Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies, with a 5-year survival rate remaining below 10%.1 This grim prognosis stems from PDAC’s inherent aggressiveness, late diagnosis, and its pronounced resistance to conventional therapies.2 A defining feature of PDAC is its unique tumor microenvironment (TME), characterized by dense desmoplastic stroma, poorly organized and dysfunctional vasculature, and significant immunosuppression. Collectively, these elements contribute to regions of oxygen deprivation or hypoxia that have emerged as critical barriers to effective treatments.

Hypoxia disrupts drug delivery via multiple interrelated mechanisms that collectively establish a formidable barrier to therapy. One such mechanism is the abnormal and poorly organized vasculature characteristic of hypoxic tumors, where vessels are tortuous, leaky, and inefficient at delivering oxygen and nutrients. This vascular dysfunction not only reduces overall perfusion but also creates heterogeneous blood flow, leaving large tumor regions under-supplied and inaccessible to circulating drugs. In parallel, PDAC is marked by an exceptionally dense extracellular matrix (ECM), composed of collagen and hyaluronan, which physically impedes molecular diffusion and increases tissue rigidity. This stiff stromal network restricts the passive spread of chemotherapeutic agents and further compresses intratumoral blood vessels, compounding perfusion deficits. Adding to this complexity is the markedly elevated interstitial fluid pressure, which arises from the combination of vascular leakage, stromal expansion, and impaired lymphatic drainage. Elevated pressure effectively counteracts convective transport, pushing therapeutic agents away from the tumor core and preventing uniform intratumoral distribution. Together, these intertwined mechanisms—vascular abnormality, stromal density, and fluid pressure elevation—create a microenvironment where drug penetration is severely compromised, resulting in poorly accessible tumor niches that sustain hypoxia and fuel resistance to multiple treatment modalities. Abnormal vasculature impairs perfusion and oxygenation, whereas excessive extracellular matrix (ECM) deposition increases interstitial pressure and creates a physical barrier to drug diffusion.3 Additionally, hypoxia-inducible factors such as HIF-1α promote pathological angiogenesis and further ECM remodeling, exacerbating these structural impediments. These changes reduce the intratumoral accumulation of conventional chemotherapeutics and nanoparticle (NP)-based agents.4 Moreover, hypoxia shapes the immunosuppressive milieu by recruiting suppressive immune cells and excluding cytotoxic T cells, thereby weakening the efficacy of immunotherapies.5

In addition to impairing drug transport, hypoxia directly promotes a more aggressive tumor phenotype. It induces epithelial-to-mesenchymal transition (EMT), a biological process whereby epithelial cells lose their cell–cell adhesion and polarity, and acquire mesenchymal and migratory properties that facilitate invasion and metastasis, along with metabolic reprogramming and the emergence of stem-like cancer cells, which are all hallmarks that drive metastasis and confer broad-spectrum therapeutic resistance.6 Together, these factors make the hypoxic TME a central contributor to therapeutic failure observed in PDAC.7

In this review, we explored how hypoxia contributes to drug resistance in PDAC by creating formidable physiological and biochemical barriers. Particular emphasis has been placed on recent advances in hypoxia-responsive NP systems, which are designed to selectively release therapeutic payloads within the hypoxic TME and overcome current limitations in drug delivery. Through this lens, we aimed to highlight the promise of nanomedicine as a transformative approach for tackling therapeutic challenges in PDAC.

Tumor Hypoxia and Therapeutic Barriers in Pancreatic Cancer

PDAC is characterized by a uniquely aggressive and hypoxic tumor microenvironment.8 Tumor hypoxia, defined as inadequate oxygen supply, is both a hallmark of PDAC and a key contributor to its resistance to therapy.9 In pancreatic tumors, oxygen tensions are markedly lower than in normal pancreatic tissue, owing to abnormal vasculature and dense stromal content.10 This section explores the origin of hypoxia in the PDAC microenvironment (section 2.1), its role in impeding drug delivery, and its effect on the efficacy of conventional therapies.(section 2.2) An in-depth understanding of these mechanisms highlights the need for new strategies to overcome hypoxia-induced therapeutic failure in pancreatic cancer. Table 1 presents a comparison of the hypoxic features of various solid tumors to contextualize PDAC. As summarized in Table 1, PDAC exhibits uniquely severe and chronic hypoxia (pO2 < 2 mmHg), extensive desmoplasia occupying more than 80% of tumor volume, dense collagen- and hyaluronan-rich ECM, and markedly elevated interstitial fluid pressure. Notably, the desmoplastic stroma in PDAC is exceptionally pronounced, often accounting for the vast majority of the tumor bulk. This stromal overgrowth is not a passive histological feature but an active pathological barrier: cancer-associated fibroblasts and pancreatic stellate cells generate abundant extracellular matrix that compresses blood vessels, elevates interstitial fluid pressure, and physically encases tumor nests. As a result, even though tumor cells may represent less than 20% of the total tumor volume, the stromal compartment dominates the microenvironment, dictating drug distribution, immune cell access, and oxygen availability. This disproportionate stromal expansion is a defining hallmark that distinguishes PDAC from other solid tumors, where tumor cell density typically outweighs stromal content. Clinically, such extreme desmoplasia underpins the poor vascular perfusion, immune exclusion, and therapeutic resistance that remain major barriers to improving outcomes in PDAC. Therefore, emphasizing the stromal predominance is crucial to understanding why pancreatic cancer remains among the most treatment-refractory malignancies.

Table 1 Comparative Characteristics of Hypoxic Tumor Microenvironments Across Major Solid Tumors

In line with these characteristics, Figure 1 schematically illustrates the major pathophysiological features of the PDAC microenvironment, including desmoplastic stroma, abnormal vasculature, tumor hypoxia, immune suppression, therapeutic resistance, and early metastasis. Together, these barriers underscore the therapeutic challenges of PDAC and provide the rationale for developing hypoxia-responsive nanoparticle strategies.

