Case 1: A 61-year-old male patient was diagnosed with upper left lung cancer in December 2020. In August 2021, he developed metastases in the nasal region, left face, and right upper limb. Subsequently, in December 2021 the patient started treatment at the Fourth Affiliated Hospital of Soochow University, where he received nasal RT in combination with natalizumab.

Case 2: A 71-year-old male patient presented with a diagnosis of squamous cell carcinoma of the upper lobe of the left lung in May 2024. Positron Emission Tomography-Computed Tomography (PET-CT) imaging revealed increased Fluorodeoxyglucose (FDG) uptake in the left hilum lymph node (station 10), indicating a potential for metastatic involvement. Following the diagnostic workup, the patient underwent Stereotactic Body Radiation Therapy (SBRT) targeting the lesion in the left upper lobe. This was subsequently followed by three cycles of combination therapy, encompassing both chemotherapy and immunotherapy.

Case 3: A 59-year-old male patient was diagnosed with squamous cell carcinoma of the upper lobe of the right lung in July 2024. PET-CT imaging demonstrated increased FDG metabolism in the right upper tracheal lymph node (station 2R), hinting at metastatic involvement. Subsequently, the patient underwent SBRT to address the lesion in the upper lobe of the right lung. This therapeutic intervention was followed by three cycles of chemotherapy combined with immunotherapy.

The human T lymphocyte Jurkat cell line and the human H1299 NSCLC cell line were purchased from Shanghai Mingjin Biolo Gy Co., Ltd. (Shanghai, China), while the human A549 NSCLC cell line was purchased from the American Type Culture Collection (ATCC). The lines and mouse lung cancer cells (LLC) mouse lung cancer cell line was provided by Nanjing University. LLC and A549 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; eallbio, China) containing 10% fetal bovine serum (FBS) and 1% penicillin−streptomycin. Jurkat and H1299 cells were cultured in RPMI 1640 medium (eallbio, China) containing 10% FBS and 1% penicillin−streptomycin. All cells were maintained at 37°C in a humidified atmosphere containing 5% CO₂.

LLC, H1299, and A549 cells were seeded into T25 culture flasks (Corning, USA). When cell density reached 75%–85%, the cells were exposed to X-ray irradiation doses of 3, 6, 9, and 12 Gy using a linear accelerator (Elekta Limited, Sweden) at a dose rate of 3 Gy per minute. During irradiation, a 1.5-cm-thick bolus was used to adjust the radiation distribution, with the T25 flask positioned on the treatment couch. The radiation properties were as follows: 6-MV photon beam energy, 100 cm source-surface distance, 10×10 cm² radiation field size, and 180° gantry angle. Dosimetry was performed using a cylindrical ionization chamber prior to irradiation.

Male C57BL/6 mice, aged six weeks and weighing 18 ± 2 g, were obtained from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China) and maintained in a pathogen-free environment. To establish bilateral tumors, 4×10⁶ LLC cells were injected subcutaneously into the left hind limb to create the primary tumor, while an additional 4×10⁶ LLC tumor cells were implanted in the right limb to form the secondary tumor. When the secondary tumor reached approximately 500 mm³ in size, mice bearing bilateral tumors were randomly assigned to various treatment groups.

Tumor progression was monitored and quantified using previously described methods. Prior to radiation therapy, each mouse was shielded with a lead plate to ensure that only the targeted tumor was exposed to radiation. An Elekta Synergy instrument (Elekta limited, Sweden) was used to administer the radiation treatment. An anti-PD-L1 antibody (Bioxcell, USA) was administered via intraperitoneal injection, with an initial dose of 200 µg followed by subsequent doses of 100 µg every three days for a total of five injections. An anti-CTLA4 (Bioxcell, USA) was also administered intraperitoneally at a dose of 100 µg per mouse every three days for a total of five injections. In accordance with ethical standards, mice were euthanized when the implanted tumors reached a volume of 2000 mm³.

