fold expansion. Flow cytometry showed CD4⁺ and CD8⁺ effector memory phenotypes with minimal exhaustion; Tregs were <10% and γδ T cells were detected. TCRseq revealed high frequency αβ clones and diverse γδ repertoires. Expanded γδ T cells had higher D50 and diversity indices, suggesting a broad, non-dominant repertoire. TIL⁺ tumors exhibited elevated IL7R and ACSS3 expression, enriched Th1/pro-inflammatory gene signatures, and lower Treg-associated transcripts. Spatial analysis showed perivascular immune clusters in TIL⁺ tumors, enriched for CD8⁺ T cells and antigen-presenting cells, with local upregulation of IL7R, MHC II, and CXCL13. These areas were associated with vascular markers (e.g., VCAM1) and T cell retention signatures. In contrast, TIL⁻ tumors displayed immune exclusion, neuronal lineage enrichment, and higher expression of immune suppressive genes like TOX and FERMT1. Tumor-connected TAMs expressing SOX2, NES, and AQP4 were more abundant in TIL⁻ tumors, suggesting a barrier to infiltration or active immune suppression. CONCLUSION TIL⁺ tumors exhibit distinct molecular and spatial features associated with successful TIL expansion, including IL7R-driven T cell responsiveness, mesenchymal metabolic programs, and structured perivascular immune niches. These findings define biomarkers of therapeutic TIL production and inform strategies to stratify patients for TIL⁻ based therapies in glioblastoma and other immune-excluded tumors. Incorporating these biomarkers into patient selection and workflows could improve clinical trial design, product consistency, and therapeutic response rates.
Authors
- Mustafa Khasraw (ORCID: https://orcid.org/0000-0003-3249-9849)
- Michael C. Brown (ORCID: https://orcid.org/0000-0002-3957-1164)
- Kristen A. Batich (ORCID: https://orcid.org/0000-0002-8699-8769)
- Jingyuan Xie (ORCID: https://orcid.org/0000-0002-5508-0664)
- Evan Calabrese (ORCID: https://orcid.org/0000-0002-1464-0354)
- Anoop P. Patel (ORCID: https://orcid.org/0000-0001-7078-9535)
- Ying Zhou (ORCID: https://orcid.org/0000-0001-8276-2735)
- Kirit Singh (ORCID: https://orcid.org/0000-0002-0419-5381)
- Kelly Hotchkiss (ORCID: https://orcid.org/0000-0001-9175-1579)
- John W. Hickey (ORCID: https://orcid.org/0000-0001-9961-7673)
- Jose Ramon Conejo-Garcia (ORCID: https://orcid.org/0000-0001-6431-4074)
- A Corcoran (ORCID: https://orcid.org/0000-0001-8890-6939)
- Pamy Noldner
- Elizabeth B. Owens (ORCID: https://orcid.org/0000-0003-2149-0320)
- Roger E. McLendon (ORCID: https://orcid.org/0000-0001-6682-4588)
- Beth H. Shaz (ORCID: https://orcid.org/0000-0002-2270-4821)
- Katayoun Ayasoufi (ORCID: https://orcid.org/0000-0002-3797-0823)
- Kenan Zhang (ORCID: https://orcid.org/0000-0001-7270-569X)
- Chelsea Railton
- Kyra Van Batavia
- Jodie Jepson
Institutions
- Duke Medical Center (US)
Publication Details
- Journal
- Neuro-Oncology
- Published
- 2025-10-01
- DOI
- https://doi.org/10.1093/neuonc/noaf193.025
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00