Figure 1 Pathophysiological features of the pancreatic tumor microenvironment Illustration of key tumor microenvironmental traits driving PDAC progression and treatment resistance. (A) Dense desmoplastic stroma impedes drug delivery. (B) Abnormal vasculature limits oxygen supply. (C) Hypoxic regions promote chemo/radioresistance. (D) Immune suppression inhibits anti-tumor responses. (E) Hypoxia-driven EMT facilitates early metastasis. (F) Combined features confer resistance to conventional therapies. These barriers highlight the therapeutic relevance of hypoxia-responsive nanoparticles.

Mechanisms Underlying Hypoxia in the TME Desmoplastic Stroma and Vascular Dysfunction

A hallmark of PDAC is its extensive desmoplastic stroma, accounting for up to 80–90% of the tumor mass.25 This fibrotic stromal tissue, composed of cancer-associated fibroblasts (notably pancreatic stellate cells), immune cells, and ECM, physically encases tumor cells and dramatically influences tumor physiology.26–28 This dense stroma results in severe vascular dysfunction, compressing blood vessels and impairing normal angiogenesis, yielding a sparse and irregular vasculature that cannot adequately perfuse the tumor.25 Studies have documented that PDAC lesions have significantly lower microvessel density and blood flow than other malignancies, resulting in chronic nutrient and oxygen deprivation within the tumor core.29

Despite these hostile conditions, PDAC cells survive and even thrive by adapting their metabolism to the hypoxic microenvironment. They shift towards glycolytic pathways, increase autophagy, and modulate redox homeostasis to withstand oxidative stress. Moreover, they engage in metabolic crosstalk with stromal cells, including pancreatic stellate cells, to access alternative nutrient sources and enhance their survival.

The remaining blood vessels are often structurally disorganized and highly permeable, leading to a heterogeneous blood supply where pockets of tumor cells receive little to no oxygen. Consequently, PDAC tumors develop regions of intrinsic hypoxia, an oxygen-starved microenvironment driven by inadequate perfusion. This hypoxic state is further exacerbated by the rapid proliferation of tumor cells and metabolically active stromal cells such as activated fibroblasts, which consume oxygen at a high rate, outpacing its delivery.8,9 The net effect of the desmoplastic stroma is a vicious cycle of vascular insufficiency and oxygen shortage: fibroblast-rich tissue elevates interstitial pressure and collapses vessels, and the ensuing hypoxia in turn promotes pro-fibrotic and pro-angiogenic signals (such as HIF-1α-mediated VEGF expression) that produce aberrant, nonfunctional vessels.30,31 Collectively, these stromal and vascular abnormalities explain why PDAC is among the most hypoxic solid tumors.

Extracellular Matrix Rigidity and Oxygen Imbalance

In addition to vessel compression, the composition and stiffness of the PDAC extracellular matrix critically contributes to tumor hypoxia. The ECM in pancreatic tumors is exceptionally stiff due to abundant collagen cross-linking and the deposition of molecules such as hyaluronic acid, which imbibe water and increase tissue turgor.32–34 This elevated matrix rigidity leads to high interstitial fluid pressure, which creates a physical barrier to blood flow and oxygen diffusion.32,34 Essentially, as the ECM stiffens, it not only squeezes vessels shut, but also limits oxygen penetration into tissues by increasing the diffusion distance and creating a gradient where oxygen is rapidly consumed at the periphery of stromal bundles and cannot reach the tumor core.10 This leads to a metabolic imbalance, in which oxygen consumption by the tumor and stromal cells vastly exceeds the supply. Even when proangiogenic factors are released under hypoxic conditions, the resulting neovasculature is abnormal and insufficient, failing to relieve the hypoxia.8,35 Moreover, the fibrotic matrix can sequester growth factors and alter cellular phenotypes, for example, by inducing a more contractile matrix-remodeling phenotype in stellate cells, which perpetuates stiffness and hypoperfusion.27,36 Recent studies have suggested that modulating ECM components, such as enzymatically degrading hyaluronan or inhibiting collagen cross-linking, can temporarily decompress blood vessels and enhance tumor oxygenation.37,38 In summary, the desmoplastic reaction and ECM remodeling in PDAC establish a microenvironment with poor vascular function and high tissue tension, resulting in pervasive hypoxia throughout the tumor bed.

Immune Suppression and Early Metastasis

In addition to the structural barriers, the hypoxic microenvironment of PDAC substantially alters the immune landscape, fostering an immunosuppressive state that impedes antitumor immunity. Hypoxia-driven expression of immunomodulatory molecules such as PD-L1, adenosine, and VEGF suppresses the recruitment and cytotoxic function of effector T cells and natural killer (NK) cells, while promoting the accumulation of regulatory T cells, myeloid-derived suppressor cells (MDSCs), and M2-polarized tumor-associated macrophages (TAMs). These suppressive cell populations not only dampen immune surveillance but also secrete growth factors and cytokines (eg, TGF-β, IL-10) that promote stromal remodeling and tumor progression. The net result is an immune-excluded “cold” tumor phenotype that contributes to resistance against immune checkpoint inhibitors and other immunotherapies.17

Moreover, hypoxia promotes early metastatic dissemination by inducing EMT, a process whereby epithelial cancer cells lose polarity and adhesion and gain migratory and invasive capabilities.7 Hypoxia-inducible factors (HIFs), particularly HIF-1α, upregulate transcription factors such as Snail, Twist, and ZEB1, which suppress E-cadherin and activate mesenchymal gene programs, enabling tumor cells to detach and invade surrounding tissues.39 Combined with extracellular matrix degradation and vascular abnormalities, this facilitates the intravasation and dissemination of cancer cells even at early stages. Clinical studies have shown that PDAC often exhibits micrometastases at diagnosis, suggesting that metastatic seeding occurs before the primary tumor becomes clinically detectable.40 This aggressive trait is strongly associated with a hypoxic and fibrotic tumor milieu.

Taken together, immune suppression and early metastasis are critical hypoxia-driven mechanisms that worsen the prognosis and limit therapeutic success in PDAC. These insights further underscore the need for therapies that can modulate the tumor microenvironment, reverse immune exclusion, and inhibit early dissemination.

