Radionuclide-labeled oxygen-self-supplying calcium reservoir for high-efficiency radioimmunotherapy via synergistic pyroptosis activation and PD-L1 downregulation Authors: Lin Liu, Kaiyan Huang, Hanjing Guo, Linping He, Xin Wang, Chao Wang, Jianqing Lin, Naihan Huang, Weibing Miao Published: 2026 ABSTRACT AbstractInternal radiation therapy is an irreplaceable modality for local tumor treatment, offering high-effective radiotherapy while sparing surrounding healthy tissues. However, current clinical non-biodegradable radionuclide implants suffer from permanent retention risks and limited efficacy against distal tumors due to inadequate immune activation. Herein, we engineered a biodegradable radionuclide-labeled oxygen-self-supplying calcium reservoir (131I@CaO2), fabricated through biomimetic mineralization followed by chloramine-T mediated 131I-radiolabeling, for high-efficiency internal-radioimmunotherapy. 131I@CaO2 effectively enhances the internalization of 131I to improve internal-radiotherapy and exhibits pH-responsive biodegradability, remaining stable under physiological condition while progressively degrading in tumor microenvironment to release H2O2, O2 and Ca2+. These bioactivators synergistically enhance 131I-induced internal-radioimmunotherapy through both radiosensitization and immune activation: H2O2 and O2 boost 131I-induced ROS generation while Ca2+ disrupted mitochondrial respiration to reduce oxygen consumption and ATP production (suppressing Ca2+ pump activity), creating a positive feedback loop of oxygen supply and ion overload, thereby achieving radiosensitization; Concurrently, O2 alleviates tumor hypoxia to downregulate PD-L1 expression and Ca2+ overload activates dual pyroptosis pathways, establishing a robust immunogenic microenvironment. As a result, 131I@CaO2 effectively eliminates both primary and distant tumors by activating immune cells. This study simultaneously resolves the long-term retention risks of non-biodegradable radionuclide and the dichotomy between local radiotherapy and comprehensive immune modulation.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12951-026-04702-w.Keywords: Immunotherapy, Pyroptosis, Radiotherapy, Hypoxia, Biodegradable Introduction Internal radiation therapy, commonly known as brachytherapy, achieves precise localized tumor ablation through direct implantation of radionuclide (such as 125I seeds or 90Y microspheres), delivering high-effective radiotherapy while minimizing damage to the surrounding healthy tissues [1–3]. Despite its outstanding localized efficacy, current non-biodegradable radionuclide implants in clinical practice pose permanent foreign-body retention risks and demonstrate extremely limited efficacy against distal or recurrent tumors [4, 5]. While studies have shown that radiotherapy-induced release of damage-related molecular patterns (DAMPs) and tumor-specific antigens (TSAs) stimulate dendritic cells (DCs) maturation and cytotoxic T cell activation, this immunogenic response remains suboptimal for distal or recurrent tumor control due to the limited immune activation ability of radiotherapy [6–8]. Consequently, developing biodegradable radionuclide carriers capable of enhancing radiation-induced immune activation represents a crucial need for realizing effective internal-radioimmunotherapy. The immune response caused by tumor treatment is largely determined by the specific mode of cell death [9–11]. Radiotherapy predominantly triggers apoptosis through limited ROS generation and DNA damage, which fails to elicit robust immune activation due to the immunologically silent nature of apoptotic cell clearance via phagocytosis in vivo [12, 13]. In marked contrast, pyroptosis - a highly inflammatory mode of programmed cell death characterized by cellular swelling, plasma membrane rupture, and consequent release of immunostimulatory molecule - potently activates anti-tumor immunity by recruiting and activating immune cells [14, 15]. Among pyroptosis induction strategies, metal ion-overload (such as Na+, K+, Zn2+, Bi3+, and Ca2+) is particularly favored due to its superior immunogenicity [16]. In particular, Ca2+ serve as a pivotal intracellular second messenger, and their dysregulated accumulation triggers profound biochemical cascades [16, 17]. Thus, achieving intracellular Ca²⁺ accumulation could amplify internal-radioimmunotherapy effects by triggering pyroptosis to enhance recruiting and