The Tumor Immune Estimation Resource (TIMER) database contains 10,897 samples from 32 cancer types sourced from the Cancer Genome Atlas (TCGA). The differential expression of CD8A, PD-L1 (CD274), and CTLA4 between tumors and normal tissues was examined using the TCGA database, and the significance of the difference was determined using the Wilcoxon test. A p-value of less than 0.05 was considered significant.

In this study, we utilized the Kaplan-Meier plotter database to investigate the relationship between the presence of CCL17 expression and survival duration in individuals with lung cancer. The database enables analysis of the impact of 54,000 genes (including mRNA, miRNA, and protein) on survival rates across 21 cancer types. Data for our analysis were obtained from the Gene Expression Omnibus (GEO), European Genome-phenome Archive (EGA), and TCGA databases, which provided comprehensive information. To thoroughly examine the association between CCL17 expression in lung cancer and overall survival (OS), we calculated hazard ratios (HRs) with corresponding 95% confidence intervals (CI) and log-rank P-values (27). This approach allowed for a robust statistical assessment of the prognostic significance of CCL17 in lung cancer.

The Tumor and Immune System Interaction Database (TISIDB) is a comprehensive resource designed to analyze relationships between tumors and various immune components, including tumor-infiltrating lymphocytes (TILs), immunomodulators, and major histocompatibility complexes (MHCs) (28). The TISIDB enables researchers to explore gene functions and their impact on tumor-immune interactions. We employed the Spearman correlation test to evaluate associations between prognostic genes and TILs. All hypothesis tests were conducted using two-sided considerations, with P < 0.05 considered statistically significant.

Tumor samples were excised, diced, and then suspended in phosphate buffered saline (PBS) to create a cell suspension. The tumor fragments were then digested with a combination of collagenase, deoxyribonuclease I, and hyaluronidase (Sigma, USA) for 50 minutes at 37°C in a temperature-controlled shaker. The resulting supernatant was filtered through a 70 µm mesh (BD Falcon, USA). Immune cells were isolated using a mouse tumor-infiltrating mononuclear cell isolation kit (Solar Bio, China) and resuspended in PBS at a concentration of 1 × 10⁷/mL. Subsequently, 100 µL of the cell suspension was stained with the fluorescently labeled anti-mouse antibodies: CD3-FITC, CD8-APC (all from Biolegend, USA) for 30 minutes on ice. After two PBS washes, the cells were resuspended for analysis using flow cytometry. These data were analyzed using NovoExpress software (Agilent, USA).

Mice were euthanized by cervical dislocation, and tumors were excised, rinsed twice with DEPC-treated water, minced into small fragments, and transferred to cryostorage tubes. These samples were then stored in liquid nitrogen and sent to Shanghai OE Biotech Co., Ltd. (Shanghai, China) for sequencing. Libraries were sequenced on an Illumina NovaSeq 6000 platform, generating 150 bp paired-end reads. Approximately 50 million raw reads were obtained per sample. Raw reads in FASTQ format were processed using fastp (29) to remove low-quality reads, resulting in approximately 45 million clean reads per sample for subsequent analyses. Clean reads were mapped to the reference genome using HISAT2 (30). FPKM values for each gene were calculated, and read counts were obtained using HTSeq-count (31). Principal Component Analysis (PCA) was performed using R (v3.2.0) to evaluate biological replication of samples. Differential expression analysis was conducted using DESeq2 (32), with a threshold of q-value < 0.05 and fold change > 1 or < 0.5 for significantly differentially expressed genes (DEGs). Hierarchical cluster analysis of DEGs was performed using R (v3.2.0) to illustrate gene expression patterns across different groups and samples. A radar map of the top 30 genes was generated using the R package ggradar to visualize up- and downregulated DEGs. Gene Ontology (GO), KEGG pathway, Reactome, and WikiPathways enrichment analyses of DEGs were performed using R (v3.2.0) based on the hypergeometric distribution to identify significantly enriched terms. Column diagrams, chord diagrams, and bubble plots were generated to visualize significant enrichment terms. Gene Set Enrichment Analysis (GSEA) was conducted using GSEA software (33). This analysis employed predefined gene sets, with genes ranked according to their differential expression between two sample types. The analysis then tested whether the predefined gene sets were enriched at the top or bottom of the ranking list.

