1. Introduction

Immune checkpoint inhibitors (ICIs), especially PD-1/PD-L1 inhibitors, have revolutionized cancer treatment in recent years^[1]. These therapies have demonstrated remarkable clinical efficacy in various advanced malignant tumors by mitigating the immunosuppressive effects of tumors on T cells. Although immunotherapy has made significant progress, it still faces many unresolved challenges. For example, only 15% to 30% of patients with solid tumors respond to immunotherapy^[2,3]. Many patients either exhibit primary resistance from the outset or rapidly develop acquired resistance after a brief period of efficacy. The low response rate has hindered the widespread adoption of immunotherapy, so researchers are investigating why tumors develop resistance to it. This limited response rate varies significantly across cancer types, with higher responsiveness observed in melanoma and non-small cell lung cancer (NSCLC), while tumors such as pancreatic cancer, prostate cancer, and colorectal cancer generally exhibit strong primary resistance to PD-1/PD-L1 blockade^[4-9]. These differences highlight the heterogeneity of PD-L1 resistance across clinical settings and emphasize the need for tumor-specific therapeutic strategies.

Accumulating evidence suggests that resistance to ICIs cannot be fully explained by tumor-intrinsic genetic alterations alone. Increasing attention has therefore been directed toward the tumor microenvironment (TME), which plays a central role in shaping antitumor immune responses^[10]. The TME is usually divided into three types based on how well immune cells can get into the tumor: hot tumors, cold tumors, and immune-rejecting tumors. “Hot” tumors are more likely to respond to ICIs than “cold” tumors and “immune-rejecting” tumors^[11].

Furthermore, regarding the TME, traditional treatments like chemotherapy and radiotherapy (RT) have their own problems when used with immunotherapy. A moderate dose of chemotherapy drugs can kill tumor cells and release “danger signals” (damage-associated molecular patterns, DAMPs), thereby “activating” the immune system and helping it recognize tumor antigens. On the other hand, radiation therapy utilizes the cGAS-STING pathway to convert DNA damage into an immune signal, thereby triggering an immune response^[12]. However, studies have shown that chemotherapy drugs may enhance the expression of PD-L1 on the surface of tumor cells by activating inflammation-related signaling pathways, including NF-κB. This may help tumor cells protect themselves from attack by the immune system^[13]. Radiation therapy also increases PD-L1 expression by releasing inflammatory mediators such as IFN-γ. This triggers feedback defense mechanisms involving immune checkpoints, leading to reduced efficacy of immunotherapy with long-term use^[14,15].

Although both chemotherapy and RT can enhance the function of the immune system, when used in combination with immunotherapy, they often encounter difficulties in clinical application due to the mismatch between the timing and site of treatment. Nanotechnology has proven in recent years to be able to solve the problem of synergistic therapy regimens that were previously impossible. By developing nanoplatforms that respond to their environment, researchers are able to precisely control when and where drugs are released. This significantly enhances the synergistic effects of different types of treatments. Additionally, controllable nanocarriers can release ICIs at the appropriate time, such as when chemotherapy and radiation therapy cause PD-L1 upregulation to compensate for the loss of immune cells. This precise control over time and space creates a perfect synergy between antigen release, immune system activation, and therapeutic intervention. This review summarizes nanomedicine strategies based on TME types and explores the potential benefits and challenges of combining these strategies with chemotherapy, radiation therapy, and immunotherapy in overcoming PD-L1 resistance and enhancing systemic immune responses (Figure 1).

Display Full Size

Download

Figure 1. A schematic diagram illustrating common nanomaterials and their applications in cancer treatment, including RT, chemotherapy, and nanotherapy. Created in BioRender. Lutf, B. (2026) https://BioRender.com/ixmdm2i. RT: radiotherapy; NP: nanoparticle; IFN: interferon; TGF: transforming growth factor; TAMs: tumor-associated macrophages; DC: dendritic cell.

2. Molecular Mechanisms of PD-L1 Resistance and Heterogeneity of TME

PD-1/PD-L1 immune checkpoint blockade (ICB) has emerged as a standard therapeutic strategy for multiple malignancies. However, substantial molecular differences between “hot” and “cold” tumors critically determine immunotherapy responsiveness^[11]. Elucidating the regulatory networks underlying PD-L1 expression across distinct TME contexts is therefore essential for understanding immune escape and guiding therapeutic design.

2.1 Molecular mechanisms regulating PD-L1 expression

PD-L1 expression is regulated through both intrinsic genetic programs and adaptive signaling responses. Intrinsic PD-L1 expression is mainly driven by oncogenic and epigenetic alterations, independent of exogenous stimuli, and is closely associated with the baseline immune phenotype of the TME. For example, PTEN loss or MYC amplification can sustain PD-L1 expression via constitutive oncogenic signaling^[16-19].

In contrast, adaptive PD-L1 expression is induced by external stressors such as immune pressure and therapeutic interventions, including chemoradiotherapy (CRT). This reflects a dynamic “adaptive immune resistance” mechanism that enhances tumor immune evasion under treatment stress^[14,20-23].

2.1.1 NF-κB in transcriptional regulation of PD-L1

NF-κB is a central transcription factor in inflammatory and stress responses and is considered a key positive regulatory hub for PD-L1 expression in tumors^[13,24]. It can directly bind to the PD-L1 promoter to promote transcription and can also indirectly enhance its expression through multiple signaling networks, thereby driving immune escape within the TME^[13].

In tumor cells, NF-κB integrates oncogenic and inflammatory signaling pathways, positioning it as a central node linking tumor progression and immune evasion. Chemotherapy-induced stress can activate NF-κB p65, thereby promoting PD-L1 transcriptional upregulation. This process is closely associated with the DNA damage response (DDR) signaling pathway and checkpoint regulation and may vary depending on the genetic background of tumors^[13,25,26].

Furthermore, the DDR serves as a key upstream mechanism linking CRT to the NF-κB/PD-L1 axis. The ataxia telangiectasia mutated/ataxia telangiectasia and RAD3-related signaling pathway can activate NF-κB and the STAT1/STAT3/IRF1 pathways, synergistically enhancing PD-L1 expression^[15,27]. Additionally, NF-κB can participate in DNA repair processes and enhance tumor cell survival, further promoting treatment resistance^[28].

2.1.2 The cGAS-STING-IFN pathway

The cGAS-STING pathway is a critical innate immune signaling axis that mediates DNA damage sensing and plays a key role in RT-induced antitumor immune activation (Figure 2)^[29,30]. DNA double-strand breaks (DSBs) generated by RT lead to cytoplasmic dsDNA accumulation, which is recognized by cGAS and triggers the STING-TBK1 cascade, subsequently inducing IRF3 and NF-κB activation and promoting the secretion of type I interferons and various pro-inflammatory cytokines^[20].

