Introduction
Cancer immunotherapy represents a transformative approach in tumor treatment, harnessing the immune system to target both local and metastatic cancer cells1. However, sustained therapeutic responses are observed in less than 25% of patients, highlighting the necessity for early and effective differentiation between responders and non-responders to guide timely treatment decisions2,3. Current standard methods, including blood analysis of lymphocytes and cytokines, as well as histological examination of biopsies, present limitations due to their static, invasive, and indirect nature, lacking immune-related spatiotemporal information4. Cytotoxic T lymphocytes (CD8+ T cells) are pivotal in the immunotherapy cycle, serving as primary effectors that directly induce cytotoxic effects on tumor cells5. The quantity, localization, and activation status of T cells significantly influence the efficacy of immunotherapy, making their dynamic assessment crucial for evaluating patient responses and optimizing treatment strategies6,7. Molecular imaging techniques offer essential support for real-time, precise, and non-invasive visualization of T cell activation processes. Notably, granzyme B (GzmB), a key serine protease involved in T cell-mediated cytotoxicity, serves as a biomarker for in vivo immune responses8, 9–10. Recent advancements have led to the development of GzmB-targeted imaging agents that facilitate non-invasive evaluations of tumor immunotherapy efficacy11,12. Despite this progress, conventional imaging modalities such as magnetic resonance imaging (MRI) and positron emission tomography (PET) face challenges, including low sensitivity, high background signals, and limitations in multi-channel imaging, alongside the need for specialized instruments and well-trained technicians8. These hurdles underscore the urgent demand for innovative imaging strategies that offer high specificity, sensitivity, and real-time monitoring of T cell activation.
Optical imaging technology (fluorescence or self-luminescence imaging) presents several advantages, including rapid imaging capabilities, high sensitivity, and excellent spatial-temporal resolution13, 14, 15–16. Various fluorescence probes have been reported for real-time tracking of immune cell activation17, 18, 19–20, primarily in the visible (Vis) and near-infrared I (NIR-I) ranges21, 22, 23–24. However, these approaches often encounter issues such as shallow tissue penetration and significant tissue autofluorescence, which hinder effective imaging. To reduce tissue autofluorescence, the self-luminescence imaging strategies without real-time light excitation (afterglow luminescence25,26 or chemiluminescence imaging27, 28–29) have been developed and show a high signal-to-noise ratio (SNR), but suffer from limited imaging time and need for exogenous substances with potential safety concerns. Correspondingly, NIR-II fluorescence imaging (1000–1700 nm) significantly improves tissue penetration and reduces background fluorescence19,30,31, making it more suitable for in vivo monitoring of T cell activation. While some GzmB-based imaging methods have been developed based on single fluorescence intensity changes, they are susceptible to confounding factors such as uneven probe distribution and environmental variability, thus compromising quantitative assessments32,33. Ratiometric fluorescence imaging, utilizing reference signal intensity for self-calibration, offers a solution by minimizing these interferences and enhancing the accuracy of in vivo quantitative measurements34,35.
Lanthanide doped down-conversion emission nanoparticles (DCNPs) exhibit unique NIR-II luminescent properties, including narrow-band emission, excellent photostability, deep tissue penetration, and low toxicity. These characteristics position them as promising candidates for NIR-II imaging applications36,37. Furthermore, the complex energy-level structures of lanthanide ions enable the design of multi-excitation or multi-spectral emission probes38, 39–40, facilitating ratiometric fluorescence analysis41,42. Despite these advantages, the application of ratiometric NIR-II imaging has been limited, primarily focusing on small molecules like reactive oxygen species and glutathione, with fewer studies addressing critical biomolecules such as proteases43. Furthermore, integrating the advantages of NIR-II fluorescence imaging and ratiometric imaging based on DCNPs for anti-tumor immune responses has not yet been achieved.
In this work, we report a ratiometric NIR-II fluorescence probe (DCGA nanoprobe) to quantitatively image T cell activation based on GzmB, which is correlated with hepatocellular carcinoma (HCC) patients’ responses to immune therapies (Supplementary Fig. 1), and further predict tumor responses to immunotherapy (Fig. 1). The DCGA comprises a dual-emission NIR-II luminescence center, a GzmB-cleavable peptide, and a NIR dye (A1094) that quenches emission at 1060 nm (emission of Nd3+). Upon GzmB activation, the fluorescence signal at 1060 nm is restored and at 1525 nm (emission of Er3+) is retained, allowing for real-time monitoring of T cell activation and enabling differentiation between immune responders and non-responders in immunomodulator evaluation models, immune checkpoint inhibitor (ICI) based immunotherapeutic models, and adoptive T cell immunotherapy in patient-derived xenograft (PDX) models. Overall, this ratiometric NIR-II fluorescence imaging strategy provides a valuable tool for non-invasive early detection of T cell activation, thereby increasing the potential for tailoring immunotherapy regimens.
Fig. 1 Schematic representation of DCGA for real-time NIR-II ratiometric fluorescence imaging of CD8+ T cell activation during tumor immunotherapy. [Images not available. See PDF.]
a Illustration of the preparation of DCGA. b Mechanism of DCGA-based real-time NIR-II ratio (F1060nm/F1525nm) fluorescence imaging of CD8+ T cell activation in vivo and for evaluating immunotherapy efficacy. c Prediction of adoptive T cell therapy efficacy in a PDX model via in vivo real-time ratio imaging using DCGA.
Results
Synthesis and characterization of DCGA
A GzmB-activatable ratiometric NIR-II nanoprobe (DCNPs@GzmB-A1094, DCGA) was constructed by conjugating GzmB peptide substrates to dual-emission down-conversion nanoparticles (DCNPs) modified with DSPE-PEG. The NIR absorber A1094 was incorporated to quench the emission at 1060 nm, while the emission at 1525 nm served as a reference (Fig. 2a). The core-shell β-NaErF4@NaYF4@NaYF4:10%Nd@NaYF4 (DCNPs) were synthesized using a layer-by-layer growth method, utilizing Er3+ and Nd3+ ions as dual activators (Supplementary Fig. 2). A carefully chosen of NaYF4 interlayer minimized energy losses and quenching (Supplementary Fig. 3). The successful synthesis of the core nanoparticles and DCNPs was confirmed by transmission electron microscopy (TEM), dynamic light scattering (DLS), X-ray diffraction (XRD), and high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) (Supplementary Figs. 4–7). Subsequently, DSPE-PEG facilitated the phase transfer of hydrophobic DCNPs to hydrophilic DCNPs@PEG, retaining dual NIR-II emission characteristics (Supplementary Fig. 8). The GzmB-cleavable peptide (Tyr(SO3)AIEFD|SGC) was conjugated to DCNPs@PEG via a thiol-maleimide addition reaction, monitored by a 2,2’-dithiobis(2-nitrobenzoic acid) (DTNB) assay (Supplementary Fig. 9). Following this, the NIR-II dye A1094 (Supplementary Figs. 10–14) was conjugated to DCNPs@GzmB through ionic bonding44, yielding the ratiometric nanoprobe DCNPs@GzmB-A1094 (DCGA) (Supplementary Fig. 15). The TEM image (Fig. 2b) indicated that DCGA was well-dispersed, with a size of approximately 80 × 50 nm. Post-modification, the hydrodynamic diameter (Dh) of the nanoparticles ranged from 68 nm to 91 nm (Fig. 2c), and zeta potential measurements revealed negative surface charges for all formulations, indicating suitability for in vivo applications (Fig. 2d). UV-Vis-NIR absorption spectra confirmed successful peptide and A1094 modifications, showing strong absorption peaks at 216 nm and 1094 nm (Fig. 2e). The overlap between A1094 absorption at 1094 nm and DCNPs@GzmB fluorescence at 1060 nm suggested the effective Förster resonance energy transfer (FRET)-based quenching (Fig. 2f). The fluorescence intensity at 1060 nm decreased by approximately 85.3% with increasing A1094 concentration (Fig. 2g, h). The average DLS size, polydispersity index (PDI), and fluorescence signal of DCGA remained stable in 10% fetal bovine serum (FBS) over 7 days, confirming its colloidal and optical stability (Fig. 2i and Supplementary Figs. 16–18). Furthermore, DCGA exhibited superior photostability compared to the small molecule NIR-II dye IR1061 under continuous 808 nm laser irradiation (Fig. 2j), indicating its potential for long-term in vivo imaging.
