Purpose:

Efforts have been devoted to select eligible candidates for PD-1/PD-L1 immune checkpoint blocker (ICB) immunotherapy. Here, we have a serendipitous finding of positron emission tomography (PET) imaging tracer 2-[18F]FDG as a potential immunomodulator. Therefore, we hypothesize that 2-[18F]FDG could induce PD-L1 expression change and create an immune-favorable microenvironment for tumor immunotherapy.

Experimental Design:

We designed a series of assays to verify PD-L1 upregulation, and tested immunotherapy regimens based on 2-[18F]FDG and anti–PD-L1 mAb, as monotherapy and in combination, in fully immunocompetent mice of MC38 and CT26 models. PD-L1 expression and tumor microenvironment (TME) changes were analyzed by Western blot, transcriptomics study, and flow-cytometric analysis.

Results:

PD-L1 was upregulated in a time- and dose-dependent manner after being induced by 2-[18F]FDG. The activation of NF-κB/IRF3 pathway and STAT1/3-IRF1 pathway play crucial parts in modulating PD-L1 expression after DNA damage and repair. Improved αPD-L1 mAb utilization rate and significant tumor growth delay were observed when the personalized therapeutic alliance of 2-[18F]FDG stimulation and ICB was used. In addition, combination of 2-[18F]FDG with αPD-L1 mAb could reprogram a TME from “cold” to “hot,” to make low immunoactivity tumors sensitive to ICB therapy.

Conclusions:

In summary, this promising paradigm has the potential to expand the traditional tumor theranostics. 2-[18F]FDG-based ICB immunotherapy is highly significant in enhancing antitumor effect. A research of 2-[18F]FDG-based ICB immunotherapy has been proposed to enhance the antitumor effect.

Translational Relevance

ICB immunotherapy benefits only a fraction of patients, with clinical trials reporting response rates less than 20% under most circumstances. Our study showed that 2-[18F]FDG could regulate PD-L1 expression levels, which is associated with therapeutic effect toward PD-L1 inhibitors in malignances. By combining 2-[18F]FDG with αPD-L1 mAb immunotherapy, we developed a promising paradigm of cancer immunotherapy that could reprogram TME from “cold” to “hot” to make low immunoactivity tumors sensitive to ICB therapy. As 2-[18F]FDG has been routinely used in cancer diagnosis, the combination of 2-[18F]FDG and αPD-L1 mAb as an effective strategy in cancer therapy should be further explored clinically.

The great success of immunotherapy has initiated a new phase in cancer treatment, and an increased understanding of mechanisms leads to the development of various immune checkpoint blockers (ICB). ICB antibodies targeting the immune checkpoint programmed death receptor 1 (PD-1) expressed on T cells and its immune regulatory ligand PD-L1 have revolutionized the oncology (1, 2). Unfortunately, immunotherapies with PD-1/PD-L1 ICBs alone do not produce enough anticancer effect in most malignancies (3, 4). There is a general consensus that targeting PD-L1 is not equally successful in PD-1/PD-L1 ICB immunotherapy and should have tumor PD-L1 expression as a prerequisite. Thereby, tumor PD-L1 expression is deemed to be a predictive biomarker for pinpointing potential candidates who might benefit most from PD-1/PD-L1 immunotherapy (5, 6).

It is known that external beam radiation triggers immunogenic tumor cell death (7, 8) and local release of inflammatory cytokines that increase immune cell infiltration and activation (9, 10). By activating inflammatory cytokine signaling, such as IFNγ, TNFα, low-dose radiation could remodel a tumor microenvironment (TME) that is beneficial to the proliferation of antitumor immune cells. If it can be combined with immunotherapy, that should achieve favorable therapeutic effect. However, for most patients with metastatic cancers, it is impractical to deliver low-dose immunomodulatory radiation to all tumors using external beam radiotherapy because of the poor targeting to occult sites and the distinct toxicity results from large-field or whole-body radiation (11). Stereotactic body radiotherapy (SBRT) is a type of radiotherapy in which special equipment is used to precisely deliver a dose of radiation to tumors. Although SBRT with potential abscopal effects on immune response (12) has been well documented in a series of clinical trials, tumor size, location, and its adjacent organs will affect therapeutic regimen formulation, which could potentially contraindicate treatment (13). For the immunomodulation following radionuclide therapy is still a novel area of investigation without prospective multi-center trial evidence at present. Hence, we aimed to combine the targeted radionuclide therapy with immunotherapy to improve the specific targeting and enhance the therapy effects of metastatic cancers.

Recent study showed that PD-L1 expression of cancer cells was transiently upregulated following ionizing radiation (IR), and when the PD-1/PD-L1 interaction was blocked with anti–PD-1 antibody during this upregulation, T-cell activity was recovered and tumor growth was delayed in an immunocompetent mouse tumor model (14). Another study demonstrated that low levels of DNA damage led to cell-cycle arrest and promoted DNA damage repair, whereas serious DNA damage could result in cell apoptosis (15). Beta-emitting radionuclides with long half-life such as 177Lu, 90Y, and 131I, are actively pursued in targeted radionuclide therapy to destroy the tumor cells. We speculated whether diagnostic radionuclide could play a role in such combination therapy. Based on the consideration of radionuclide characteristics, if 18F-based PET tracers can be applied for radionuclide therapy, there would expand the traditional tumor theranostic mode, as glucose analogue [18F]fluorodeoxyglucose (2-[18F]FDG) is the most commonly prescribed PET radiotracer in nuclear medicine.

In this study, we explored the potential of 2-[18F]FDG, as an immunomodulator, to increase T cell infiltration, to reconstitute TME, and meanwhile, to upregulate PD-L1 expression. We demonstrated for the first time to our knowledge, across multiple tumor cell lines and mouse models including patient-derived xenografts (PDX), that 2-[18F]FDG could result in temporary upregulation of PD-L1 expression in vitro and in vivo. With growing evidences (16, 17) that upregulation of PD-L1 expression is beneficial to PD-1/PD-L1 immune modulation therapy, we would like to conclude that 2-[18F]FDG is a potential coagent for building an immune-favorable microenvironment for enhancing the efficacy of anti–PD-L1.

General remarks

All chemicals were obtained commercially. The murine colorectal cancer cell lines (MC38 and CT26) were purchased from the China National Infrastructure of Cell Line Resource. 4T1 breast cancer and B16F10 melanoma cell lines were purchased from the American Type Culture Collection (ATCC). The authenticity of these cell lines was ensured using short tandem repeat analysis. Aliquots of cell culture supernatants from cells in active culture were evaluated for Mycoplasma contamination using a PCR-based method. Cells were maintained in RPMI1640 medium supplemented with 10% fetal bovine serum (FBS) in 5% CO2 incubator at 37°C. All experiments were done with cells acquired and frozen within 1 month to maintain the phenotype of the cell lines, and were used within 6 months of testing. InvivoPlus anti-mouse PD-L1 mAb (BP0101) was purchased from Bio X Cell Co. Ltd. Anti-human PD-L1 (SHR-1316) was provided from Hengrui Co. Ltd. Antibodies used for Western blot assays were purchased from Abcam or Cell signaling Technology Inc. PET imaging studies were performed by a microPET device (Siemens Inveon PET). The radioactivity was measured with γ-counter (Wizard 2480, Perkin-Elmer) and CRC-25R dose calibrators (CAPIN-TEC Inc. USA). Cell immunofluorescence was performed by laser scanning confocal microscope (Olympus FV1200). IHC of tissues were detected by microscope of Leica DM4 B (Leica). Flow cytometry was detected with Beckman Coulter CytoFLEX (Beckman Coulter).

Animal models

BALB/c mice, BALB/c nude mice, and C57BL/6 mice (female, 6–8 weeks age) were used in our studies. All animals were randomly assigned to the experimental groups, obtained from Vital River Laboratory Animal Technology Co., Ltd and housed with a 12-hour light/dark cycle at 22°C and food and water ad libitum. All animal protocols were approved by the Institutional Animal Care and Use Committee of Laboratory Animals Center for Xiamen University (ID XMULAC20190150).

Murine subcutaneous tumor models

The right rear flanks of female BALB/c mice were given a suspension of CT26 colorectal tumor cells (2 × 106 tumor cells in 100 µL PBS) subcutaneously. For female C57BL/6 mice, subcutaneous tumors were generated by inoculating MC38 colorectal cancer cells (2 × 106 tumor cells in 100 µL PBS) suspension. The tumor volume (mm3) was using a digital vernier caliper and calculated as length × width2/2.

NSCLC PDX models

In this study, fresh tumor samples from patients with NSCLC were implanted subcutaneously into BALB/c nude mice to establish NSCLC-PDX models. The PDX model has been validated to retain the morphology and molecular signatures of the corresponding parental tumor in previous study (18). The studies involving human participants were approved by the Clinical Research Ethics Committee of the First Affiliated Hospital of Xiamen University (ID KYZ2017–001). All procedures involving human participants were carried out in accordance with the ethical standards of the national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from patients before the utilization of all clinical samples and data. Briefly, NSCLC specimens were surgically removed and immediately placed in DMEM supplemented with 2% antibiotics. To establish PDX models, the right upper limbs of female BALB/c immune-deficient mice were implanted subcutaneously fresh tumor specimens within an average of two hours following the patient's surgery. The mouse was euthanized to remove the tumor until it reached a volume of 1,000 mm3. For investigational purposes, the tumor fragments were implanted in BALB/c nude mice to expand a higher number of PDXs to obtain statistically relevant results. Models were subjected to imaging study when the tumor volume reached 100 mm3.

Cellular uptake of Cy5.5-αPD-L1

For flow cytometry experiments, the cells were incubated with 2-[18F]FDG (1.85 MBq/mL) overnight and then replaced with fresh medium (without serum) containing Cy5.5-αPD-L1. After further incubation for different amounts of time (10, 30, 60, 120, and 240 minutes) at 37°C, the cells were washed twice with cold PBS and collected, then resuspended in 200 µL PBS for flow cytometric analysis (BD Biosciences). Untreated cells were used as control. A total of 10,000 events were collected for each sample.

RNA isolation, first-strand cDNA synthesis, and qRT-PCR

To quantify PD-L1 mRNA expression, CT26 and MC38 colorectal tumor cells were seeded in 6-well plates to grow overnight. Then the cells were incubated with 2-[18F]FDG (1.85 MBq/mL) for different times (2, 4, and 24 hours) at 37°C. Total RNA was isolated from CT26 and MC38 tumor cells using the RNeasy Mini Plus Kit (Sangon Biological). The cDNA was obtained by reverse transcription using the Maxima First-Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific), according to the manufacturer's instructions. iScript and SsoAdvanced SYBR Green SuperMix (Bio-Rad) were used for RT-PCR gene expression analysis on ABI StepOne Plus Real-Time PCR system. Amplification of specific transcripts was confirmed by melting curve profiles generated at the end of the PCR program. Expression levels of target genes were normalized to the housekeeping gene (β-actin) and were calculated based on the comparative cycle threshold method (2–ΔΔCt) and shown by heat map with GraphPad prism 7.0 software. All the measurements were performed in triplicate.