Display Full Size

Download

Figure 2. Schematic illustration of RT-induced cGAS-STING signaling and immune activation in the tumor microenvironment. RT induces DNA DSBs, leading to micronuclei formation and subsequent micronuclear rupture, which releases dsDNA into the cytoplasm. Cytosolic dsDNA activates the cGAS-STING-TBK1 signaling pathway, promoting IRF3 and NF-κB activation and inducing type I interferon production. Type I interferon signaling further activates the JAK1/2-STAT1/2/3-IRF1 axis, contributing to PD-L1 upregulation in the tumor microenvironment. These events enhance DC activation, CXCL9/CXCL10-mediated CXCR3⁺ CD8⁺ T-cell recruitment, vascular permeability, immune-cell infiltration, and the response to immune checkpoint therapy. In contrast, high-dose radiation may induce TREX1 overexpression, resulting in cytosolic dsDNA degradation and inhibition of STING signaling. Created in BioRender. Lutf, B. (2026) https://BioRender.com/qx3l09z. RT: radiotherapy; cGAS: cyclic GMP-AMP synthase; STING: stimulator of interferon genes; DSBs: double-strand breaks; dsDNA: double-stranded DNA; TBK: TANK-binding kinase; IRF: interferon regulatory factor; NF-κB: nuclear factor kappa B; STAT: signal transducer and activator of transcription; DC: dendritic cell; CXCL: C-X-C motif chemokine ligand; CXCR: C-X-C motif chemokine receptor; TREX: three-prime repair exonuclease.

This pathway can exert context-dependent effects within the TME. It may enhance antitumor immunity by promoting antigen presentation and dendritic cell (DC) activation, while also contributing to adaptive immune resistance under sustained activation conditions^[30-34]. The downstream IFN-JAK-STAT axis further links STING activation to PD-L1 transcription. STAT1/STAT2/STAT3 activation via JAK1/2 induces IRF1-mediated PD-L1 promoter binding, thereby enhancing PD-L1 expression^[35,36].

2.1.3 The HIF-1α pathway: PD-L1 regulation in the hypoxic microenvironment

HIF-1α is a master regulator of cellular adaptation to hypoxia and is widely activated in solid tumors^[37,38]. In the hypoxic TME, HIF-1α not only drives metabolic and angiogenic reprogramming but also contributes to immune evasion.

Specifically, HIF-1α can directly promote PD-L1 transcription and enhance its expression in both tumor and immune cells, establishing a hypoxia-driven immunosuppressive program^[39-41]. In addition, HIF-1α regulates angiogenic factors and metabolic pathways that collectively reshape the immune microenvironment, leading to reduced T cell and NK cell activity and facilitating tumor immune escape^[42,43]. Hypoxia is also closely associated with therapy resistance and may create a feedback loop that further stabilizes HIF-1α signaling^[44,45].

2.1.4 Integrative analysis of shared and context-specific mechanisms

The regulatory networks governing PD-L1 expression exhibit strong context dependency across different TME states. The NF-κB and JAK-STAT pathways function as shared signaling hubs activated by both chemotherapy and RT^[13,36]. The cGAS-STING pathway is primarily associated with DNA damage induced by RT, while HIF-1α signaling predominates under hypoxic conditions^[30,39,46].

Given this complexity, a single therapeutic strategy is unlikely to effectively overcome PD-L1-mediated immune escape across all tumor contexts. Instead, context-specific intervention strategies targeting NF-κB, JAK-STAT, or HIF-1α signaling may provide more effective immunotherapeutic outcomes^[47-50].

2.2 PD-L1 resistance mechanisms based on TME subtypes

Immunophenotyping of the TME represents a significant advancement in recent immuno-oncology research. Immune cell infiltration and stromal characteristics profoundly influence tumor immune escape mechanisms, treatment response, and drug resistance^[11]. With the ongoing advancement of tumor immunotherapy, identifying tumors with distinct immune subtypes not only helps predict treatment outcomes but also provides a theoretical basis for designing personalized treatment strategies^[51-53].

The distinct characteristics of these immune subtypes influence tumor response to ICIs therapy and the development of resistance mechanisms. In clinical practice, treatment strategies should be tailored to the specific tumor immune subtype. Table 1 below summarizes the immune features, core resistance mechanisms, and ICB response predictions for each TME immune subtype.

2.2.1 Hot tumors: Adaptive immune resistance

Hot tumors (inflammatory type) typically exhibit high levels of immune cell infiltration and PD-L1 expression, and thus often demonstrate favorable clinical responses during the initial stages of ICIs therapy^[11]. However, as treatment continues, metabolic inhibitors (such as lactate and adenosine) accumulate in the TME, gradually hindering immune activity. Research indicates that high concentrations of tumor-derived lactic acid acidify the microenvironment and suppress effector T cell function, thereby correlating closely with resistance to anti-PD-1/PD-L1 therapy^[59-61]. Meanwhile, adenosine acts as a key immunosuppressive factor in inflammatory tumors, weakening antitumor immune responses through the A2A/A2B receptor signaling pathway. This may also counteract the fundamental immune advantage of hot tumors^[62].

2.2.2 Cold tumors: Multidimensional immunosuppression

In contrast, “cold tumors” are typically in a state of immune silencing or an “immune desert”, meaning that immune cells rarely penetrate the tumor, antigens are not effectively presented, and the immune response is weak^[11,63]. This immunosuppressive effect is particularly pronounced in certain types of tumors, including pancreatic cancer and prostate cancer. Studies have shown that restoring the function of DCs is crucial for activating potent anti-tumor immunity. Traditional chemotherapy and radiation therapy promote the release of tumor antigens and the activation of immune cells by inducing immunogenic cell death (ICD) and releasing DAMPs^[64-66]. Therefore, enhancing the recruitment and activation of immune cells is considered key to improving the efficacy of immunotherapy for cold tumors.

2.2.3 Immune-excluded tumors: The dual challenge of physical barriers and immunosuppression

Immune-excluded tumors typically present with dense extracellular matrix (ECM) deposition and elevated interstitial fluid pressure (IFP), thereby restricting immune cell and therapeutic drug penetration into the tumor parenchyma, thereby restricting immune cell and therapeutic drug penetration into the tumor parenchyma^[67-70]. Among these, cancer-associated fibroblasts (CAFs) are important contributors to this barrier formation; they not only remodel and deposit the ECM but also secrete immunosuppressive factors, inhibit CD8⁺ T cell infiltration, and promote tumor resistance to ICB therapy^[67]. Therefore, for immune-excluded tumors, strategies such as degrading or remodeling the ECM, reducing stromal barriers, modulating abnormal vasculature, blocking TGF-β signaling pathways, and achieving targeted drug delivery can promote T cell re-entry into the tumor parenchyma and thereby improve immunotherapy efficacy^[71,72].

2.3 Exosome PD-L1 (Exo-PD-L1)

Recent studies have shown that PD-L1 not only exists on the surface of tumor cells, but can also be released into the bloodstream via exosomes. Exosomes are tiny vesicles secreted by cells that encapsulate a variety of molecules derived from their parent cells, including PD-L1. The PD-L1 on these exosomes (i.e., Exo-PD-L1) is identical to the PD-L1 on the surface of tumor cells and can travel throughout the body via the bloodstream^[73,74]. This mechanism enables tumors to exert immunosuppressive effects within the local immune microenvironment, thereby influencing the systemic immune system.

The immune evasion mechanism of exosomal PD-L1 manifests primarily in two aspects: First, Exo-PD-L1 acts as a “decoy” by binding to PD-1 on the surface of peripheral T cells, prematurely inducing T cell exhaustion. This prevents effector immune cells from effectively responding to the tumor^[75,76]. Second, the secretion of Exo-PD-L1 diminishes the efficacy of anti-PD-L1 antibodies at the tumor site, as it spreads through systemic circulation to induce systemic immune suppression^[77,78].