Fig. 2 Synthesis and characterization of DCGA. [Images not available. See PDF.]
a Illustration of the preparation of DCGA. b TEM image of DCGA probe (scale bar: 100 nm). Inset: High-magnification TEM image of DCGA (scale bar: 50 nm). The experiments in (b) were repeated independently three times with similar results. c, d DLS results (c) and surface zeta potential (d) results of DCNPs@PEG, DCNPs@GzmB and DCNPs@GzmB-A1094 (termed DCGA) (n = 3 independent samples). e Absorption spectra of GzmB peptide, DCNPs@PEG, A1094 and DCGA. f Overlap of fluorescence emission spectra of DCNPs@GzmB and absorption spectra of A1094. g Fluorescence spectra of DCGA with varying concentrations of A1094 (0–80 μg/mL) under 808 nm laser excitation. h Plot of fluorescence ratio (F1060nm/F1525nm) changes as a function of A1094 concentration (0–80 μg/mL) (n = 3 independent samples). i Average size and PDI changes of DCGA (n = 3 independent samples) probe in medium containing 10% FBS. j Normalized fluorescence changes of DCGA (F1525nm and F1060nm) and small molecule dye IR1061 (F1060nm) under irradiation with 808 nm laser (0.4 W/cm2) for 20 min (n = 3 independent samples). In the figure, the “a.u.” means “arbitrary units”. The data are presented as mean ± SD. Source data are provided as a Source Data file.
Ratiometric NIR-II fluorescence detection of GzmB activity by DCGA
Upon interaction with GzmB, the substrate peptide undergoes cleavage, leading to the detachment of A1094 from DCNPs and restoration of the 1060 nm fluorescence (Fig. 3a and Supplementary Fig. 19). The emission at 1525 nm remained stable, serving as a reference. Thus, the fluorescence ratio (F1060nm/F1525nm) changes can effectively detect GzmB activity. Unlike traditional probes that rely on single fluorescence intensity changes, our NIR-II ratiometric probe provides precise quantitative capabilities by eliminating interference from factors such as probe distribution and instrumental variation. We investigated the response of DCGA to GzmB by assessing NIR-II fluorescence changes in the presence or absence of GzmB. Incubation with GzmB resulted in a significant enhancement of the 1060 nm signal, while the 1525 nm fluorescence remained stable (Fig. 3b and Supplementary Fig. 20a). The fluorescence ratio increased 5.1-fold after GzmB incubation (Supplementary Fig. 20b). Fluorescence spectroscopy further confirmed that the intensity at 1060 nm increased significantly, indicating successful hydrolysis of A1094-(Tyr(SO3)AIEFD|SGC) by GzmB (Fig. 3c and Supplementary Fig. 19). Next, the sensitivity of DCGA for GzmB was evaluated, revealing a gradual restoration of fluorescence at 1060 nm with increasing GzmB concentrations (0–0.3 μM), while the 1525 nm signal remained constant (Fig. 3d). The fluorescence ratio (tested by fluorescence spectra) was linearly correlated with GzmB concentration (y = 8.237x + 0.6653, R2 = 0.9758), with a detection limit of 41.6 nM, confirming the feasibility of DCGA as an activatable NIR-II ratiometric probe for GzmB activity (Fig. 3e). The response kinetics of DCGA to GzmB confirmed its rapid responsive behavior, which is beneficial for further real-time imaging (Supplementary Figs. 21, 22). The DCGA also showed good responsive specificity towards GzmB (Fig. 3f, g and Supplementary Fig. 23), demonstrating its potential for specific in vivo detection of GzmB.
Fig. 3 Detection of DCGA response to GzmB. [Images not available. See PDF.]
a Schematic illustration of the fluorescence spectral changes of DCGA probe in response to GzmB. b, c The NIR-II fluorescence images (single channel) and corresponding ratio images (b) and fluorescence emission spectra (c) response of DCGA when incubated without or with GzmB (0.2 μM). The “a.u.” means “arbitrary units”. d, e NIR-II fluorescence images (single channel) and corresponding ratio images (d), and ratio values (e) obtained from fluorescence spectra of DCGA incubated with different concentrations of GzmB (0–0.3 μM) (n = 4 independent samples). f NIR-II fluorescence images and corresponding ratio images of DCGA after incubation with different substrates (hydrogen peroxide (H2O2), alkaline phosphatase (ALP), γ-glutamyl transferase (GGT), tyrosinase (TYR), cathepsin C (CTSC), neutrophil elastase (NE) and GzmB). g Ratio values were calculated from the images shown in (f) using ImageJ software (n = 4 independent samples). The data are presented as mean ± SD. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test (****p < 0.0001). Source data are provided as a Source Data file.
In vitro detection of GzmB at the cellular level by DCGA
To explore the cellular detection capability of the DCGA probe for GzmB activity, immature CD8+ T cells were isolated from the spleens of C57BL/6 mice and then activated using anti-CD3ε/CD28 antibodies45 (Fig. 4a, b and Supplementary Figs. 24–26). The high expression of GzmB in activated CD8+ T cells was further confirmed by Western blot analysis (Supplementary Fig. 27). The cytotoxicity of DCGA was assessed through a Cell Counting Kit-8 (CCK-8) assay on normal mouse liver cells (BNL CL.2), mouse hepatocellular carcinoma cells (Hepa 1-6), CD8+ T cells, mouse breast cancer cells (4T1) and mouse embryonic fibroblast cells (3T3). The results demonstrated negligible cytotoxicity up to 0.6 mg/mL of DCGA over 24 h (Fig. 4c and Supplementary Fig. 28), indicating its good biocompatibility on cells.
Fig. 4 In vitro detection of DCGA response to GzmB at cellular level. [Images not available. See PDF.]
a Schematic representation of the isolation and activation of CD8+ T cells from mouse. b The quantification results of the expression of CD69 (left) and CD25 (right) on CD8+ T cells from flow cytometry (n = 4 independent experiments). c Relative cell viability of BNL CL.2 cells, Hepa 1-6 cells and CD8+ T cells after incubation with DCGA (n = 5 independent experiments). d NIR-II fluorescence images and corresponding ratio images of DCGA after incubation with Hepa1-6 cells, unactivated CD8+ T cells, activated CD8+ T cells, and activated CD8+ T cells pre-treated with the GzmB inhibitor (Ac-IEPD-CHO). e Ratio values calculated from the images shown in (d) using ImageJ software (n = 5 independent experiments). f Schematic representation of DCGA responsive imaging in the co-culture system of activated CD8+ T cells/Hepa1-6 cells. g NIR-II fluorescence images and corresponding ratio images of DCGA after incubation with effector cells (E, activated CD8+ T cells) and target cells (T, Hepa 1-6 cells) at various E:T ratios. h Ratio values calculated from the images shown in (g) using ImageJ software (n = 4 independent experiments). i Secreted GzmB concentrations from cells under different treatments (n = 4 independent experiments). j Correlation between GzmB concentration in cell secretions and the ratio signal (F1060nm/F1525nm) of DCGA in cell pellets under different treatments, analyzed using a linear regression model. k Cell death rate under different treatments was determined by measuring the LDH concentration in cell secretions (n = 4 independent experiments). l The correlation between the cell death rate and the ratio signal of DCGA with different treatments. The treatments included different E:T cell ratios of 0:1, 1:1, 5:1, and 10:1. The data are presented as mean ± SD. Statistical analysis was performed using a two-tailed paired Student’s t-test (for b) and one-way ANOVA followed by Tukey’s multiple comparisons test (for e, h, i, k) (**p < 0.01, ****p < 0.0001). Source data are provided as a Source Data file.
To investigate DCGA’s selective response to intracellular GzmB, experiments were conducted on activated CD8+ T cells (GzmB-positive), GzmB-negative Hepa 1-6 cells, and unactivated CD8+ T cells. Following co-incubation with the nanoprobe, fluorescence changes were monitored using NIR-II imaging (Fig. 4d, e). Enhanced F1060nm intensity and ratiometric signals were observed in activated CD8+ T cells (G3) compared to both unactivated CD8+ T cells and Hepa 1-6 cells, confirming specific probe responsiveness towards GzmB. Furthermore, pretreatment with Ac-IEPD-CHO, a GzmB inhibitor, eliminated fluorescence signal changes (G4), underscoring the role of GzmB in restoring DCGA fluorescence (Fig. 4d, e). Quantitative analysis of GzmB concentrations across different treatment groups further confirmed the consistency between fluorescent signals and GzmB levels (Supplementary Fig. 29).
Afterwards, we further assessed the ability of DCGA to image GzmB-mediated cytotoxic T lymphocyte (CTL) responses in a co-culture system. Hepa 1-6 cells, pre-incubated with DCGA for 12 h, were co-cultured with varying numbers of activated CD8+ T cells for 24 h, during which lactate dehydrogenase (LDH) activity and GzmB concentrations in the supernatant were quantified (Fig. 4f). The F1060nm signal increased with increasing numbers of activated CD8+ T cells, while the F1525nm signal remained unchanged, leading to significantly enhanced ratiometric signals (Fig. 4g). Quantitative analysis indicated that the ratiometric signal (F1060nm/F1525nm) was 2.45-fold higher post-CTL treatment at an E:T ratio of 10:1 (Fig. 4h). ELISA results also confirmed that the GzmB concentrations in the supernatant increased along with the increasing of CTL numbers (Fig. 4i). A linear regression analysis illustrated a positive correlation between the ratiometric signals and GzmB concentrations (R2 = 0.8974, r = 0.9473) (Fig. 4j). Additionally, LDH assay results indicated a correlation between Hepa 1-6 cell mortality and DCGA ratiometric signals (R2 = 0.9066, r = 0.9521) (Fig. 4k, l). These findings reveal that DCGA effectively detects GzmB levels and holds promise for predicting CTL-mediated cytotoxicity.
