Introduction

Human serum albumin (HSA) is among the most extensively studied carriers in cancer therapy because of its unique characteristics, such as biocompatibility and biodegradability.1 Albumin is also actively scavenged by cancer cells due to its high capacity for fatty acid binding and its glutamine content, which makes up approximately 10% of its amino acid composition. This makes albumin a valuable source for fueling the Krebs cycle and lipid biosynthesis, as well as providing nitrogen necessary for amino acid and nucleotide synthesis.2–7 Interestingly, hypoalbuminemia in cancer is frequently associated with cancer chemoresistance due to reduced drug transportation (reviewed in).8

Albumin structure is amenable to modifications, also allows for the efficient design of active targeting strategies, specifically directed at particular receptors. Thus, it enables both passive and active (via receptor-mediated pathways) targeting of cancer cells.

Passive targeting results from the prolonged circulation time of human serum albumin (HSA). The growth of cancerous tissue causes local swelling (due to a reduction in lymphatic flow), increasing the accumulation of albumin in the interstitial space. This accumulation is caused by a significant increase in vascular permeability called the enhanced permeability and retention (EPR) effect, reported by Maeda and Matsumura in 1986. Hoogenboezem and Duvall provided a comprehensive review of the relationship between the EPR effect and albumin accumulation in tumor tissues, highlighting its role as a drug carrier in oncology.9 Despite passive drug targeting, clinical evidence suggests that tumor accumulation of organic compounds, including albumin-based nanoparticles, typically only reaches less than 1% of the administered dose, and satisfactory outcomes are seldom achieved solely through the enhanced permeability and retention effect strategy.10,11

This is because mere accumulation at the target site is insufficient for effective drug delivery. The issue of drug penetration into tumor tissue remains complex and is dependent on factors such as the tumor’s vascularization. Furthermore, a critical question is – is it possible for the nanoparticle or drug to enter the cancer cell to excerpt therapeutic effect? To precisely address this question and tackle it effectively, it is essential to understand how a cancer cell internalizes a given particle and whether the process could be enhanced or bypassed. Targeting and uptake can be improved by incorporating ligands into the nanoparticle that are known to effectively bind to receptors on the cancer cell surface. On the other hand, resistance mechanisms can actively expel the particle from the cell, preventing it from interfering with the cancer cell’s life pathways.

It is estimated that therapy resistance is responsible for almost 90% of deaths caused by cancer.12 The resistance can be pre-existing, also described as primary, innate or intrinsic. This means that the patient is characterized with a certain quality of cancer cells or tissues that causes therapy to be ineffective. Also, in terms of chemotherapy, such resistance is present before the treatment. On the other hand, due to the extraordinary plasticity, cancer cells can get resistant during the treatment. Such resistance is called acquired, secondary or extrinsic.13 One of the factors that can emerge as a result of both primary and secondary resistance is perturbation of cell influx/efflux mechanisms. In 1976, Juliano and Ling reported a specific membrane protein of 170 kDa, which was the reason for colchicine resistance in cell culture.14 They called it P-glycoprotein (P-gp, multiple drug resistance type 1 – MDR1). MDR1 belongs to the ATP-binding cassette (ABC) transporters, responsible for the distribution, absorption and secretion of unspecific substrate repertoire.15,16 Once MDR1 is expressed on a cancer cell, its nonspecificity enables cross-resistance to various drugs, a phenomenon known as multidrug resistance. MDR1 is not only a concern confined to cytostatic therapy; evidence has demonstrated that certain targeted therapeutics, such as tyrosine kinase inhibitors, are able to interact with MDR pumps.16 Using nanoparticles that prevent expulsion through MDR, enable drug release within the cancer cell and are readily internalized by the cancer cell, like albumin, a seemingly essential nutrient, may serve as a “wolf in sheep’s clothing”—a strategy that deceives cancer cells and significantly hinders the development of drug resistance.