Transcriptomics study

Transcriptomics study was performed by HaploX Genomics Center, Ltd. MC38 and CT26 tumor cells were seeded in 6-well plates overnight and then incubated with 2-[18F]FDG (1.85 MBq) for 24 hours. After that, these tumor cells were collected and high-throughput sequencing was performed. 2-[18F]FDG-untreated cells were used as control.

MC38 tumor tissue samples of different groups were collected for high-throughput sequencing. Reference genome and gene model annotation files were downloaded from the genome website directly. Paired-end clean reads were aligned to the reference genome using HISAT2 v2.1.0 (hierarchical indexing for spliced alignment of transcripts), which is a highly efficient system for aligning reads from RNA sequencing experiments.

Immunofluorescence, histology, immunohistochemistry, and microscopy

For immunofluorescence staining, CT26, MC38, 4T1, and B16F10 tumor cells were seeded in confocal dishes. Each sample was fixed with 100 µL of 4% paraformaldehyde for 10 minutes. After that, the cells were washed three times with PBS and incubated with 10% goat serum for 30 minutes to reduce nonspecific binding. Cells were stained with the first antibody: anti–PD-L1 antibody (clone: EPR19759, Abcam, ab213524) overnight, its concentration was set to 1/200, then rinsed three times with PBS. After cells were stained for 1 hours with secondary antibody Alexa Fluor Plus 488–conjugated IgG and washed three times with PBS, cell nuclei were stained blue with DAPI (Invitrogen Molecular Probes). To identify DNA double-strand break (DSB) and DNA repair level, MC38 cells were stained with γH2AX and EdU after irradiation with 2-[18F]FDG. Anti-γH2AX (phospho S139) antibody (clone: EP854(2)Y, Abcam, ab81299) was used to identify DNA double-strand break, its concentration was set to 1/200. Edu staining was performed used BeyoClick Edu-594 Cell Proliferation Detection Kit (Product ID: C0078S), Edu working solution concentration was set to 1/1,000. For histologic analysis, tissue specimens were fixed with 10% buffered formalin, dehydrated in ethanol, embedded with paraffin, and stained with H&E. Immunohistochemistry on frozen or paraffin-embedded mouse tissues was performed using antibodies directed against PD-L1, CD4 (clone: EPR19514, Abcam, ab183685), IFNγ (clone: R4–6A2, Santa Cruz Biotechnology, sc-53700), CD8 (polyclone, Abcam, ab203035). For paraffin-embedded samples (PD-L1, CD4, IFNγ, CD8), samples were dewaxed in ethanol, followed by antigen retrieval with 0.01 mol/L sodium citrate with 0.05% Tween. Immunofluorescence staining on frozen mouse tissues was performed using antibodies against PD-L1, ki67 (polyclonal, Abcam, ab15580), caspase3 (polyclonal, Abcam, ab13847) and DAPI. Immunofluorescence images were acquired using the Zeiss LSM880 confocal microscope with ZEN 2010 software. Histologic and Immunohistochemistry images were acquired using the Leica DM4 B upright digital research microscopes (Leica) with Leica Application Suite X (LAS X). All the images were analyzed with ImageJ 7.0 software, the operation for relative quantitation of immunofluorescent images as follows (19): (i) measurement of mean fluorescence intensity in a region of interest; (ii) automated cell counting from tissue sections; (iii) automated determination of mean fluorescence intensity of cells PD-L1 protein; (iv) processing and analysis the data with GraphPad prism 7.0 software.

Western blot analysis

Western blot was performed as described previously (20), with minor modifications. Briefly, cell lysates were made in ice-cold RIPA buffer containing complete protease inhibitor cocktail and phosphatase inhibitor cocktail (Sigma). Total protein was quantified using the BCA Assay according to the manufacturer's instructions (Thermo Fisher Scientific). Ten percent of Bis-Tris polyacrylamide gels was loaded with 20 μg of protein, electrophoresed at 120 V, and electro-transferred to PVDF membranes. After blocking with 5% BSA, membranes were probed with primary antibodies to PD-L1, IRF-1 (clone: D5E4, CST, #8478), STAT1 (clone: D1K9Y, CST, #14994), phospho-STAT1 (Phospho Tyr701, clone: D4A7, CST, #7649), STAT3 (clone: 79D7, CST, #4904), phospho-STAT3 (phospho Y705, clone: EP2147Y, Abcam, ab76315), NF-κB p65 (clone: E379, Abcam, ab32536), phospho-NF-κB p65 (clone: 93H1, CST, #3033), IRF3 (clone: EPR2418Y, Abcam, ab68481), phospho-IRF-3 (Phospho Ser379, clone: E6F7Q, CST, #79945), β-actin (clone: 13E5, CST, #4970), and tubulin (clone: 9F3, CST, #2128). Blots were developed by ECL (Thermo Fisher Scientific).

Flow cytometry of cell lines and organs

Tumor cells were seeded in 6-well plates overnight and treated with 2-[18F]FDG. Saline-treated cells were used as control. After incubation for different times (0.5, 2, 4, 8, and 24 hours), cells were harvested and washed with cold PBS, then stained with mouse anti–PD-L1 mAb (MIH6; Abcam). The secondary antibody labelled with fluorescein isothiocyanate (FITC) was used for detecting the primary antibodies. Fluorescence intensities of the stained cells were analyzed using a FACS Aria III flow cytometer (BD Biosciences). Data were assessed and analyzed quantitatively by FlowJo software version 10 (FlowJo).

Single-cell suspensions of mouse tumors were prepared for flow cytometry as described previously (21). In brief, samples were harvested and cut into small fragments (1–2 mm3), and placed in DMEM containing Collagenase IV (1 mg/mL; Gibco, USA), trypsin inhibitor (1 mg/mL; EMD Millipore), and DNase I (2 U/mL; Promega). The fragments were then incubated at 37°C for 60 minutes with gentle shaking every 10 min. Specimens were passed through a 70-µm mesh and centrifuged at 350 g for 5 minutes. Red blood cells were eliminated from the samples with a hypo-osmotic red blood cell lysis buffer (Solarbio). Each sample was fixed with 100 µL of 4% paraformaldehyde for 10 minutes. After that, the cells were collected and washed three times with PBS and incubated with 10% goat serum to reduce nonspecific binding. For detecting PD-L1 expression and T-cell alteration, preprocessed cells were stained as follows: anti-PD-L1; Teff T cells: CD45+ (clone: 30-F11, BioLegend, #103106), CD4+, IFNγ+ (clone: XMG1.2, BioLegend, #505805); Treg cells: CD45+, CD4+, Foxp3+ (clone: 150D, BioLegend, #320007); CTLs: CD45+, CD8+, IFNγ+. In addition, other immune cells were defined as follows: M1 macrophages: iNOS+ (clone: CXNFT, eBioscience, #12–5920–82), CD11b+ (clone: M1/70, BioLegend, #101211), F4/80+ (clone: BM8, BioLegend, #123120); M2 macrophages: CD206+ (clone: C068C2, BioLegend, #141706), CD11b+, F4/80+; dendritic cells (DC): CD80+ (clone: 16–10A1, BioLegend, #104705), CD86+ (clone: GL-1, BioLegend, #105007); myeloid-derived suppressor cells (MDSC): CD45+, CD11b+, Gr-1+ (clone: RB6–8C5, BioLegend, #108406), respectively. Effector memory T cells of spleen were defined as follows: CD44high (clone: IM7, BioLegend, #103012), CD4+, CD62Llow (clone: MEL-14, BioLegend, #104405) or CD44high, CD8+, CD62Llow. PDX model tumors were dissected and stained with anti-human PD-L1 mAb [EPR19759] to examine PD-L1 expression change.

Cytokine analysis

Serum samples were isolated from mice after various treatments and diluted for analysis. The proinflammatory cytokines including TNFα, IFNγ, and IL6 were determined by using enzyme-linked immune sorbent assay (ELISA) kits according to vendors’ protocols (Dakewe Biotech). The data were calculated and shown in a bar chart with GraphPad prism 7.0 software.

In vivo anticancer efficacy

As the tumor volume reached about 50 mm3, the CT26, MC38 or 4T1 tumor-bearing mice were randomly divided into different groups (n = 8/group) and treated with different schemes. An additional therapeutic course was scheduled on day 4. After initiation of radiotracer-related therapy, the feeding surroundings were shielded with lead bricks to protect them from any contact with extrinsic radiation. The tumor volume and body weight were monitored at the given time points. Mice were euthanized if the tumor volume exceeded 1,500 mm3. The percent survival of mice in each group was measured until all the mice had been sacrificed.

PET imaging

Small animal PET imaging studies were performed with an Inveon small-animal PET scanner (Siemens Preclinical Solution) at the given time points under the approved guidelines. The injected activities were identical to that in therapeutic trials. All the mice underwent 10-minute static PET scans at different time points after injection with 2-[18F]FDG. During the scan procedure, anesthesia was induced with 2% isoflurane/air mixture to maintain spontaneous breathing of mice. The Inveon Research Workplace (IRW) 2.0 software was used to reconstruct and acquire PET image data. PET images were reconstructed using three-dimensional ordered-subset expectation-maximization (3D OSEM) algorithm and with a Maximum a Posteriori (MAP) method. Injected activity and body weight were input before imaging to accomplish normalized and decay corrected radioactivity concentration. For quantitative comparisons, the tissue uptake was acquired by selecting the regions of interest (ROI) on images. For each scan, ROIs were drawn using vendor software (ASI Pro 5.2.4.0; Siemens Medical Solutions) on decay-corrected whole-body coronal images. These values were divided by the administered activity to obtain an image-ROI-derived percentage injected dose per gram (%ID/g).

Statistical analysis

For flow cytometry, immunofluorescence and immunohistochemistry analysis, WB, tumor growth curves and transcriptomics study, the statistical significance of observed differences between groups was assessed using the multiple comparison and a two-way analysis of variance (ANOVA) followed by a Tukey post hoc comparisons test with GraphPad Prism 7.0 software. Survival curve data were analyzed with the Kaplan–Meier method followed by the log-rank test, and analyzed with Bonferroni correction. Data are presented as the mean ± standard deviation. Statistical significance is defined at *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; and ****, P ≤ 0.0001 level, ns = not significant.