In vivo NIR-II ratiometric fluorescence imaging of CD8+ T cell activation by DCGA
Before in vivo imaging, the biosafety of DCGA was first evaluated. Hemolysis assays demonstrated that DCGA did not induce significant hemolysis at concentrations as high as 0.6 mg/mL (Supplementary Fig. 30). Healthy mice were subsequently used for in vivo biosafety tests, where body weight was monitored after intravenous injections of either PBS (control) or DCGA. No significant weight loss was observed in the nanoprobe-treated group compared to controls (Supplementary Fig. 31). Additionally, serum biochemical indices and liver and kidney function markers tests showed no significant abnormalities in the nanoprobe-treated mice (Supplementary Fig. 32). Histological (H&E staining) analysis of major organs further confirmed the biosafety of DCGA (Supplementary Fig. 33). Biodistribution studies indicated that DCGA nanoprobe primarily accumulated in the liver, aligning with previous findings that nanomaterials are first cleared by the liver and also transported by immune cells to the spleen46 (Supplementary Fig. 34). The half-life of DCGA in blood circulation was found to be about 2 h (Supplementary Fig. 35), and the clearance of DCGA was mainly mediated by the liver and biliary excretion (Supplementary Fig. 36). Collectively, these results indicate that DCGA exhibits low systemic toxicity and good biocompatibility, supporting its potential for further in vivo biomedical applications.
The complex in vivo physiological environment and tumor immune microenvironment pose significant challenges for developing imaging agents with high specificity for CTL detection47. The ability of DCGA to specifically image the activity of GzmB released by CTLs was evaluated in a Hepa 1-6 tumor-bearing mouse model (Fig. 5a). Following injection of DCGA, NIR-II fluorescence imaging was performed 12 h post-injection to assess T cell activation. Mice were then treated with a GzmB inducer (phorbol myristate acetate (PMA) and ionomycin, via intratumoral injection)48, a GzmB activity inhibitor (Ac-IEPD-CHO) 4 h before the injection of GzmB inducer, and taken PBS as the control. The mechanism of GzmB inducer for T cells activation49 was shown in Fig. 5b. Longitudinal imaging at 24 h post-injection revealed that while F1060nm and F1525nm signal intensities varied among mice, the ratio (F1060nm/F1525nm) remained consistent, indicating no significant difference in GzmB activity at 12 h (Fig. 5c and Supplementary Figs. 37, 38). Notably, the GzmB inducer-treated group exhibited enhanced fluorescence recovery and significantly higher ratiometric signals at 24 h with 1.58-fold and 1.69-fold increases compared to the inhibitor and control groups, respectively (Fig. 5c, d and Supplementary Fig. 37). In addition, there were no significant differences between PBS and the GzmB inhibitor treated group, indicating the specificity of our imaging strategy (Supplementary Fig. 39). GzmB concentrations in tumor tissues corroborated these findings, showing marked increases in the inducer-treated group compared to controls (Supplementary Fig. 40). Linear regression analysis indicated a strong positive correlation between DCGA ratio signals and GzmB concentrations (R2 = 0.8733, r = 0.9366) (Fig. 5e). Thus, DCGA demonstrates potential for NIR-II ratiometric imaging of T cell activation in vivo.
Fig. 5 In vivo NIR-II fluorescence imaging of CD8+ T cell activation with DCGA. [Images not available. See PDF.]
a Schematic illustration of in vivo real-time imaging of tumor CD8+ T cell activation in subcutaneous Hepa 1-6 tumor model. b Schematic illustration of the mechanism of PMA and ionomycin-induced CD8+ T cell activation and GzmB release. c NIR-II fluorescence images and ratio plots of tumor sites of individual mice under different treatments. d Ratio values of tumor sites treated with G1: PBS, G2: GzmB inducer (combination of PMA and ionomycin) and G3: GzmB inhibitor (Ac-IEPD-CHO, administered 4 h before the inducer treatment) at 24 h post-injection of DCGA (n = 4 mice per group). e The correlation between GzmB concentration in isolated tumors and the ratio signals of DCGA in tumors with different treatments. The data are presented as mean ± SD. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test (**p < 0.01). Source data are provided as a Source Data file.
DCGA enables early prediction of immune therapy responsiveness
The capability of DCGA to visualize CTL activity in vivo was validated in a Hepa 1-6 tumor-bearing mouse model, utilizing direct intratumoral T cell activation. To further assess DCGA’s potential for in vivo imaging and early prediction of immune therapy responsiveness, we administered three immunotherapeutic agents with distinct mechanisms: NLG919, BMS-1, and S-(2-boronethyl)-L-cysteine hydrochloride (BEC) (Fig. 6a). NLG919 inhibits indoleamine 2,3-dioxygenase (IDO), alleviating immune suppression in the TIME by reducing tryptophan metabolism50. BMS-1 functions as an immune checkpoint inhibitor, blocking the interaction between PD-1 and PD-L1, thereby restoring T cell activity and enhancing anti-tumor immune responses51. BEC, an arginase inhibitor, promotes T cell proliferation by decreasing arginine metabolism52. These agents target the TIME through different pathways, facilitating the infiltration of CTLs or activating CTLs for tumor immunotherapy (Fig. 6b). However, suitable ratiometric NIR-II imaging tools for in vivo evaluation of the therapeutic responses of these immunomodulators remain scarce.
Fig. 6 In vivo real-time NIR-II fluorescence imaging of DCGA in Hepa 1-6 tumor model with immunotherapy. [Images not available. See PDF.]
a Schematic illustration of timeline for immunotherapy and real-time imaging. b Chemical structures and targets of different immunomodulators for T cell activation and GzmB release. c Representative NIR-II fluorescence images and ratio plots at different time points after i.v. injection of DCGA in tumor-bearing mice treated with different immunomodulators. d, e In vivo NIR-II fluorescence images and ratio plots (d) and the corresponding ratio signal (F1060nm/F1525nm) values (e) of tumor-bearing mice following different treatments at 24 h post-injection of DCGA. f Correlation between the relative tumor volume on day 22 and the ratio signal values of DCGA in tumors on day 10 with different treatments. g Schematic illustration of the correlation between the relative tumor volume and the ratio signal values. The different treatments applied were G1: PBS, G2: NLG919, G3: BMS-1, and G4: BEC, respectively. Source data are provided as a Source Data file.
Following three rounds of immunotherapy, mice were intravenously injected with DCGA for NIR-II fluorescence imaging and quantitative analysis at designated time points. Longitudinal fluorescence imaging of representative mice from each group is presented in Fig. 5c. DCGA exhibited gradual accumulation in tumors, with F1060nm and F1525nm signals progressively increasing, peaking at 24 h post-injection. Subsequently, fluorescence signals decreased, reflecting DCGA accumulation in the tumor and subsequent metabolism (Supplementary Fig. 41). In immunotherapy groups, GzmB activity was higher than in controls, indicated by stronger fluorescence recovery on the F1060nm channel (Fig. 6c and Supplementary Fig. 41). Quantitative analysis revealed that the F1060nm/F1525nm fluorescence ratio increased over time, plateauing at 24 h, suggesting ongoing GzmB accumulation and immune response (Supplementary Fig. 42). At 24 h post-injection of DCGA (day 10 after treatment), in vivo imaging revealed significant differences in fluorescence ratios among treatment groups (Fig. 6d and Supplementary Fig. 43). The self-calibrated ratio signal analysis provided more accurate imaging than that of single channel (F1060nm)53, with fluorescence ratios of 1.2, 1.41, and 1.89 times higher in the NLG919, BMS-1, and BEC treated groups than in the control group, respectively (Supplementary Fig. 44).
The therapeutic effects of various immunomodulators on tumor growth were evaluated by monitoring tumor volume (Supplementary Fig. 45). Although no significant differences in relative tumor volume changes were observed across groups on day 10, notable differences in ratiometric signals were detected between treated and control groups (Supplementary Figs. 44, 46a). By day 22, significant tumor volume suppression was evident in treated groups but not in controls (Supplementary Fig. 46b). These findings indicate that DCGA facilitates early, non-invasive monitoring of immune responses, detecting immunotherapy efficacy prior to observable changes in tumor volume. Due to individual differences, precisely analyzing each mouse’s response to the immunotherapeutic agents is more meaningful than evaluating the entire treatment group. To assess individual responses to immunotherapy, a cut-off value of 0.729 (mean control + 2-fold of standard deviations54) was established to predict immune response occurrence. Fluorescence ratio analysis indicated immune response rates of 25%, 75%, and 100% for NLG919, BMS-1, and BEC treated groups, respectively (Fig. 6e). Correlation analysis between the fluorescence ratio and relative tumor volume on day 22 revealed a negative correlation (R2 = 0.8535, r = –0.9239) (Fig. 6f and Supplementary Fig. 47), indicating that CTL activation serves as a positive prognostic indicator of treatment outcomes55,56. Thus, DCGA provides an early, non-invasive, and accurate method for predicting treatment efficacy (Fig. 6g), allowing differentiation between responders and non-responders across various immunomodulator treatments, addressing individual response variability that has been underexplored.
