Dynamic evolution and antitumor mechanisms of CXCR6+CD8+ T cells in small cell lung cancer treated with low-dose radiotherapy and immunotherapy

Abstract

Background: Patients with small-cell lung cancer (SCLC) have the poor prognosis. Current research suggested that low-dose radiotherapy (LDRT) combined with immunotherapy can enhance the immunogenicity of tumor cells, thereby improving antigen presentation and promoting the intratumoral infiltration of CD8⁺ T cells, which significantly extends the survival of patients. However, the change trajectory of T cells, and the mechanisms underlying the promotion of intratumoral infiltration of CD8⁺ T cells, and the enhancement of their cytotoxic functions remain to be elucidated.

Methods: To delineate the dynamic changes of T cells, we collected tumors from Kaede tumor-bearing mice that had undergone radioimmunotherapy. Using flow cytometry, we sorted intratumoral-infiltrating immune cells, which were required for single-cell RNA sequencing, at various time points (Kaede Red: derived from tumor-draining lymph node [TDLN]). The results obtained from the sequencing analysis were further validated through experiments, such as flow cytometry, immunofluorescence, and analysis of clinical cohort data.

Results: Here, we observed stem-like T cells migrating from the TDLN to the tumor site and differentiating into effector phenotypes within the tumor. Dendritic cells (DCs) are the key cluster that induces the differentiation of stem-like T cell into effector phenotypes. Moreover, SCLC patients with a high infiltration of tumor-specific CXCR6⁺CD8⁺ T cells exhibited a supportive TME and longer survival time (P < 0.001).

Conclusions: This study delineates the change trajectory of CD8⁺ T cells, identifies the crucial role of DCs in T cell differentiation, and highlights the significance of tumor-specific CXCR6⁺CD8⁺ T cells in anti-tumor immunity. Future therapeutic strategies for SCLC could focus on enhancing the infiltration of activated DCs and CXCR6⁺CD8⁺ T cells within the tumor microenvironment to improve treatment efficacy.

Keywords: CD8-positive T-lymphocytes; Immunotherapy; Radiotherapy; Small-cell lung cancer; Tumor-infiltrating.

Fig. 1

The dynamic migration of immune cells between TDLN and tumor site in SCLC-bearing mice. A Schematic of SCLC-bearing mouse model experiment design and treatment plan. B, C The tumor volume curve (B) and body weight curve (C) of the PRM-SCLC model with different treatments (n = 4–5). D, E Representative cell smears (D) and flow cytometry plots (E) of TDLN before photoconversion and post-photoconversion. F Representative flow cytometry plots (left) and the proportion (right) of “Kaede Red” CD45⁺cells in tumors resected on day 6, 7, and 8 post-treatment (n = 3). G Representative images of mIF staining indicating “Kaede Red” CD3⁺CD8⁺T cells in tumors resected on day 6, 7, and 8 post-treatment. Statistical analyses were conducted using two-way ANOVA (B and C), and one-way ANOVA (F). P values are shown, and error bars indicate the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, s ****P < 0.0001

Fig. 2

Single-cell RNA sequencing revealed the differences between TDLN-derived and tumor-resident cells. A Schematic of single-cell sequencing. B UMAP plot of 16,890 CD45⁺cells from scRNA-seq of “Kaede-Green” and “Kaede-Red” on day 6 and 8 post-treatment, colored by cell type. C Bar plot showing the distribution proportion of each cell subtype. D UMAP plot of all CD8⁺ T cells from scRNA-seq of “Kaede-Green” and “Kaede-Red” on day 6 and 8 post-treatment, colored by cell type. E Pie charts showing the distribution proportion of each CD8⁺T cells subtype from four samples. F The expression level of Ki-67 across all CD8⁺ T cells. G Violin plot of signature scores in each CD8⁺T cell clusters. H UMAP showing pseudotime trajectory of CD8⁺ T cells based on PAGA. Cells are color coded for their corresponding pseudotime. I Heatmaps showing the expression of chemokine and cytokine receptors with the differentiation trajectory of CD8⁺T cells. The gradient from blue to white to red on the heatmap represents an increase in the above expression levels scores

Fig. 3

CXCR6⁺CD8⁺T cells were identified as crucial and conferred better prognosis in patients with SCLC. A Heatmap plot showing the expression levels of chemokine receptors across CD8⁺ T cell subtypes. B UMAP plots displaying CXCR6 and PD-1 expression levels across all CD8⁺ T cells, with cells are color coded by estimated density. C Flow cytometry analysis of CXCR6⁺CD8⁺T cells % CD45⁺ cells in SCLC-tumor from different treatments on Day8 (n = 3). D Flow cytometry analysis of CXCR6⁺CD8⁺T cells % CD45⁺ cells in various organs of SCLC-bearing treatment by LDRT combined with αPD1 on Day8 (n = 3). E Kaplan–Meier curves of overall survival (OS) based on CXCR6⁺CD8⁺T cells infiltration in Peng-Zhang cohort. F–H The enrichment of DEGs between high and low CXCR6⁺CD8⁺T cells infiltration via GO analyses and GSEA analyses in Peng-Zhang cohort. I, J Heatmaps for CIBERSORT analysis showing the correlation between CXCR6⁺CD8⁺T cells signature levels and immune cells in Peng-Zhang cohort. The statistical analyses were performed using one-way ANOVA (C and D), and log rank test (E). P values are indicated, with error bars indicate the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 4

DCs promote stem-like T cells differentiated into effector CD8⁺T cells. A, B The proportion of CXCR6⁺CD8⁺T cells % CD45⁺ cells (A) and PD-L1⁺DCs % DCs (B) in SCLC-tumor from different treatments on Day8 post-treatment (n = 3). C, D The correlation between CXCR6⁺CD8⁺T cells and DCs based on ZP cohort and SJ cohort. E The correlation between CXCR6⁺CD8⁺T cells and CXCL16 expressed by DCs based on two cohorts. F The UMAP plot showed the CXCL16 expression level across DC subtypes. G This chord diagram showed specific connections between CD8T4 and DCs. The color represents different cell types, and the thickness of the line represents the strength of the connection. H Schematic of co-culture. I The TNF-α, GZMB and IFN-γ level in the co-culture of DCs and stem-like T cells (Tsl). The statistical analyses were performed using one-way ANOVA (A, B and I), and Spearman correlation analyses (C–E). P values are shown, and error bars indicate the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001