Data availability statements

The sequence data generated in this study are publicly available in Genome Sequence Archive (GSA) Family at CRA004993, and within the article and its Supplementary Data files. The sequence data were also deposited in an approved INSDC database: NCBI BioProject (Accession: PRJNA816563). Other data generated in this study are available within the article and its Supplementary Data files.

Additional information

The details of flow cytometry analysis, histological identification, therapeutic courses, and allocations are described in the Supplementary File.

Tumor PD-L1 expression is upregulated after 2-[18F]FDG stimulation in vitro

First, we validated the 2-[18F]FDG-induced PD-L1 upregulation in multiple tumor cell lines (melanoma, breast, and colorectal cancer cells). This stimulation was embodied prominently through the flow cytometric analysis. As shown in Fig. 1A and B, the proportions of PD-L1–positive cells in CT26, MC38, 4T1, and B16F10 tumor cells were significantly increased after co-incubation with 2-[18F]FDG. The percentages of PD-L1–positive cells increased from 23.3% to 96.5%, 54.3% to 98.7%, 21.8% to 60.6%, and 61.4% to 96.2% in CT26, MC38, 4T1 and B16F10 tumor cells after 24 hours, respectively. The immunofluorescence assay was also performed at 24 hours after 2-[18F]FDG treatment (Fig. 1C and D), which revealed that 2-[18F]FDG upregulated PD-L1 expression in different tumor cells. Heatmaps generated from reverse transcription-quantitative real-time PCR (qRT-PCR) analysis (Fig. 1E) found that 2-[18F]FDG increased the expression of PD-L1 mRNA on tumor cells. After 8-hour coincubation, 2-[18F]FDG increased PD-L1 mRNA on MC38 and CT26 cells by 53-fold and 17-fold, respectively. The expression of PD-L1 was elevated to a greater extent by a higher amount of 2-[18F]FDG activity, which was further confirmed by flow cytometric analysis in Fig. 1F, clearly indicating that within a reasonable range of 2-[18F]FDG radioactivity, PD-L1 was upregulated in a dose-dependent manner.

Figure 1.

PD-L1 expression of tumor cells is significantly upregulated after stimulation with 2-[18F]FDG. A and B, The increased PD-L1 expression on multiple tumor cell lines (CT26, MC38, 4T1, and B16F10 tumor cells) after 2-[18F]FDG (370 kBq) irradiation at different time points was assessed and analyzed by flow cytometry (A). Representative histograms were used to present the upregulation of PD-L1 after radionuclide stimulation (B). Data were expressed as mean ± SD (n = 3/group). Each experimental group was compared with the control (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). C and D, Confocal images of PD-L1 immunofluorescence staining in multiple tumor cell lines at 24 hours after coincubation with 2-[18F]FDG (370 kBq; C). The cells with nonradioactive treatment were set to controls. Representative histograms showed the upregulation of PD-L1 after radionuclide stimulation (D). Data were expressed as mean ± SD (n = 5/group). Each experimental group was compared with the control. E, Bar charts generated from quantitative RT-PCR were used to analyze the mRNAs encoding PD-L1 in MC38 and CT26 tumor cells after irradiation with 2-[18F]FDG. Data were expressed as mean ± SD (n = 6/group). Untreated cells were used as controls. F, The increase in PD-L1 expression on MC38 and CT26 tumor cells after coincubation with different amounts of 2-[18F]FDG was quantified by flow cytometry. G, Volcano plots of DEGs with MC38 (left) and CT26 (right) tumor cells 24 hours after coincubation with 2-[18F]FDG. H, KEGG bubble map of the functional pathways involved in the biological effect of MC38 tumor cells induced by 2-[18F]FDG. The sizes of dots represent the counts of DEGs in the corresponding pathway.

Figure 1.

PD-L1 expression of tumor cells is significantly upregulated after stimulation with 2-[18F]FDG. A and B, The increased PD-L1 expression on multiple tumor cell lines (CT26, MC38, 4T1, and B16F10 tumor cells) after 2-[18F]FDG (370 kBq) irradiation at different time points was assessed and analyzed by flow cytometry (A). Representative histograms were used to present the upregulation of PD-L1 after radionuclide stimulation (B). Data were expressed as mean ± SD (n = 3/group). Each experimental group was compared with the control (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001). C and D, Confocal images of PD-L1 immunofluorescence staining in multiple tumor cell lines at 24 hours after coincubation with 2-[18F]FDG (370 kBq; C). The cells with nonradioactive treatment were set to controls. Representative histograms showed the upregulation of PD-L1 after radionuclide stimulation (D). Data were expressed as mean ± SD (n = 5/group). Each experimental group was compared with the control. E, Bar charts generated from quantitative RT-PCR were used to analyze the mRNAs encoding PD-L1 in MC38 and CT26 tumor cells after irradiation with 2-[18F]FDG. Data were expressed as mean ± SD (n = 6/group). Untreated cells were used as controls. F, The increase in PD-L1 expression on MC38 and CT26 tumor cells after coincubation with different amounts of 2-[18F]FDG was quantified by flow cytometry. G, Volcano plots of DEGs with MC38 (left) and CT26 (right) tumor cells 24 hours after coincubation with 2-[18F]FDG. H, KEGG bubble map of the functional pathways involved in the biological effect of MC38 tumor cells induced by 2-[18F]FDG. The sizes of dots represent the counts of DEGs in the corresponding pathway.

Close modal

Transcriptomic analysis was performed to explore the potential mechanism of PD-L1 upregulation stimulated by 2-[18F]FDG. From the volcano plot (Fig. 1G), there were a total of 2,002 DEGs, which had changed in 2-[18F]FDG–treated MC38 cells compared with the control group, with 1,223 upregulated genes and 779 downregulated genes (|log2(FC)| > 1.0, P value < 0.05). For 2-[18F]FDG–treated CT26 tumor cells, the changed number was 2167 (1357 upregulated genes and 810 downregulated genes). DEGs were mapped into the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database to further explain the individual function analysis. Herein, transcriptome analysis using the RNA-seq technology was applied to compare DEGs between 2-[18F]FDG and saline-treated MC38 cells. As shown in Fig. 1H, many receptor-interaction signaling pathways and metabolic pathways were significantly enhanced, including the cytokine–cytokine receptor interaction, the NOD-like receptor signaling pathway, necroptosis, TNF signaling pathway, IL17 signaling pathway, and the Jak–STAT signaling pathway. According to the literature reports, the NF-κB signaling pathway is one of the major NOD-like receptor signaling pathways, that Nod1 and Nod2 stimulation induces NF-κB activation (22). Besides, TNF (also known as TNFα) could induce the expression of NF-κB target genes and trigger the activation of NF-κB signaling pathway, which indirectly upregulated PD-L1 expression (23, 24). These results suggest that multiple inflammatory signaling pathways and metabolic processes participate in 2-[18F]FDG induced MC38 cells.

DNA damage repair upregulates PD-L1 expression by NF-κB and IRF3 activation

Colony formation assay was performed to identify cell viability after incubation with 2-[18F]FDG for 24 hours, and no MC38 cell death was observed in the 3.7 MBq or 0.37 MBq group at all (Fig. 2A). To study DNA damage repair, γH2AX foci analysis and EdU staining were carried out to examine DSB levels and detect DNA repair after damage, respectively. As shown in Fig. 2B, the results demonstrated that 3.7 MBq of 2-[18F]FDG caused obvious DNA damage but the cell proliferation was still active, indicating that DNA replication and self-repair capacity was still maintained. To further study the mechanism, we performed γH2AX and caspase-3 staining of tumor tissues after treatment with 2-[18F]FDG in MC38 tumor mice. The data revealed that injection of 2-[18F]FDG into mice has more distinct DNA damage to tumor tissues than that of saline group (Supplementary Fig. S1A), but there was no significant difference in the tumor apoptosis between saline and 2-[18F]FDG treatment groups (Supplementary Fig. S1B).

Figure 2.

2-[18F]FDG-induced PD-L1 upregulation is mediated through NF-κB/IRF3 pathway. A, Colony formation assay was performed in MC38 cells to examine cell viability after coincubation with 2-[18F]FDG for different times. B, MC38 cells were irradiated with 2-[18F]FDG (3.7 MBq) for 24 hours and stained with γH2AX and EdU to identify DNA DSB and DNA repair level. C and D, NF-κB and IRF3 signaling pathways were activated by 2-[18F]FDG. Depletion of Rela and IRF3 weakens the upregulation of PD-L1 after 2-[18F]FDG induction. MC38 cells were exposed to Rela-435, IRF3–526 siRNA for knockdown of these genes. PD-L1 and a series of related proteins were examined after 2-[18F]FDG (3.7 MBq/mL) treatment. Quantification of the WB bands is shown at the bottom with heatmaps.

Figure 2.

2-[18F]FDG-induced PD-L1 upregulation is mediated through NF-κB/IRF3 pathway. A, Colony formation assay was performed in MC38 cells to examine cell viability after coincubation with 2-[18F]FDG for different times. B, MC38 cells were irradiated with 2-[18F]FDG (3.7 MBq) for 24 hours and stained with γH2AX and EdU to identify DNA DSB and DNA repair level. C and D, NF-κB and IRF3 signaling pathways were activated by 2-[18F]FDG. Depletion of Rela and IRF3 weakens the upregulation of PD-L1 after 2-[18F]FDG induction. MC38 cells were exposed to Rela-435, IRF3–526 siRNA for knockdown of these genes. PD-L1 and a series of related proteins were examined after 2-[18F]FDG (3.7 MBq/mL) treatment. Quantification of the WB bands is shown at the bottom with heatmaps.