Correlation between DCGA signal response and immune activation in tumors
To further investigate the relationship between DCGA signal response and immune activation post-treatment, additional immune-related evaluations were performed. Following three treatment cycles, tumor-bearing mice received intravenous injections of DCGA for imaging. Tumors were isolated 24 h post-injection for ex vivo NIR-II fluorescence imaging and subsequent immune mechanism analysis (Fig. 7a). Treatment groups exhibited enhanced fluorescence recovery in the F1060nm signal compared to the PBS-treated group, with NIR-II fluorescence ratios in tumors from the NLG919, BMS-1, and BEC treated groups increasing by 1.28, 1.54, and 1.87 times, respectively (Fig. 7b, c). The low scattering and deep tissue penetration capability of DCGA facilitated a strong correlation between in vivo and ex vivo fluorescence ratio data (Fig. 7c and Fig. 6e, respectively), consistent with previous studies57.
Fig. 7 Ex vivo characterization of NIR-II fluorescence imaging of DCGA and immune analysis in isolated tumors with immunotherapy. [Images not available. See PDF.]
a Schematic illustration of the timeline of immunotherapy and immune analysis. b, c NIR-II fluorescence images and ratio plots (b), and the ratio signal values (c) of DCGA in tumors from individual mice following different treatments. d, e Typical FCM plots (d) and corresponding quantification (e) results of CD3+CD8+ T cells in tumors from individual mice after different treatments. f, g Correlation between the proportion of CD3+CD8+ cells and the ratio signals (f), and between GzmB concentrations and the ratio signals (g) of DCGA in tumors with different treatments. h Multiple immunofluorescence images display immunostaining of CD8+ (green) and GzmB (red) within tumor tissues after different treatments. i TUNEL, (j) H&E, and (k) Ki67 staining of typical tumor slices after different treatments. Scale bar: 50 μm. The different treatments include G1: PBS, G2: NLG919, G3: BMS-1, and G4: BEC. The experiments in (i–k) were repeated independently three times with similar results. Source data are provided as a Source Data file.
During the adaptive immune response, CD8+ T cells migrate to tumor sites to target and eliminate tumor cells, while CD4+ T cells enhance CD8+ T cell cytotoxicity through cytokine secretion58. To assess whether DCGA signal response correlates with the infiltration of CD8+ and CD4+ T cells in TME, flow cytometry was conducted to quantify infiltrating T cells. Treatment groups demonstrated increased infiltration of CD3+CD4+ T cells compared to controls (Supplementary Figs. 48–50). The intratumoral infiltration of CD3+CD8+ T cells in the NLG919, BMS-1, and BEC groups increased by 1.39, 1.95, and 2.26 times as compared to PBS-treated group, respectively (Fig. 7d, e and Supplementary Fig. 51). A linear regression model revealed a positive correlation between the F1060nm/F1525nm ratio signals and the infiltration of both CD3+CD4+ T cells (R2 = 0.8417, r = 0.9174) and CD3+CD8+ T cells (R2 = 0.8991, r = 0.9482) (Fig. 7f and Supplementary Fig. 50b). To determine whether probe activation at the tumor site correlated with increased GzmB release due to CD8+ T cell infiltration, the linear regression analysis was performed and showed a positive correlation between the DCGA ratio signals and GzmB concentrations in isolated tumors (R2 = 0.8758, r = 0.9539) (Fig. 7g and Supplementary Fig. 52). These results indicate a strong alignment between immune-related analyses and DCGA imaging results, confirming that the nanoprobe’s ratio signal effectively evaluates immune activation in vivo.
Immunofluorescence staining of tumor tissues from representative mice revealed significantly enhanced green fluorescence signals for CD8+ T cells and red fluorescence signals for GzmB in the treatment groups (Fig. 7h and Supplementary Fig. 53), consistent with DCGA fluorescence ratio changes. Higher CD8+ T cell infiltration and increased GzmB levels in treated groups led to significant tumor cell apoptosis and necrosis, as evidenced by TUNEL and H&E staining, alongside reduced tumor cell proliferation (Ki67 staining) (Fig. 7i–k and Supplementary Fig. 54).
Immune checkpoint inhibitors, such as anti-PD-1 antibody or anti-PD-L1 antibody, have been recognized as the most successful and the most widely applied first-line immunotherapy strategy in a wide range of solid tumors, but lack available methods to predict its effectiveness in advance. Therefore, the feasibility of early prediction of immunotherapy responsiveness to the clinically widely used anti-PD-L1 antibody by DCGA was further investigated in a Hepa 1-6 tumor-bearing mouse model. Consistently, the results also clearly demonstrated that the DCGA could effectively evaluate T cell activation and differentiate between responders and non-responders after anti-PD-L1 treatment (Supplementary Figs. 55–58). The NIR-II imaging results were also well matched to the tumor growth and immunofluorescence staining of tumor tissues. These observations underscore the potential of DCGA as a non-invasive imaging tool for assessing CTL activity and predicting therapeutic effects in cancer immunotherapy, which will provide a non-invasive imaging tool for developing new immunotherapy drugs.
DCGA evaluates and predicts the efficacy of adoptive T cell therapy in the PDX model
Building on the successful application of DCGA for monitoring immune responses and CTL activity in Hepa 1-6 tumor model, we further investigated the probe’s applicability and sensitivity in a PDX mouse model for adoptive T cell therapy, aiming to facilitate clinical translation (Fig. 8a). The PDX model more accurately mimics human tumor growth and allows for the assessment of various immunotherapy strategies. Early monitoring of tumor responses to adoptive T cell therapy using DCGA may inform clinical decisions regarding patient eligibility for this therapy. The PDX model was established by subcutaneously implanting freshly resected primary tumor tissues from an HCC patient into immunocompromised NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NCG) mice, generating the first generation (P1) of PDX tumors. Subsequent passages utilized P1 tumors by similar steps (Fig. 8b and Supplementary Fig. 59). Histopathological examination, including Hep Par 1, H&E staining, Ki67 and GPC3 immunohistochemical analyses, confirmed that the PDX tumors retained characteristics of the primary tumors (Fig. 8c). Concurrently, autologous peripheral blood mononuclear cells (PBMCs) were isolated and cultured ex vivo to activate and expand CD8+ T cells for adoptive T cell therapy (Fig. 8b, d, and Supplementary Fig. 60).
Fig. 8 Prediction of adoptive T cell therapy efficacy in a PDX model via in vivo real-time ratio imaging using DCGA. [Images not available. See PDF.]
a Schematic representation of the therapeutic timeline for adoptive T cell therapy and DCGA-based imaging in the PDX tumor model. b Schematic illustration of PDX model construction and adoptive T cell therapy. c Histological sections of primary HCC tissues (from patient) and PDXs stained with H&E and immunohistochemistry. The HCC markers Hep Par1, Ki67, and GPC3 were detected by immunohistochemistry. Scale bar: 100 μm. The experiments in (c) were repeated independently three times with similar results. d The quantification results by flow cytometry for the expression of CD69 (left) and CD25 (right) on CD8+ T cells isolated from PBMC of the HCC patient (n = 3 independent experiments). e, f In vivo NIR-II fluorescence images and ratio plots (e), and the ratio signal (F1060nm/F1525nm) values (f) at tumor sites from mice with different treatments (n = 3 mice per group). g Tumor growth kinetics of individual mice with different treatments. h, i Relative tumor volume comparison on day 24 (h) and day 9 (i) following different treatments (n = 3 mice per group). j The ratio signal values from tumors of individual mice on day 9 following different treatments. k Correlation between the relative tumor volume on day 24 and the ratio signal values of DCGA in tumors on day 9 with different treatments. The different treatments applied were G1: PBS, G2: unactivated CD8+ T cells, and G3: activated CD8+ T cells, respectively. The data are presented as mean ± SD. Statistical analysis was performed using a two-tailed paired Student’s t-test (for d) and one-way ANOVA followed by Tukey’s multiple comparisons test (for f, h and i) (**p < 0.01, ***p < 0.001, ****p < 0.0001). Source data are provided as a Source Data file.
Following three rounds of therapy, PDX mice received intravenous injections of DCGA and underwent in vivo imaging 24 h later. Imaging results showed that F1060nm and F1525nm signal intensities in the non-activated CD8+ T cell therapy group (G2) varied among individual mice but exhibited similar F1060nm/F1525nm ratios compared to the control group (G1), indicating no significant T cell activation or GzmB activity (Fig. 8e and Supplementary Fig. 61). In contrast, tumors from the activated CD8+ T cell therapy group (G3) displayed enhanced ratiometric signals, showing a 1.76-fold increase compared to G2 (Fig. 8f), reflecting effective DCGA response and sustained GzmB release. Tumor volume changes were assessed to evaluate therapeutic efficacy (Fig. 8g and Supplementary Fig. 62). No significant tumor growth inhibition occurred in G1 and G2, while G3 demonstrated substantial tumor growth inhibition by day 18 and at the end of treatment (day 24) (Fig. 8h, i and Supplementary Figs. 63, 64).