Tumor samples were fixed in 10% neutral buffered formalin, embedded in paraffin wax, and subsequently stained with hematoxylin and eosin (H&E) or subjected to immunohistochemistry. For immunostaining, tumor sections were deparaffinized, rehydrated, and subjected to antigen retrieval. Sections were then incubated overnight at 4°C with primary antibodies: rabbit anti-CD8 (1:100, Bioss, China), mouse anti-PD-L1 (1:200, Protein, China), mouse anti-CTLA-4 (1:50, Santa Cruz, China), and rabbit anti-CCL17 (1:200, Abcam, UK). Following primary antibody incubation, sections were treated with corresponding secondary antibodies (Genentech, USA). Immunoreactivity was then visualized using diaminobenzidine (DAB) as the chromogen. For quantification, multiple high-power fields were randomly selected from each section, and positively stained cells were counted per unit area.

A549 cells, irradiated and unirradiated at various doses, were resuspended and plated in 24-well plates with 300 µL of complete medium. After 3 days, the supernatant was collected and CCL17 expression levels were evaluated using ELISA kits (Mlbio, China) according to the manufacturer’s instructions.

Total RNA was extracted using TRIzol reagent (TaKaRa, Japan). Complementary DNA (cDNA) was synthesized using the Prime-Script RT reagent kit (CWBIO, China). Radiation-induced mRNA expression was detected using UltraSYBR mixture (CWBIO, China). Actin levels were used as a reference for normalization. The experiment was performed in triplicate, with RT-PCR reactions repeated independently three times. Primer sequences are provided in the Supplementary Table .

Jurkat cells (2 × 10⁵ cells/mL) were plated in 200 μL of serum-free medium (HyClone, USA) in the upper chamber of a 5 µm Transwell insert (Corning, USA). The lower chamber contained 500 μL of medium supplemented with 10% FBS (Albino, China) and either CCL17 chemokines (5 µg/mL) (Sangon, China) or the cell supernatant from irradiated (9 Gy) A549 and H1299 cells, with or without anti-CCL17 mAb (R&D Systems, USA). After a 3-hour incubation, migrated cells were fixed with 4% paraformaldehyde, stained with 0.1% crystal violet for 10 minutes, and washed thrice with PBS. Quantification was performed by analyzing five randomly selected fields.

Data were analyzed using SPSS 22.0 and GraphPad Prism 9.4. Results are expressed as mean ± standard deviation (SD). Two-tailed Student’s t-test was used for comparisons between two groups, while one-way analysis of variance (ANOVA) was employed for comparisons among multiple groups. Results with P < 0.05 were considered statistically significant.

Case 1 included a 61-year-old man with squamous cell carcinoma of the lung with nasal metastasis. The patient presented to the hospital with primary lung cancer of the left upper lung (December 2020). Later, the patient developed nasal metastasis (August 2021). Nasal radiotherapy (60 Gy/30 f) (December 2021) was performed on the patient. During radiotherapy, the patient’s left face and left jaw developed new tumors from the previous ones ( Figure 1A ). Due to the large irradiation range of the patient’s face, radiotherapy was not performed for the new tumors considering the patient’s facial injuries and other problems. Low-dose radiotherapy (≈20 Gy/30 f) was administered to the left facial tumor in the irradiation path of the nasal tumor, and the nasal tumor significantly shrank after the end of radiotherapy, but the left facial tumor did not shrink. Two cycles of anti-PD-L1 (Etrilizumab) therapy (January 2022) were subsequently performed, and the new tumors on the left face significantly subsided (April 2022) ( Figure 1B ). However, no significant shrinkage was observed in the lung and left mandibular areas that were not irradiated ( Figures 1C, D ).