activating immune cells. In addition to mode of cell death, the dynamic change of hypoxic tumor microenvironment (HTME) during radiotherapy represents a critical factor limiting the success of internal-radioimmunotherapy [18–20]. Studies have demonstrated that radiotherapy consumes O2 at the tumor site and damages tumor vasculature, collectively exacerbating local tumor hypoxia and promote the upregulation of hypoxia inducible factor-1α (HIF-α), reducing the sensitivity to radiotherapy [21–23]. Importantly, HIF-1α initiates downstream gene transcription, including the expression of programmed cell death ligand 1 (PD-L1), an immune checkpoint protein whose interaction with programmed cell death protein 1 (PD-1) on CD8+ T cells, resulting in T cell exhaustion and suppressing anti-tumor immune responses [24]. Based on the above discussion, it is urgently needed to develop biodegradable radionuclide carrier capable of releasing Ca2+ and O2 within tumor cells, presenting a promising solution that addresses the drawbacks of long-term retention risks and limited immune activity of internal radiation therapy. CaO2 fulfills this requirement, as it degrades gradually and releases both Ca2+ and O2 [25–27]. Therefore, labeling 131I onto CaO2 would combine radiotherapy with Ca2+ and O2 delivery, offering a potential solution to overcome the permanent retention and insufficient immune activation associated with internal radiation therapy. However, the method for efficiently labeling 131I onto CaO2 and the therapeutic efficacy of this combination against tumors remain to be investigated. Herein, we developed a biodegradable radionuclide-labeled oxygen-self-supplying calcium reservoir (131I@CaO2) for high-efficiency internal-radioimmunotherapy, fabricated through biomimetic mineralization and subsequent chloramine-T mediated 131I radiolabeling approach (Scheme 1a). 131I@CaO2 enhances the internalization of 131I to improve internal-radiotherapy and exhibits pH-responsive biodegradation characteristics, maintaining stability under physiological conditions while undergoing progressively degradation in weakly acidic tumor microenvironments to release H2O2, O2 and Ca2+. These bioactivators synergistically enhance 131I-induced internal-radioimmunotherapy through both radiosensitization and immune activation: H2O2 and O2 boost 131I-induced ROS generation while Ca2+ disrupted mitochondrial respiration to reduce oxygen consumption and ATP production (suppressing Ca2+ pump activity), creating a positive feedback loop of oxygen supply and ion overload, realizing radiosensitization; Concurrently, O2 alleviates tumor hypoxia to downregulate PD-L1 expression and Ca2+ overload activates dual pyroptosis pathways (caspase-1/GSDMD and caspase-3/GSDME), establishing a robust immunogenic microenvironment. As a result, 131I@CaO2 drives DCs maturation, enhances T cells infiltration and proliferation, and establishes immunological memory to effectively eliminate tumors, as well as this effect is further amplified when combined with PD-L1 blockade (Scheme 1b). In a word, this study simultaneously resolves the long-term retention risks of non-biodegradable radionuclide and the dichotomy between local radiotherapy and comprehensive immune modulation. (a) The preparation of biodegradable radionuclide-labeled oxygen-self-supplying calcium reservoir. (b) Mechanism of biodegradable radionuclide-labeled oxygen-self-supplying calcium reservoir inducing pyroptosis and PD-L1 downregulation for tumor immunotherapy Results and discussion The 131I@CaO2 was synthesized through combining biomimetic mineralization and chloramine-T mediated radiolabeling approach (Fig. 1a). Transmission electron microscopy (TEM) characterization revealed that the 131I@CaO2 adopted a well-dispersed spherical morphology (Figs. 1b). Dynamic light scattering (DLS) analysis showed a hydrodynamic size distribution ranging from 50 to 200 nm with a polydispersity index (PDI) of ~ 0.324 (Fig. 1c). X-ray diffraction (XRD) pattern revealed that the diffraction peaks of 131I@CaO2 matched well with the standard CaO2 reference (PDF#085–0514), confirming its dominant CaO2 crystal structure (Fig. 1d). Elemental mapping coupled with energy-dispersive X-ray spectroscopy (EDS) analyses provided compelling evidence for the uniform distribution and successful labeling of 131I (Figs. 1e, f). The result was further corroborated by X-ray photoelectron spectroscopy (XPS) and