Individual data analysis revealed that all mice in G3 exhibiting immune response based on the above defined cut-off ratio value. Correlation analysis between the ratio value (day 9) and relative tumor volume (day 24) indicated a negative correlation (R2 = 0.8347, r = –0.9136) (Fig. 8j, k and Supplementary Fig. 65), suggesting T cell infiltration and GzmB release as positive prognostic indicators. Post-treatment, tumors were isolated, photographed, and weighed (Supplementary Fig. 66). A linear regression analysis confirmed a negative correlation between the ratio signal and tumor weight (R2 = 0.8262, r = –0.909) (Supplementary Fig. 67). Throughout treatment, no significant body weight loss was observed, indicating the safety of both adoptive T cell therapy and DCGA (Supplementary Fig. 68). In vivo imaging conducted on day 9 provided early, non-invasive monitoring of immune responses, significantly preceding measurable tumor volume changes. The PDX model’s clinical relevance supports the potential application of GzmB-responsive probes in personalized immunotherapy, laying a reliable foundation for future clinical translation.
Discussion
The variability in immune responses among patients undergoing immunotherapy underscores the critical need for early identification of individuals who may benefit from tumor immunotherapy59,60. Imaging T cell activation with granzyme B release offers a precise means to capture the dynamic immune response, assess immunotherapy efficacy, and inform early treatment predictions and personalized therapeutic adjustments61. However, conventional visible or NIR-I fluorescence imaging probes, either “always on” or “turn-on” types, face challenges such as limited tissue penetration, strong autofluorescence, and high background noise. Additionally, probes relying solely on changes in fluorescence intensity are easily affected by factors such as uneven probe distribution, nonspecific accumulation, and microenvironmental conditions (e.g., pH, polarity, temperature), thus compromising precise quantification of target molecules33. We tackled these challenges with an activatable ratiometric nanoprobe, DCGA, utilizing NIR-II fluorescence ratiometric rare-earth-based nanomaterials to quantitatively visualize in vivo CTL activation and facilitate early predictions of immunotherapy outcomes. By harnessing the NIR-II and dual-emission capabilities of lanthanide-doped DCNPs and the self-calibrating nature of ratiometric signals, DCGA displayed enhanced tissue penetration depth, improved imaging contrast, and greater quantification accuracy in the NIR-II window. The ratio signal of DCGA showed positive correlation with GzmB concentration in Hepa 1-6 tumor models with intratumoral T cell activation (Fig. 5e) and negative correlation with relative tumor volume growth changes in different immunomodulator or ICIs treated models (Fig. 6f). The earlier in vivo prediction of T cell activation efficacy was also further verified in adoptive T cell immunotherapy in PDX models (Fig. 8k).
In summary, our experimental findings demonstrate that DCGA can effectively predict immune responses early and non-invasively in both mouse xenograft and PDX models with different immunotherapy strategies, enabling robust differentiation between responders and non-responders at earlier stages of treatment than conventional tumor volume-based methods. This predictive ability provides a valuable framework for assessing immunotherapy efficacy and optimizing personalized treatment regimens, presenting promising prospects for new immunotherapy strategy development. Future improvements can be directed toward optimizing the multifunctionality and imaging sensitivity of probes, potentially incorporating response modules for multiple targets and multimodal imaging technologies to facilitate dynamic imaging in deep tissues. Such advances will contribute to a deeper understanding of tumor immunotherapy mechanisms while providing innovative support for drug development and precision diagnostics. Further investigation into the long-term biocompatibility and metabolic profiles of DCGA, along with its application to additional disease models, will strengthen its foundation for clinical translation.
Methods
Materials
Neodymium (III) acetate hydrate (Nd(CH3COO)3, 99.9%) and erbium (III) chloride (ErCl3, 99.9%) were purchased from Sigma-Aldrich (St Louis, MO, USA). BEC hydrochloride (98%), BMS-1 (99.5%), NLG919 (99.9%) and anti-mouse PD-L1 antibody (HY-P99145) were purchased from MedChemExpress (Monmouth Junction, NJ, USA). Recombinant mouse granzyme B (C765) was brought from Novoprotein Scientific Inc. (Shanghai, China). The granzyme B substrate peptides (Tyr(SO3)AIEFD|SGC, or Tyr(SO3)AIEPD|SGC) were synthesized by Sangon Biotech (Shanghai, China). Phorbol 12-myristate 13-acetate (PMA, S7791) and ionomycin (S7074) were brought from Selleck (Houston, TX, USA). Other regents can be found in the Supplementary Information for details.
Synthesis of DCNPs@GzmB-A1094 (DCGA) nanoprobe
Conjugation of DCNP@PEG with GzmB-substrate peptide (DCNPs@GzmB)
DCNPs@PEG NPs (0.1 mmol) and GzmB substrate peptide (2 mg) were dispersed in 10 mL of deionized water and reacted under stirring at 500 rpm for 2 h. The resulting mixture was isolated and centrifuged at 21,000×g for 10 min, and the unbound peptide was removed by washing the pellets three times with deionized water. After that, the DCNPs@GzmB were obtained and re-dispersed in 10 mL of deionized water, kept in the dark and stored at 4 °C before further use. For testing the responsiveness towards human GzmB, the peptide was replaced by Tyr(SO3)AIEPD|SGC.
Immobilizing A1094 to DCNPs@GzmB (DCGA)
A1094 molecules were immobilized on the surface of DCNPs@GzmB via strong ionic bonding between the quaternary ammonium groups of A1094 and the sulfonate groups of peptides. Briefly, 40 μL of A1094 (2 mg/mL dissolved in DMF) was mixed with 1 mL of the prepared DCNPs@GzmB. The mixture solution was stirred at 500 rpm at room temperature for 1 h. After reaction, the nanoparticles were centrifuged, washed three times and re-dispersed in 1 mL of deionized water to obtain the DCNPs@GzmB-A1094 (DCGA) probe.
In vitro characterization of the response of DCGA to GzmB
Different concentrations of GzmB (0.01, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.3 μM) were incubated with 2 mM DCGA probe in HEPES buffer (50 mM HEPES, 10 mM CaCl2, pH 7.4) at 37 °C for 2 h. Granzyme B was pre-activated in HEPES buffer with cathepsin C (GzmB:CTSC = 2.5:1, w/w) for 4 h, followed by incubation with DCGA. After incubation, fluorescence imaging of the sample solutions was recorded using the NIR-II fluorescence imaging system, and the fluorescence intensity of solutions was also measured on a fluorescence spectrometer (FLS-1000, Edinburgh Instruments, U.K.). All experiments were repeated three times.
The limit of detection (LOD) was calculated based on the linear relationship between the blank sample standard deviation and the slope of the fitting curve. The LOD of DCGA for GzmB was determined according to the commonly used criterion of 3δ/S rule, where δ represents the standard deviation of the blank sample and S is the slope of the fitting curve.
NIR-II imaging of DCGA in cytotoxicity assay of cytotoxic T lymphocytes (CTLs)
Hepa 1-6 cells were seeded at a density of 4 × 105 cells per well in 6-well plates and cultured overnight. After removing the medium, the cells were washed three times with PBS. The cells were then incubated with fresh medium containing 0.2 mg/mL DCGA for 24 h. After incubation, the cells were washed again with PBS and detached using trypsin. After centrifugation at 300×g for 3 min, the cells were collected, counted, and subsequently re-seeded into 12-well plates. Afterwards, different ratios of effector cells (E, activated CD8+ T cells) to target cells (T, Hepa 1-6 cells) were used for co-culture (E:T = 0:1, 1:1, 5:1, and 10:1). After 24 h of co-incubation, the culture medium was collected for detecting GzmB levels using the Mouse Granzyme B ELISA Kit. The release of lactate dehydrogenase (LDH) from dead Hepa 1-6 cells was measured using an LDH Assay Kit. In addition, the cells were further collected by centrifugation for NIR-II fluorescence imaging.
In vivo NIR-II imaging on Hepa 1-6 tumor model
Animal studies were conducted using female C57BL/6 mice (6–8 weeks old, 18–22 g) and male NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NCG) mice (6–8 weeks old, 18–22 g). All procedures were approved by the Animal Ethics Committee of Mengchao Hepatobiliary Hospital of Fujian Medical University (No. MCHH-AEC-2023-17). All mice were housed under Specific Pathogen Free (SPF) conditions with free access to food and water, in a temperature-controlled (18–22 °C), humidity-controlled (50 ± 10%) environment with a 12 h light/dark cycle. NCG mice received sterilized bedding, autoclaved water, and irradiated feed. The maximal tumor volume permitted by the ethics committee was 1500 mm3, and this threshold was not exceeded in any experiment. Sex was not considered a biological variable in the study design or data analysis. For the PDX model, fresh tumor tissue was obtained from an 81-year-old male patient with hepatocellular carcinoma, and written informed consent was obtained prior to sample collection.