Case 2 involved a 71-year-old male who was diagnosed with squamous cell carcinoma of the upper left lung lobe in May 2024. PET-CT imaging revealed increased FDG uptake in the left hilar lymph node (station 10), suggesting tumor metastasis to this location ( Supplementary Figure S1A ). Subsequently, in June 2024, the patient underwent SBRT with the upper left lung lobe tumor receiving a dose of 24 Gy/3f, while the adjacent 10L lymph node received a low dose of 5 Gy ( Supplementary Figure S1B ). After completing three cycles of chemotherapy combined with immunotherapy, a CT scan showed a reduction in the size of the upper left lung tumor and alleviation of atelectasis ( Supplementary Figure S2 ). Following this, the patient underwent radical surgery, and postoperative pathology confirmed a complete pathological response in the hilar lymph nodes at station 10.

In Case 3, the patient is a 59-year-old male who was diagnosed with squamous cell carcinoma of the upper right lung lobe in July 2024. PET-CT imaging revealed increased FDG metabolism in the right upper paratracheal lymph node (station 2R), suggesting a possibility of metastasis ( Supplementary Figure S3A ). Subsequently, starting from August 2024, the lesion in the upper right lung lobe underwent SBRT with a prescribed dose of 24 Gy delivered in 3 fractions. The adjacent 2R lymph node received a low dose of 5 Gy for the majority of its volume, while the more distant cephalic portion of the lesion received only a dose of 3 Gy ( Supplementary Figure S3B ). Following three cycles of chemotherapy combined with immunotherapy, a CT scan demonstrated a partial response of the tumor lesion ( Supplementary Figure S4 ). Later, in October 2024, a lymph node biopsy was performed, and postoperative pathology indicated a complete pathological response in the lymph nodes of station 2R.

These cases suggest that the combination therapy of HDRT + LDRT + ICI may further activate the immune response and enhance anti-tumor efficacy. Therefore, we intend to explore whether this combination therapy mode can be used in some special cases of patients with multiple tumors or multiple metastases that are not suitable for large-area irradiation.

To investigate the effects of diverse irradiation doses on tumor growth inhibition and immune cell recruitment, we developed a bilateral LLC lung cancer model in C57BL/6J mice. To minimize radiation-induced damage to normal tissues, tumor cells were subcutaneously implanted into both lower limbs. The left limb received a uniform dose of 20 Gy, designated as the “primary tumor,” while the right limb was irradiated with varying doses of 3, 6, 9, or 12 Gy, termed the “secondary tumor” ( Figure 2A ). Tumor size was measured every other day following the initiation of irradiation. Our results demonstrated that the curative dose of 20 Gy effectively suppressed the growth of the “primary tumor” across all groups, with no significant differences in tumor size ( Figures 2B, C ). By day 9 post-irradiation, the average tumor volume in the “secondary tumor” groups receiving 3, 6, 9, and 12 Gy was reduced compared to the non-irradiated group, with the 20 + 6 Gy group exhibiting the most pronounced tumor inhibition ( Figures 2B, D ). To assess treatment-related toxicities, we monitored changes in mouse body weight and found no significant differences among the groups ( Figure 2E ). In conclusion, combined treatments using various irradiation dose regimens demonstrated inhibitory effects on tumor growth in mice, with the 20 + 6 Gy group exerting the most substantial influence. Notably, this treatment approach did not induce significant toxic side effects in the mice.