zeta potential measurements (Figures S1, S2). Furthermore, radio-thin layer chromatography (radio-TLC) analysis also demonstrated the successful labeling of 131I and high labeling efficiency, as evidenced by detection of the negligible free 131I (Figure S3). Meanwhile, the 131I@CaO2 exhibited excellent radio-stability across various physiological media (Fig. 1g). Notably, the 131I@CaO2 showed pH-responsive behavior, maintaining structural integrity at neutral conditions (pH 7.4), but undergoing progressive degradation under weakly acidic conditions (pH 5.5–6.5), with degradation levels showing an inverse correlation with environmental pH (Fig. 1k). When exposed to acidic conditions, CaO2 reacts with H+ to release Ca2+ and H2O2. The H2O2 can further decompose to generate O2 and H2O. Thus, this acid-triggered degradation capability of 131I@CaO2 facilitated the controlled release of therapeutic Ca2+, O2 and H2O2 (Figs. 1h-j). This degradation behavior establishes a unique multimodal therapeutic platform that combines radiotherapy (131I) with calcium overload (Ca2+), hypoxia alleviation (O2) and oxidative stress amplification (H2O2). These well-characterized physicochemical properties and pH-responsive behavior of 131I@CaO2 provide a robust foundation for us to subsequent investigation of intracellular therapeutic mechanisms. The preparation and characterization of 131I@CaO2. (a) Schematic representation for the preparation process of 131I@CaO2. (b) TEM images of 131I@CaO2. (c) Size distribution of 131I@CaO2 measured by DLS. (d) XRD pattern of 131I@CaO2. (e) Element mapping and (f) EDS of 131I@CaO2. (g) Radio-stability of 131I@CaO2 NPs across various physiological media (n = 3). Release of (h) Ca2+, (i) O2 and (j) H2O2 from 131I@CaO2 under different pH (n = 3). (k) TEM of 131I@CaO2 dispersed in aqueous solutions of different pH values Next, we investigated the cellular internalization of 131I@CaO2 in 4T1 cells. Quantitative flow cytometry analysis demonstrated a time-dependent intracellular accumulation of 131I@CaO2 (Fig. 2a). Cellular uptake was also directly quantified by measuring intracellular radioactivity using a gamma counter, which showed significantly higher radioactivity in 131I@CaO2 treated cells compared to free 131I (Fig. 2b), suggesting the potential of 131I@CaO2 to enhance radiation effects. Subsequently, we assayed the intracellular degradation behavior of 131I@CaO2. Following cellular internalization, 131I@CaO2 released H2O2 (quantified using H2O2 assay kit, Fig. 2c), O2 (measured by RDPP probe, Fig. 2d and Figures S4, S5) and Ca2+ (detected by Fluo-4 probe, Figs. 2e, f) within 4T1 cells. Notably, the H2O2 concentration in 131I@CaO2 is lower than that in CaO2. This is attributed to the effect of ionizing radiation from 131I on H2O2 stability. The released components of H2O2, O2 and Ca2+ were expected to disrupt intracellular ion homeostasis, alleviate hypoxia, and elevate ROS generation, thereby potentiating 131I-induced radiotherapeutic efficacy. As expected, the 131I@CaO2 demonstrated stronger inhibition of 4T1 cell proliferation than either free 131I or CaO2 alone (Fig. 2g) and show a synergistic effect (Figure S6). The enhanced radiation damage capacity of 131I@CaO2. (a) Flow cytometry analysis of 4T1 cells treated with FITC-labeled 131I@CaO2. (b) Radioactivity within 4T1 cells treated with free 131I and 131I@CaO2 (n = 3). (c) H2O2 level in 4T1 cells after different treatment (n = 3). (d) Flow cytometry analysis of Intracellular O2 content after different treatment. (e, f) Intracellular Ca2+ content and corresponding CLSM images after different treatment (n = 3, scale bar: 40 μm). (g) 4T1 cell viability after different treatment (n = 5). (h, i) Intracellular ROS content and corresponding CLSM images after different treatment (n = 3, scale bar: 200 μm). (j, k) DNA damage level and corresponding CLSM images after different treatment (n = 3, scale bar: 100 μm). CLSM images of (l) JC-1, (m) cleaved caspase-3 and (n) cleaved PARP in 4T1 cells after different treatment (scale bar: 100 μm). (o) Flow cytometry analysis of apoptosis after different treatment. (p) MR rate, (q) MOC rate, (r) intracellular ATP content and (s) Ca2+ pump activity after different treatment To further elucidate the mechanisms underlying the enhanced cytotoxicity, two main indicators of radiotherapeutic efficacy, ROS level and DNA damage, were assessed. Using the DCFH-DA fluorescent probe (ROS indicator), we observed the elevation of ROS level in 4T1 cells treated with 131I@CaO2 (Figs. 2h, i). The sequential enhancement of ROS across control, free 131I, CaO2, and 131I @CaO2 groups arises from the additive effects of radiation-induced water radiolysis, CaO2-derived H2O2, and Ca2+ overload, with the combination group achieving the highest oxidative stress. Concurrently, immunofluorescence staining of γ-H2AX (a sign of DNA double-strand breakage) showed the increase of DNA damage in the 131I@CaO2 treatment group (Fig. 2j, k), demonstrating the ability of CaO2 to potentiate radiation-induced DNA damage. The CaO2-mediated DNA damage enhancement is related to the intracellular Ca2+ homeostasis imbalance, which profoundly impair cellular physiological processes, especially the change in mitochondrial membrane potential (ΔΨm) and further triggering apoptosis (prevention of DNA repair and enhancement of DNA damage) [17]. Thus, we investigated the process of apoptosis associated with mitochondrial dysfunction. The 5,5′,6,6′-tet-rachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine iodide (JC-1) probe analysis, which emits red fluorescence at normal ΔΨm and green fluorescence when ΔΨm collapses, demonstrated the mitochondria dysfunction in the 4T1 cells treated with 131I@CaO2 (Fig. 2l, Figure S7). Mechanistically, Ca2+ acts as a dual regulator, supporting energy production at moderate levels while triggering ΔΨm loss under overload conditions, which explains the observed mitochondrial dysfunction. This mitochondrial dysfunction triggered caspase-3 activation (Fig. 2m, Figure S8) and subsequent poly adenosinediphosphate-ribose polymerase (PARP, a DNA repair enzymes) cleavage (Fig. 2n, Figure S9), indicating the occurrence of caspase3-dependent apoptosis. The flow cytometry analysis further confirmed significantly apoptosis in the 4T1 cells treated with 131I@CaO2 (Fig. 2o). These results indicate 131I@CaO2 activates apoptosis to enhance the DNA damage. Notably, the apoptosis level induced by 131I@CaO2 was superior to the free 131I or CaO2 alone. In addition, the mitochondria damage induced by 131I@CaO2 also caused the reduction of mitochondrial respiration (MR, Fig. 2p) and mitochondrial oxygen consumption (MOC, Fig. 2q). This further reduce the oxygen consumption within the 4T1 cells, thereby enhance the radiotherapy effect of 131I. The mitochondrial oxygen consumption suggests the damage of mitochondrial respiratory chain, which is related to cellular oxidative phosphorylation. This is reflected in that the ATP content within the 4T1 cells treated with 131I@CaO2 decreased significantly (Fig. 2r). As an energy donor, ATP plays a crucial role in maintaining the activity of ion transporters on the cell membrane. The insufficient ATP supply observably inhibited the activity of the main ion transporters Ca2+ pump (Ca2+ ATPase, Fig. 2s) in 4T1 cells treated with 131I@CaO2, weakening the buffering ability of 4T1 cells against intracellular Ca2+ and potentially enhancing to Ca2+ overload. These results indicate that 131I@CaO2 not only directly delivered H2O2, O2 and Ca2+ to enhance radiation effects, but also disrupted mitochondrial to further create a positive feedback loop that enhances the oxygen production and ions overload. Pyroptosis represents a gasdermin-dependent programmed cell death pathway characterized by plasma membrane permeabilization and subsequent release of cellular content, eliciting potent anti-tumor immune response [28]. Following ¹³¹I@CaO₂ treatment, characteristic pyroptosis features including cellular swelling and large bubble formation were clearly observed (Fig. 3a), suggesting the occurrence of pyroptosis. According to previous reports, the intracellular Ca2+ overload can activate NLRP3 inflammasome and subsequent pyroptosis [29]. The release of Ca2+ from 131I@CaO2 in 4T1 cells effectively induced intracellular Ca2+ overload (Fig. 2e). Thus, the expression of NLRP3 inflammasome and subsequent pyroptosis pathway in 4T1 cells treated with 131I@CaO2 was measured. As shown in Fig. 3b and Figure S10, the 131I@CaO2 treatment significantly upregulated NLRP3 inflammasome expression in 4T1 cells. This NLRP3 inflammasome activation induced the cleavage of caspase-1 (Fig. 3c, Figure S11), which subsequently processed gasdermin-D (GSDMD) to generate its active N-terminal fragment GSDMD-N (execution proteins of pyroptosis, Fig. 3d, Figure S12), thereby