Close modal

To identify the influence factors for the upregulation of PD-L1 induced by 2-[18F]FDG, we carried out a screening study through the depletion of NF-κB P65, IRF3, IRF1, STAT1, and STAT3, respectively. Targets were knocked-down by siRNAs (siRela-435, siIRF3–526, siIRF1–701, siSTAT3–1962, siSTAT1–1407) in MC38 cells with verified on-target effect and maximum knocked-down efficiency (Supplementary Fig. S2). The increased PD-L1 transcription by 2-[18F]FDG treatment was diminished by siRela-435 transfection into MC38 cells (Fig. 2C), suggesting that 2-[18F]FDG-activated NF-κB P65 pathway could boost PD-L1 transcription. Interestingly, as shown in Fig. 2D, we also found that 2-[18F]FDG-induced PD-L1 upregulation was obviously reduced in siIRF3–526 treated MC38 cells, in which NF-κB P65 activation was not influenced. The results indicated that 2-[18F]FDG-activated NF-κB P65 pathway may not singularly act on PD-L1 upregulation, and IRF3 may also play a vital role in mediating PD-L1 upregulation through 2-[18F]FDG. In previous researches, IRF3 was found to interact with NF-κB P65 (25), and the IRF3–P65 complex was enriched on the NF-κB–binding site in the PD-L1 promoter (26). According to these, we speculated that 2-[18F]FDG–induced upregulation of PD-L1 was influenced by IRF3–P65 complex. Taken together, these results demonstrated that the expression of PD-L1 was upregulated by DNA damage signaling pathway in MC38 cells. In the experimental process, we also found that the depletion of NF-κB P65 and IRF3 through siRNAs could diminish IRF1 and STAT1/3 phosphorylation (Fig. 2C and D). Previous studies demonstrated that NF-κB and IRF3 activation could induce the secretion of IFNα, TNFα, and other inflammatory cytokines (27), which could activate STAT1/3–IRF1 pathway and contribute to PD-L1 upregulation. Therefore, we also studied whether STAT1/3–IRF1 pathway could be activated by 2-[18F]FDG.

PD-L1 upregulation after DNA damage is mediated via STAT1/3-IRF1 pathway

A recent study demonstrated that the JAKs–STATs–IRF1 pathway primarily regulates PD-L1 expression (28). In our study, a total of 21,725 genes and 21,144 genes were identified in 2-[18F]FDG treated MC38 cells and CT26 cells, respectively. As shown in Fig. 3A, Stat1, Stat3, Fos, Nfkbia, Nfkbib, Nfkbie, and Cd274 (PD-L1) genes in 2-[18F]FDG treated MC38 cells were significantly upregulated compared with the untreated cells. Based on these, we tested whether 2-[18F]FDG could induce the activation of STAT1, STAT3, and IRF1 signaling pathways. As shown in Fig. 3BD, first, we found that phosphorylated PD-L1, IRF1, and STAT1/3 were significantly upregulated in MC38 cells after being induced by 2-[18F]FDG. Second, the depletion of IRF1, STAT1, and STAT3 notably reduced PD-L1 upregulation after 2-[18F]FDG treatment, suggesting that 2-[18F]FDG–induced PD-L1 upregulation is mediated by STAT1/3–IRF1 pathway. Moreover, we found that depletion of STAT1 and STAT3 not only inhibited self-phosphorylation, but also reduced IRF1 upregulation, indicating the direct action on IRF1. Also, both STAT1 and STAT3 phosphorylation were diminished by siIRF1–701 in MC38 cells. It might suggest that there is a feedback effect from IRF1 to STAT1 and STAT3 phosphorylation. In addition, we found that the depletion of IRF1, STAT1, and STAT3 could reduce NF-κB P65 phosphorylation but not affect IRF-3 and its phosphorylation. Taken together, these results suggested that 2-[18F]FDG induced DSB-dependent signaling pathway mediates PD-L1 upregulation, and that PD-L1 upregulation also relies on the activation of STAT1/3–IRF1 pathway.

Figure 3.

2-[18F]FDG–induced PD-L1 upregulation is related to canonical STAT1/3-IRF1 pathway. A, Heatmap of DEGs in tumor cells 24 hours after coincubation with 2-[18F]FDG. The intensity of the color represents log2 fold-change (2-[18F]FDG–treated vs. untreated cells). Some upregulated genes (e.g., Stat1, Stat3, NF-κB) are particularly associated with PD-L1. B–D, 2-[18F]FDG activates STAT1/3 and IRF1 signaling. Depletion of IRF1, STAT3, and STAT1 weakens the upregulation of PD-L1 after 2-[18F]FDG induction. MC38 cells were exposed to IRF1–701 (B), STAT3–1962 (C), and STAT1–1407 (D) siRNA for knockdown of these genes. PD-L1 and series of related proteins were examined after 2-[18F]FDG (3.7 MBq/mL) treatment. Quantifications of the WB bands are shown at the bottom with heatmaps.

Figure 3.

2-[18F]FDG–induced PD-L1 upregulation is related to canonical STAT1/3-IRF1 pathway. A, Heatmap of DEGs in tumor cells 24 hours after coincubation with 2-[18F]FDG. The intensity of the color represents log2 fold-change (2-[18F]FDG–treated vs. untreated cells). Some upregulated genes (e.g., Stat1, Stat3, NF-κB) are particularly associated with PD-L1. B–D, 2-[18F]FDG activates STAT1/3 and IRF1 signaling. Depletion of IRF1, STAT3, and STAT1 weakens the upregulation of PD-L1 after 2-[18F]FDG induction. MC38 cells were exposed to IRF1–701 (B), STAT3–1962 (C), and STAT1–1407 (D) siRNA for knockdown of these genes. PD-L1 and series of related proteins were examined after 2-[18F]FDG (3.7 MBq/mL) treatment. Quantifications of the WB bands are shown at the bottom with heatmaps.

Close modal

2-[18F]FDG causes PD-L1 upregulation and increases αPD-L1 uptake in tumors

We first investigated the biodistribution profile of 18.5 MBq 2-[18F]FDG in CT26 and MC38 tumors at different time points using a small-animal PET scanner (Fig. 4A). PET quantification data showed in Fig. 4B, demonstrating that the signal in the tumor region was well delineated from that of other tissues and the high accumulation of 2-[18F]FDG to tumor lesion. Through immunohistochemical (IHC) staining, PD-L1 expression in tumor tissue was compared between different groups. Strikingly, as shown in Fig. 4C and D, PD-L1 expression levels in the tumor region with 2-[18F]FDG uptake were more strongly positive compared to control samples within 24 hours. As shown in Supplementary Fig. S3, compared with the saline group and αPD-L1 group, the increased tumor PD-L1 level following 2-[18F]FDG alone or 2-[18F]FDG plus αPD-L1 mAb strongly reflected response to radionuclide stimulus. Then PD-L1 expression showed a tendency to decrease over time. Similar to the IHC results, tumor PD-L1 expression measured by flow cytometry showed a decreasing trend from day 1 to day 7 in groups containing 2-[18F]FDG radiotracer (Fig. 4E and F).

Figure 4.

PD-L1 expression of tumor was significantly increased after intravenous injection of 2-[18F]FDG. A, PET images of 2-[18F]FDG in tumor-bearing mice (CT26 and MC38 tumor) at different time points. Tumor areas are indicated by yellow arrows. Images were adjusted to the same maximum value to show the clearance of 2-[18F]FDG. B, PET quantification data of 2-[18F]FDG in CT26 and MC38 tumor-bearing mice (expressed in percentage injected dose per gram, %ID/g). C, IHC was performed to determine the PD-L1 expression on different tumor biopsies. Each CT26/MC38 tumor-bearing mouse was injected with 18.5 MBq of 2-[18F]FDG via the tail vein. Afterward, tumors were harvested at 0.5, 2, 4, and 24 hours p.i. (n = 3 per time point). The tumor biopsy samples of mice without radiotracer injection served as controls. D, Quantification of IHC for PD-L1 expression via ImageJ 7.0. E and F, The expression of PD-L1 in MC38 tumor treated with saline, 400 µg αPD-L1 mAb, or 37 MBq 2-[18F]FDG. The tumors of different groups were collected on days 1, 3, 5, and 7 after the mice received different injections. G, Representative PET images of 2-[18F]FDG in PDX tumor models evaluated at 0.5 and 4 hours p.i. Tumors are indicated by yellow arrows. H, Tumor uptakes and T/M ratios derived from PET images by drawing regions of interest (ROIs). I and J, Flow cytometry analysis of tumor PD-L1 expression at 4 hours p.i. The tumor samples of PDX mice injected with saline served as controls. Data were expressed as mean ± SD (n = 3/group). K, IHC was performed to determine the PD-L1 expression on different tumor biopsy samples from PDXH-FDG and PDXL-FDG groups. Tumors were harvested at 4 hours p.i.

Figure 4.

PD-L1 expression of tumor was significantly increased after intravenous injection of 2-[18F]FDG. A, PET images of 2-[18F]FDG in tumor-bearing mice (CT26 and MC38 tumor) at different time points. Tumor areas are indicated by yellow arrows. Images were adjusted to the same maximum value to show the clearance of 2-[18F]FDG. B, PET quantification data of 2-[18F]FDG in CT26 and MC38 tumor-bearing mice (expressed in percentage injected dose per gram, %ID/g). C, IHC was performed to determine the PD-L1 expression on different tumor biopsies. Each CT26/MC38 tumor-bearing mouse was injected with 18.5 MBq of 2-[18F]FDG via the tail vein. Afterward, tumors were harvested at 0.5, 2, 4, and 24 hours p.i. (n = 3 per time point). The tumor biopsy samples of mice without radiotracer injection served as controls. D, Quantification of IHC for PD-L1 expression via ImageJ 7.0. E and F, The expression of PD-L1 in MC38 tumor treated with saline, 400 µg αPD-L1 mAb, or 37 MBq 2-[18F]FDG. The tumors of different groups were collected on days 1, 3, 5, and 7 after the mice received different injections. G, Representative PET images of 2-[18F]FDG in PDX tumor models evaluated at 0.5 and 4 hours p.i. Tumors are indicated by yellow arrows. H, Tumor uptakes and T/M ratios derived from PET images by drawing regions of interest (ROIs). I and J, Flow cytometry analysis of tumor PD-L1 expression at 4 hours p.i. The tumor samples of PDX mice injected with saline served as controls. Data were expressed as mean ± SD (n = 3/group). K, IHC was performed to determine the PD-L1 expression on different tumor biopsy samples from PDXH-FDG and PDXL-FDG groups. Tumors were harvested at 4 hours p.i.

Close modal

IFNγ and active CD4+/CD8+ T cells in TME play important roles in mediating antitumor immunity. For this reason, we used 2-[18F]FDG as a stimulus to observe the responses of TME and determine the optimized immunotherapy time window for the administration of αPD-L1 mAb. From the IHC in Supplementary Fig. S4, we could see that 2-[18F]FDG gradually upregulated the expression of IFNγ, and enhanced the infiltration of CD4+ and CD8+ T cells over time.

Prior studies have demonstrated that the expression of PD-L1 has been an inclusion criterion for selecting patients of non-small cell lung cancer (NSCLC) for anti–PD-L1 treatment (29,30). To date, PD-1/PD-L1 ICBs have shown promise in advanced NSCLC without driver oncogene mutations, but wider use is restricted due to the low objective response rate (31,32). In this study, the high versus low uptake was defined through comparison of PET imaging in patients after administration of 2-[18F]FDG as described in our previous study (18). The tissue samples of patients with high and low 2-[18F]FDG uptakes were obtained to establish NSCLC PDX tumor models and found that 2-[18F]FDG uptakes in PDX tumor are consistent with the parental NSCLC.