This review focuses on the unique molecular features of human serum albumin that make it a versatile carrier for drug delivery, particularly in overcoming cancer cell resistance. While much has been written about the general effectiveness and methods for synthesizing albumin-based delivery systems,9,11,17–20 our review takes an interdisciplinary approach, integrating insights from chemistry, cell biology, and translational medicine. We highlight how specific chemical modifications to albumin can improve its ability to bypass resistance mechanisms, especially those related to MDR1 (Figure 1). By examining the interplay between albumin’s chemical properties, its cellular interactions, and the potential for manipulating these to overcome specific oncological barriers, we emphasize the translational relevance of this work. Rather than focusing solely on how the carrier reaches cancer cells, we investigate how the choice of receptor can influence the success of overcoming resistance mechanisms. Drawing on key studies, including the clinically used formulation Abraxane and new formulations – Pazenir and Fyarro, we identify areas where current knowledge is lacking and propose strategies to enhance the therapeutic impact of albumin-based drug delivery systems. Our goal is to shed light on how albumin can be optimized to break through cancer resistance and improve treatment outcomes. Although various sources of albumin are commonly used (as reviewed extensively in21), our attention will be given to human origin, as there is some evidence that the source of albumin carrier might influence the risk of allergic reactions,20 which is highly important for potential use in immunotherapy.9

Figure 1 Human serum albumin as a modifiable carrier to target cancer cells. Folate receptor, gp60 (glycoprotein 60), CD44 (cluster of differentiation 44), and FcRn (neonatal Fc receptor) are involved in the uptake and transport of albumin-based carriers.

Basic Properties of Human Serum Albumin Human Serum Albumin Function

Human albumin is the most abundant, with an average concentration of 42 grams per liter (equivalent to 632 µM)22 and a half-life of 19 days, which allows it to maintain a prolonged presence in circulation.23 During its life of 27 days, an average albumin molecule makes about 15000 trips around the circulatory system plus about 15 trips through extravascular spaces.24,25 Approximately 120 grams of albumin are present within blood vessels, and due to its relatively low molecular weight compared to other plasma proteins, it plays a crucial role in maintaining about 80% of the colloid osmotic pressure. This function facilitates the movement of fluids between tissues and the bloodstream, helping to retain fluid within the blood vessels.19,26,27

The highly organized structure and multiple binding sites of HSA enable it to transport various compounds, including ions, fatty acids, hormones, and other molecules, serving as a high-capacity reservoir that stabilizes the concentration of free ligands. Albumin is indeed a crucial protein in the human body functions primarily as a carrier protein, which is vital for transporting various substances in the bloodstream and controlling their active concentration, making them an important part of drug delivery.28–30 The transport of metabolites involves uptake by albumin at one organ and release to the cells of another. However, the failure of protein therapy in clinical trials has been attributed to poor knowledge of protein pharmacokinetics, as well as the uncontrolled spread of proteins when they are administered systemically.

Furthermore, HSA extends beyond its traditional functions in transport, osmotic regulation, and detoxification, playing a crucial role in cellular defense through its significant antioxidant properties. HSA is central to combating oxidative stress, a process closely linked to various diseases and aging. It contributes to cellular antioxidant defense by scavenging free radicals, binding metals, participating in redox reactions, and preventing lipid peroxidation.31–33 Additionally, HSA serves as a biomarker for several diseases, including cancer, rheumatoid arthritis, ischemia, and obesity, and is clinically used to treat conditions such as blood loss, hemorrhage, chronic and acute liver failure, and hypoalbuminemia.34

Human Serum Albumin Structure

The liver synthesizes HSA, a non-glycosylated protein chain of 585 amino acids with a molar mass of 66,439 Daltons.35 Human serum albumin and its derivatives, including recombinant HSA (rHSA) and the albumin dimer (dHSA), have well-known primary and secondary chemical structures. These molecules have a globular shape and are relatively stable under normal physiological conditions.

The crystalline structure of HSA is predominantly composed of α-helix formations, which make up 69% of its overall structure. The sequence of albumins is characterized by a unique arrangement of subunits linked by disulfide bonds. This configuration includes 35 cysteine residues that are integral in forming 17 disulfide bridges, which are crucial for maintaining the molecule’s spherical stability.31,33

The molecule is organized into three repeating domains: I (residues 1 to 195), II (residues 196 to 383), and III (residues 384 to 585), each displaying a similar structural motif (Figure 2).

Figure 2 Modular domain structure of the HSA molecule. The panel illustrates the structure of the HSA molecule, with its subdomains distinctly color-coded. The molecule is organized into three repeating domains, each exhibiting a similar structural motif: Domain I (IA: residues 1–112, IB: residues 112–195), Domain II (IIA: residues 196–303, IIB: residues 303–383), and Domain III (IIIA: residues 384–500, IIIB: residues 500–585).