Hypoxia-Mediated Barriers to Drug Delivery Impaired Perfusion and Physical Exclusion

Most systemic chemotherapies rely on adequate blood flow to reach tumor cells; however, in hypoxic pancreatic tumors, perfusion is patchy and often insufficient.41,42 The compressed and collapsed vasculature reduces the intratumoral concentration of drugs, such as gemcitabine, as much of the administered dose never effectively penetrates the tumor mass.43

Stromal barriers that contribute to hypoxia hinder effective drug distribution. The extensive extracellular matrix forms a dense meshwork that impedes the convection and diffusion of therapeutic molecules, particularly large compounds and NPs, effectively excluding them from hypoxic tumor regions.32 For example, dense collagen bundles and hyaluronan-rich zones can slow the transit of drugs, creating a concentration gradient where peripheral tumor areas receive more drug exposure than the poorly perfused, hypoxic interior.43,44 Owing to this heterogeneity, cells residing in the most hypoxic regions often escape adequate drug exposure because therapeutic agents fail to reach them at sufficient concentrations.

Empirical evidence from PDAC models has shown that the enzymatic ablation of stromal components, such as using PEGylated hyaluronidase to degrade hyaluronic acid, can increase intratumoral vascular perfusion, thereby improving drug uptake in previously inaccessible regions.45 Conversely, when hypoxia is severe, interstitial fluid pressure remains high and vessels remain nonfunctional; therefore, even diffusible drugs have difficulty penetrating the tumor.

Hypoxia marks and reinforces a physical barrier to therapy; it indicates regions of minimal drug penetration and reflects a microenvironment that shields cancer cells from treatment.37,43 Thus, the impaired perfusion associated with hypoxia serves as a critical barrier, allowing a fraction of tumor cells to survive the initial chemotherapy owing to insufficient drug penetration.

Hypoxia-Induced Cellular Resistance and Chemo-Therapy Failure

In addition to impeding drug delivery, hypoxia induces cellular adaptations that confer resistance to both chemotherapy and radiotherapy in pancreatic ductal adenocarcinoma (PDAC). Oxygen-deprived cancer cells activate hypoxia-inducible factors (HIFs), particularly HIF-1α, which trigger transcriptional programs enabling survival under low oxygen conditions.46,47 These adaptive pathways include metabolic reprogramming toward anaerobic glycolysis, angiogenesis, and stress response mechanisms, all of which reduce cellular susceptibility to cytotoxic therapies.48,49 For instance, hypoxia can drive tumor cells into a quiescent or slow-cycling state, limiting the efficacy of cell cycle-dependent chemotherapeutics.50 In PDAC, HIF-1α activation has been associated with the upregulation of drug efflux transporters, such as P-glycoprotein and other multidrug resistance pumps, that actively expel chemotherapeutic agents and lower intracellular drug accumulation.47 In addition, hypoxia promotes autophagy as a survival mechanism, which has been implicated in gemcitabine resistance by enabling tumor cells to recycle nutrients and maintain energy homeostasis.51 Collectively, these adaptations explain why hypoxic PDAC tumors often show poor response rates to standard chemotherapies such as gemcitabine-based regimens and FOLFIRINOX.52,53 Clinical studies have demonstrated that patients with highly hypoxic PDAC exhibit shorter progression-free survival intervals, highlighting the prognostic impact of hypoxia.54,55

Radiotherapy is also compromised by tumor hypoxia. Radiation-induced DNA damage relies on oxygen to generate free radicals that stabilize DNA breaks.56,57 In hypoxic regions, reduced oxygen availability results in fewer free radicals and diminished DNA damage, such that PDAC cells can tolerate significantly higher radiation doses than well-oxygenated cells—doses often not feasible due to normal tissue toxicity limits.58 As a consequence, radiotherapy in pancreatic cancer has historically achieved only marginal improvements in local tumor control, even with advanced modalities such as stereotactic body radiotherapy. In summary, hypoxia not only conceals tumor cells behind a barrier of poor perfusion but also fortifies them biologically, enabling a dual mode of therapeutic evasion.59,60 These clinical limitations underscore how the hypoxic microenvironment conceals tumor cells from drug penetration while simultaneously fortifying them biologically, creating a dual barrier to therapeutic efficacy. Together, these mechanisms contribute to the persistently poor outcomes observed in PDAC, where the five-year survival rate remains in the single digits despite aggressive multimodal treatment.61–63

Prognostic Biomarkers and Need for Targeted Strategies

Tumor hypoxia not only hinders real-time therapy but also functions as a prognostic marker of pancreatic cancer aggressiveness. Hypoxic biomarkers, such as HIF-1α overexpression or elevated levels of carbonic anhydrase IX (CAIX, a hypoxia-regulated enzyme), have been associated with poor survival in PDAC patients.64 Likewise, gene expression signatures reflecting hypoxic stress correlate with higher rates of metastasis and treatment failure, suggesting that measuring tumor hypoxia could inform prognosis and guide therapy selection.65 Recognizing the central role of hypoxia and stromal barriers in PDAC, researchers have actively explored targeted strategies to mitigate these effects. One such approach is microenvironment normalization, that uses agents to decompress blood vessels and increase perfusion. A notable strategy tested in clinical trials involved pegylated hyaluronidase (PEGPH20) to degrade hyaluronan in PDAC, which in a Phase II study, improved drug delivery and progression-free survival in patients with high stromal hyaluronan levels.66,67 While subsequent trials showed inconsistent efficacy, this line of research supports the rationale that alleviating physical barriers can enhance the clinical response. Another approach is to exploit hypoxia using hypoxia-activated prodrugs, which are activated into toxic agents under low-oxygen conditions. One such agent, evofosfamide (TH-302), showed promise in early phase trials by selectively killing hypoxic PDAC cells when added to chemotherapy, although Phase III results were inconclusive.67 Additionally, the direct targeting of hypoxia signaling pathways is under investigation. Inhibition of HIF-1α or downstream effectors (for instance, using acriflavine or other HIF inhibitors) has demonstrated the potential to sensitize pancreatic tumors to chemotherapy in preclinical models by reversing hypoxia-induced resistance programs.68,69 There is also a growing interest in repurposing drugs such as angiotensin inhibitors or modulators of nitric oxide to “normalize” tumor vessels and oxygenation (an approach pioneered in other solid tumors) for pancreatic cancer.70,71 In summary, the clinical message is clear: hypoxia is a key driver of therapeutic failure in PDAC, and overcoming it may be key to significantly improving outcomes. While conventional therapies are limited by the biological and physical barriers imposed by hypoxia, emerging strategies targeting the hypoxic tumor microenvironment or using hypoxia for treatment represent promising avenues. Ongoing clinical trials and research efforts are focused on integrating such strategies with standard treatments, with the goal of dismantling hypoxia-driven resistance, which has long hindered progress in pancreatic cancer treatment.