After 2-[18F]FDG PET imaging (Fig. 4G and H), the PDXs were divided into high and low 2-[18F]FDG uptake (denoted as PDXH-FDG and PDXL-FDG) groups to evaluate the in vivo biological behavior of 2-[18F]FDG and predict the PD-L1 response of NSCLC to the radionuclide. Flow cytometry analysis revealed that the PD-L1 expression change in PDXL-FDG (24.9% ± 3.5%) was lower than that in PDXH-FDG (76.9% ± 4.9%) at 4 hours p.i. (Fig. 4I and J). Consistent with the flow cytometry results, PD-L1 IHC showed more prominent expression in post-tracer PDXH-FDG tumor biopsies (Fig. 4K). Hence, as we have observed, the immunomodulatory effects of 2-[18F]FDG could reasonably explain the PD-L1 upregulation in the tumors.

Intuitively, the PD-L1 upregulation caused by 2-[18F]FDG would increase αPD-L1 mAb uptake in the tumor. We confirmed this with a fluorescent αPD-L1 probe (Supplementary Fig. S5A). The flow cytometric analysis showed that 2-[18F]FDG groups had higher uptake of fluorescent probe Cy5.5-αPD-L1 which further increased over time, presumably due to the upregulated PD-L1 levels in CT26 and MC38 tumor cells. In striking contrast, much lower Cy5.5-αPD-L1 uptakes were observed in the control tumors without 2-[18F]FDG treatment. Representative histograms of the PD-L1 expression after 2-[18F]FDG stimulation were shown in Supplementary Fig. S5B.

Coupling 2-[18F]FDG with αPD-L1 mAb enhances the antitumor effect and immunologic memory

To further explore the immunomodulatory effect of 2-[18F]FDG for enhancing immunotherapy, we subsequently investigated the effect of αPD-L1 mAb on MC38 tumor growth delay in cooperation with 2-[18F]FDG. As shown in Fig. 5A; Supplementary Fig. S6A, tumor models were treated with 2-[18F]FDG, αPD-L1 mAb, or their combination in specific treatment sequences. In the combination groups, αPD-L1 mAb was tail vein injected into the tumor-bearing mice at different intervals (simultaneous injection, 4-h and 24-h; hereinafter denoted as @ 4 h and @ 24 h) after administration of the radiotracer. Fig. 5B and C; Supplementary Fig. S6B and S6C illustrated the tumor volumes, time-dependent tumor growth curves, weight changes and survival curves for each group. In the control groups of αPD-L1 mAb, isotype mAb and saline alone, the tumor sizes developed uncontrollably. Also, single administration of 2-[18F]FDG or the same dose of nonradioactive 2-[19F]FDG did not significantly alter MC38 tumor growth. We then compared the therapeutic effect of αPD-L1 mAb which was administered simultaneously, 4 or 24 hours post radiotracer injection. Notably, the 4-hour interval turned out to be the most optimal treatment sequence, and administration of 18.5 MBq or 37 MBq 2-[18F]FDG + 400 μg αPD-L1 mAb at 4 hours resulted in the maximum therapeutic efficacy (4/8 or 5/8 mice were tumor free), clearly indicating that the immunotherapy of 2-[18F]FDG combined with αPD-L1 mAb was regulated by administration doses and time windows.

Figure 5.

Combination of 2-[18F]FDG with PD-L1 ICB immunotherapy results in significant tumor growth delay and overall survival improvement. A, Schematics of the procedures and timelines of treatment for MC38 tumor-bearing mice. As the tumor volume reached about 50 mm3, the course of treatment began on day 0. The mice were randomly divided into several groups (n = 8 per group), including saline, αPD-L1 mAb alone, 2-[18F]FDG alone, and simultaneous and sequential combined therapy with 2-[18F]FDG plus αPD-L1 mAb (4- or 24-hour interval). An additional therapeutic course was scheduled on day 4. B, Individual tumor growth profiles as well as survival rate on day 90 of MC38 tumor-bearing mice from different treatment groups as indicated. C, Time-dependent tumor growth curves (top) and survival curves (bottom) of the MC38 tumor-bearing mice treated with 2-[18F]FDG induced immunotherapy. See Supplementary Fig. S6 for additional information. Data were expressed as mean ± SD (n = 8/group). Different activities of 2-[18F]FDG and αPD-L1 mAb are designated by the pound sign. Dose of αPD-L1 mAb: 10 mg/kg; 20 mg/kg (##). Activity of 2-[18F]FDG: 925 MBq/kg; 1,850 MBq/kg (##). All radiotracers and αPD-L1 were administered by intravenous injection. D, Dynamics of cytokine levels in blood. The serum was separated and the concentrations of IFNγ (left), TNFα (middle), and IL6 (right) were determined by ELISA. The groups were as follows: (I) saline group, (II) separate 2-[18F]FDG injection, (III) αPD-L1 mAb group, and (IV) sequential injection of 2-[18F]FDG plus αPD-L1 mAb at 4 hours. Data were expressed as mean ± SD (n = 3/group). Each experimental group was compared with the saline group. E, Flow cytometry analysis of the infiltration of memory T cells (CD4+CD44highCD62Llow and CD8+CD44highCD62Llow) in the spleen at day 60 after various treatments. See Supplementary Fig. S8 for additional information. F, Quantification of CD4+CD44highCD62Llow cells in total cells (left) and CD8+CD44highCD62Llow cells in total cells (right) in the spleen with FlowJo v10 software. Data were expressed as mean ± SD (n = 4/group). Each group was compared with the IV group. G, Images of MC38 tumor models were acquired during the monitoring period. The left rear flanks of cured mice were challenged with MC38 cells at day 91 and monitored until day 150. H, Combining 2-[18F]FDG with αPD-L1 mAb enhanced the long-lasting immunologic memory.

Figure 5.

Combination of 2-[18F]FDG with PD-L1 ICB immunotherapy results in significant tumor growth delay and overall survival improvement. A, Schematics of the procedures and timelines of treatment for MC38 tumor-bearing mice. As the tumor volume reached about 50 mm3, the course of treatment began on day 0. The mice were randomly divided into several groups (n = 8 per group), including saline, αPD-L1 mAb alone, 2-[18F]FDG alone, and simultaneous and sequential combined therapy with 2-[18F]FDG plus αPD-L1 mAb (4- or 24-hour interval). An additional therapeutic course was scheduled on day 4. B, Individual tumor growth profiles as well as survival rate on day 90 of MC38 tumor-bearing mice from different treatment groups as indicated. C, Time-dependent tumor growth curves (top) and survival curves (bottom) of the MC38 tumor-bearing mice treated with 2-[18F]FDG induced immunotherapy. See Supplementary Fig. S6 for additional information. Data were expressed as mean ± SD (n = 8/group). Different activities of 2-[18F]FDG and αPD-L1 mAb are designated by the pound sign. Dose of αPD-L1 mAb: 10 mg/kg; 20 mg/kg (##). Activity of 2-[18F]FDG: 925 MBq/kg; 1,850 MBq/kg (##). All radiotracers and αPD-L1 were administered by intravenous injection. D, Dynamics of cytokine levels in blood. The serum was separated and the concentrations of IFNγ (left), TNFα (middle), and IL6 (right) were determined by ELISA. The groups were as follows: (I) saline group, (II) separate 2-[18F]FDG injection, (III) αPD-L1 mAb group, and (IV) sequential injection of 2-[18F]FDG plus αPD-L1 mAb at 4 hours. Data were expressed as mean ± SD (n = 3/group). Each experimental group was compared with the saline group. E, Flow cytometry analysis of the infiltration of memory T cells (CD4+CD44highCD62Llow and CD8+CD44highCD62Llow) in the spleen at day 60 after various treatments. See Supplementary Fig. S8 for additional information. F, Quantification of CD4+CD44highCD62Llow cells in total cells (left) and CD8+CD44highCD62Llow cells in total cells (right) in the spleen with FlowJo v10 software. Data were expressed as mean ± SD (n = 4/group). Each group was compared with the IV group. G, Images of MC38 tumor models were acquired during the monitoring period. The left rear flanks of cured mice were challenged with MC38 cells at day 91 and monitored until day 150. H, Combining 2-[18F]FDG with αPD-L1 mAb enhanced the long-lasting immunologic memory.

Close modal

ELISA assays were performed to measure the levels of immunostimulatory cytokines in the serum of mice. The combination of 2-[18F]FDG and αPD-L1 mAb increased the production of IFNγ, TNFα, and IL6, and maintained for a longer period of time than single treatment or saline group, which might also account for the synergistic anticancer efficacy (Fig. 5D). However, no obvious side effects were observed in the fully recovered mice (Supplementary Fig. S7), indicating that the antitumor therapeutic strategy was well tolerated.

To verify the immunologic memory of 2-[18F]FDG combined immune checkpoint therapy, the infiltration of the effector memory T (TEM) cells (CD8+CD44highCD62Llow and CD4+CD44highCD62Llow) in the spleen were detected and analyzed through flow cytometry (Fig. 5E and F; Supplementary Fig. S8). As expected, the levels of splenic TEM cells gradually increased between day 3 and day 7 in the best-performing groups (37 MBq 2-[18F]FDG + 400 μg αPD-L1 at 4 hours), which was higher than the saline group, αPD-L1 mAb or 2-[18F]FDG group. Furthermore, for 2-[18F]FDG-induced immunotherapy, the splenic TEM cells remained at a high level until 60 days after the combined treatment. In addition, rechallenging on the left rear flanks of cured mice at day 91 with MC38 tumor cells did not observe tumor recurrence for at least 2 months (Fig. 5G and H). These results demonstrated that the prevention of tumor recurrence by 2-[18F]FDG-induced immunotherapy was credited to the activation of immunological memory effect.