Across these three domains, a total of ten helices are arranged in antiparallel, presenting a surface of hydrophobic residues oriented toward the interior of each domain. These helices are further grouped into two subdomains: Subdomain A consists of six helices (h1 to h6), whereas Subdomain B contains four helices (h7 to h10).5,30,36,37

The amino acid composition of albumins is distinguished by a low abundance of tryptophan, glycine, and methionine, along with a high concentration of cystine and charged amino acids, such as aspartic acid, glutamic acid, lysine, and arginine.25 The albumin molecule is highly polar, possessing approximately 100 negative and 82 positive charges, with the net negative charge being most pronounced in the N-terminal domain and lowest in the C-terminal domain, resulting in an asymmetric charge distribution. Human albumin’s high solubility and stability are attributed to its overall net charge of −15 at neutral pH and the numerous disulfide bonds within its structure. Therefore, developing an HSA-based delivery system requires determining the electrokinetic charge of HSA, as this information is crucial for understanding drug adsorption mechanisms and ensuring optimal functionality. It is important to note that both electrokinetic and surface charges should be understood as net (averaged) values representative of the entire molecule, forming regions of positive and negative charges across a broad pH range.Various techniques are employed to estimate or predict a protein’s isoelectric point (IEP), with one approach involving computational methods that consider factors such as crystallographic structure, electrostatic interactions, and amino acid properties to determine the IEP of human serum albumin, which ranges from 5.67 to 5.9.38,39 When dispersed in aqueous solutions, HSA has a positive electrokinetic charge at pH values below 5 because of specific amino acid residues, such as cysteine (35), tyrosine (19), lysine (59), and arginine (23). However, when the pH of these solutions reaches 5, the electrokinetic charge of the molecule becomes negative, primarily because of the presence of glutamic acid (59) and aspartic acid (41) residues.40 This shift establishes the molecule’s isoelectric point at 5.1.41 Depending on the methodology used, the isoelectric point values for human albumin can range from 4.7 to 5.8.38,40,42,43 The differences observed when comparing experimental and computational methods for determining IEP can be attributed to variations in amino acid composition, the presence of fatty acids, or potential errors in measurement techniques.44–46

Ligands Binding and Functional Sites

A defining characteristic of albumin is its ability to reversibly bind a wide range of endogenous and exogenous ligands. Most of these interactions rely on hydrophobic bonds that affect albumin’s structural flexibility, enabling it to adapt its shape to accommodate various ligands, which likely explains its extensive binding capacity. The majority of ligands bind reversibly, with typical association constants (Ka) ranging from 104 to 106 M−1.

Studying the pH-dependent conformational isomerization of HSA offers valuable insights into its properties and interactions with various molecules, including drugs. These pH-induced changes in HSA are particularly significant in pharmacology and biochemistry, as shifts in pH lead to notable alterations in protein shape and structure. The abundance of acidic and basic amino acid residues in HSA allows it to undergo reversible conformational changes across different pH values (Figure 3).

Figure 3 Proposed configurations and interrelation of the N, F, and E isomeric forms of HSA by Carter and Ho (1994) and Peters (1985).

At pH values below 2.7, HSA takes on an elongated structure known as the “E-form.” Interestingly, between pH 2.7 and 4.3, HSA adopts a different form, termed the “F-form”, which is characterized by increased viscosity, reduced solubility, and a decrease in the alpha-helix content. From pH 4.3 to 8.0, HSA primarily exists in the “N-form”, recognized by its heart-shaped structure. When the pH exceeds 8.0, HSA transitions to the “B-form”, which is characterized by a reduced alpha-helix content and a greater affinity for certain ligands than the “N-form”.30,34 The spherical arrangement of domains in HSA gives each HSA monomer allosteric properties, enabling it to interact effectively with a diverse array of ligands. This allosteric nature allows HSA to undergo conformational changes, which alter its binding sites and enhance its ability to bind and transport various molecules efficiently. This structural flexibility and adaptability in ligand interactions make HSA essential for transporting substances within the bloodstream.