Research indicates that the level of PD-L1 in exosomes is closely correlated with the degree of tumor immune escape. Consequently, targeted therapy against PD-L1 in exosomes has emerged as a novel direction in current immunotherapy^[74]. By downregulating the secretion of Exo-PD-L1 or intercepting its interaction with T cells, it is anticipated that the efficacy of ICIs therapy can be enhanced^[79].

2.4 MHC-I downregulation

Antigen presentation is fundamental for T cells to recognize and attack tumor cells, and effective immune killing depends on tumor cells presenting antigens to CD8⁺ T cells via MHC-I molecules. However, many tumor cells evade T cell recognition by downregulating MHC-I expression through genetic mutations or epigenetic modifications. MHC-I downregulation leads to APD, in which tumor cells cannot effectively display their surface antigens, preventing T cells from recognizing and killing them^[80,81].

This process is particularly pronounced in “cold tumors” and certain tumors that inhibit the infiltration of immune cells. In cold tumors, the expression of MHC-I on the surface of tumor cells is significantly reduced, thereby inhibiting the infiltration of T cells into the tumor^[82]. For example, epigenetic modifications such as B2M gene mutations or DNA methylation have been shown to be important causes of MHC-I downregulation^[83]. Recent experimental data indicate that restoring MHC-I expression can enhance the immune response of tumors and improve the efficacy of immunotherapy^[84].

In clinical practice, ICIs, such as anti-PD-1 and PD-L1 antibodies, are important treatment strategies for various tumors. However, in tumors with downregulated MHC I, the efficacy of ICIs is often suppressed because the absence of MHC I leads to impaired antigen presentation and weakens CD8⁺ T cell-mediated immune responses^[85,86]. Based on this mechanism, researchers are working to restore MHC I expression through epigenetic modifications or drug regulation. For example, some small molecule drugs or epigenetic interventions can reverse MHC I downregulation and improve the sensitivity of tumors to immunotherapy^[85]. In addition, a strategy of targeting and demethylating MHC I antigen-presenting transcription factors (such as NLRC5) has increased MHC I expression in animal models and enhanced the efficacy of anti-PD-1 therapy, suggesting that this approach has potential clinical application value^[87].

3. The Immunomodulatory Role of Chemotherapy in TME Remodeling

Chemotherapy has long been regarded as a cytotoxic strategy that inhibits tumor proliferation^[88]. However, accumulating evidence demonstrates that its role extends beyond direct tumor killing, as it can also remodel the TME and modulate host immune responses. In particular, chemotherapy-induced ICD serves as a key mechanism linking tumor cell death to immune activation. However, chemotherapy can simultaneously induce adaptive immune resistance, including PD-L1 upregulation through NF-κB-dependent signaling networks (as discussed in Section 2), thereby complicating the therapeutic synergy between chemotherapy and immunotherapy^[89,90].

3.1 Dual immunomodulatory effects of chemotherapy

Although chemotherapy can activate robust anti-tumor immunity through ICD, it also induces compensatory immunosuppressive mechanisms within both tumor and immune compartments^[91-93]. These effects include modulation of myeloid cell function, activation of immunosuppressive populations, and upregulation of immune checkpoint molecules, collectively contributing to therapy resistance^[94-96]. Table 2 summarizes the immunomodulatory benefits and potential immune negative feedback risks of different chemotherapy drugs.

3.1.1 Chemotherapy-induced reprogramming of tumor-associated macrophages (TAMs)

TAMs play a central role in tumor progression, metastasis, and post-treatment relapse^[106,107]. TAMs exhibit high plasticity along the M1-M2 polarization axis, where M1-like macrophages are generally tumoricidal, whereas M2-like macrophages promote tumor progression and immunosuppression^[108,109]. Within the TME, TAMs predominantly exhibit an M2-like phenotype that supports angiogenesis, immune suppression, and metastatic dissemination^[110,111]. Chemotherapy-induced systemic inflammatory responses can further drive TAM polarization toward the M2 phenotype, thereby reinforcing immunosuppressive conditions. NF-κB signaling acts as a context-dependent regulator of immune responses within tumor-infiltrating immune cells^[80,112]. It is required for the activation and effector functions of T cells, NK cells, and NKT cells^[113], while also sustaining the immunosuppressive activity of regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs)^[94]. Importantly, chemotherapy can partially reprogram TAMs toward an M1-like phenotype under specific conditions^[114]. However, it may also promote the expansion of immunosuppressive TAM subsets (e.g., CXCL7⁺ TAMs), which contribute to chemoresistance through CXCL7-CXCR2 signaling and downstream tumor-promoting pathways^[115].

3.1.2 Reactive oxygen species (ROS)

ROS play a context-dependent dual role in chemotherapy-induced tumor immune regulation^[116,117]. Moderate ROS levels promote endoplasmic reticulum stress and the release of DAMPs (calreticulin, adenosine triphosphate (ATP), high mobility group box 1 (HMGB1)), thereby enhancing DC activation and T cell priming through ICD^[118]. Ferroptosis-associated lipid peroxidation can further amplify this immunogenic effect^[119,120]. In contrast, sustained oxidative stress is associated with immunosuppression and immune escape. Increased ROS levels correlate with upregulation of PD-L1 in tumor and immune cells, contributing to adaptive resistance^[121,122]. Chemotherapy-induced oxidative stress, such as that triggered by paclitaxel (PTX) or gemcitabine, has been linked to transient PD-L1 elevation in tumor cells and macrophages^[121]. Overall, the effects of ROS are highly dose- and duration-dependent, promoting either ICD-mediated immune activation or immune suppression depending on stress intensity^[123-125].

3.1.3 Positive effect of inducing ICD

ICD is a regulated form of cell death characterized by the emission of DAMPs, which convert dying tumor cells into endogenous vaccines that activate anti-tumor immunity^[11,126-129]. Chemotherapeutic agents such as cisplatin, oxaliplatin, and PTX can induce ICD by triggering cellular stress responses, thereby promoting the transition of the TME from an immune-cold to an immune-hot state (Figure 3).

Display Full Size

Download

Figure 3. Illustration of chemotherapy-induced ICD and its role in converting cold tumors into hot tumors. Created in BioRender. Lutf, B. (2026) https://BioRender.com/93z6ui0. ICD: immunogenic cell death; INF: interferon; DC: dendritic cell; TCR: T-cell receptor; TLR: toll-like receptor; LRP: low-density lipoprotein receptor-related protein; HMGB: high mobility group box; CRT: calreticulin; ATP: adenosine triphosphate; DAMPs: damage-associated molecular patterns.

During ICD, CRT exposure acts as an “eat-me” signal that promotes DC recognition and phagocytosis of tumor antigens. Subsequently, HMGB1 released from dying cells interacts with pattern recognition receptors such as TLR4 to enhance antigen-presenting cell (APC) maturation and cytokine production^[129-131]. In parallel, extracellular ATP functions as a chemotactic and inflammasome-activating signal that promotes T cell priming and expansion^[132,133].

Activated DCs further stimulate CD4⁺ and CD8⁺ T cell responses through co-stimulatory signaling pathways such as CD28-CD80/86^[134,135]. Although PD-1/PD-L1 signaling can modulate this process, ICD overall enhances tumor antigen presentation and T cell clonal expansion, thereby improving the efficacy of subsequent immunotherapy^[136].