Overview of Hypoxia-Targeted NPs as a Therapeutic Strategy

The TME of PDAC is characterized by dense stroma and pronounced hypoxia, making it a particularly challenging therapeutic target. Hypoxia, defined as regions of low oxygen tension, occurs in over 50% of solid tumors due to abnormal, leaky vasculature and rapid tumor growth.72 In healthy tissues, the partial pressure of oxygen is approximately 40–60 mmHg, whereas in tumors, it often falls below 10 mmHg.9 PDAC is among the most hypoxic cancers, with some tumor regions measuring < 3 mmHg O2 (over 10-fold lower than normal tissue).73 Such oxygen deprivation contributes to aggressive tumor behavior, therapeutic resistance, and poor patient outcomes.72,73 Consequently, strategies that specifically target the hypoxic tumor microenvironment have gained increasing interest as a means of improving PDAC treatment.25 One promising approach is the use of hypoxia-targeted NP drug delivery systems engineered to sense and respond to low-oxygen conditions in tumors. In addition to conventional engineering approaches, recent advances in artificial intelligence (AI) are expected to play an increasingly crucial role in optimizing nanoparticle design and overcoming current limitations, thereby complementing the strategies discussed in this section.

This section provides an overview of the mechanisms by which these hypoxia-responsive NP systems operate, how they release drugs under hypoxic conditions, the common materials and structures used in their construction, and a classification of targeting strategies (passive vs active) and stimulus triggers (endogenous vs exogenous). Each subsection highlights recent PDAC-specific examples to illustrate design principles and their clinical relevance.

General Mechanisms of Hypoxia-Responsive NP Systems

Hypoxia-responsive NP systems are designed with built-in triggers that undergo chemical or structural changes in low-oxygen environments, thereby conferring selectivity to hypoxic tumors. The general design principles, representative chemical strategies, and nanocarrier structures of hypoxia-responsive nanoparticles are summarized in Table 2. This table highlights key oxygen-sensitive groups, bioreductive activation pathways, drug release mechanisms, and different carrier types, providing a concise framework for the detailed examples discussed in the following subsections. Most commonly, these triggers are hypoxia-sensitive moieties and functional groups that are stable under normoxia, but become activated or cleaved in hypoxia.74 Typical examples include nitroaromatic groups (eg, 2-nitroimidazole derivatives), quinone N-oxides, and azo linkers, all of which can be bioreduced by cellular enzymes that are more active under oxygen-starved conditions.75 Incorporating such groups into a nanocarrier (for instance, as linkers or caps on a prodrug) renders the entire construct hypoxia-responsive, remains stable under normoxic conditions, and is selectively activated in hypoxic tumor zones.76

Table 2 Hypoxia-Responsive NP Design: Core Mechanisms, Structures,and Materials

Hypoxia-responsive NPs achieve targeted drug delivery in hypoxic tumor environments such as those found in PDAC. Following systemic administration, NPs circulate through the bloodstream and accumulate in tumor tissues via the enhanced permeability and retention (EPR) effect. Once inside the tumor microenvironment, low-oxygen conditions trigger structural or chemical changes in the NPs, such as disassembly or cleavage of hypoxia-sensitive linkers, leading to the controlled release of the encapsulated drug. This strategy enables precise drug release in hypoxic regions of the tumor, enhances therapeutic efficacy, and reduces systemic toxicity. The general mechanism of hypoxia-responsive nanoparticle-mediated drug release is schematically illustrated in Figure 2, which depicts systemic administration, tumor accumulation via the EPR effect, hypoxia-triggered transformations, and controlled release of the therapeutic payload within oxygen-deprived tumor regions.

Figure 2 Mechanism underlying hypoxia-responsive nanoparticle-mediated drug release in the tumor microenvironment The process by which hypoxia-responsive nanoparticles (NPs) achieve targeted drug delivery within hypoxic tumor environments, such as those found in PDAC. Following systemic administration, the nanoparticles circulate through the bloodstream and accumulate in the tumor tissue via the enhanced permeability and retention (EPR) effect. Once inside the tumor microenvironment, the low oxygen conditions trigger a structural or chemical change in the nanoparticles—such as disassembly or cleavage of hypoxia-sensitive linkers—leading to the controlled release of the encapsulated drug. This strategy enables precise drug release in hypoxic regions of the tumor, enhancing therapeutic efficacy and reducing systemic toxicity.

Under normoxic conditions, many bioreductive reactions are reversed by oxygen, which prevents premature activation. For example, 2-nitroimidazole residues attached to an NP are enzymatically reduced inside cells; however, in the presence of oxygen, they are re-oxidized back to their initial state, maintaining the nanocarrier stable.74 However, in hypoxic cells, 2-nitroimidazole is irreversibly converted to 2-aminoimidazole, which breaks the linkage or alters the polarity of the carrier and triggers drug release by disrupting the NP structure.84 Azo bonds behave similarly; under low oxygen conditions, azo linkers are reduced (often by azoreductase enzymes), causing the bond to cleave, thereby activating the NP or payload.85 Through such mechanisms, hypoxia serves as an intrinsic “switch” that ensures the nanocarrier deploys its therapeutic cargo predominantly in the targeted oxygen-depleted tissue and not in normoxic healthy tissue.