To investigate gene expression in the “secondary tumor” post-irradiation, we collected mouse tumor samples for RNA sequencing. Transcriptomic analysis was performed on the sequencing data, integrating gene detection values to generate expression box plots and conduct PCA ( Figure 3A ). While reproducibility across groups was generally satisfactory, occasional biased data points were observed. Consequently, we selected data points exhibiting reasonable repeatability between sample groups for subsequent analysis. Our findings revealed that the 20 + 6 Gy treatment regimen induced the most pronounced variations in gene expression within the “secondary tumor.” KEGG pathway enrichment analysis of DEGs across groups ( Supplementary Figure S5A ) indicated that the majority of these genes were primarily implicated in immune system functions. Given that CD8⁺ T cells are central to antineoplastic immunity, we focused our analysis on CD8 expression across various tumors. Using the TIMER database, we assessed CD8 expression between TCGA tumors and adjacent normal tissues. Notably, CD8 expression was significantly reduced in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) tissues compared to adjacent normal tissues ( Figure 3B ). Additionally, CTLA4 expression was increased, while PD-L1 expression decreased in LUAD and LUSC tissues ( Supplementary Figures S5B, D ). CD8⁺ T-cell infiltration is frequently associated with favorable prognosis in solid tumors (34). Using the Kaplan-Meier plotter database, we investigated the correlation between these factors and lung cancer patient survival, finding that PD-L1 and CTLA4 did not significantly affect OS ( Supplementary Figures S5C, E ). Integrating transcriptomic analysis results, we performed pathway enrichment on immune system-related DEGs across groups ( Supplementary Figure S5F ). The 20 + 6 Gy group exhibited the highest number of upregulated differential genes. Further examination via the GEPIA database revealed that the majority of these upregulated genes were expressed at higher levels in normal lung tissues, suggesting that changes in the 6 Gy irradiated group might potentially improve lung cancer prognosis. Immunohistochemical staining for CD8 on tumor samples demonstrated that CD8⁺ T-cell infiltration was most pronounced in the 20 + 6 Gy group ( Figures 3C, D ). These findings confirm that the 20 + 6 Gy regimen represents the optimal combined irradiation dose for effectively recruiting CD8⁺ T cells.

To further validate the efficacy of combined RT and immunotherapy in tumor treatment, we incorporated two ICIs, anti-PD-L1 and anti-CTLA4, with RT in the 20 + 6 Gy group ( Figure 4A ). Nine days post-irradiation, the average tumor volume of the “secondary tumor” was significantly reduced in the 20 + 6 Gy+αPD-L1 group (293 mm³) and the 20 + 6 Gy+αPD-L1+αCTLA4 group (235 mm³) compared to the 20 + 6 Gy group (395 mm³) ( Figures 4B, C ). Notably, the combination of both ICIs resulted in more pronounced tumor suppression, with statistically significant differences observed. These findings demonstrate that LDRT combined with ICI treatment further enhances the inhibitory effect on mouse tumors without inducing apparent adverse effects. To assess tumor-immune infiltration, we analyzed CD8⁺ cell populations within CD3⁺ cells as characteristic markers for infiltrating T-cell populations. In the 20 + 6 Gy, 20 + 6 Gy+αPD-L1, and 20 + 6 Gy+αPD-L1+αCTLA4 groups, the percentages of CD8⁺ T cells were 2.12%, 11.13%, and 20.69%, respectively ( Figures 4D, E ). Immunohistochemical staining of mouse tumors corroborated these findings, revealing increased infiltration of CD8⁺ T cells in tumors treated with combined RT and ICIs compared to those treated with RT alone ( Figures 4F, G ). These results suggest that combined treatment with 20 + 6 Gy irradiation and ICIs can further enhance tumor inhibition and immune activation within the tumor microenvironment in mice.

