Cargando…

Impact of CD4 T cells on intratumoral CD8 T-cell exhaustion and responsiveness to PD-1 blockade therapy in mouse brain tumors

BACKGROUND: Glioblastoma is a fatal disease despite aggressive multimodal therapy. PD-1 blockade, a therapy that reinvigorates hypofunctional exhausted CD8 T cells (T(ex)) in many malignancies, has not shown efficacy in glioblastoma. Loss of CD4 T cells can lead to an exhausted CD8 T-cell phenotype,...

Descripción completa

Detalles Bibliográficos
Autores principales: Khan, Saad M, Desai, Rupen, Coxon, Andrew, Livingstone, Alexandra, Dunn, Gavin P, Petti, Allegra, Johanns, Tanner M
Formato: Online Artículo Texto
Lenguaje:English
Publicado: BMJ Publishing Group 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9772691/
https://www.ncbi.nlm.nih.gov/pubmed/36543376
http://dx.doi.org/10.1136/jitc-2022-005293
Descripción
Sumario:BACKGROUND: Glioblastoma is a fatal disease despite aggressive multimodal therapy. PD-1 blockade, a therapy that reinvigorates hypofunctional exhausted CD8 T cells (T(ex)) in many malignancies, has not shown efficacy in glioblastoma. Loss of CD4 T cells can lead to an exhausted CD8 T-cell phenotype, and terminally exhausted CD8 T cells (T(ex)(term)) do not respond to PD-1 blockade. GL261 and CT2A are complementary orthotopic models of glioblastoma. GL261 has a functional CD4 T-cell compartment and is responsive to PD-1 blockade; notably, CD4 depletion abrogates this survival benefit. CT2A is composed of dysfunctional CD4 T cells and is PD-1 blockade unresponsive. We leverage these models to understand the impact of CD4 T cells on CD8 T-cell exhaustion and PD-1 blockade sensitivity in glioblastoma. METHODS: Single-cell RNA sequencing was performed on flow sorted tumor-infiltrating lymphocytes from female C57/BL6 mice implanted with each model, with and without PD-1 blockade therapy. CD8(+) and CD4(+) T cells were identified and separately analyzed. Survival analyses were performed comparing PD-1 blockade therapy, CD40 agonist or combinatorial therapy. RESULTS: The CD8 T-cell compartment of the models is composed of heterogenous CD8 T(ex) subsets, including progenitor exhausted CD8 T cells (T(ex)(prog)), intermediate T(ex), proliferating T(ex), and T(ex)(term). GL261 is enriched with the PD-1 responsive T(ex)(prog) subset relative to the CT2A and CD4-depleted GL261 models, which are composed predominantly of the PD-1 blockade refractory T(ex)(term) subset. Analysis of the CD4 T-cell compartments revealed that the CT2A microenvironment is enriched with a suppressive T(reg) subset and an effector CD4 T-cell subset that expresses an inhibitory interferon-stimulated (Isc) signature. Finally, we demonstrate that addition of CD40 agonist to PD-1 blockade therapy improves survival in CT2A tumor-bearing mice. CONCLUSIONS: Here, we describe that dysfunctional CD4 T cells are associated with terminal CD8 T-cell exhaustion, suggesting CD4 T cells impact PD-1 blockade efficacy by controlling the severity of exhaustion. Given that CD4 lymphopenia is frequently observed in patients with glioblastoma, this may represent a basis for resistance to PD-1 blockade. We demonstrate that CD40 agonism may circumvent a dysfunctional CD4 compartment to improve PD-1 blockade responsiveness, supporting a novel synergistic immunotherapeutic approach.