confirming that the 131I@CaO2 triggers the caspase-1/GSDMD-dependent pyroptosis. In addition to the canonical caspase-1/GSDMD pathway, pyroptosis can also be initiated through caspase-3-induced cleavage of gasdermin-E (GSDME). Given the established ability of intracellular Ca2+ overload induced by 131I@CaO2 to activate caspase-3 (Fig. 2m), we examined its downstream effects on GSDME processing. CLSM analysis demonstrated significant elevation of GSDME-N levels in the 4T1 cells treated with 131I@CaO2 (Fig. 3e, Figure S13), providing evidence that the 131I@CaO2 simultaneously triggers the non-canonical caspase-3/GSDME-dependent pyroptosis pathway. In addition, the GSDMD-N/GSDME-N oligomerized and then ruptured the cell membrane integrity, as assessed by lactate dehydrogenase (LDH) release (Fig. 3h). Simultaneously, the 131I@CaO2 NPs induced the release of the inflammatory factor interleukin-1 (IL-1, Fig. 3i) and interleukin-18 (IL-18, Fig. 3j), which was associated with pyroptosis. These results confirm that the 131I@CaO2 can effectively induce pyroptosis. In addition to pyroptosis induction, the hypoxia alleviation is anticipated to enhance anti-tumor immunity by downregulating HIF-1α and PD-L1 expression. Given the confirmed O2-releasing capacity and the reduction in mitochondrial oxygen consumption induced by 131I@CaO2 in 4T1 cells (Fig. 2d and q), we investigated their regulatory effects on HIF-1α and PD-L1 expression under hypoxic conditions. The O2-generating capacity of 131I@CaO2 led to significant downregulation of HIF-1α expression (Fig. 3f, Figure S14), which consequently suppressed PD-L1 expression (Fig. 3g, Figure S15). This reduction in PD-L1 could reverse CD8⁺ T cell exhaustion, potentially restoring their cytotoxic function and improving tumor immune surveillance [21]. The pyroptosis induction and hypoxia alleviation mediated by 131I@CaO2. (a) The morphology of 4T1 cells after different treatment (scale bar: 100 and 20 μm). Western blotting analysis of (b) NLRP3, (c) cleaved caspase-1 and (d) GSDMD-N in 4T1 cells after different treatment. (e) CLSM images of GSDME-N in 4T1 cells after different treatment (scale bar: 100 μm). Western blotting analysis of (f) HIF-α and (g) PD-L1 in 4T1 cells after different treatment. The release of (h) LDH, (i) IL-1β and (j) IL-18 in 4T1 cells after different treatment (n = 5) During pyroptosis, dendritic cells (DCs) are activated by the released cellular content, specifically adenosine triphosphate (ATP, a chemoattractant for antigen-presenting cells), high mobility group box 1 (HMGB1, promoting the combination of DCs and dying tumor cells) and calreticulin (CRT, stimulating the phagocytosis of DCs) [30]. Thus, we examined these damage-associated molecular patterns (DAMPs) from 4T1 cells treated with 131I@CaO2. Immunofluorescence analysis revealed significantly enhanced calreticulin (CRT) exposure in 131I@CaO2-treated 4T1 cells, as evidenced by intensified green fluorescence signal (Fig. 4a, Figure S16). Concurrently, quantitative assays demonstrated markedly increased extracellular release of ATP (Fig. 4b) and HMGB1 (Fig. 4c) from 131I@CaO2 NP-treated 4T1 cells. Subsequently, we evaluated DC maturation in vitro using a transwell co-culture system (Fig. 4d). Flow cytometry analysis demonstrated that 131I@CaO2-treated 4T1 cells significantly enhanced DCs maturation, as evidenced by elevated expression of co-stimulatory molecules CD80 and CD86 (Fig. 4e, Figure S17). To further characterize DCs activation, we quantified cytokine secretion profiles using enzyme-linked immunosorbent assay (ELISA). When DCs were co-cultured with 131I@CaO2-pretreated 4T1 cells, we observed marked increases in key immunostimulatory cytokines, including interleukin-12 (IL-12, Fig. 4f), interleukin-6 (IL-6, Fig. 4g), and tumor necrosis factor-α (TNF-α, Fig. 4h) - all critical mediators of T cell priming and anti-tumor immunity [31]. These results suggest that 131I@CaO2 can activate immune response. The 131I@CaO2-induced DCs maturation. (a) CLSM images of CRT exposure in 4T1 cells after different treatment (scale bar: 50 μm). The extracellular levels of (b) ATP and (c) HMGB1 after different treated-4T1 cells (n = 5). (d) The diagram illustrating DCs maturation process using a transwell co-culture system. (e) Flow cytometry analysis of DCs maturation after different treatment in vitro. The release of (f) IL-12, (g) IL-6 and (h) TNF-α from DCs co-cultured with