Coupling 2-[18F]FDG with αPD-L1 mAb reprograms TME

The impact of 2-[18F]FDG to PD-L1 ICBs is multifaceted. The flow cytometric results in Fig. 6A and B; Supplementary Figs. S9 and S10 showed that intratumoral CD4+ Th1 (CD45+CD4+IFNγ+ T cells) and CD8+ CTLs (CD45+CD8+IFNγ+ T cells) became exhausted on day 3 and day 7 in single 2-[18F]FDG or αPD-L1 mAb groups. However, the combination of 2-[18F]FDG + αPD-L1 mAb at 4 hours significantly increased the numbers of CD4+ Th1 and CD8+ CTLs compared to the other groups. As shown in Fig. 6C and D, the absolute numbers of tumor-infiltrating CD45+CD8+IFNγ+ T lymphocytes and CD45+CD4+IFNγ+ T lymphocytes in MC38 mice received 2-[18F]FDG + αPD-L1 mAb at 4 hours treatment were approximately 5-fold of those in the saline group. However, as one type of CD4+ T cells, the fraction of immunosuppressive CD45+CD4+FOXP3+ regulatory T cells (Treg) in tumors showed a decrease in 2-[18F]FDG + αPD-L1 mAb at 4 hours group (Fig. 6E and F; Supplementary Fig. S11). Specifically, further comparative analysis showed a significant increase of CD4+ Teff/Treg and CD8+ CTLs/Treg ratios in the combination group. 2-[18F]FDG + αPD-L1 mAb at 4 hours group elicited a 36-fold increase in the ratio of CD8+ CTLs/Treg and a 24-fold increase in the ratio of CD4+ Teff/Treg in the TME, respectively, compared with saline group (Fig. 6C and D). Additionally, we also observed increased level of IFNγ for 2-[18F]FDG-induced immunotherapy through IHC (Supplementary Fig. S12). Further results indicating both CD4+ Th1 and CD8+ cytotoxic T lymphocytes in the TME were enhanced from day 1 to 7 in the group of 2-[18F]FDG + αPD-L1 mAb at 4 hours, whereas the levels of these indicators were unaltered in the saline group. Meanwhile, tumor samples were harvested for detecting proliferation and apoptosis by immunofluorescence staining of Ki67 and caspase-3. As depicted in Supplementary Fig. S13, the dynamic change of PD-L1 expression from day 1 to 7 in the combination therapy groups were further validated. As expected, at the corresponding time points, the Ki67 indexes were significantly higher in the saline group. While the positive rate of caspase-3 expression in the combination group was obviously higher than that of saline group.

Figure 6.

Combination of 2-[18F]FDG with PD-L1 ICB immunotherapy results in antitumor immune response. A and B, Flow cytometric analysis of CD8+ CTL (CD8+IFNγ+; A) and CD4+ Teff (CD4+IFNγ+; B) cells in MC38 tumor gating on CD45+ cells. See Supplementary Fig. S9 and Supplementary Fig. S10 for additional information. C and D, Quantification of CD45+CD8+IFNγ+ cells in total cells and ratio of CD8+ CTL to Treg (C), CD45+CD4+IFNγ+ cells in total cells, and ratio of CD4+ Teff to Treg (D) in MC38 tumor tissues with FlowJo v10 software. E, Flow cytometric analysis of CD4+ Treg (CD4+Foxp3+) cells in MC38 tumor gating on CD45+ cells. See Supplementary Fig. S11 for additional information. F, Quantification of CD45+CD4+Foxp3+ cells in total cells in MC38 tumor tissues. G and H, Flow cytometry analysis of the infiltration of M1-like macrophages (iNOS+F4/80+) and M2-like macrophages (CD206+F4/80+) in MC38 tumor gating on CD11b+ tissues after various treatments. See Supplementary Fig. S14 for additional information. I and J, Flow cytometry analysis of the infiltration of dendritic cells (CD86+CD80+) and MDSCs (CD45+CD11b+Gr-1+) in tumor tissues after various treatments. See Supplementary Fig. S15 for additional information. K, Quantification of iNOS+F4/80+CD11b+ and CD206+F4/80+CD11b+ cells in total cells in MC38 tumor tissues. L, Quantification of CD80+CD86+ and CD45+Gr-1+CD11b+ cells in total cells in MC38 tumor tissues. The tumor samples were excised on day 7 after different injection strategies as indicated. The groups were as follows: (I) saline group, (II) separate 2-[18F]FDG injection, (III) αPD-L1 mAb group, and (IV) sequential injection of 2-[18F]FDG plus αPD-L1 mAb at 4 hours. Data were expressed as mean ± SD (n = 4/group).

Figure 6.

Combination of 2-[18F]FDG with PD-L1 ICB immunotherapy results in antitumor immune response. A and B, Flow cytometric analysis of CD8+ CTL (CD8+IFNγ+; A) and CD4+ Teff (CD4+IFNγ+; B) cells in MC38 tumor gating on CD45+ cells. See Supplementary Fig. S9 and Supplementary Fig. S10 for additional information. C and D, Quantification of CD45+CD8+IFNγ+ cells in total cells and ratio of CD8+ CTL to Treg (C), CD45+CD4+IFNγ+ cells in total cells, and ratio of CD4+ Teff to Treg (D) in MC38 tumor tissues with FlowJo v10 software. E, Flow cytometric analysis of CD4+ Treg (CD4+Foxp3+) cells in MC38 tumor gating on CD45+ cells. See Supplementary Fig. S11 for additional information. F, Quantification of CD45+CD4+Foxp3+ cells in total cells in MC38 tumor tissues. G and H, Flow cytometry analysis of the infiltration of M1-like macrophages (iNOS+F4/80+) and M2-like macrophages (CD206+F4/80+) in MC38 tumor gating on CD11b+ tissues after various treatments. See Supplementary Fig. S14 for additional information. I and J, Flow cytometry analysis of the infiltration of dendritic cells (CD86+CD80+) and MDSCs (CD45+CD11b+Gr-1+) in tumor tissues after various treatments. See Supplementary Fig. S15 for additional information. K, Quantification of iNOS+F4/80+CD11b+ and CD206+F4/80+CD11b+ cells in total cells in MC38 tumor tissues. L, Quantification of CD80+CD86+ and CD45+Gr-1+CD11b+ cells in total cells in MC38 tumor tissues. The tumor samples were excised on day 7 after different injection strategies as indicated. The groups were as follows: (I) saline group, (II) separate 2-[18F]FDG injection, (III) αPD-L1 mAb group, and (IV) sequential injection of 2-[18F]FDG plus αPD-L1 mAb at 4 hours. Data were expressed as mean ± SD (n = 4/group).

Close modal

In addition to T cells, other alterations of immune cells were also notable. M1-like macrophages, M2-like macrophages, dendritic cells (DC), and MDSCs were detected through flow cytometry (Fig. 6GJ; Supplementary Figs. S14 and S15). The MC38 tumor-bearing mice that received 37 MBq of 2-[18F]FDG + 400 μg αPD-L1 at 4 hours decreased the fraction of M2-like macrophages (CD206+CD11b+F4/80+), but significantly increased the fraction of M1-like macrophages (iNOS+CD11b+F4/80+), further indicating the repolarization of M2-like macrophages or recruitment of M1-like macrophages, implying the reduced immunosuppression and enhanced antitumor immunity (Fig. 6K). Similarly, activated DCs (CD80+CD86+) in the combination group generated a 2.5-fold increase over the saline group in the TME (Fig. 6I and L). Moreover, 2-[18F]FDG-induced PD-L1 immunotherapy generated about 10-fold decrease in the fraction of MDSCs (CD45+CD11b+Gr-1+) compared with the saline group or 2-[18F]FDG alone, which suggested that the immunosuppressive microenvironment was relieved (Fig. 6J and L).

KEGG enrichment analysis was performed to identify the detailed immune activation associated pathways and inflammatory signaling pathways mediated by the therapeutic strategy of 2-[18F]FDG + αPD-L1 at 4 hours. Transcriptomics analysis focused on the DEGs during the treatment process. As show in Supplementary Fig. S16A, compared with the saline group, the CD274 gene (PD-L1) in the tumor of 2-[18F]FDG or 2-[18F]FDG + αPD-L1 at 4 hours groups was upregulated on day 1. A few days later, this indicator showed a noticeable decline. In addition, several representative pathways were shown in Supplementary Fig. S16B. The most significant differences between the saline group and 2-[18F]FDG + αPD-L1 at 4 hours group were found in the antigen processing and presentation, phagosome, cell adhesion molecules, the NOD-like receptor signaling pathway, cytokine-cytokine receptor interaction and Th17 cell differentiation. Moreover, we went a step further to confirm that the 2-[18F]FDG radiotracer would affect the expression of PD-1, another key component of immune checkpoint blockade (Supplementary Fig. S16C). Together, these profiles further confirmed that the radionuclide-induced PD-L1 ICB immunotherapy could inflame the TME and activate the immune system.

To test the merit of this combination treatment strategy for the tumors with low expression of PD-L1, we investigated the antitumor efficacy of 2-[18F]FDG plus αPD-L1 mAb in CT26 (murine colorectal tumor) and 4T1 (murine breast cancer) tumor models, which are originally less sensitive to PD-L1 immune checkpoint blockade. Although to a less extent as compared with the MC38 section, the growth of CT26 tumors was still greatly suppressed in the groups of 18.5 MBq or 37 MBq 2-[18F]FDG + 400 μg αPD-L1 mAb at 4 hours, resulting in prolonged overall survivals (Supplementary Figs. S17 and S18A-S18E). Dynamic changes of PD-L1 in CT26 tumor and cytokine levels in blood were found in the group of 2-[18F]FDG + αPD-L1 at 4 hours during the therapy period (Supplementary Fig. S18F–S18H). Therapeutically, the activation of CD4+ and CD8+ T cells in the tumor and increased CD4+ Th1/Treg and CD8+ CTLs/Treg levels highlighted the potential of the effective coordination to enhance antitumor immunity (Supplementary Fig. S19). As shown in Supplementary Fig. S20, 37 MBq 2-[18F]FDG + 400 μg αPD-L1 mAb at 4 hours treatment also led to 4T1 tumor growth delay and prolonged survival compared with the saline, single 2-[18F]FDG or αPD-L1 mAb groups, similar to what was observed in the CT26 tumor model.

In this study, we demonstrated that a single dose of 2-[18F]FDG could cause obvious DNA damage but not cell apoptosis, and DNA damage repair vitality was still maintained. The PD-L1 expression in MC38 cells was upregulated by the DNA damage signaling following DSBs induced by 2-[18F]FDG (Figs. 2 and 3). Such transient upregulation of PD-L1 in DNA-damaged cells might be used to prevent overactivation of the immune system surrounding the tumors. Previous researches indicated that IRF1 is a key factor for PD-L1 expression (26), and that the upregulation of PD-L1 is mediated through IRF1 binding to the PD-L1 promoter in the JAK-STAT-IRF1 signaling pathway (33). As shown in Fig. 7A, 2-[18F]FDG induces DNA damage and repair, which on the one hand activates NF-κB and IRF3 pathways, promotes NF-κB P65–IRF3 complex formation, leading to PD-L1 transcription increase and PD-L1 upregulation, and on the other hand, 2-[18F]FDG-induced DNA damage directly activates STAT1/3-IRF1 pathway, resulting in IRF1 transcription enhancement to further upregulate PD-L1 expression. Furthermore, after treatment with 2-[18F]FDG, DNA damage signaling within the fraction of live injured cells is vital to the activation of antitumor T-cell responses (34). Our results showed that upregulation of PD-L1 expression was mediated by both NF-κB/IRF3 and STAT1/3-IRF1 pathways, which was related to DNA damage induced by 2-[18F]FDG. In addition, from Supplementary Fig. S4, we can see that CT26 and MC38 tumor tissues treated with 2-[18F]FDG can lead to infiltration of T lymphocyte cells and secretion of immunomodulatory molecules, such as IFNγ, which in turn also upregulate PD-L1 expression in the tumor.