3.2 Optimization of chemotherapy efficacy under nano-intervention

To overcome the adverse effects of traditional chemotherapy in immunotherapy, particularly immune resistance caused by upregulation of the PD-L1/PD-1 axis, nanotechnology emerges as a novel therapeutic strategy, providing critical support for enhancing chemotherapy efficacy while reducing toxicity^[137-139]. To overcome this problem, researchers designed and developed a hypoxia-responsive nanoprodrug (DJNP nanoparticles), which combines PTX with the BRD4 inhibitor JQ1, and utilizes the hypoxia microenvironment to trigger the activation of the prodrug, precisely releasing the therapeutic drug (Figure 4)^[140]. JQ1, as a BRD4 inhibitor, reduces PD-L1 expression by inhibiting the binding of BRD4 to the PD-L1 promoter. Studies have shown that DJNP nanoparticles can reverse chemotherapy-induced PD-L1 upregulation, effectively enhancing the tumor’s immune response by reducing PD-L1 expression and blocking tumor cell immune escape. In mouse models, DJNP nanoparticles exhibited higher tumor accumulation and prolonged blood circulation time, and compared to the traditional chemotherapy drug Abraxane, they more effectively inhibited tumor growth without showing significant systemic toxicity. In another study, Hu et al. constructed a polydopamine nanocarrier that could synergistically load doxorubicin (DOX) and the BRD4 inhibitor JQ1, enabling JQ1 to inhibit PD-L1 expression at the molecular level and significantly enhance the anti-tumor synergistic effect in vivo^[141]. Similarly, Liu et al. designed a hormone-reversing eggshell liposome system that co-loaded the PD-L1 inhibitor JQ1 and the chemotherapeutic drug DOX, and enhanced the chemotherapeutic immunotherapy effect on tumors through this nanoplatform (Figure 5)^[142]. These eggshell liposomes exhibited good drug loading stability in the TME and achieved drug release in the acidic TME through a pH-responsive mechanism. JQ1 inhibited PD-L1 expression, thereby blocking the tumor immune escape pathway; simultaneously, DOX-induced ICD further enhanced the anti-tumor immune response. In vivo experiments showed that JQ1/DOX eggshell liposomes not only significantly inhibited tumor growth but also enhanced the immune system’s recognition and clearance of tumors by increasing CD8⁺ T cell infiltration and inhibiting PD-L1 expression.

Display Full Size

Download

Figure 4. (A) Schematic illustration of the preparation of hypoxia-responsive self-immolative DJNP NPs; (B) Their application in chemo-immunotherapy via PTX-induced ICD and JQ1-mediated PD-L1 downregulation. Reproduced from reference^[140]. CC BY 4.0. DJNP NPs: prodrug nanoparticles; PTX: paclitaxel; ICD: immunogenic cell death; DCs: dendritic cells; BRD4: bromodomain-containing protein 4; DSPE: 1,2-distearoyl-sn-glycero-3-phosphoethanolamine; PEG: polyethylene glycol.

Display Full Size

Download

Figure 5. The illustration shows how charge-reversed eggshell liposomes can be loaded with DOX and JQ1 to improve chemotherapy-immunotherapy by blocking the PD-L1 pathway. Reproduced with permission from reference^[142]. Copyright © 2023 Elsevier. DOX: doxorubicin; ICD: immunogenic cell death; BRD4: bromodomain-containing protein 4; ROS: reactive oxygen species; DC: dendritic cell; CTLs: cytotoxic T lymphocytes; PLGA: poly(lactic-co-glycolic acid); PEG: polyethylene glycol; HMGB: high mobility group box; CRT: calreticulin; DSPE: 1,2-distearoyl-sn-glycero-3-phosphoethanolamine.

4. Local Effects of RT

RT, as a traditional local treatment, has long primarily reduced tumor volume through physical means. However, with the rapid development of tumor immunology, the role of RT has gradually transcended simple tumor control, becoming an important “igniter” in tumor immunotherapy^[143,144]. RT can not only directly kill tumor cells, but more importantly, it can reshape the TME, transforming physical damage into strong immune activation signals, thereby stimulating the host’s immune response, overcoming the tumor’s immune escape mechanism, and effectively reversing the state of immune resistance^[32].

4.1 RT-induced immune activation and adaptive resistance

RT exerts potent immunomodulatory effects by inducing DNA DSBs, which represent the primary trigger of radiation-driven immune activation (see Section 2.1.2 for detailed cGAS-STING signaling mechanisms)^[145]. These DNA damage events ultimately convert irradiated tumors into highly immunogenic sites, promoting anti-tumor immune responses^[143,146].

Importantly, RT-induced immune activation is tightly regulated by dose and temporal dynamics. While moderate radiation doses enhance immune priming, excessive irradiation may attenuate immune stimulation through negative feedback mechanisms, highlighting the importance of optimized fractionation strategies^{[143,147-149]}.

However, RT simultaneously induces adaptive immune resistance. Radiation-triggered interferon signaling can promote upregulation of immune checkpoint molecules such as PD-L1 in both tumor and immune cells^[20]. This dynamic feedback represents a key limiting factor in sustained anti-tumor immunity and underscores the importance of timing in RT-immunotherapy combinations.

4.2 Application of nanotechnology in radiosensitization and immunomodulation

Traditional RT still faces several challenges, such as severe side effects and limited local control rates due to insufficient absorption of radiation by cancerous tissue and exposure to high-dose ionizing radiation, which ultimately affect patient prognosis. Therefore, it is crucial to avoid residual cancer cells, as these cells may develop RT resistance. It has been reported that pretreatment with low-dose radiation (LDR) before RT can significantly reduce tumor growth and invasiveness^[150].

Advances in nanotechnology have significantly broadened the application of RT in cancer treatment. To enhance the effectiveness of RT, radiosensitizers are widely used in the process^[151,152]. Radiation therapy using High-Z nanoparticles as radiosensitizers is a key method for increasing the radiation dose delivered to the target area during treatment. Gold (Au), hafnium (Hf), and bismuth (Bi) are examples of high-atomic-number elements that absorb X-rays very well. When these substances accumulate in tumor tissue in the form of nanoparticles, they enhance the photoelectric effect, generating large numbers of secondary electrons and Auger electrons. This exacerbates localized DNA damage and makes tumor cells more sensitive to radiation therapy^[150,153]. For example, hafnium oxide nanoparticles NBTXR3 based on high-Z elements not only significantly enhance the local efficacy of RT but also promote immune cell infiltration at tumor sites and induce abscopal effects in distant tumors. This demonstrates that high-Z nanoscale sensitization technology can also promote RT-associated antitumor immune responses^[154]. Recently, a copper-doped HfO₂ nanocage was developed as an RT sensitizer (Figure 6)^[155]. In this system, Cu ions are doped into the HfO₂ nanocapsule shell, while elesclomol, a copper ionophore, is loaded into the hollow mesoporous structure. Upon X-ray irradiation, the nanocapsules release Cu ions, which may undergo intracellular redox conversion between Cu(II) and Cu(I). Since Cu(I) is generally considered the bioactive copper species associated with cuproptosis, the released copper can promote mitochondrial lipoylated protein aggregation and iron-sulfur cluster protein loss, thereby enhancing tumor cell death and radiosensitization.