An emerging hypoxia-responsive strategy employs agents that are inherently cytotoxic only under low-oxygen conditions, effectively triggering the drug itself. One notable example in PDAC involves encapsulating a bioreductively activated cytotoxin (a BE-43547A2 analog) in a sulfonated azocalixarene carrier to create a “dual” hypoxia-responsive supramolecular complex.73 Azocalixarenes contain azo groups that are cleaved under hypoxic conditions, whereas the prodrug requires reductive activation. This design ensures that the drug is released and becomes fully active only inside the hypoxic PDAC cells. In their study, hypoxia-triggered activation led to selective toxicity in oxygen-deprived pancreatic tumor cells and significantly suppressed tumor growth in vivo with minimal systemic side effects.73 This illustrates the general mechanism by which hypoxia-targeted NPs achieve spatially selective therapy by harnessing the reductive chemistry unique to hypoxic tumor microenvironments to trigger selective drug release or activation.

Drug Release Mechanisms Under Hypoxic Conditions

Hypoxia-targeted NPs are engineered such that once they accumulate in the tumor, the low-oxygen milieu activates drug release through one or more specific mechanisms. In many designs, the drug is tethered to the NP via a hypoxia-labile linker or encapsulated by a hypoxia-sensitive matrix. Under well-oxygenated conditions, these linkers remain intact (or are continually re-stabilized by O2), keeping the therapeutic payload securely bound.86 When the nanoparticle enters a hypoxic region, however, the lack of oxygen allows reductive cleavage of the linker or degradation of the matrix, liberating the drug at the target site.87 In essence, the sharp contrast in chemical conditions between normoxic blood/tissues and hypoxic tumor zones serves as a trigger that unleashes the drug only where needed. This spatial control can greatly increase the drug concentration in hypoxic tumor cores while minimizing premature leakage in circulation or oxygenated normal tissues.88

Hypoxia-triggered release can take different forms, depending on the design. In some cases, the covalent bond between the drug (or prodrug) and carrier is broken, directly releasing the active drug molecule. In other cases, the entire NP may undergo a structural change or disassembly, leading to its release. For example, Chen et al89 developed an aptamer-targeted dendrimer NP for PDAC, in which hypoxia caused the NP to fall apart into ultrasmall fragments, dumping its payload in the process. Under normoxia, this NP (made of a dendritic poly-lysine core) remains large and intact, but upon exposure to the reducing conditions of the hypoxic tumor, it sheds its surface coating and reduces in size, releasing gemcitabine-phosphate payloads in the form of tiny dendrimer pieces.89 The released sub-50 nm fragments penetrate deeply into the tumor tissue, enabling drug delivery to poorly accessible tumor regions. This hypoxia-triggered burst release and size transformation results in enhanced drug uptake by cancer cells in the oxygen-starved tumor interior and improved therapeutic efficacy in PDAC models.89

Another release mechanism involves hypoxia-activated prodrugs carried by NPs. Instead of physically releasing the drug, the drug molecules remain inactive until they experience a hypoxic environment, where they are chemically converted into potent toxic agents. The PDAC-directed complex developed by Guo et al73 is an example; the NP releases a prodrug of BE-43547A2 inside hypoxic cells, where the prodrug then undergoes bioreduction to its active cytotoxic form. Thus, even if some drugs are released into normoxic tissue, they remain in an innocuous prodrug state, adding a second layer of specificity. Overall, by cleaving a linker, degrading a carrier, or biochemically activating a prodrug, hypoxia-responsive NPs leverage the distinct redox chemistry of hypoxic tumor niches to achieve site-specific drug release. This targeted-release mechanism is critical for achieving high drug concentrations in resistant hypoxic tumor regions, such as PDAC, while sparing healthy tissues from exposure.

Conventional Materials and Structures Used in Hypoxia-Targeted NP Systems

Hypoxia-responsive NPs have been designed using various carrier materials and nanostructures. In practice, any nanocarrier functionalized with hypoxia-sensitive groups can serve as a platform for hypoxia-targeted delivery.90 Polymeric NPs are especially common; for example, polymer micelles, nanospheres, and dendrimers constructed from biocompatible polymers (such as dextrans, polyesters, and polypeptides) have been equipped with nitroimidazole or azo linkers to create hypoxia-sensitive drug conjugates.85,91 In a recent PDAC study, a dendrimer-based nanocarrier (built from a dendritic polylysine core) was co-loaded with gemcitabine and a STAT3 inhibitor and decorated with targeting aptamers.89 This polymeric NP was designed to remain stable in circulation, alter its surface charge, and degrade in size under hypoxic conditions, demonstrating the tunability of polymeric architectures for hypoxia-responsive PDAC therapy.89

Lipid-based nanostructures such as liposomes have also been adapted for hypoxia targeting, often by incorporating hypoxia-cleavable lipids or prodrug phospholipids into the bilayer. Upon entering the low-O2 tumor zone, these liposomes destabilize and release their drug load. Protein-based NPs are another important class of NPs; for instance, albumin has been used as a carrier for hypoxia-responsive delivery. Yao et al (2025) reported an albumin-based supramolecular NP modified with azobenzene groups that co-delivered a drug pair and released it selectively in hypoxic tumors.72 In this system, the albumin scaffold provides biocompatibility and a long circulation time, whereas the azobenzene cross-linkers in the matrix are cleaved under hypoxia to unload the therapeutic cargo.72 These albumin–azo constructs illustrate the versatility of natural macromolecules in hypoxia-targeted nanomedicine.

In addition to conventional carriers, researchers have explored hybrid supramolecular structures for hypoxia-responsive delivery. Hybrid lipid–polymer NPs combine the stability of polymer cores with the biomimicry of a lipid shell and have been formulated with hypoxia-sensitive elements for multimodal therapy (eg, chemo-phototherapy combinations).92 Host–guest supramolecular assemblies have also been employed; for example, azobenzene-functionalized calixarene macrocycles have served as hosts to encapsulate hydrophobic prodrugs, forming NPs that disassemble under hypoxic conditions.73 In PDAC models, a sulfonated azocalixarene carrying a BE-43547A2 prodrug remained stable in the blood, but released the drug inside cancer cells, leveraging both host–guest chemistry and azo bond sensitivity to target hypoxia.73

Classification by Targeting Strategy and Trigger Type

Hypoxia-targeted NPs can be delivered to tumors either by passive accumulation or active targeting prior to hypoxia-triggered drug release. Likewise, hypoxia-responsive systems can be classified by their trigger type as endogenous (eg, hypoxia, pH, enzymes) or exogenous (eg, light, ultrasound, magnetic fields). Figure 3, schematically summarizes these classifications, providing a conceptual overview that frames the specific examples described in the following subsections.