different-pretreated 4T1 cells (n = 5) in vitro Given the promising in vitro results demonstrating the 131I@CaO2-induced effective cell killing and immune activation, we next evaluated its anti-tumor performance and immune response efficacy in vivo. BALB/c mice were inoculated with 4T1 cells and randomly divided into 4 groups (tumor sizes are about 70 mm3, n = 5), followed by therapy and safety assessment (Fig. 5a). Following treatment with 131I@CaO2, tumor growth in mice was significantly inhibited compared with other groups, as demonstrated by reductions in both tumor weight and volume (Figs. 5b-e). Hematoxylin and eosin (H&E) staining of tumor slides also validated the efficient anti-tumor efficacy of 131I@CaO2 NPs, revealing extensive tumor cell destruction (Fig. 5f). Furthermore, the immunofluorescence slides exhibited the significant up-regulation of GSDMD-N (green fluorescence, Fig. 5g, Figure S18) and GSDME-N (red fluorescence, Fig. 5h, Figure S19), confirming the occurrence of pyroptosis in tumor. Concurrently, we observed significant down-regulation of HIF-α (green fluorescence, Fig. 5i, Figure S20) and PD-L1 (green fluorescence, Fig. 5j, Figure S21), verifying hypoxia amelioration in tumor. The occurrence of pyroptosis and the improvement of hypoxia further verified the anti-tumor ability of 131I@CaO2. Whereafter, we monitored mice survival outcome following tumor inoculation and observed significantly prolonged survival in 131I@CaO2-treated mice compared to control groups (Fig. 5k). Next, the biodistribution profiles of 131I@CaO2 in 4T1 tumor-bearing mice were acquired by ex vivo gamma counting of various organ tissues and metabolites at different time points (Figure S22). Due to local intratumoral injection, 131I@CaO2 was primarily retained at the tumor site. As degradation proceeded, the radiation dose in the tumor decreased, and radioactive signal was detected in metabolic samples, confirming that 131I@CaO2 can gradually be eliminated from the body. Throughout the treatment period, no significant changes in mouse body weight were observed (Fig. 5l). In addition, no significant changes were observed in major organs post-treatment (Fig. 5m and Figure S23). The blood biochemical analysis revealed normal levels of hepatic function markers (Fig. 5n), including alanine aminotransferase (ALT), aspartate aminotransferase (AST) and alkaline phosphatase (ALP), and renal function markers (Figure S24), including blood urea nitrogen (BUN) and creatinine (CREA), indicating preserved liver and kidney function. These results indicate the effective anti-tumor effect and biosafety of 131I@CaO2. The anti-tumor performance and biosafety of 131I@CaO2 in unilateral 4T1 tumor-bearing mice. (a) Schematic of the treatment procedure in unilateral 4T1 tumor-bearing mice. (b, e) Volumes, (c) images (scale bar: 1 cm) and (d) weight of tumor tissues from 4T1 tumor-bearing mice after different treatment (n = 5). (f) H&E slides of tumor tissues from 4T1 tumor-bearing mice after different treatment. Immunofluorescence staining of (g) GSDMD-N, (h) GSDME-N, (i) HIF-α and (j) PD-L1 in tumor tissues from 4T1 tumor-bearing mice after different treatment. (k) Mice survival rate after different treatment. (l) Mice body weight after different treatment (n = 5). (m) Mice major organ index after different treatment (n = 5). (n) Mice hepatic function markers after different treatment (n = 3) Following confirmation of the anti-tumor efficacy, we systematically evaluated the immunostimulatory effects of 131I@CaO2 by assessing the DCs maturation in tumor-draining lymph nodes, the CD4+ T and CD8+ T cell proliferation in spleen and infiltration in tumor (Fig. 6a). The 131I@CaO2 treatment significantly enhanced DCs maturation in tumor-draining lymph nodes (Fig. 6b). These activated DCs subsequently promoted both proliferation in spleen and infiltration in tumor of T lymphocytes, as evidenced by significantly increased CD4⁺ and CD8⁺ T cell populations in spleen (Figures S25, S26) and tumor tissue (Fig. 6c). Among CD4⁺ T cells, regulatory T cells (Tregs) are known to suppress immune activation [32]. We therefore quantified Treg proportions within the CD4⁺ T cell population and found that 131I@CaO2 NPs NP treatment significantly reduced tumor-infiltrating Treg frequencies (Fig. 6d). Granzyme B (GrzB) serves as a key cytotoxic effector molecule during CD8⁺ T cell-mediated tumor cell killing [33]. Quantitative analysis