Display Full Size

Download

Figure 6. Schematic illustration of ES@HM-HfO₂ tumor radiosensitization. Upon X-ray irradiation, the HfO₂-based nanocapsules release and reduce Cu ions, which are transported to mitochondria by encapsulated ES. The accumulated Cu⁺ disrupts Fe-S cluster proteins and promotes DLAT aggregation, triggering cuproptosis and enhancing radiotherapeutic efficacy. Reproduced with permission from reference^[155]. Copyright © 2025 Elsevier. ES: elesclomol; HM: hollow mesoporous; DLAT: dihydrolipoamide S-acetyltransferase; TEOS: tetraethyl orthosilicate.

In nanotechnology applications, triphenylphosphine-derived metformin (Tpp-Met)@MnO₂@Alb nanoparticles demonstrate significant potential as a novel radiosensitizer for enhancing the efficacy of tumor RT (Figure 7)^[156]. Research indicates that Tpp-Met@MnO₂@Alb enhances RT efficacy through a dual mechanism: On one hand, it inhibits oxidative phosphorylation in tumor cells via Tpp-Met, effectively reversing the tumor’s hypoxic environment and amplifying DNA damage induced by RT; on the second hand, it improves the tumor immune microenvironment by suppressing PD-L1 and TGF-β1 expression, thereby overcoming post-RT immune tolerance. Specifically, this therapeutic strategy enhances tumor-specific immune responses by activating the cGAS-STING pathway. This pathway stimulates immune cells by sensing tumor-derived dsDNA induced by RT, further promoting DC maturation and activating T cells, thereby enhancing antitumor immunity. Therefore, Tpp-Met@MnO₂@Alb nanoparticles not only enhance the efficacy of localized RT but also activate RT-induced antitumor immune responses through immunomodulation. Specifically, by activating the cGAS-STING signaling pathway, they provide more persistent and extensive immune memory. As nanotechnology continues to advance in cancer treatment, it holds promise for playing a greater role in enhancing RT sensitization and immunomodulation in the future.

Display Full Size

Download

Figure 7. Schematic illustration of Tpp-Met@MnO₂@Alb nanoparticles for enhanced radio-immunotherapy. The nanoparticles were prepared by biomineralization and improve therapeutic efficacy through PD-L1 and TGF-β1 downregulation, cGAS-STING pathway activation, and two-stage oxygen modulation. Reproduced with permission from reference^[156]. Copyright © 2024 John Wiley & Sons. TGF: transforming growth factor; cGAS: cyclic GMP-AMP synthase; STING: stimulator of interferon genes; ADP: adenosine diphosphate; ATP: adenosine triphosphate; AMPK: AMP-activated protein kinase; dsDNA: double-stranded DNA; DCs: dendritic cells.

5. The Mechanism of Synergy Between Chemotherapy and Radiation Therapy

The therapeutic synergy between chemotherapy and RT extends beyond direct cytotoxicity and is closely associated with the restoration of anti-tumor immune surveillance. This process is characterized by the release of immunogenic signals, including CRT exposure, extracellular ATP, and HMGB1 release, which collectively promote DC activation and adaptive immune priming^[157]. In parallel, RT-induced DNA damage activates the cGAS-STING pathway, serving as a central hub that integrates genotoxic stress with type I interferon signaling, thereby enhancing anti-tumor immunity and immune surveillance^[158]. Importantly, the combination of chemotherapy and RT amplifies tumor immunogenicity and creates a more favorable immune microenvironment for immunotherapy. However, this synergy is often limited by systemic toxicity and suboptimal drug delivery, highlighting the need for advanced therapeutic strategies.

5.1 Synergistic application of nanotechnology in chemotherapy and RT

As mentioned earlier, chemotherapy is a commonly used and effective treatment method in clinical practice, while CRT is an organic combination of chemotherapy and RT. Combined CRT is more effective at controlling local tumors than single therapy because it works synergistically. However, it still has many problems, such as the inability to deliver enough drugs to the tumor site and the serious side effects on healthy tissues^{[151,159,160]}. Nanotechnology plays a crucial role in the combined application of CRT and immunotherapy. Researchers can utilize multifunctional nanoplatforms that not only deliver drugs and immunomodulatory molecules to tumor sites with high precision, ensuring their proper delivery, but also leverage the unique physicochemical properties of nanomaterials to significantly enhance therapeutic synergy, thereby overcoming the limitations of conventional therapies^[161-163].

Platinum-based drugs are not only widely used in chemotherapy, but also serve as highly effective radiosensitizers^[151,164]. For example, lactose-modified self-targeting nanoprodrugs utilize ASGPR-mediated endocytosis to precisely enrich hepatocellular carcinoma cells, and the released cisplatin significantly enhances radiosensitivity by inducing DNA damage and arresting cells in the G2/M phase (Figure 8)^[165]. Similarly, the AgNPs@TEOS-MTX platform developed by Soroushnasab et al. utilizes MTX-mediated active targeting to achieve highly efficient drug enrichment in glioblastoma (GBM). This system combines the radiosensitizing effect of silver nanoparticles with the cytotoxicity of MTX, further enhancing the synergistic therapeutic effect by inducing large-scale apoptosis^[166].

Display Full Size

Download

Figure 8. Schematic illustration of platinum (IV) amphiphilic Pt STNA and its radiosensitizing mechanism for synergistic CRT of hepatocellular carcinoma. Reproduced with permission from reference^[165]. Copyright © 2022 Elsevier. Pt STNA: prodrug-based self-targeting nanoassembly; CRT: chemoradiotherapy; ASGPR: asialoglycoprotein receptor; ROS: reactive oxygen species; HCC: hepatocellular carcinoma.

In addition, in response to the challenges of blood-brain barrier blockage and significant side effects in the treatment of GBM, another study developed an X-ray activated nanoprodrug platform (Figure 9)^[167]. This system uses downconversion nanoparticles as the core, coated with X-ray sensitive materials, which can release chemotherapeutic drugs in situ under LDR excitation. This design not only realizes imaging-guided CRT synergy, but also allows researchers to assess the dynamic changes in the microvascular environment in real time through visualization, thereby providing precise morphological evidence for clinical decision-making. This shift from “blind treatment” to “visualized diagnosis and treatment” represents an important direction of evolution in nanomedicine.

Display Full Size

Download

Figure 9. DCNP@P(Se-DOX)@ANG performs thetargeted penetration of BBB, and NIR-II FL imaging maps the microvascular structure of glioma and guides the mechanism and application of synergistic antiglioma radiation-chemotherapy. Reproduced with permission from reference^[167]. Copyright © 2024 American Chemical Society. BBB: blood-brain barrier; NIR: near-infrared; FL: fluorescence; DOX: doxorubicin; DCNP: downconversion nanoparticle; PEG: polyethylene glycol; ROS: reactive oxygen species; RT: radiotherapy.

6. Finding the Optimal Intervention Window for Time-Series Strategies

In this section, “sequential nanoregulation” refers to the coordinated control of treatment timing, therapeutic sequence, and stimulus-responsive drug release. It integrates three levels of regulation: (1) temporal control of CRT and immunotherapy administration, (2) sequential drug release enabled by nanocarriers, and (3) combination strategies that align biological response windows with therapeutic intervention.