Figure 4 Modification and ligand binding sites on human serum albumin molecules: Tryptophan Residue 214 (Purple), Lysine Residues and N-terminal Primary Amine (Yellow) as Potential mPEG/PEG-NHS Binding Sites, and FA Binding Site (Light Green) with Structural Illustration (Orange) and Labeled Surrounding Amino Acids.37

Tryptophan in the HSA Structure

The tryptophan residue in HSA (Figure 4) is particularly noteworthy because of its strategic position near the hydrophobic cavity of subdomain IIA, the principal active binding region of HSA (eg aspirin, warfarin, diazepam, ibuprofen).25 This area also contains two binding sites, Sudlow I and II, known for their significant binding affinities classified based on studies on molecular interactions with fluorescent probes. The binding locus of large heterocyclic compounds possessing a negative charge was identified as Site I, with Site II showing a preference for small aromatic carboxylic acids. Consequently, the tryptophan residue is often utilized in spectroscopic studies and is regarded as a valuable tool in this area of research.36,47,48

Lysines in the HSA Structure

HSA contains 59 lysine residues with free N-termini (primary amino groups) that have become a key focus for structural modifications of the protein (Figure 4). Lys-199 interacts with drugs of Sudlow’s Site I, corresponds to the hydrophobic pocket in subdomain IIA, and his Site II to the pocket within IIIA. The reactivity of Lys-199 may be attributed to its unusually low pKa of 7.9,24,25 which can be rationalized based on its close interaction with His-242. These residues readily react with N-hydroxysuccinimide (NHS) esters and propionaldehydes, allowing significant alterations to the protein structure without causing denaturation. One common modification involves conjugating polyethylene glycol (PEG) or polyethylene glycol monomethyl ether (mPEG) through reactions between the primary amines of lysine residues and NHS esters or propionaldehydes.49,50

Folic Acid Binding Site in the HSA Structure

Human albumin has a binding site for folic acid (FA) located within domain I (Figure 4). The ligand fits into the protein’s structure through noncovalent interactions, primarily hydrogen bonds. Structural analysis revealed that folic acid binds to HSA through a combination of hydrophilic and hydrophobic contacts, with a total binding constant in the range of 104 to 105 M−1.51 In the folic acid-HSA complex, folic acid is surrounded by Ile-142, Leu-185, Tyr-138, and Trp-161.51 These findings indicate that HSA can effectively bind and transport folic acid, highlighting its potential for further exploration in the context of folic acid utilization in complex with HSA.

Albumin in Cancer

Despite more than 30 years since identification of albumin receptors, and numerous studies and scientific reviews on their expression and function,1,9,19,34,52,53 the precise mechanisms by which they operate remain poorly understood.54,55 This knowledge is however crucial to identify cancer types that respond efficiently to albumin-based drug delivery systems and also to improve efficacy of existing therapies, also in the context of drug resistance.56

Mechanisms of Albumin Uptake in Healthy and Cancer Cells

Albumin is not merely a passive protein circulating in bodily fluids, it plays a dynamic role in cellular physiology, tailored to the needs of the environment. Albumin uptake operates through two general mechanisms: endocytosis and micropinocytosis. Endocytosis is specific to substrate-receptor binding, which involves frequently auxillary role of clathrin or caveolin-1 (CAV1). Clathrin-mediated endocytosis operates as a stochastic process, suggesting a more random and ubiquitous mechanism for internalizing molecules, whereas CAV1-dependent endocytosis is spatially restricted to lipid raft regions, indicating a more organized and compartmentalized pathway. This distinction highlights the potential for differential targeting strategies in drug delivery, depending on the desired endocytic route and cellular localization.57,58 Albumin uptake in healthy cells is primarily mediated by caveolin-1 (CAV-1), however clathrin-mediated endocytosis and macropinocytosis-like processes have been described in pulmonary epithelium or skeletal muscle primary cells.59,60 Macropinocytosis is not specific to the substrate, but it involves cytoskeleton movement it is actin-dependent. This process enables sampling of large volumes of extracellular fluid and is considered a preferred mechanism for albumin scavenging by cancer cells.61 It is a form of nutrient sensing and may facilitate survival in nutrient-poor environments.2,4,7