revealed that 131I@CaO2 treatment significantly increased the frequency of GrzB⁺CD8⁺ T cells (Fig. 6e). This increase is attributed to the 131I@CaO2-induced pyroptosis, which releases DAMPs to promote DC maturation and CD8+ T cell activation, while the alleviation of hypoxia downregulates PD-L1 expression to reverse T cell exhaustion, collectively enhancing the cytotoxic function of CD8+ T cells. Taken together, these results clearly illustrate that the 131I@CaO2 treatment triggered a strong immune response for effective antitumor. The immune response efficacy of 131I@CaO2 in unilateral 4T1 tumor-bearing mice. (a) Schematic of the immune assessment procedure in unilateral 4T1 tumor-bearing mice. Flow cytometry analysis of (b) DCs maturation in tumor-draining lymph nodes from 4T1 tumor-bearing mice after different treatment (n = 3). (c) CD4+ T and CD8+ T cell proportion in tumor from 4T1 tumor-bearing mice after different treatment (n = 3). (d) Tregs proportion and (e) GrzB+CD8⁺ T cells proportion in tumor from 4T1 tumor-bearing mice after different treatment (n = 3) Inspired by its strong immune activation observed in the unilateral model, we then performed the bilateral tumor experiment to further understand whether 131I@CaO2 could achieve an abscopal effect. In brief, 4T1 tumor-bearing BALB/c mice (primary tumors: right back, ~ 70 mm3) received contralateral tumor inoculation (left back) to establish distal metastases and randomly divided into 2 groups (n = 5), followed by treatment and immune assessment (Fig. 7a). As demonstrated in Figs. 7b-f, control group mice exhibited progressive growth of both primary and distal tumors. Remarkably, 131I@CaO2 treatment of right back tumor mediated significant regression of primary tumor while concurrently suppressing distal tumor growth. The H&E staining of tumor slides also confirmed the efficient anti-tumor effect of 131I@CaO2 (Figure S27). These results demonstrate that 131I@CaO2 stimulate immune response in mice, producing both direct tumor regression at the treatment site and abscopal effects at distal tumor sites. To elucidate the mechanism of distal tumor inhibition without direct treatment of 131I@CaO2, we performed immune profiling of primary and distal tumor tissue using flow cytometry and immunohistochemistry. After treated with 131I@CaO2, we observed DCs maturation (Fig. 7g) and T lymphocyte infiltration (Fig. 7h) not only in primary tumor site, but more importantly, in distant tumor sites, demonstrating systemic immune activation beyond the treatment location. Furthermore, the proliferation of T lymphocytes in spleen (Figures S28) also plays important role in the inhibition of distal tumor. The results establish that local 131I@CaO2 treatment initiates systemic immune activation. Following therapy-induced tumor cell death, released tumor-associated antigens are captured by DCs. These antigen-loaded DCs subsequently migrate to lymph nodes where they activate T cells through antigen presentation, ultimately driving T cell infiltration and immune-mediated suppression of distant tumors. The anti-tumor and immune response efficacy of 131I@CaO2 in bilateral 4T1 tumor-bearing mice. (a) Schematic of the treatment and immune assessment procedure in bilateral 4T1 tumor-bearing mice. (b) Images (scale bar: 1 cm), (c) weight and (d-f) volume of tumor tissue from 4T1 tumor-bearing mice after different treatment (n = 5). Flow cytometry analysis of (g) DCs maturation in tumor-draining lymph nodes from 4T1 tumor-bearing mice after different treatment (n = 3) and (h) CD4+ T and CD8+ T cell proportion in tumor from 4T1 tumor-bearing mice after different treatment (n = 3) Given that 131I@CaO2 effectively suppressed both primary and distant tumors, and that aPD-L1 is known to enhance antitumor immunity, we tested whether combining the two could further improve therapeutic outcomes and immune memory effects. Mice were therefore subcutaneously reinoculated with 4T1 cells after treatment with 131I@CaO2, aPD-L1 or their combination (Fig. 8a). The results revealed striking differences in reinoculated tumor growth between treatment groups (Figs. 8b-d). Control and aPD-L1 groups exhibited rapid growth of the reinoculated tumor, whereas both 131I@CaO2 monotherapy and combination therapy (131I@CaO2 + α-PD-L1) showed more effective suppression of the reinoculated tumor growth, suggesting the induction of immune memory. Notably, the combined treatment