6.1 Dynamic evolution of PD-L1 expression and its time-dependent characteristics

PD-L1/PD-1 expression appears to follow a dynamic temporal pattern that may influence the synergy between ICIs and chemotherapy. Multiple studies suggest that a 3-5 days window may represent a favorable period for combining immunotherapy with chemotherapy in certain settings - though this conclusion warrants a stratified interpretation: (1) In chemotherapy-dominant regimens, particularly among lung cancer patients, the evidence supports a 3-5 day window. Yao et al. found that administering anti-PD-1 therapy on day 3 after chemotherapy (e.g., cisplatin) was associated with reduced chemotherapy-induced lymphocyte toxicity, and its clinical efficacy was superior to concurrent or first-line immunotherapy. A retrospective analysis of patients with advanced lung cancer further confirmed that administration of ICIs within 1-10 days after chemotherapy (especially days 3-5) was associated with significantly improved overall survival (OS) compared to other time arrangements (17.3 months vs. 12.7 months)^[168]. From an immunological perspective, Huo et al.’s research revealed the biological basis of this phenomenon: the third day after chemotherapy is a period of increased or peak-like PD-1 expression on CD8⁺ T cells. Clinical data show that, compared with concurrent treatment, this 3-day delayed sequential treatment regimen significantly improved the objective response rate (68% vs. 37%) and disease control rate (98% vs. 81%), while effectively prolonging progression-free survival (PFS)^[169]. (2) In the case of RT alone or combined CRT, the relevant evidence is highly heterogeneous, and the timing of ICI administration after RT appears to be more variable and may extend beyond the early post-treatment phase depending on experimental or clinical conditions. Similar to the rapid induction induced by chemotherapy, RT-induced PD-L1 expression exhibits a significant “hysteresis effect”, indicating a critical therapeutic window. Studies have shown that the 3rd to 5th day after RT is a favorable time to introduce ICIs. Lee et al. demonstrated that injecting anti-PD-L1 antibodies on day 5 after mice received a 10 Gy radiation dose most effectively inhibited tumor growth, because at this time, the activity of cytotoxic T lymphocytes (CTLs) in the tumor is closely related to the peak expression of PD-L1, and precise blocking time can interrupt the inhibitory signal, thereby maintaining the killing ability of CTLs^[170]. Yi et al. used dynamic SPECT/CT imaging to identify a critical window period of PD-L1 upregulation following RT, demonstrating that anti-PD-1 intervention during this window significantly enhanced anti-tumor immune responses^[171].

However, in complex clinical practice, the process of tumor immune microenvironment remodeling induced by CRT exhibits more dynamic and individualized characteristics. Pouw et al. used whole-body positron emission tomography imaging to track ICI uptake in NSCLC patients during CRT and came to an unexpected conclusion: in the first week after treatment, local tumor drug uptake did not increase as expected, but remained stable or even decreased, while uptake in peripheral immune organs such as the bone marrow and spleen fluctuated significantly^[172]. This finding highlights that extrapolating from a single point in time may not be applicable to all clinical situations. Similarly, Zhong et al. combining immune cell and PD-L1 imaging techniques, found that in some combined CRT regimens, the optimal window for immunotherapy can be extended to day 14 after RT, at which point intervention produces the most significant immune enhancement effect^[173].

In addition to macroscopic regulation based on the CRT induction cycle, the “circadian rhythm of infusion” at the microscopic level has gradually become an important dimension affecting the timing of intervention. Landré et al. conducted a meta-analysis on the impact of immunotherapy infusion time on the survival of patients with advanced cancer^[174]. Specifically, patients who received early infusion before 13:00 (including NSCLC, melanoma, etc.) had significantly better OS and PFS than those who received late infusion, with hazard ratios as low as 0.50 and 0.51, respectively. This suggests that early infusion may optimize the rhythmic function of immune cells and produce a better synergistic effect with the immunomodulatory effect of PD-L1, further enhancing the overall efficacy of immunotherapy. This suggests that early infusion may optimize the rhythmic function of immune cells and produce a better synergistic effect with the immunomodulatory effect of PD-L1, further enhancing the overall efficacy of immunotherapy. This finding may be explained by circadian regulation of immune cell trafficking and activation, as immune cell proliferation, cytokine secretion, and T cell responsiveness exhibit time-of-day-dependent fluctuations governed by systemic circadian clocks.

Overall, the current evidence suggests that the optimal immunotherapy intervention window is not a fixed time point but rather a dynamic and context-dependent period shaped by tumor type, treatment modality, dosing schedule, and host immune status. Therefore, dynamic imaging technology for monitoring systemic immune patterns has irreplaceable value in determining personalized dosing regimens and optimizing clinical treatment strategies.

6.2 Nanocarrier platforms for spatiotemporally controlled drug delivery

Due to the significant spatiotemporal mismatch between chemotherapy/RT and immunotherapy in inducing antitumor immune responses (Section 6.1), traditional systemic combination therapies struggle to precisely capture the optimal intervention window. Nanotechnology, through meticulous functionalization design, provides an ideal carrier platform for the clinical translation of sequential strategies^{[139,152,175]}.

The multiresponsive nature of nanocarriers enables sequential drug release, allowing precise coordination between chemotherapy-induced immune activation and the desensitization process of immunotherapy. For instance, Wang et al. developed a pH/reduction dual-responsive self-assembling nanoparticle that first releases DOX in the acidic TME to induce ICD, followed by the release of immunostimulants within DCs. The sequential release mechanism of “ignition first, acceleration later” greatly improves the response rate of tumor-infiltrating CD8⁺ T cells^[176]. Similarly, Reda et al. successfully delivered PLK1 inhibitors and PD-L1 antibodies simultaneously using a mesoporous silica nanomaterial system (Antigen Release Agent and Checkpoint inhibitor). This strategy not only enhanced the immune response to lung cancer but also significantly reduced systemic toxicity, fully demonstrating the unique advantages of nanocarriers in sequential synergistic therapy^[177].

In addition to endogenous environmental responses, external physical stimuli can also enhance the dynamic regulation capabilities of drug delivery systems. Studies have shown that polymer nanosystems responsive to NIR-II (second near-infrared window) can immediately induce a strong ICD effect upon photoexcitation, thereby promoting the controlled release of siPD-L1 and the reconstruction of the immune microenvironment^[178]. This light-controlled release method has high spatiotemporal resolution, making it easy to pinpoint the exact time and location for immune intervention. In summary, carrier systems utilizing nanomaterials can simultaneously deliver CRT and immunotherapy, and the drug release time can be controlled very precisely through responsive structural design. This “on-demand release” characteristic breaks through the inherent time limitations of traditional drug delivery, thus helping to achieve precise timing of immune interventions^[179].

7. Multimodal Embedding and Spatiotemporal Controlled Release in Nanotechnology

Given the complexity and heterogeneity of the TME, single treatment methods are often insufficient to effectively overcome PD-L1 resistance. Therefore, the development of precise and personalized therapeutic strategies is of paramount importance. By combining multi-target nanocarriers with PD-L1 inhibitors and adding them to chemotherapy, RT and other treatments, immune responses can be significantly improved^[139,180]. Recent studies have further demonstrated that nanosystems can enhance tumor immunotherapy through multiple coordinated interactions rather than a single delivery function. For example, pH-responsive and cascade-responsive nanoplatforms have been designed to integrate tumor targeting, antigen capture, PD-L1 silencing, DC modulation, and T cell reprogramming within a single therapeutic system. These findings suggest that multifunctional nanosystems may improve antitumor immunity by simultaneously regulating tumor cells, APCs, and the immunosuppressive microenvironment^[181,182]. Table 3 summarizes the comprehensive applications and therapeutic effects of various nanomaterials in a variety of diseases. This section will explore various approaches to treating different types of tumors that affect the immune system, with a focus on real-time treatment options for “cold tumors”, “hot tumors”, “immune-rejecting tumors”, and tumors that cannot effectively present antigens. Furthermore, it will investigate intelligent controlled release technology, the role of PD-L1 in exosomes, and methods for dynamically evaluating the TME.