Cancer patients often experience hypoalbuminemia, a condition associated with cachexia.62 This reduction in albumin levels is thought to stem, at least in part, from the disruption of albumin recycling mechanisms that normally maintain its balance in bodily fluids. Under physiological conditions, this recycling is orchestrated by glycoprotein 60 (gp60), the neonatal fragment crystallizable (Fc) receptor (FcRn) and the cubulin/megalin complex in kidneys. However, in the tumor microenvironment, these mechanisms face significant challenges. Driven by the demands of unchecked growth, tumor cells scavenge albumin from their environment. They employ micropinocytosis and upregulate receptor-mediated pathways to access essential nutrients, such as fatty acids and iron. Notably, receptors like glycoprotein 18/30 (gp18/30), Cluster of Differentiation 36 (CD36) or Transferrin receptor protein 1 (TfR1) play key roles in this process, making them attractive targets for therapeutic strategies (Table 1). A deeper understanding of these pathways is essential to justify modifications that circumvent resistance mechanisms, paving the way for more effective interventions.

Table 1 The Ability of Selected Albumin-Recognizing Receptors to Improve Cancer Cell’s Targetability and Nanoparticle Design

gp60

The first and most described receptor responsible for native albumin binding is gp60, also known as albondin. It serves as a key mediator of native albumin transcytosis across endothelial cells, which is essential for maintaining vascular oncotic pressure and delivering nutrients bound to albumin to tissues where they are needed, including tumor microenvironment63 (Table 1). Binding of albumin to gp60 initiates a CAV-1-dependent formation of caveolae – a structure that facilitates transcellular transport (Figures 5 and 6).64 CAV-1 correlates with a favorable response to Abraxane, suggesting that gp60 might be involved in Abraxane uptake, it is unclear whether gp60 would be able to recognize albumin-based nano carriers.64 Despite its importance, research on gp60 remains limited by the lack of specific tools to study its mechanisms in detail.56,65 Several sources reference a 1984 paper by Schnitzer et al, describing a sequence of gp60 derived from bovine leukemia virus envelope proteins,66 later identified as the gp60 albumin receptor.53 However, up to date this sequence is not deposited in any database. Some groups have reported the use of antibodies for immunofluorescent labeling directed against gp60/110 isolated from human herpesvirus 6,56,67 but it is not suggested by the antibody producer that it can be used to visualize human gp60. Another discrepancy in the literature concerns the localization of the gp60 receptor. While it is widely accepted that gp60 receptors are located on endothelial cells, some sources, perhaps erroneously, report their presence on the surface of cancer cells.68 This possibility cannot be entirely excluded, given the lack of appropriate investigative tools and limited data on gp60 expression. Understanding the nuances of gp60-mediated albumin transcytosis could reveal new therapeutic targets, particularly in diseases where albumin transport is co-opted for pathological purposes.

Figure 5 Gp60 recycles undamaged albumin. Albumin is recognized by gp60 glycoprotein receptors on the cell surface (I). After binding, albumin is internalized via caveolin-1-mediated endocytosis (II) and becomes entrapped in an endosome (III). Because pH changes due to H+ ATPase activity, the FcRn receptor located in the endosome subsequently recognizes albumin (IV), which prevents it from further lysosomal degradation. Endosomes merge with the cellular membrane, and the pH increases, resulting in albumin release (V).

Figure 6 Pathways of albumin acquisition by cancer cells. Albumin or albumin-based nanoparticles present in the bloodstream (A) can enter the tumor interstitial space through the leaky vasculature or via transcytosis mediated by the gp60/caveolin-1/FcRn pathway. (B) Alternatively, albumin may be internalized by macropinocytosis, where the invagination of the cell membrane encloses both albumin and the FcRn receptor within an endosome. This process protects albumin from degradation, allowing it to be released into the extracellular milieu. A detailed description of this process is provided in Figure 5. In the interstitial space, albumin or albumin-based nanoparticles can interact with the SPARC (secreted protein acidic and rich in cysteine) protein and be endocytosed in a clathrin-dependent manner, potentially entering the lysosomal degradation pathway. (C) An additional pathway involves the gp18/30 system, where albumin or albumin-based nanoparticles are internalized and degraded within the lysosomal compartment. This process is further detailed in Figure 7.