showed a more significant efficacy compared to 131I@CaO2 group. Flow cytometry analysis of tumor tissue demonstrated markedly enhanced T cell infiltration following treatment, with the combination group showing the most pronounced effect (Fig. 8e and Figure S29). Furthermore, spleen tissue analysis revealed a substantial expansion of effector memory T cell populations in both 131I@CaO2 and combination treatment groups, with the combination group again showing the most pronounced effect (Fig. 8f). In addition, the levels of IFN-γ and TNF-α (Figure S30) were significantly elevated after treatment for a period of time, indicating that functional memory T cells had been primed after the initial treatment. These findings collectively demonstrate that 131I@CaO2 not only induce robust and durable immunological memory capable of preventing tumor recurrence but also synergize effectively with PD-L1 blockade to amplify antitumor immunity. Immune memory effects and synergistic potential with PD-L1 blockade of 131I@CaO2 in 4T1 tumor-bearing mice. (a) Schematic of the treatment and immune assessment procedure in 4T1 tumor-bearing mice. (b) Images, (c) weight and (d) volume of reinoculated tumor tissue from 4T1 tumor-bearing mice after different treatment (n = 5). Flow cytometry analysis of (e) CD4+ T and CD8+ T cell proportion in reinoculated tumor from 4T1 tumor-bearing mice after different treatment (n = 3). (f) Effector memory T cell populations in spleen from 4T1 tumor-bearing mice after different treatment (n = 3) Conclusion In summary, we developed a biodegradable radionuclide-labeled oxygen-self-supplying calcium reservoir (131I@CaO2) for high-efficiency internal-radioimmunotherapy. 131I@CaO2 shows pH-responsive degradation, maintaining stability under physiological conditions while undergoing progressively in weakly acidic tumor microenvironments to release H2O2, O2 and Ca2+. These bioactivators synergistically enhance 131I-induced internal-radioimmunotherapy through both radiosensitization and immune activation: H2O2 and O2 boost 131I-induced ROS generation while Ca2+ disrupted mitochondrial respiration to reduce oxygen consumption and ATP production (suppressing Ca2+ pump activity), creating a positive feedback loop of oxygen supply and ion overload, realizing radiosensitization; Concurrently, O2 alleviates tumor hypoxia to downregulate PD-L1 expression and Ca2+ overload activates dual pyroptosis pathways (caspase-1/GSDMD and caspase-3/GSDME), establishing a robust immunogenic microenvironment. As a result, 131I@CaO2 drives DCs maturation, enhances CD4+/CD8+ T cells infiltration and proliferation, and establishes immune environment to effectively eliminate both primary and distant tumors while suppressing recurrence. Our approach overcomes key limitations of permanent retention risks and severely limited efficacy against distal or recurrent tumors in internal radiation therapy by combining biodegradable carriers with enhanced immune activation. Nevertheless, the applicability of this approach to other tumors requires further investigation. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 : Supplementary figures. Acknowledgements This work is financially supported by the National Natural Science Foundation of China (No. 82472012), the Fujian Provincial Clinical Key Specialty Construction Project (2023SZDZK-HYXK), the Science and Technology Innovation Joint Fund Project of Fujian Province (2023Y9249, 2025Y9164) and the Natural Science Foundation of Fujian Province (2025J08172). Author contributions L. L. and KY. H. designed and performed the experiments, and wrote the paper with support from JQ. L., NH. H. and WB. M. All authors contributed to the general discussion. Funding This work is financially supported by the National Natural Science Foundation of China (No. 82472012), the Fujian Provincial Clinical Key Specialty Construction Project (2023SZDZK-HYXK), the Science and Technology Innovation Joint Fund Project of Fujian Province (2023Y9249, 2025Y9164) and the Natural Science Foundation of Fujian Province (2025J08172). Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Lin Liu and Kaiyan Huang contributed equally to this work. Contributor Information Kaiyan Huang, Email: kaiyan_huang@126.com. Jianqing Lin, Email: dr_jianqinglin@163.com. Naihan Huang, Email: hnh@fjmu.edu.cn. Weibing Miao, Email: miaoweibing@126.com. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.