7.1 Treatment decisions based on TME subtyping

7.1.1 Cold tumors typically

This type of tumor typically presents with insufficient infiltration of immune cells and impaired antigen presentation mechanisms^[198,199]. To address the problem of “cold” tumors, scientists have created a pH-responsive peptide-based nanorobot (NRb)^[200]. It can recognize and bind to PD-L1 on the surface of colorectal cancer cells. In an acidic TME (pH 6.5), NRb transforms from nanoparticles into nanofibers, mechanically disrupting cell membranes and inducing ICD, promoting the release of DAMPs (ATP, HMGB1, CRT) and activating DCs, thereby significantly enhancing CD3⁺, CD4⁺, and CD8⁺ T cell infiltration in PMMR colorectal cancer. Simultaneously, the formed fibrous network can sustainably block the PD-1/PD-L1 pathway for over 120 hours, achieving a synergistic effect of antigen release and ICB.

In addition, other researchers have proposed a TPP-TAM@Alb nanosystem^[201]. This system enhances the enrichment of PTX@Alb in tumors and improves the ICD effect by activating the AMP-activated protein kinase (AMPK) pathway to inhibit TGF-β expression and reduce collagen deposition, while inhibiting PD-L1 upregulation and promoting the transition from a “cold” microenvironment to a “hot” microenvironment. Additionally, Sun et al. employed nano-delivered PDLIM2 expression plasmids (nanoPDLIM2) in a mouse lung cancer model. When combined with PD-1 blockade and chemotherapy drugs, this approach significantly increased MHC-I expression, enhanced T-cell infiltration, and successfully converted cold tumors into hot tumors. This strategy markedly amplified the synergistic effects of PD-1 blockade combined with chemotherapy. In most mice, this treatment regimen even achieved complete tumor regression^[202].

7.1.2 Immune-exclusion tumors typically

Manifesting as immune cells becoming trapped within the tumor’s peripheral stroma, these tumors are unable to effectively penetrate into the tumor’s core region^[203-205]. The convective pressure generated by high IFP significantly restricts the migration of nanoparticles across blood vessels^[206]. To overcome this problem, nanotechnology has achieved a systematic reshaping of the tumor biophysical environment through a “physical dismantling” effect. Theoretical simulations and case studies have confirmed that nano-intervention-induced tumor vascular normalization is key to restoring the intravascular and extravascular pressure gradient. The mathematical model developed by Sirin Yonucu et al. demonstrates that by pruning dysfunctional dilated microvessels, nanoscale systems can drive a fundamental shift in drug delivery patterns, transforming the traditional peripheral accumulation around tumors of micelles, nanoprobes, or liposomes into deep penetration into the tumor core^[207]. Zhou et al. used the IR-TAM nanoplatform to target mitochondria and activate the AMPK pathway, which not only improved hypoxia but also reduced PD-L1 and TGF-β expression, thereby reducing ECM collagen deposition and enhancing T cell infiltration. Simultaneously, it improved radiosensitivity and inhibited radiation-induced pulmonary fibrosis (Figure 10)^[208]. Further studies confirmed that combined blockade of PD-L1 and TGF-β not only downregulates immunosuppressive factors by activating the AMPK pathway but also reshapes the T cell landscape at the cellular differentiation level. Specifically, it induces the expansion of stem-like CD8⁺ T cells and promotes their differentiation into highly effective effector cells. Concurrently, CAFs maintain the immunosuppressive microenvironment of immune-rejecting tumors by secreting factors like TGF-β. The combined blockade of PD-L1/TGF-β reverses this suppression. Coupled with the IFN-γ-triggered “interferon-licensing” effect across the entire microenvironment, this significantly enhances T-cell infiltration dynamics. Ultimately, immune-rejecting tumors successfully transition to an inflammatory phenotype, markedly improving clinical outcomes^[209].

Display Full Size

Download

Figure 10. The Preparation Process of IR-TAM@Alb Nanoparticles and Mechanism for Enhancing Radioimmunotherapy and Attenuating RIPF Through Simultaneous Targeting of TGF-β and PD-L1. Reproduced from reference^[193]. CC BY 4.0. RIPF: radiation-induced lung fibrosis.

7.1.3 Hot tumors

Thermal tumors exhibit high levels of immune cell infiltration, but are affected by PD-L1 and Exo-PD-L1-mediated immunosuppression^[11]. The apatinib-ExoaPD-L1 exosome platform enhances targeting through aPD-L1 modification and synergistically induces ICD and promotes DC maturation with apatinib, while reducing the proportion of Tregs and MDSCs, thereby enhancing CD8⁺ T cell infiltration and improving anti-tumor immune responses^[210]. Furthermore, the IR-FFA@Alb nanosystem activates the AMPK/mTOR pathway to simultaneously downregulate PD-L1 and CD276, and regulates lipid metabolism (promoting FAO and inhibiting glycolysis) to reduce lactate levels, thereby enhancing CD4⁺ and CD8⁺ T cell infiltration, promoting long-term immune memory formation, and reducing the risk of tumor recurrence^[211].

7.2 Cellular remodeling

To overcome the immunosuppressive TME, nanotechnology-enabled strategies have been developed to reprogram immune cells and restore anti-tumor immune activity.

For TAMs, biomimetic nanoplatforms provide an effective approach to promote M1-like polarization and enhance immune activation. For example, Wang et al. used natural signaling molecules on the membrane surface to encapsulate PLGA nanoparticles within the membrane of M1 macrophages, thereby accurately reprogramming the phenotype of TAMs^[212]. Experimental results show that reprogrammed M1 macrophages can reactivate exhausted CD8⁺ T cells by releasing IL-12 and TNF-α. This can inhibit the growth of primary breast tumors and reduce the likelihood of distant metastasis. Besides directly delivering agonists, changing the pathways that process intracellular miRNA has become a new strategy. Zhang et al. found that targeting and upregulating Dicer expression can effectively reverse the TAM phenotype in colorectal cancer liver metastases^[213]. Activating Dicer promotes the processing and maturation of tumor-suppressing miRNAs and blocks M2 polarization signals. This triggers a robust antitumor immune response and prolongs survival. To further enhance precision and reduce systemic side effects, bioorthogonal chemistry has been introduced into microenvironment remodeling. Zhang et al. employed gold nano-enzymes to catalyze the in situ release of “locked” toll-like receptor (TLR 7/8) agonists within macrophages. This in situ triggering mechanism not only enabled polarized directional induction but also significantly enhanced macrophage phagocytic activity toward tumors, offering a novel approach to overcome the toxic side effects of traditional immune agonists^[214]. Additionally, the nanoplatform demonstrates exceptional multidimensional synergistic potential. For instance, the combination of TLR 7/8 agonist-loaded nanovaccines with sunitinib and PD-L1 inhibitors, where sunitinib reduces infiltration to optimize the metabolic microenvironment while ICB unleashes T-cell efficacy, successfully elicited durable antitumor immune memory^[215].