Figure 3 Classification of hypoxia-targeted nanoparticle strategies Schematic illustration summarizing the classification of delivery strategies and trigger types for hypoxia-responsive nanoparticles. Delivery can occur through passive targeting via the enhanced permeability and retention (EPR) effect, or active targeting using surface-functionalized ligands (eg, antibodies, peptides) that recognize tumor-associated markers. Therapeutic release is further controlled by endogenous triggers (eg, hypoxia, acidic pH, redox potential, enzyme activity) or exogenous triggers (eg, light, ultrasound, magnetic field). Together, these classifications provide a conceptual framework for the rational design of hypoxia-targeted nanomedicines in pancreatic cancer.

Passive Targeting

Passive targeting relies on the natural propensity of NPs to accumulate in tumor tissue via the EPR effect; leaky tumor vasculature and poor lymphatic drainage allow nanoscale carriers to concentrate in tumors over time.11,93 This approach does not require any special homing ligands. The NPs simply circulate and gradually extravasate into the tumor, including its hypoxic regions, driven by the abnormal anatomy of the tumor and compromised physiology. Many first-generation nanomedicines (eg, albumin-bound paclitaxel or liposomal irinotecan) function through passive targeting. Hypoxia-responsive NPs can passively reach the tumor and be activated in situ. However, in PDAC, passive delivery often proves inadequate owing to the dense stromal architecture and highly irregular perfusion, which hinders uniform NP distribution.

Active Targeting

Active targeting strategies have been employed to improve NP delivery to tumors and specific cell populations. Active targeting typically involves functionalizing the NP surface with a ligand (such as an antibody, peptide, or aptamer) that binds to a receptor overexpressed on cancer cells or in the tumor microenvironment. In the context of hypoxia-targeted systems, active targeting can enhance the accumulation of therapeutic payloads before hypoxia-triggered release. For example, Chen et al used an aptamer-decorated NP that recognizes a PDAC-associated surface marker, thereby improving the uptake of NPs by pancreatic tumor cells and stroma.73 Once actively delivered to the tumor, the NP’s built-in hypoxia sensor (in this case, a degradable coating) then kicks in to release the drugs intracellularly. Other ligands for active targeting include antibodies against hypoxia-inducible factors, surface proteins overexpressed under hypoxia, or small molecules like 2-nitroimidazole itself—which, after reduction, can covalently bind intracellular macromolecules in hypoxic cells, “trapping” the carrier in those cells. The combination of active targeting with a hypoxia-responsive trigger can yield a highly selective delivery system; the active ligand guides the NP to the vicinity of the tumor, and the hypoxia trigger ensures activation only in the poorly oxygenated tumor core. This two-tier targeting approach has been validated by aptamer-functionalized NPs in PDAC, which demonstrated enhanced tumor accumulation and hypoxia-triggered drug release.73 In summary, passive targeting leverages tumor physiology (EPR effect) to accumulate NPs in hypoxic regions, whereas active targeting uses molecular recognition to home in on tumors, which can be combined with hypoxia-sensitive release mechanisms for greater precision.

Trigger Type (Endogenous Vs Exogenous)

Stimuli-responsive nanomedicines can be broadly classified based on the origin of the triggering stimulus: endogenous, arising from the tumor microenvironment, or exogenous, applied externally.74 Hypoxia is an endogenous trigger, which is a naturally occurring condition within the tumor that the NPs can exploit. Other endogenous triggers used in drug delivery include low pH (acidity), high levels of glutathione or reactive oxygen species, and enzymes overexpressed in the tumor; in each case, the NP is engineered to sense a biochemical cue present in the cancer milieu.94 Endogenous triggers have the advantage of automatic activation: the NP releases the drug as soon as it encounters the right environment, without requiring any external action. Hypoxia-responsive systems fall into this category, activating deep in the tumor, where direct external intervention might be difficult. The previous sections focused on internally triggered designs.

In contrast, exogenous triggers are stimuli deliberately applied from outside the patient to activate the NP after it reaches the tumor. Examples include light (eg, UV or infrared laser), ultrasound, magnetic fields, and heat applied externally.88 NPs responsive to exogenous triggers typically contain components such as photosensitizers, thermosensitive polymers, or magnetic cores that respond when an external stimulus is administered, allowing a high degree of temporal control over drug release. Importantly, exogenous and endogenous triggering mechanisms can be combined to achieve synergistic effects. A pertinent example is the work of Yao et al72 who developed a hypoxia-responsive albumin NP that also carried a photosensitizer for laser-activated phototherapy. In their design, the NP accumulates in the tumor and releases autophagy inhibitor drugs under hypoxic conditions, while a light stimulus is applied to activate the photosensitizer component—which was engineered to generate tumor-killing reactive oxygen species independent of oxygen.72 This dual-trigger approach tackles the hypoxic tumor on two fronts: the endogenous hypoxia trigger releases one drug, and the exogenous light trigger activates another, overcoming the limitations of each single stimulus. Such multi-modal systems illustrate the creative strategies being explored to improve PDAC outcomes (detailed discussions of multiple stimuli and advanced targeting designs are provided in the next section).

In summary, hypoxia-targeted NPs can be categorized by how they reach the tumor via passive accumulation or active ligand-mediated targeting, and by how they are triggered by the tumor microenvironment or external stimuli. The key takeaway is that hypoxia, as an endogenous trigger, offers a powerful route for tumor-specific drug delivery, especially in hypoxia-prevalent cancers such as PDAC, and it can be augmented with active targeting and/or external triggers to further refine the delivery profile. Building on this conceptual foundation, the next section delves into specific design strategies for hypoxia-responsive NPs in pancreatic cancer, examining how these principles are implemented in the current research.