In vivo NIR-II imaging of granzyme B activity
Hepa 1-6 tumor-bearing mice (tumor volume reached ~200–300 mm3) were divided into three groups (n = 3): Group 1 (G1) served as the PBS control, Group 2 (G2) received the granzyme B inducer, and Group 3 (G3) received the granzyme B inhibitor (Ac-IEPD-CHO, administrated 4 h before the inducer treatment). Initially, all mice were intravenously injected with 100 μL of DCGA (0.4 mM in saline based on Er3+). At 12 h post-injection, the first round of NIR-II fluorescence imaging on mice was performed. Subsequently, the mice in the G3 group were intratumorally (i.t.) injected with the granzyme B inhibitor Ac-IEPD-CHO (10 μg). Four hours later, mice in both G2 and G3 groups received i.t. injections of T cell activators: PMA (100 ng) and ionomycin (5 μg) in 100 μL saline. Finally, at 24 h post-injection of DCGA, the second round of NIR-II fluorescence imaging was performed. After imaging, granzyme B levels in the excised tumors were determined using a Mouse Granzyme B ELISA kit. Additionally, a negative control group receiving only the granzyme B inhibitor (Ac-IEPD-CHO) without granzyme B inducer (PMA and ionomycin) was included as a control to confirm nonspecific effects.
In vivo real-time imaging of GzmB activity during tumor immunotherapy
Hepa 1-6 tumor-bearing mice (tumor volume ~80–120 mm3) were randomly divided into 4 groups for immunotherapy (n = 4). As follows: Group 1 (G1) received PBS as control, Group 2 (G2) received NLG919 (15 mg/kg), Group 3 (G3) received BMS-1 (15 mg/kg), and Group 4 (G4) received BEC (15 mg/kg). The immunotherapeutic agents were administered via intravenous injection every three days for a total of three doses. Throughout the treatment period, tumor volumes were recorded every two days. Tumor volume (V) was calculated using the formula: V = a × b2 / 2, where “a” is the longest diameter and “b” is the shortest diameter of the tumor. On day 9, 100 μL of DCGA (0.4 mM in saline, based on Er3+) was intravenously injected, and real-time NIR-II fluorescence imaging was performed for continuous monitoring over a 48-h period.
To further evaluate the applicability of DCGA in established immunotherapy regimens, an anti-PD-L1 antibody treatment was further included. Mice received intraperitoneal injections of anti-PD-L1 antibody (100 μg in 100 μL PBS) every three days for a total of three doses. On day 9, DCGA (100 μL, 0.4 mM based on Er3+) was administered via tail vein injection, followed by NIR-II fluorescence imaging to assess immune response.
In vivo NIR-II imaging of GzmB activity on PDX model
Establishment of the PDX tumor model on mice
To establish the PDX tumor model, 6-week-old male NCG mice were used. Within 1–2 h post-surgery, fresh tumor samples from a Stage II patient were cut into small pieces (2 × 2 × 2 mm3). Using sterile instruments, tumor fragments were implanted into the right axilla of the mice (hair was removed with depilatory cream before implantation). The incision was then sutured and disinfected, and this model was labeled as P1 generation PDX. Tumor size was measured every three days using a caliper, and tumor growth was monitored closely for three months to confirm the successful establishment of the P1 PDX model. When the tumor volume reached 1200 mm3, euthanasia was performed, and the tumor was excised. The tumor was then cut into fragments and re-implanted into the right axilla of a new mouse for the P2 generation. Subsequently, when the model was expanded to the P3 generation, adoptive T cell therapy and in vivo imaging were performed.
In vivo real-time imaging during adoptive T cell therapy
PDX tumor-bearing mice (tumor volume ~80–120 mm3) were randomly divided into three groups for adoptive T cell therapy (n = 3). As follows: Group 1 (G1) served as the PBS control, Group 2 (G2) received unactivated CD8+ T cells (1 × 106), and Group 3 (G3) received activated CD8+ T cells (1 × 106). T cells were administered via intravenous injection every three days for a total of three doses. Tumor volumes were measured every three days during treatment. On day 8, 100 μL of DCGA (0.4 mM in saline, based on Er3+) was intravenously injected. At 24 h post-injection, in vivo NIR-II imaging was performed as described above.
Additional experimental details can be seen in the Supplementary Information.
Statistics and reproducibility
All data are expressed as the mean ± SD with individual data points indicated. All experiments were independently repeated three times with similar results. Statistical analysis was performed using one-way ANOVA for comparisons among multiple groups, or the two-tailed paired Student’s t-test for comparisons between two groups. All statistical analyses were conducted using GraphPad Prism 8.0.2. Statistical significance was defined as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Pearson’s correlation coefficient was used to assess the correlation matrices. All flow cytometry data were analyzed using FlowJo 10.8.1. For immunofluorescence imaging, CaseViewer 2.4 was used to observe and analyze the images. Mass spectrum, 1H NMR, and 13C NMR data were processed using MestReNova 14.
Ethics statement
All animal experiments were conducted in accordance with relevant ethical regulations and approved by the Animal Ethics Committee of Mengchao Hepatobiliary Hospital of Fujian Medical University (No. MCHH-AEC-2023-17). The use of patient-derived tumor tissue for establishing the PDX model was approved by the Ethics Review Committee of Mengchao Hepatobiliary Hospital of Fujian Medical University (KESHEN-2021-100-02), and written informed consent was obtained from the donor prior to tissue collection.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Acknowledgements
The work was supported by the National Natural Science Foundation of China (62275050, Y.Z.; 62175031, X.L.), the Joint Funds for the Innovation of Science and Technology of Fujian Province (2023Y9268, Y.Z.), the Major Research Projects for Young and Middle-aged Talent of Fujian Provincial Health Commission (2021ZQNZD013, Y.Z.), the Natural Science Foundation of Fujian Province (2024J011224, X.Zh.), the Fujian Provincial Clinical Research Center for Hepatobiliary and Pancreatic Tumors (2020Y2013, Y.Z.), and the Scientific Foundation of Fuzhou Municipal Health Commission (2021-S-wp1, M.W.).
Author contributions
X.Zh., L.D. and X.L. conceived the idea; X.Zh., Y.Z., M.H. and X.L. supervised the study; X.Zh., L.D. and P.W. designed the probe and experiments; Y.Zh., J.K. and L.D. constructed the PDX and other tumor models; Z.W., Y.C. and H.L. synthesized the A1094; X.Zh., L.D., X.L. and M.W. analyzed the data; G.Ch. and G.Zh. analyzed the patients’ GzmB expression data; L.D., X.Zh., Y.Z., M.H. and X.L. discussed the results, wrote and revised the manuscript. L.D. and X.Zh. contributed equally to this work and acted as the joint first authors.
Peer review
Peer review information
Nature Communications thanks Qingqing Miao and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Data availability
The experimental data generated in this study are provided in the Supplementary Information and Source Data file. No datasets require deposition in public repositories, and no access restrictions apply. are provided with this paper.
Competing interests
The authors declare no competing interests.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1038/s41467-025-63311-7.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1. Mellman, I; Coukos, G; Dranoff, G. Cancer immunotherapy comes of age. Nature; 2011; 480, pp. 480-489.2011Natur.480.480M1:CAS:528:DC%2BC3MXhs1GksLjO [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22193102][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3967235]
2. Chen, DS; Mellman, I. Elements of cancer immunity and the cancer-immune set point. Nature; 2017; 541, pp. 321-330.2017Natur.541.321C1:CAS:528:DC%2BC2sXht1Oksr8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28102259]
3. Sharma, P; Allison, JP. The future of immune checkpoint therapy. Science; 2015; 348, pp. 56-61.2015Sci..348..56S1:CAS:528:DC%2BC2MXls1Wmurg%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25838373]
4. Kazim, M; Yoo, E. Recent advances in the development of non-invasive imaging probes for cancer immunotherapy. Angew. Chem. Int. Ed. Engl.; 2024; 63, e202310694.1:CAS:528:DC%2BB3sXit1Ogt7bM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37843426]
5. Oliveira, G; Wu, CJ. Dynamics and specificities of T cells in cancer immunotherapy. Nat. Rev. Cancer; 2023; 23, pp. 295-316.1:CAS:528:DC%2BB3sXns1yru7Y%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37046001][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10773171]
6. Lu, J et al. Nanomedicine-induced programmed cell death enhances tumor immunotherapy. J. Adv. Res.; 2024; 62, pp. 199-217. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37743016]
7. Rizvi, NA et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science; 2015; 348, pp. 124-128.2015Sci..348.124R1:CAS:528:DC%2BC2MXls1Wmtbg%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25765070][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4993154]
8. Krekorian, M et al. Imaging of T-cells and their responses during anti-cancer immunotherapy. Theranostics; 2019; 9, pp. 7924-7947.1:CAS:528:DC%2BB3cXhtFamsLnK [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31656546][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6814447]
9. Chen, Y; Pei, P; Yang, Y; Zhang, H; Zhang, F. Noninvasive early diagnosis of allograft rejection by a granzyme B protease responsive NIR-II bioimaging nanosensor. Angew. Chem. Int. Ed. Engl.; 2023; 62, 1:CAS:528:DC%2BB3sXpt1Kntbo%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37052894]e202301696.