Figure 7.

Schematics of 2-[18F]FDG–mediated antitumor immune response. A, Model for PD-L1 upregulation in response to DNA double-strand breaks. 2-[18F]FDG induces DNA damage and repair pathway. During the repair process, radiation-dependent NF-κB/IRF3 pathway is activated. Furthermore, PD-L1 upregulation requires IRF1, suggesting that DSB-dependent PD-L1 upregulation is also induced by the canonical STAT–IRF1 pathway. B, 2-[18F]FDG in combination with αPD-L1 mAb synergistically enhances antitumor immunity by reinvigorating the TME.

Figure 7.

Schematics of 2-[18F]FDG–mediated antitumor immune response. A, Model for PD-L1 upregulation in response to DNA double-strand breaks. 2-[18F]FDG induces DNA damage and repair pathway. During the repair process, radiation-dependent NF-κB/IRF3 pathway is activated. Furthermore, PD-L1 upregulation requires IRF1, suggesting that DSB-dependent PD-L1 upregulation is also induced by the canonical STAT–IRF1 pathway. B, 2-[18F]FDG in combination with αPD-L1 mAb synergistically enhances antitumor immunity by reinvigorating the TME.

Close modal

Subsequent studies provided evidence on the necessity and timeliness of the participation of αPD-L1 mAb. Taking MC38 tumor model receiving 2-[18F]FDG + αPD-L1 for example, of all the groups, the overall therapeutic outcomes of the 4-hour interval were most impressive. Compared with 0 and 24 hours, 4-hour interval is an optimal time window. At 0-hour interval, there was no time for 2-[18F]FDG to modulate PD-L1 expression and immune cell infiltration, thus, led to limited αPD-L1 mAb uptake. At 24-hour interval, although PD-L1 expression had been upregulated after injection of 2-[18F]FDG, however, it would take time (24–48 hours) for αPD-L1 mAb to accumulate in the tumor, meanwhile the short-lived 2-[18F]FDG has already decayed thus won't be able to continuously upregulate PD-L1. Therefore, 4-hour interval not only is sufficient to upregulate PD-L1 expression but also contributes to increased αPD-L1 mAb uptake and immune cell infiltration in the tumors, resulting in the optimal therapeutic effect.

In previous studies, high-activity 2-[18F]FDG (5.55–11.1 GBq/kg) was used for the cancer therapy in mouse models and modest therapeutic response (incomplete cure) was observed (35–37). Meanwhile, radiotoxicity of 2-[18F]FDG (11.1 GBq/kg) was not found (37). On this basis, with the synergy of αPD-L1 mAb and properly scheduled time window, it is possible to further drive 2-[18F]FDG dosage lower, which could be favorable for clinical application. Kaplan–Meier curves found no significant differences in survival between 1850 and 925 MBq/kg dosages in the study. The US Food and Drug Administration's current guidance is an empirical approach based on normalization of dose to body surface area (38). Under the guidance, if we apply 925 MBq/kg 2-[18F]FDG in MC38 tumor model, the human equivalent dose would be 75.2 MBq/kg. For an adult of 60 kg body weight, about 4.4 GBq 2-[18F]FDG will be needed. Srinivasan and colleagues (39) reported that the mean absorbed dose of total body after intravenous administration of 2-[18F]FDG was 0.0103 mSv/MBq. After administration of 4.4 GBq 2-[18F]FDG, the mean absorbed dose of total body would be 45.7 mSv. According to the suggestion by the US National Research Council, there is barely any risk to human health when short-term radiation intake is less than 100 mSv (40). The dose of our administration strategy is thus considered well tolerated.

Because the pharmacokinetic differences of drug between human and mouse, the half-life and elimination rate of αPD-L1 mAb in vivo are different, more experiments of administration dosage and time window should be explored before this paradigm could directly translate to human. Theoretically, for 2-[18F]FDG, the distribution differences in human and mouse would lead to the timeframe of upregulation of PD-L1 in human to be within 7 days posttreatment. In this study, the paradigm based on 2-[18F]FDG and anti–PD-L1 mAb combination therapy in MC38 model is efficient, it significantly improved the therapeutic effect of cancer immune checkpoint blockade. When anti–PD-L1 mAb was administered twice, it could achieve more than half the cure rate and complete tumor remission. Therefore, the treatment cycles of patients with anti–PD-L1 antibodies may be reduced from 4 to 2 cycles and achieve better therapeutic effect when combined with 2-[18F]FDG.

Studies have shown that one of the main challenges for ICB immunotherapy lies in “cold” tumor with limited T-cell responses (41,42). Fortunately, as shown in Fig. 7B, we found that radionuclide-based ICB immunotherapy synergistically enhanced antitumor immunity by increasing the effector T cells, DCs, and M1-like macrophages infiltration, and that Tregs, M2-like macrophages and MDSCs, which were defined as immunosuppressors for maintaining immunological tolerance, had obvious decrease in the tumor. In addition, inflammatory cytokines including TNFα, IFNγ significantly increased. It appears that such a combination therapy strategy is an attractive option to eliminate treatment resistance of patients with low or no PD-L1 expression in pretreatment tumors and improve clinical outcomes. After treatment with 2-[18F]FDG combined with anti-PD-L1 immunotherapy, the TME might be changed from “cold” to “hot”, remodeling the whole immune status to make it sensitive to PD-L1 ICB therapy.

MC38 and CT26 cell lines have been widely applied in the studies of immune checkpoint blockade (43) and radiotherapy (44). Although both MC38 and CT26 are colorectal cancer cell lines, MC38 is high immunogenic (45) while CT26 is less immunogenic (43). The results in this study proved that the combination paradigm is effective to tumor models regardless of their immunogenicity. Furthermore, the similarity of MC38 and CT26 tumor models to the clinical observations of the response of patients with colorectal cancer treated with immune checkpoint blockade further supports the rationality of the selected models. We also evaluated PD-L1 upregulation in 4T1 and B16F10 tumor cells after 2-[18F]FDG exposure (Fig. 1AD), and further performed the combination therapy of 2-[18F]FDG and αPD-L1 mAb in 4T1 tumor model (Supplementary Fig. S20). The therapeutic effect differences between 4T1 and MC38 may be attributed to a number of factors, including species difference of BALB/c and C57BL/6 mice, cell line specific characteristics, immunogenic disparity, original PD-L1 expression level, and so on. These possible influences will be further studied by more experiments and the application will be explored with more tumor types in the next work, which is important for clinical translation.

In summary, we have demonstrated that 2-[18F]FDG induces significant PD-L1 upregulation and remodels the TME. Understandably, these results highlight a promising paradigm of cancer immunotherapy based on 2-[18F]FDG and anti–PD-L1 mAb combination.

X. Zhang reports grants from National Natural Science Foundation of China during the conduct of the study, as well as a patent for CN202010697679 pending. No disclosures were reported by the other authors.

X. Wen: Conceptualization, data curation, investigation, writing–original draft. C. Shi: Investigation. X. Zeng: Investigation. L. Zhao: Investigation. L. Yao: Investigation. Z. Liu: Investigation. L. Feng: Investigation. D. Zhang: Investigation. J. Huang: Investigation. Y. Li: Investigation. Q. Lin: Funding acquisition, investigation. H. Chen: Investigation. R. Zhuang: Investigation. X. Chen: Conceptualization, supervision, writing–review and editing. X. Zhang: Resources, supervision, funding acquisition. Z. Guo: Resources, supervision, funding acquisition, investigation, writing–review and editing.