Each of the above strategies offers distinct advantages, and hybrid approaches are common in practice. Therefore, the design of hypoxia-targeted NPs is a multidisciplinary balancing act that integrates tumor biology (oxygen levels and expression of targetable markers) with smart chemistry (stimuli-responsive linkers/materials) and nanotechnology (optimal size, surface, and payload properties).

Strategies for Hypoxia-Responsive NPs in Pancreatic Cancer

PDAC is characterized by a profoundly hypoxic TME, dense stromal barriers, and poor vascularization, all of which hinder the efficacy of conventional therapies.25,95 Hypoxia-responsive NPs are emerging as a promising solution, engineered to exploit the low-oxygen conditions of PDAC for site-specific drug delivery, improved penetration, and reduced off-target toxicity.25 This section reviews the major design strategies for such NPs, organized by trigger mechanisms, targeting ligands, structural optimization, and multi-stimuli responsiveness, with a focus on mechanistic insights, practical examples, and clinical relevance. To provide a visual overview, the four major strategies for designing hypoxia-responsive nanoparticles in pancreatic cancer—trigger mechanisms, targeting ligands, multi-stimuli responsiveness, and physicochemical optimization—are schematically illustrated in Figure 4. This framework sets the stage for the detailed subsections that follow.

Figure 4 Strategies for hypoxia-responsive nanoparticles in pancreatic cancer therapy. This schematic illustrates four core strategies for designing hypoxia-responsive nanoparticles tailored for PDAC. (A) Trigger Mechanisms: Hypoxia-sensitive groups (e.g., azo, nitroimidazole) and prodrugs enable localized drug release under low oxygen. (B) Targeting Ligands: Ligands like antibodies, peptides, or aptamers enhance tumor selectivity and cellular uptake in hypoxic PDAC. (C) Multi-Stimuli Responsiveness: Integration of hypoxia with other triggers (e.g., acidic pH, enzymes, light) offers precise control of drug release. (D) Physicochemical Optimization: Core–shell structures, size shrinkage, and charge switching improve penetration into dense tumor tissue. Together, these strategies enable smart nanoparticle systems to overcome PDAC’s hypoxic microenvironment.

Hypoxia-Triggered Drug Release Systems

One of the key strategies for designing hypoxia-responsive NPs for PDAC is the incorporation of oxygen-sensitive drug release mechanisms. These systems exploit the characteristically low oxygen levels of the TME to trigger the selective release of therapeutic agents, thereby maximizing drug accumulation at tumor sites and minimizing systemic toxicity. Central to this approach is the use of hypoxia-sensitive linkers, prodrugs, and structural transformation mechanisms that remain stable in normoxic tissues, but are activated under hypoxic conditions.

Such oxygen-dependent systems can be broadly categorized into four main subtypes: (i) bioreductively cleavable linkers that undergo bond cleavage via enzymatic reduction in hypoxia, (ii) hypoxia-activated prodrugs that release active cytotoxins only after enzymatic conversion, (iii) NPs that undergo physical disassembly under hypoxic conditions to facilitate deep tumor penetration and co-release of payloads, and (iv) matrix degradation mechanisms in which the NP structure itself becomes destabilized under low oxygen tension. The following subsections provide a detailed analysis of each mechanism, including the representative materials and their applications in PDAC models.

Bioreductively Cleavable Linkers (Eg Azo, Nitroimidazole)

Bioreductively cleavable linkers are chemical bonds, such as azo, nitroaromatic, or quinone groups, that remain stable in oxygen-rich tissues but are selectively cleaved in hypoxic environments due to enzymatic reduction.86 In normoxic conditions, oxygen serves as an electron acceptor and prevents the reduction of these linkers, keeping the nanocarrier intact as it circulates through healthy tissues. However, when NPs reach hypoxic tumor regions, the lack of oxygen allows reductase enzymes, which are upregulated in these environments, to reduce the linkers (for example, converting azo to aniline or nitro to amine), resulting in the cleavage of the bond and subsequent release of the conjugated drug.73,86 This mechanism functions as a molecular switch governed by local oxygen levels.

A major advantage of this strategy is that it enables highly selective drug release within hypoxic tumor zones, thereby improving the therapeutic index and minimizing damage to healthy tissues.25 The chemistry of these linkers is versatile, allowing their integration into a variety of nanocarrier platforms—such as polymers, liposomes, and dendrimers—without significantly compromising their stability in circulation. This approach also makes it feasible to use potent cytotoxic agents that would otherwise be too toxic for systemic administration.

However, the effectiveness of bioreductive linkers is highly dependent on the degree of hypoxia present in the tumor.96 Regions with only moderate hypoxia may not trigger complete linker cleavage, leaving part of the drug payload inactive and some untreated tumor cells.97 In addition, the expression of the necessary reductase enzymes can be heterogeneous within tumors, leading to inconsistent drug activation. There is also a risk that normal tissues with physiologically low oxygen levels, such as the bone marrow, could inadvertently activate the linker, causing off-target effects. For example, the nitroimidazole-linked prodrug evofosfamide (TH-302) has shown efficacy in targeting hypoxic regions of PDAC, but its effectiveness is limited to the deepest tumor pockets where drug activation is incomplete;25 additional strategies, such as inducing acute hypoxia, have been explored to enhance its performance.97

HAPs and Enzyme-Mediated Release

HAPs are designed as inactive drug derivatives that become cytotoxic only after enzymatic reduction in low-oxygen environments.9 These prodrugs exploit hypoxia-induced enzymatic pathways, such as one-electron reductases, to unmask potent drugs specifically within the tumor. When formulated into NPs, HAPs can be delivered more efficiently to tumors, further reducing systemic toxicity.98 In oxygenated tissues, oxygen inhibits the reduction cascade by re-oxidizing the radical intermediate, thereby preventing prodrug activation and release. In contrast, in hypoxic tumor regions, reduction generates a free radical or an active metabolite that binds to or damages cellular targets. Similarly, enzyme-mediated release strategies take advantage of tumor-associated enzymes, which are often more active under hypoxia, to cleave bonds or degrade NP matrices, thereby releasing the drug payload.99

The key benefit of this approach is its ability to selectively kill hypoxic tumor cells, which are often resistant to conventional therapies. As hypoxia is a common feature of many solid tumors, HAP-based NPs can target a broad range of cancers without the need for a unique cell surface receptor. This strategy also complements therapies that are more effective in well-oxygenated tumor regions, helping to overcome tumor heterogeneity.