10. Fu, J et al. Development of granzyme B-targeted smart positron emission tomography probes for monitoring tumor early response to immunotherapy. ACS Nano; 2024; 18, pp. 18910-18921.1:CAS:528:DC%2BB2cXhsFeht7fE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39001856]
11. Larimer, BM et al. Granzyme B PET imaging as a predictive biomarker of immunotherapy response. Cancer Res.; 2017; 77, pp. 2318-2327.1:CAS:528:DC%2BC2sXmvFartrg%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28461564][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5474226]
12. Goggi, JL et al. Granzyme B PET imaging of combined chemotherapy and immune checkpoint inhibitor therapy in colon cancer. Mol. Imaging Biol.; 2021; 23, pp. 714-723.1:CAS:528:DC%2BB3MXhtFKltbrI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33713000][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8410722]
13. He, S; Li, J; Lyu, Y; Huang, J; Pu, K. Near-infrared fluorescent macromolecular reporters for real-time imaging and urinalysis of cancer immunotherapy. J. Am. Chem. Soc.; 2020; 142, pp. 7075-7082.1:CAS:528:DC%2BB3cXlt1CrsbY%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32196318]
14. Wang, R; Deutsch, RJ; Sunassee, ED; Crouch, BT; Ramanujam, N. Adaptive design of fluorescence imaging systems for custom resolution, fields of view, and geometries. BME Front.; 2023; 4, 0005. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37849673][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10521686]
15. Zhang, P et al. Molecular imaging of tumour-associated pathological biomarkers with smart nanoprobe: from “seeing” to “measuring”. Exploration; 2023; 3, 20230070.1:CAS:528:DC%2BB2cXhtVGgs7%2FP [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38264683][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10742208]
16. Chen, L et al. Molecular imaging: design mechanism and bioapplications. Sci. China Chem.; 2023; 66, pp. 1336-1383.2023spri.book...C1:CAS:528:DC%2BB3sXkslyht7k%3D
17. Wang, Y et al. Ultrabright and ultrafast afterglow imaging in vivo via nanoparticles made of trianthracene derivatives. Nat. Biomed. Eng.; 2024; 9, pp. 656-670. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39472533]
18. Ji, Y et al. Near-infrared fluorescence imaging in immunotherapy. Adv. Drug Deliv. Rev.; 2020; 167, pp. 121-134.1:CAS:528:DC%2BB3cXht1Ons7%2FE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32579891][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8450770]
19. Zhong, Y et al. In vivo molecular imaging for immunotherapy using ultra-bright near-infrared-IIb rare-earth nanoparticles. Nat. Biotechnol.; 2019; 37, pp. 1322-1331.1:CAS:528:DC%2BC1MXhvVOqu77M [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31570897][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6858548]
20. Xia, W; Singh, N; Goel, S; Shi, S. Molecular imaging of innate immunity and immunotherapy. Adv. Drug Deliv. Rev.; 2023; 198, 114865.1:CAS:528:DC%2BB3sXhtVKhsrrI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37182699]
21. Nguyen, A et al. Granzyme B nanoreporter for early monitoring of tumor response to immunotherapy. Sci. Adv.; 2020; 6, eabc2777.2020SciA..6.2777N1:CAS:528:DC%2BB3cXitlWitbrI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33008894][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7852386]
22. Patel, MR et al. Longitudinal, noninvasive imaging of T-cell effector function and proliferation in living subjects. Cancer Res.; 2010; 70, pp. 10141-10149.1:CAS:528:DC%2BC3cXhsFGrt7jN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21159636][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3057959]
23. Zhang, Y et al. Activatable polymeric nanoprobe for near-infrared fluorescence and photoacoustic imaging of T lymphocytes. Angew. Chem. Int. Ed. Engl.; 2021; 60, pp. 5921-5927.1:CAS:528:DC%2BB3MXivVWhur8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33305425]
24. Dobrenkov, K et al. Monitoring the efficacy of adoptively transferred prostate cancer-targeted human T lymphocytes with PET and bioluminescence imaging. J. Nucl. Med.; 2008; 49, pp. 1162-1170. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18552144]
25. Jiang, Y et al. Acidity-activatable upconversion afterglow luminescence cocktail nanoparticles for ultrasensitive in vivo imaging. Nat. Commun.; 2024; 15, 2024NatCo.15.2124J1:CAS:528:DC%2BB2cXlsFGqtL4%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38459025][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10923940]2124.
26. Huang, J et al. Molecular radio afterglow probes for cancer radiodynamic theranostics. Nat. Mater.; 2023; 22, pp. 1421-1429.2023NatMa.22.1421H [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37667071]
27. Zhu, J et al. A self-sustaining near-infrared afterglow chemiluminophore for high-contrast activatable imaging. Angew. Chem. Int. Ed. Engl.; 2024; 63, 1:CAS:528:DC%2BB2cXjsVGksb0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38247345]e202318545.
28. Blau, R; Shelef, O; Shabat, D; Satchi-Fainaro, R. Chemiluminescent probes in cancer biology. Nat. Rev. Bioeng.; 2023; 1, pp. 648-664.1:CAS:528:DC%2BB2cXis12gurnN
29. Ke, J et al. Activatable chemiluminescence probe based on four-arm PEG-conjugated-pyropheophorbide-a for in vivo autofluorescence-free imaging of peroxynitrite. Sci. China Chem.; 2025; 68, pp. 2125-2137.1:CAS:528:DC%2BB2MXhslGrurs%3D
30. He, A; Xia, F; Han, D; Yang, Q; Tan, W. Self-assembled amphiphilic NIR-II emissive nano-micelles for imaging-guided photothermal therapy of colorectal cancer. Sci. China Chem.; 2024; 67, pp. 2767-2774.1:CAS:528:DC%2BB2cXhtlyqs7nK
31. Chen, Y; Wang, S; Zhang, F. Near-infrared luminescence high-contrast in vivo biomedical imaging. Nat. Rev. Bioeng.; 2023; 1, pp. 60-78.1:CAS:528:DC%2BB2cXis1yns7vF
32. Tang, Y et al. NIR-II-excited off-on-off fluorescent nanoprobes for sensitive molecular imaging in vivo. Nat. Commun.; 2025; 16, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39747854][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11696168]278.
33. Huang, X et al. Ratiometric optical nanoprobes enable accurate molecular detection and imaging. Chem. Soc. Rev.; 2018; 47, pp. 2873-2920.1:CAS:528:DC%2BC1cXlslSktLY%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29568836][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5926823]
34. Chen, Z et al. Molecular engineering of direct activated NIR-II chemiluminescence platform for In Vivo chemiluminescence-fluorescence duplex imaging. Nat. Commun.; 2025; 16, 1:CAS:528:DC%2BB2cXhtlCksLzN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39747091][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11695737]238.
35. Ge, L et al. A light-activatable theranostic combination for ratiometric hypoxia imaging and oxygen-deprived drug activity enhancement. Nat. Commun.; 2024; 15, 2024NatCo.15.153G1:CAS:528:DC%2BB2cXkt1Skuw%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38167737][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10762052]153.
36. Naczynski, DJ et al. Rare-earth-doped biological composites as in vivo shortwave infrared reporters. Nat. Commun.; 2013; 4, 2013NatCo..4.2199N1:STN:280:DC%2BC3sfislSrsA%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23873342]2199.