We thank Mingyan Xu, Xuanqin Wu, Qiannan Tong, Shifu Chen, and Chao Lu at HaploX Biotechnology for technical assistance with transcriptome and signal-net analysis. This study was financially supported by the National Natural Science Foundation of China (81901805, 21976150, 21906135), Major Research Plan of the National Natural Science Foundation of China (91959122), Joint Fund of the National Natural Science Foundation of China - China National Nuclear Corporation for Nuclear Technology Innovation (U1967222), Fundamental Research Funds for the Central Universities of China (20720210115), the National University of Singapore Start-up Grant (NUHSRO/2020/133/Startup/08), NUS School of Medicine Nanomedicine Translational Research Programme (NUHSRO/2021/034/TRP/09/Nanomedicine), and the National Research Foundation, Singapore, and National Medical Research Council, Singapore under its NMRC Centre Grant Programme (CG21APR1005).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Sanmamed
MF
,
Chen
L
.
A paradigm shift in cancer immunotherapy: from enhancement to normalization
.
Cell
2018
;
175
:
313
26
.
2.
Topalian
SL
,
Taube
JM
,
Pardoll
DM
.
Neoadjuvant checkpoint blockade for cancer immunotherapy
.
Science
2020
;
367
:
eaax0182
.
3.
Chiriva-Internati
M
,
Bot
A
.
A new era in cancer immunotherapy: discovering novel targets and reprogramming the immune system
.
Int Rev Immunol
2015
;
34
:
101
3
.
4.
Yang
Y
.
Cancer immunotherapy: harnessing the immune system to battle cancer
.
J Clin Invest
2015
;
125
:
3335
7
.
5.
Kluger
HM
,
Zito
CR
,
Turcu
G
,
Baine
MK
,
Zhang
H
,
Adeniran
A
, et al
.
PD-L1 studies across tumor types, its differential expression and predictive value in patients treated with immune checkpoint inhibitors
.
Clin Cancer Res
2017
;
23
:
4270
9
.
6.
Zhou
Z
,
Zhao
J
,
Pan
Q
,
Chen
C
,
Liu
Y
,
Tang
Y
, et al
.
PD-L1 expression is a predictive biomarker for CIK cell-based immunotherapy in postoperative patients with breast cancer
.
J Immunother Cancer
2019
;
7
:
228
.
7.
Golden
EB
,
Apetoh
L
.
Radiotherapy and immunogenic cell death
.
Semin Radiat Oncol
2015
;
25
:
11
7
.
8.
Gardai
SJ
,
McPhillips
KA
,
Frasch
SC
,
Janssen
WJ
,
Starefeldt
A
,
Murphy-Ullrich
JE
, et al
.
Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte
.
Cell
2005
;
123
:
321
34
.
9.
Liu
S
.
Nonlinear dose-response relationship in the immune system following exposure to ionizing radiation: Mechanisms and implications
.
Nonlinearity Biol Toxicol Med
2003
;
1
:
71
92
.
10.
Rodriguez-Ruiz
ME
,
Garasa
S
,
Rodriguez
I
,
Solorzano
JL
,
Barbes
B
,
Yanguas
A
, et al
.
Intercellular adhesion molecule-1 and vascular cell adhesion molecule are induced by ionizing radiation on lymphatic endothelium
.
Int J Radiat Oncol Biol Phys
2017
;
97
:
389
400
.
11.
Ellsworth
SG
.
Field size effects on the risk and severity of treatment-induced lymphopenia in patients undergoing radiation therapy for solid tumors
.
Adv Radiat Oncol
2018
;
3
:
512
9
.
12.
Seiwert
TY
,
Kiess
AP
.
Time to debunk an urban myth? The “abscopal effect” with radiation and anti-PD-1
.
J Clin Oncol
2021
;
39
:
1
3
.
13.
Kole
TP
,
Tong
M
,
Wu
B
,
Lei
S
,
Obayomi-Davies
O
,
Chen
LN
, et al
.
Late urinary toxicity modeling after stereotactic body radiotherapy (SBRT) in the definitive treatment of localized prostate cancer
.
Acta Oncol
2016
;
55
:
52
58
.
14.
Wu
CT
,
Chen
WC
,
Chang
YH
,
Lin
WY
,
Chen
MF
.
The role of PD-L1 in the radiation response and clinical outcome for bladder cancer
.
Sci Rep
2016
;
6
:
19740
.
15.
d'Adda di Fagagna
F
.
Living on a break: cellular senescence as a DNA-damage response
.
Nat Rev Cancer
2008
;
8
:
512
22
.
16.
He
C
,
Duan
X
,
Guo
N
,
Chan
C
,
Poon
C
,
Weichselbaum
RR
., et al
.
Core-shell nanoscale coordination polymers combine chemotherapy and photodynamic therapy to potentiate checkpoint blockade cancer immunotherapy
.
Nat Commun
2016
;
7
:
12499
.
17.
Deng
L
,
Liang
H
,
Burnette
B
,
Beckett
M
,
Darga
T
,
Weichselbaum
RR
., et al
.
Irradiation and anti-PD-L1 treatment synergistically promote antitumour immunity in mice
.
J Clin Invest
2014
;
124
:
687
95
.
18.
Zhao
L
,
Chen
H
,
Guo
Z
,
Fu
K
,
Yao
L
,
Fu
L
, et al
.
Targeted radionuclide therapy in patient-derived xenografts using 177Lu-EB-RGD
.
Mol Cancer Ther
2020
;
19
:
2034
43
.
19.
Shihan
MH
,
Novo
SG
,
Marchand
SJ
,
Wang
Y
,
Duncan
MK
.
A simple method for quantitating confocal fluorescent images
.
Biochem Biophys Rep
2021
;
25
:
100916
.
20.
Liu
Y
,
Zhang
X
,
Han
C
,
Wan
G
,
Huang
X
,
Ivan
C
, et al
.
TP53 loss creates therapeutic vulnerability in colorectal cancer
.
Nature
2015
;
520
:
697
701
.
21.
Seifert
L
,
Werba
G
,
Tiwari
S
,
Giao Ly
NN
,
Alothman
S
,
Alqunaibit
D
, et al
.
The necrosome promotes pancreatic oncogenesis via CXCL1 and Mincle-induced immune suppression
.
Nature
2016
;
532
:
245
9
.
22.
Keestra-Gounder
AM
,
Byndloss
MX
,
Seyffert
N
,
Young
BM
,
Chávez-Arroyo
A
,
Tsai
AY
, et al
.
NOD1 and NOD2 signalling links ER stress with inflammation
.
Nature
2016
;
532
:
394
7
.
23.
Spiotto
M
,
Fu
YX
,
Weichselbaum
RR
.
The intersection of radiotherapy and immunotherapy: mechanisms and clinical implications
.
Sci Immunol
2016
;
1
:
EAAG1266
.
24.
Webb
LV
,
Ley
SC
,
Seddon
B
.
TNF activation of NF-κB is essential for development of single-positive thymocytes
.
J Exp Med
2016
;
213
:
1399
407
.
25.
Wietek
C
,
Miggin
SM
,
Jefferies
CA
,
O'Neill
LA
.
Interferon regulatory factor-3-mediated activation of the interferon-sensitive response element by Toll-like receptor (TLR) 4 but not TLR3 requires the p65 subunit of NF-kappa
.
J Biol Chem
2003
;
278
:
50923
31
.
26.
Wang
W
,
Chapman
NM
,
Zhang
B
,
Li
M
,
Fan
M
,
Laribee
RN
, et al
.
Upregulation of PD-L1 via HMGB1-activated IRF3 and NF-κB contributes to UV radiationinduced immune suppression
.
Cancer Res
2019
;
18
:
3134
.
27.
Wang
H
,
Hu
S
,
Chen
X
,
Shi
H
,
Chen
C
,
Sun
L
., et al
.
cGAS is essential for the antitumor effect of immune checkpoint blockade
.
Proc Natl Acad Sci U S A
2017
;
114
:
1637
42
.
28.
Garcia-Diaz
A
,
Shin
DS
,
Moreno
BH
,
Saco
J
,
Escuin-Ordinas
H
,
Rodriguez
GA
, et al
.
Interferon receptor signaling pathways regulating PD-L1 and PD-L2 expression
.
Cell Rep
2017
;
19
:
1189
201
.
29.
Fehrenbacher
L
,
Spira
A
,
Ballinger
M
,
Kowanetz
M
,
Vansteenkiste
J
,
Mazieres
J
, et al
.
POPLAR Study Group, Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial
.
Lancet
2016
;
387
:
1837
46
.
30.
Gong
X
,
Li
X
,
Jiang
T
,
Xie
H
,
Zhu
Z
,
Zhou
F
., et al
.
Combined radiotherapy and anti-PD-L1 antibody synergistically enhances antitumour effect in non-small cell lung cancer
.
J Thorac Oncol
2017
;
12
:
1085
97
.
31.
Mok
TSK
,
Wu
YL
,
Kudaba
I
,
Kowalski
DM
,
Cho
BC
,
Turna
HZ
, et al
.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial
.
Lancet
2019
;
393
:
1819
30
.
32.
Rizvi
NA
,
Hellmann
MD
,
Snyder
A
,
Kvistborg
P
,
Makarov
V
,
Havel
JJ
, et al
.
Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer
.
Science
2015
;
348
:
124
8
.
33.
Shin
DS
,
Zaretsky
JM
,
Escuin-Ordinas
H
,
Garcia-Diaz
A
,
Hu-Lieskovan
S
,
Kalbasi
A
, et al
.
Primary resistance to PD-1 blockade mediated by JAK1/2 mutations
.
Cancer Discov
2017
;
7
:
188
201
.
34.
Sriram
G
,
Milling
LE
,
Chen
JK
,
Kong
YW
,
Joughin
BA
,
Abraham
W
, et al
.
The injury response to DNA damage in live tumor cells promotes antitumor immunity
.
Sci Signal
2021
;
14
:
eabc4764
.
35.
Fang
S
,
Wang
J
,
Jiang
H
,
Zhang
Y
,
Xi
W
,
Zhao
C
, et al
.
Experimental study on the therapeutic effect of positron emission tomography agent [18F]-labeled 2-deoxy-2-fluoro-D-glucose in a colon cancer mouse model
.
Cancer Biother Radiopharm
2010
;
25
:
733
40
.
36.
Moadel
RM
,
Weldon
RH
,
Katz
EB
,
Lu
P
,
Mani
J
,
Stahl
M
, et al
.
Positherapy: Targeted nuclear therapy of breast cancer with 18F-2-deoxy-2-fluoro-D-glucose
.
Cancer Res
2005
;
65
:
698
702
.
37.
Moadel
RM
,
Nguyen
AV
,
Lin
EY
,
Lu
P
,
Mani
J
,
Blaufox
MD
, et al
.
Positron emission tomography agent 18F-2-deoxy-2-fluoro-d-glucose has a therapeutic potential in breast cancer
.
Breast Cancer Res
2003
;
5
:
199
205
.
38.
US FDA
.
Guidance for industry: Estimating the maximum safe starting dose in adult healthy volunteer
.
Rockville, MD
:
US Food and Drug Administration
2005
.
39.
Srinivasan
S
,
Crandall
JP
,
Gajwani
P
,
Sgouros
G
,
Mena
E
,
Lodge
MA
., et al
.
Human radiation dosimetry for orally and intravenously administered 18F-FDG
.
J Nucl Med
2020
;
61
:
613
9
.
40.
National research council
.
health risks from exposure to low levels of ionizing radiation: BEIR VII phase 2
.
Washington, DC
:
The National Academies Press
. p
7
,
2006
.
41.
Huang
L
,
Li
Y
,
Du
Y
,
Zhang
Y
,
Wang
X
,
Ding
Y
, et al
.
Mild photothermal therapy potentiates anti-PD-L1 treatment for immunologically cold tumors via an all-in-one and all-in-control strategy
.
Nat Commun
2019
;
10
:
4871
.
42.
Galon
J
,
Bruni
D
.
Approaches to treat immune hot, altered and cold tumours with combination immunotherapies
.
Nat Rev Drug Discov
2019
;
18
:
197
218
.
43.
Efremova
M
,
Rieder
D
,
Klepsch
V
,
Charoentong
P
,
Finotello
F
,
Hackl
H
, et al
.
Targeting immune checkpoints potentiates immunoediting and changes the dynamics of tumor evolution
.
Nat Commun
2018
;
9
:
32
.
44.
Grapin
M
,
Richard
C
,
Limagne
E
,
Boidot
R
,
Morgand
V
,
Bertaut
A
, et al
.
Optimized fractionated radiotherapy with anti-PD-L1 and anti-TIGIT: a promising new combination
.
J Immunother Cancer
2019
;
7
:
106
.
45.
Yadav
M
,
Jhunjhunwala
S
,
Phung
QT
,
Lupardus
P
,
Tanguay
J
,
Bumbaca
S
, et al
.
Predicting immunogenic tumour mutations by combining mass spectrometry and exome sequencing
.
Nature
2014
;
515
:
572
8
.

Supplementary data