To elucidate the mechanism by which the 20 + 6 Gy group maximally activates the immune system, we conducted a comprehensive analysis of tumor RNA sequencing results. Among the DEGs related to the immune system in the 20 + 6 Gy group, 23 genes were upregulated and two were downregulated. We analyzed the expression of these genes in cancerous and adjacent normal tissues using the TIMER database, categorizing upregulated genes highly expressed in adjacent normal tissues as “up genes,” suggesting their potential to improve patient prognosis. Through rigorous screening, we identified 19 “up genes” ( Figure 5A ). Analysis of differences in expression levels across experimental and control groups ( Figure 5B ) revealed CCL17 as the most significantly upregulated gene in the 20 + 6 Gy group, prompting its selection as a potential target for antitumor effects in subsequent studies. Using the TIMER database, we analyzed CCL17 expression across all TCGA tumor types, focusing on the contrast between tumor tissues and adjacent normal tissues ( Figure 5C ). CCL17 expression was significantly lower in LUAD and LUSC tissues compared to adjacent normal tissues (p<0.0001). Investigation of the relationship between CCL17 expression levels and survival outcomes in lung cancer patients using the Kaplan-Meier mapper database demonstrated that high CCL17 expression was significantly associated with favorable prognosis (OS, HR=0.79 (0.70-0.89), p=0.00012) ( Figure 5D ). Furthermore, exploration of the correlation between CCL17 expression and key pathways in NSCLC patients ( Figures 5E, F ) revealed a positive correlation with pro-inflammatory pathways and a significant negative correlation with tumor proliferation pathways. These findings suggest that in patients with high expression and favorable prognosis, CCL17 likely exerts dual effects of promoting inflammation and inhibiting tumor proliferation.

When investigating the correlation between CCL17 expression and immune cell infiltration in tumors, initial analysis using the TISIDB database revealed a positive correlation between CCL17 expression and CD4 and CD8 expression ( Supplementary Figure S6A ). To validate these findings, we performed immunohistochemistry and multiple immunofluorescence staining for CD8 and CCL17 on tumor sections from the initial mouse group ( Figure 6A-C , 3G , Supplementary Figure S6B ). The 20 + 6 Gy group showed the highest expression of CD8 and CCL17 compared to other groups, with an increased CD8⁺ T-cell population. These results suggest that increased CCL17 production may facilitate CD8⁺ T-cell infiltration in the tumor-immune microenvironment. We hypothesize that CCL17 exerts a chemotactic function on T cells within the tumor-immune microenvironment, attracting activated CD8⁺ T cells to the tumor site to elicit antitumor effects. Previous in vivo studies have demonstrated that CCL17 enhances T-cell infiltration into the tumor microenvironment. We subsequently validated these findings through in vitro experiments using two common NSCLC cell LLC. Irradiated progeny cells were exposed to doses of 3, 6, 9, and 12 Gy, and their survival was monitored ( Supplementary Figures S6C–E ). ELISA analysis of irradiated progeny cell supernatants revealed an overall increase in CCL17 secretion compared to unirradiated tumor cells. However, this increase peaked at a specific dose before decreasing, indicating that CCL17 expression is not linearly dependent on radiation dose. Notably, CCL17 levels in LLC supernatants peaked at 6 Gy, which is consistent with in vivo data, while A549 and H1299 cell supernatants showed peak CCL17 levels at 9 Gy. To investigate T-cell activation, Jurkat cells were cultured with varying concentrations of CCL17 cytokines. After 72 hours, qPCR analysis of T-cell activation markers (IL-2, granzyme, and perforin) revealed a significant increase in Jurkat cell activation, up to 7-fold, following CCL17 introduction ( Figures 6D-F ). Co-culture of Jurkat cells with irradiated and unirradiated lung cancer cells for 72 hours further enhanced T-cell activation ( Figures 6G-I ). Moreover, transwell assays demonstrated that CCL17 enhanced Jurkat cell migration ( Figure 6J ), while the addition of a CCL17 neutralizing antibody to A549/9 Gy and H1299/9 Gy supernatants inhibited this migration ( Figures 6K, L ).

To summarize, bioinformatic analysis revealed a positive correlation between CD4 and CD8 expression and CCL17. In addition, immunohistochemistry and multiple immunofluorescence studies demonstrated that increased CCL17 expression in tumors corresponded with increased CD8 infiltration. Further, in vitro studies confirmed substantial CCL17 chemokine secretion from precursor cells, which stimulated T-cell activation and migration.