37. Pei, P et al. X-ray-activated persistent luminescence nanomaterials for NIR-II imaging. Nat. Nanotechnol.; 2021; 16, pp. 1011-1018.2021NatNa.16.1011P1:CAS:528:DC%2BB3MXht12jt7rE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34112994]
38. Fan, Y et al. Lifetime-engineered NIR-II nanoparticles unlock multiplexed in vivo imaging. Nat. Nanotechnol.; 2018; 13, pp. 941-946.2018NatNa.13.941F1:CAS:528:DC%2BC1cXhsVGhtb7E [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30082923]
39. Wang, T et al. A hybrid erbium(III)-bacteriochlorin near-infrared probe for multiplexed biomedical imaging. Nat. Mater.; 2021; 20, pp. 1571-1578.2021NatMa.20.1571W1:CAS:528:DC%2BB3MXhs1OhtLnL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34326504]
40. Yang, Y et al. Fluorescence-amplified nanocrystals in the second near-infrared window for in vivo real-time dynamic multiplexed imaging. Nat. Nanotechnol.; 2023; 18, pp. 1195-1204.2023NatNa.18.1195Y1:CAS:528:DC%2BB3sXht12hu7rM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37349506]
41. Ding, S; Lu, L; Fan, Y; Zhang, F. Recent progress in NIR-II emitting lanthanide-based nanoparticles and their biological applications. J. Rare Earths; 2020; 38, pp. 451-463.1:CAS:528:DC%2BB3cXit1CisbY%3D
42. Wang, S et al. In vivo high-resolution ratiometric fluorescence imaging of inflammation using NIR-II nanoprobes with 1550 nm emission. Nano Lett.; 2019; 19, pp. 2418-2427.2019NanoL.19.2418W1:CAS:528:DC%2BC1MXltFeru7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30883136]
43. Zhu, X; Zhang, H; Zhang, F. Expanding NIR-II lanthanide toolboxes for improved biomedical imaging and detection. Acc. Mater. Res.; 2023; 4, pp. 536-547.1:CAS:528:DC%2BB3sXpsFWlsbk%3D
44. Zhan, Y et al. Rapid unperturbed-tissue analysis for intraoperative cancer diagnosis using an enzyme-activated NIR-II nanoprobe. Angew. Chem. Int. Ed. Engl.; 2021; 60, pp. 2637-2642.1:CAS:528:DC%2BB3cXisVygu7rJ [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33015947]
45. Ding, L et al. Genetically engineered nanovesicles mobilize synergistic antitumor immunity by ADAR1 silence and PDL1 blockade. Mol. Ther.; 2023; 31, pp. 2489-2506.1:CAS:528:DC%2BB3sXhtVSgs7fK [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37087570][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10422002]
46. Wilhelm, S et al. Analysis of nanoparticle delivery to tumours. Nat. Rev. Mater.; 2016; 1, 16014.2016NatRM..116014W1:CAS:528:DC%2BC2sXhtVert70%3D
47. Galli, F et al. Relevance of immune cell and tumor microenvironment imaging in the new era of immunotherapy. J. Exp. Clin. Cancer Res.; 2020; 39, 89. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32423420][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7236372]
48. Chatila, T; Silverman, L; Miller, R; Geha, R. Mechanisms of T cell activation by the calcium ionophore ionomycin. J. Immunol.; 1989; 143, pp. 1283-1289.1:CAS:528:DyaL1MXltlagtbc%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/2545785]
49. Nurieva, R et al. T-cell tolerance or function is determined by combinatorial costimulatory signals. EMBO J.; 2006; 25, pp. 2623-2633.1:CAS:528:DC%2BD28XltlGksrk%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16724117][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1478197]
50. Peng, J et al. Photosensitizer micelles together with IDO inhibitor enhance cancer photothermal therapy and immunotherapy. Adv. Sci.; 2018; 5, 1700891.
51. Li, K; Tian, H. Development of small-molecule immune checkpoint inhibitors of PD-1/PD-L1 as a new therapeutic strategy for tumour immunotherapy. J. Drug Target.; 2019; 27, pp. 244-256.1:CAS:528:DC%2BC1cXjtVagsLs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29448849]
52. Pudlo, M; Demougeot, C; Girard-Thernier, C. Arginase inhibitors: a rational approach over one century. Med. Res. Rev.; 2017; 37, pp. 475-513.1:CAS:528:DC%2BC2sXmt1als7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27862081]
53. Wu, L; Huang, J; Pu, K; James, TD. Dual-locked spectroscopic probes for sensing and therapy. Nat. Rev. Chem.; 2021; 5, pp. 406-421.1:CAS:528:DC%2BB3MXhvVGgsrjJ [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37118027]
54. Hajian-Tilaki, K. The choice of methods in determining the optimal cut-off value for quantitative diagnostic test evaluation. Stat. Methods Med. Res.; 2018; 27, pp. 2374-2383.3825913 [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28673124]
55. Galon, J et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science; 2006; 313, pp. 1960-1964.2006Sci..313.1960G1:CAS:528:DC%2BD28XhtVSnsrnK [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17008531]
56. Kaneko, M et al. Elevated neutrophil to lymphocyte ratio predicts poor prognosis in advanced colorectal cancer patients receiving oxaliplatin-based chemotherapy. Oncology; 2012; 82, pp. 261-268.1:CAS:528:DC%2BC38Xnt1aqtr8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22538399]
57. Pei, P et al. NIR-II ratiometric lanthanide-dye hybrid nanoprobes doped bioscaffolds for in situ bone repair monitoring. Nano Lett.; 2022; 22, pp. 783-791.2022NanoL.22.783P1:CAS:528:DC%2BB38XltlOiuw%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35005958]
58. Galon, J; Bruni, D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat. Rev. Drug Discov.; 2019; 18, pp. 197-218.1:CAS:528:DC%2BC1MXlslKlsrw%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30610226]
59. Gibney, GT; Weiner, LM; Atkins, MB. Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol.; 2016; 17, pp. e542-e551.1:CAS:528:DC%2BC28XhvF2gsL7M [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27924752][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5702534]
60. Du, S; Yan, J; Xue, Y; Zhong, Y; Dong, Y. Adoptive cell therapy for cancer treatment. Exploration; 2023; 3, 20210058.1:CAS:528:DC%2BB2cXhtVGgsrfP [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37933232][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10624386]
61. Li, X et al. Imaging and monitoring of granzyme B in the immune response. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol.; 2024; 16, 1:CAS:528:DC%2BB3sXhvFOrsL3E [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37715320]e1928.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the "License"). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Non-invasive optical imaging tools for early detecting anti-tumor immune responses are crucial for precision cancer immunotherapy. However, current probes often suffer from low imaging depth, single imaging channel, and inadequate quantification, hindering their in vivo applications. Here we develop a rare-earth-based NIR-II fluorescence ratiometric nanoprobe (DCGA) for in vivo real-time, precise, and non-invasive visualization of granzyme B (GzmB) activity, a key effector in T cell-mediated antitumor immunity, for early prediction of immunotherapy efficacy. The Nd/Er co-doped DCGA nanoprobe features NIR-II dual-emission ratiometric detection with self-calibrated target response signals, addressing challenges like uneven probe distribution and nonspecific signal interference. In vivo NIR-II ratiometric imaging reveals that GzmB activity well correlates with cytotoxic T cell responses and tumor growth, and can effectively distinguish responders from non-responders in both Hepa 1-6 tumor xenograft models and patient-derived xenograft models. Our DCGA probe shows promise for dynamic, real-time, non-invasive molecular imaging of T cell activation in deep tissues, offering effective support for tumor immunotherapy studies, precision medicine, and personalized diagnostics.
Non-invasive optical imaging tools for early detection of anti-tumor immune responses are essential for precision cancer immunotherapy. Here, the authors report a rare-earth-based NIR-II fluorescence ratiometric nanoprobe (DCGA) for in vivo real-time, precise, and non-invasive visualization of granzyme B activity for early prediction of immunotherapy efficacy.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
; Wang, Peiyuan 3
; Ke, Jianmei 1 ; Zhou, Yang 4 ; Wu, Ming 4
; Wei, Zuwu 4 ; Cao, Yanbing 4 ; Li, Hongsheng 4 ; Chen, Geng 4
; Zheng, Guangwei 4 ; Zeng, Yongyi 2 ; Hong, Maochun 5 ; Liu, Xiaolong 4
1 School of Rare Earths, University of Science and Technology of China, Hefei, P. R. China (ROR: https://ror.org/04c4dkn09) (GRID: grid.59053.3a) (ISNI: 0000000121679639); Ganjiang Innovation Academy, Chinese Academy of Sciences, Ganzhou, P. R. China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000 0001 1957 3309); The United Innovation of Mengchao Hepatobiliary Technology Key Laboratory of Fujian Province, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, P. R. China (ROR: https://ror.org/029w49918) (GRID: grid.459778.0)
2 The United Innovation of Mengchao Hepatobiliary Technology Key Laboratory of Fujian Province, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, P. R. China (ROR: https://ror.org/029w49918) (GRID: grid.459778.0); Liver Disease Center, The First Affiliated Hospital of Fujian Medical University, Fuzhou, P. R. China (ROR: https://ror.org/030e09f60) (GRID: grid.412683.a) (ISNI: 0000 0004 1758 0400)
3 The United Innovation of Mengchao Hepatobiliary Technology Key Laboratory of Fujian Province, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, P. R. China (ROR: https://ror.org/029w49918) (GRID: grid.459778.0); State Key Laboratory of Structure Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, P. R. China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309)
4 The United Innovation of Mengchao Hepatobiliary Technology Key Laboratory of Fujian Province, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, P. R. China (ROR: https://ror.org/029w49918) (GRID: grid.459778.0)
5 School of Rare Earths, University of Science and Technology of China, Hefei, P. R. China (ROR: https://ror.org/04c4dkn09) (GRID: grid.59053.3a) (ISNI: 0000000121679639); Ganjiang Innovation Academy, Chinese Academy of Sciences, Ganzhou, P. R. China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000 0001 1957 3309); State Key Laboratory of Structure Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, P. R. China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309)