Nonetheless, there are some limitations. Partial activation of prodrugs at the hypoxic margin can result in drug release into nearby normoxic cells, causing off-target toxicity. Conversely, insufficient activation deep within the tumor core may leave some cancer cells viable. The efficiency of enzymatic activation can also be influenced by tumor metabolism and the availability of reducing equivalents such as NADPH. For instance, the prodrug AQ4N (banoxantrone) is converted to a DNA-binding agent only in hypoxic cells, demonstrating effective hypoxia-selective chemotherapy in PDAC models,100 but the extent of activation depends on the redox environment of the tumor.9

Structural Disassembly Under Hypoxia

Some NPs have been engineered to undergo physical transformations or disassembly, specifically in hypoxic zones. These structural changes, such as core breakdown, shell detachment, and size reduction, are triggered by hypoxia and can release the drug payload and facilitate deeper penetration into the dense stroma of PDAC. For example, micelles crosslinked via hypoxia-labile bonds remain stable during circulation but disassemble into smaller subunits within hypoxic tissue, whereas protein-based NPs stabilized by hypoxia-sensitive crosslinks dissolve under low oxygen.73,101

The main advantage of this strategy is that it enables large NPs to break into much smaller fragments (often sub-50 nm) within the tumor, allowing them to penetrate more deeply into the pancreatic cancer tissue. Additionally, this approach allows the co-release of multiple therapeutic agents (such as drugs and siRNA) if they are co-encapsulated, enabling precise combined action in the hypoxic tumor core.

However, achieving strict specificity for hypoxic sites remains challenging. If the disassembly trigger lacks strict hypoxia selectivity, premature fragmentation can occur in circulation or normoxic organs, leading to off-target drug release. Furthermore, controlling the rate and extent of disassembly can be difficult. Once triggered, the process may result in a rapid, uncontrolled burst of drug release. For example, lipid NPs with hypoxia-sensitive components have been shown to destabilize and release gemcitabine specifically in hypoxic PDAC spheroids, improving penetration and efficacy; however, the design must be carefully tuned to avoid premature leakage.88

Hypoxia-Targeting Ligands and Surface Functionalization

To enhance the precision of NP delivery to hypoxic regions in PDAC, surface functionalization with hypoxia-targeting ligands has emerged as a critical strategy.102 This approach leverages molecular markers uniquely overexpressed in hypoxic TME, enabling NPs to actively home in on oxygen-deprived zones rather than relying solely on passive accumulation via the enhanced permeability and retention effect.103 Three primary ligand classes—aptamers/peptides, monoclonal antibodies (mAbs), and hypoxia-trapping moieties—have shown promise in guiding NPs to hypoxic niches, though each comes with distinct advantages and challenges.103

Aptamers and Peptides Targeting Hypoxia-Induced Markers

Aptamers (short DNA/RNA sequences) and tumor-homing peptides are engineered to bind hypoxia-associated proteins such as CAIX and p32, which are upregulated under low-oxygen conditions.103–105 For example, LyP-1, a cyclic peptide, targets the p32 protein expressed on tumor cells and macrophages in hypoxic regions. By conjugating NPs with LyP-1, researchers achieved selective accumulation in poorly oxygenated tumors, triggering apoptosis in hypoxic cancer cells. Similarly, RNA aptamers specific for CAIX have been used to functionalize nanocarriers, enhancing their targeting efficiency in hypoxic PDAC regions both in vitro and in vivo.106

The strength of aptamers and peptides lies in their high molecular specificity, which minimizes off-target effects, while allowing repeated dosing owing to their low immunogenicity.107 Their small size enables dense surface conjugation, fostering multivalent interactions with target cells. However, these ligands have limitations: hypoxia-induced markers, such as CAIX, are not uniformly expressed across all hypoxic cells, resulting in heterogeneous targeting efficacy.108 Additionally, peptides and aptamers are susceptible to degradation in vivo and often require chemical modifications (eg, PEGylation) to improve stability. The dense stromal barrier of PDAC further complicates ligand efficacy, as targeted NPs may struggle to penetrate deeply into the hypoxic cores.109

Antibody Conjugates Against Hypoxia-Associated Proteins

mAbs offer high-affinity targeting of extracellular hypoxia markers such as CAIX, a clinically validated marker overexpressed in PDAC.104,110,111 Antibody fragments or nanobodies can reduce size while retaining specificity.104 For instance, gold nanorods conjugated with anti-CAIX antibodies accumulated preferentially in hypoxic tumor regions, enabling focused photothermal therapy (PTT) upon near-infrared (NIR) laser irradiation.104 This approach not only localized treatment to resistant hypoxic zones but also synergized with thermal ablation to reduce tumor growth.104

The advantages of antibody-based targeting include unparalleled specificity and potential for multifunctionality, which can simultaneously deliver therapeutic payloads and block pro-tumor signaling pathways.110 However, their large size (~150 kDa) limits penetration beyond the perivascular regions, reducing their efficacy in deeply hypoxic zones.104 Immunogenicity is another concern because non-humanized antibodies may trigger immune responses or rapid clearance. Furthermore, the complexity and cost of antibody production pose scalability challenges compared with smaller ligands.110

Hypoxia-Trapping Moieties (Eg, 2-Nitroimidazole Binding)

A distinctive approach involves “trapping” molecules such as 2-nitroimidazole derivatives, which covalently bind to cellular macromolecules under hypoxia.97,112,113 For example, pimonidazole analogs attached to polymeric NPs accumulate in hypoxic tumor regions via covalent binding after