7.3 Smart controlled-release

As described in Section 3.1.2, ROS exhibit a dose-dependent role in tumor immune regulation. Taking advantage of the unique TME, nanoplatforms can utilize elevated ROS levels and acidic pH as endogenous triggers to achieve spatiotemporally controlled drug release. These stimuli-responsive systems enable sequential and programmed delivery of therapeutic agents within the TME, thereby providing a material basis for combination and staged therapeutic strategies. For example, a study developed a pH/ROS dual-responsive autocatalytic release system that enables deep-tissue delivery of immunotherapeutic drugs under acidic and high ROS conditions, significantly enhancing immunotherapy efficacy and prolonging survival in tumor model mice^[216]. Another study engineered a pH/ROS dual-responsive co-delivery nanosystem for liver cancer therapy, achieving synergistic controlled release under TME conditions. This demonstrates that responsive controlled release facilitates the implementation of combination treatment strategies^[217]. Additionally, by modulating the pH and redox responsiveness of nanocarriers, ROS generation and immune activity can be enhanced, thereby boosting the synergistic antitumor effects of photodynamic therapy and immunotherapy^[218]. Furthermore, researchers have engineered a dual ROS-responsive nanogel platform capable of sequentially releasing immune agonists (e.g., OX40 agonists) to activate T cells, followed by ICIs (e.g., PD-L1 antibodies) to block tumor immune escape. This achieves a precise sequential strategy of “antigen release - immune activation - blockade intervention”, providing concrete experimental evidence for integrating activation and blockade mechanisms within nanoplatforms for immunotherapy^[194]. This highly controllable delivery system ensures spatiotemporal co-localization of multimodal chemoprophylactic therapy while minimizing circulating drug concentrations, avoiding immune-related adverse events (irAEs), and enhancing treatment safety and efficacy.

7.4 The role of exosomal PD-L1 and dynamic assessment of the TME

Exosomes are present in biological fluids such as blood and urine and can be used to non-invasively monitor the tumor immune microenvironment. Researchers have developed the NanoEPIC nanomagnetic sorting platform to address issues such as exosome heterogeneity and low detection sensitivity^[219]. The platform uses magnetic nanoparticles to specifically label small extracellular vesicles and uses magnetic gradient sorting technology on a microfluidic chip to rapidly sort them based on PD-L1 expression levels. NanoEPIC scores showed a strong correlation between tumor volume and PD-L1 burden when observing the treatment window and efficacy. Clinical analysis results showed that patients who responded to immunotherapy had significantly lower exosomal PD-L1 levels, while patients who did not respond to immunotherapy had higher PD-L1 levels. This discovery provides an important biomarker for accurately identifying the treatment response window. In addition to static detection, real-time dynamic monitoring is also crucial for accurately identifying the treatment window. To this end, researchers have developed a microfluidic in vivo detection platform that integrates a perfusion system to continuously monitor fluctuations in the concentration of exosomes carrying PD-L1 in the blood^[220]. Experimental results show that changes in PD-L1 concentration can accurately predict the time when tumors transition from slow growth to rapid expansion, thus laying a key foundation for real-time adjustment of clinical treatment plans. Furthermore, the discovery of PD-L1 in exosomes suggests its great potential for application in the precise molecular subtyping of tumors. Another researcher made a very sensitive chemiluminescence immunosensor out of nickel-cobalt hydroxide nanoflowers that can accurately measure the amount of PD-L1-positive exosomes in serum^[221]. Clinical studies have shown that exosomal PD-L1 levels are closely associated with the invasiveness of lung adenocarcinoma. This platform can not only accurately diagnose lung adenocarcinoma, but it can also tell the difference between microinvasive and invasive adenocarcinoma. This platform can not only accurately identify lung adenocarcinoma with complete specificity, but also effectively distinguish between microinvasive adenocarcinoma and invasive adenocarcinoma. It can also serve as a reliable biomarker to assess the progression of the TME through liquid biopsy and to develop personalized treatment plans.

8. Clinical Translation and Future Outlook

In the next decade, sequential nanomedicine strategies for overcoming PD-1/PD-L1 resistance are expected to move from additive combinations toward more controllable and predictable treatment systems, where outcomes can be measured and adjusted in real time.

First, TME classification should move beyond qualitative descriptions and become a reproducible quantitative framework. This will require standardized immune features across patient groups, together with spatial omics and digital pathology to reduce bias from intratumoral heterogeneity.

Second, chemotherapy and RT induce time-dependent changes in PD-L1 expression, but these patterns differ across tumor types and treatment regimens. More longitudinal studies and modeling work are needed to define when PD-L1 and effector T cells peak, and how their timing differs. This will help make the “optimal intervention window” more precise and usable in practice.

In combined RT and immunotherapy, the cGAS-STING pathway and TREX1-related dose effects show that fractionation plays an important role in immune activation. Future strategies should therefore consider both radiation dose design and immune dynamics. At the same time, nanomedicine-based sequential delivery systems need to be tested in vivo to ensure that drug release matches the biological timing (Figure 11). For clinical use, these systems also need to meet basic manufacturing requirements such as consistent particle size, stable drug loading, and reproducible release profiles.

Display Full Size

Download

Figure 11. Schematic diagram illustrating the current status and future direction of nanomaterials in cancer treatment; currently, CRT and nanotherapy are combined, while in the future, personalized treatment and real-time immune monitoring will be used to optimize treatment outcomes. Created in BioRender. Lutf, B. (2026) https://BioRender.com/co4atey. CRT: chemoradiotherapy; RT: radiotherapy; NPs: nanoparticles.

Companion diagnostics will be essential for translating these strategies into practice. Tools such as exosomal PD-L1 detection, imaging, and microfluidic monitoring can help track immune status in real time. These markers should be integrated into clinical trials to guide patient selection, treatment timing, and early response evaluation. In the long term, a closed-loop system linking tumor subtype, timing window, drug delivery, monitoring, and treatment decisions may improve outcomes in patients who do not respond to current immunotherapy.

Despite encouraging progress in preclinical and translational studies, clinical evidence for sequential nanomedicine strategies remains limited. Most studies are still in early-stage clinical trials or preclinical models, and reliable randomized clinical data are scarce. Furthermore, as shown in Table 3, different nanoplatforms present specific safety concerns, including long-term storage stability of liposomes, off-target effects, and the risk of long-term bioaccumulation of nanomaterials, which are major obstacles to clinical application^[47,222,223]. From a regulatory and CMC perspective, the complex structure of multifunctional nanoplatforms increases the difficulty of standardization, quality control, large-scale production, and approval processes, especially when multiple active ingredients and stimulus-response mechanisms are involved. Complex designs can reduce batch-to-batch consistency and reproducibility; therefore, future nanomedicine systems should be simpler, more modular, and easier to manufacture, while striking a balance between controlled delivery and therapeutic functionality.

Authors contribution

Lutf B: Conceptualization, writing-original draft.

Li Y: Writing-review & editing.

Shen J: Supervision, writing-review & editing.

Conflicts of interest

Jianliang Shen is an Editorial Board Member of BME Horizon. The other authors declare no conflicts of interest.

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

Not applicable.

Funding

None.

Copyright

© The Author(s) 2026.