Neonatal Fc Receptors

Albumin recycling is possible because of FcRn receptors, that are widely expressed in various tissues.69–75 In mice, FcRn deficiency caused rapid albumin turnover and 50% reduction in the half-life in blood serum.76,77 This effect was irrespective of increased albumin production in those animals.26,76

FcRn-mediated albumin recycling is an elegant, pH-dependent mechanism (Figure 5) Figure 3. FcRn does not interact with albumin at neutral pH (pH 7.0), but binds it efficiently at acidic pH (6.5–5.5). When albumin is internalized either by gp60-receptor or by micropinocytosis, it enters a cell with an endosome at neutral pH. When pH within endosome decreases, it activates FcRn to bind albumin. This event prevents albumin degradation in the lysosome. The endosome then releases its contents into the serum, where the pH is approximately 7.0–7.4. The binding affinity between albumin and FcRn decreases again, allowing the release of albumin back into the extracellular space.27,77,78 Notably, the tumor extracellular microenvironment is characterized by intense acidification, with pH values ranging between 6.4 and 7.0.79 While this acidic milieu theoretically may influence the release of albumin into the extracellular space, this aspect remains unexplored in the literature. The literature is also inconsistent in the meaning of the FcRn expression level in cancer. There is a significant reduction in FcRn expression, associated with poor prognosis in many cancer types.80,81 Swiercz et al has also demonstrated that increased FcRn expression prevents albumin scavenging and instead promotes albumin recycling, which reduces glutamine levels in cancer cells and decreases the number of metastases.80 In contrast, Larsen et al reported that FcRn overexpression in human cancer cells promotes albumin recruitment and cellular growth in breast and colorectal cancer cell lines. As proof of principle, human FcRn knockout in a mouse xenograft significantly decreased the tumor volume. Additionally, the authors reported that FcRn was significantly overexpressed at the protein level in eight out of ten investigated cancer types, including breast, cervical, pancreatic, lung, colorectal, head and neck and renal cancer. In the discussion, the authors underline the need for further research in terms of albumin acquisition by cancer cells in the context of future targeted albumin-based therapies.82 The evidence points to a deeper story, where the interaction between FcRn and cancer emerges as a pivotal factor in shaping the success of albumin-based therapies. This discrepancy hints that the therapeutic outcome may depend significantly on FcRn expression levels, varying across cancer types and individual patients.

SPARC

Research conducted from the late 1990s revealed that secreted protein acidic and rich in cysteine (SPARC, osteonectin, BM-40) is highly expressed in stromal tissue and various advanced cancers.83–85 Indeed, SPARC plays a significant role in neovascularization,53,86,87 metastasis, and cancer cell survival, underscoring its critical contribution to tumor progression.53,88–90 SPARC regulates cancer interactions with the extracellular matrix (ECM) and has a strong affinity for albumin, which can lead to albumin accumulation in inflamed and tumorous tissues.91 SPARC-bound albumin facilitates the gp60/CAV-1-mediated transcytosis and simultaneously protects and directs albumin to cancer cells (Figure 6). Main SPARC-albumin receptor on the cancer cell surface is α5β1 integrin complex.60,92 The internalization of SPARC depends on clathrin, and the subsequent fate of albumin remains uncertain.60,93 SPARC regulates transport from early endosomes to late endosomes, suggesting that albumin enters the lysosomal pathway.3,52,60,94 However, some studies report that albumin may become trapped in endolysosomal or autophagosomal compartments.95 Interestingly, Abraxane appears to be rescued from lysosomal degradation by SPARC, which inhibits the cytotoxic effect of Abraxane but not paclitaxel.65

gp18/gp30

In contrast to gp60, gp18/30 are membrane receptors that bind modified or damaged albumin,96–98 for instance ischemia-modified albumin (IMA).99 These receptors were identified via agarose gel electrophoresis98 and first described by Schnitzer JE & Bravo J in 1986.98 After receptor recognition, albumin is internalized through macropinocytosis and directed to the lysosomal degradation pathway, making it a key route used by cancer cells for albumin scavenging (Figure 7). Damaged vasculature and the pressure exerted by the growing tumor on surrounding tissues, leads to inadequate nutrient supply through blood vessels. As a result, tissues resort to alternative means for essential amino acids to suffice cellular anabolism. It is unknown to what extent are those receptors involved in HSA nanoparticles intake, however Abraxane, initially hypothesized to be internalized through the gp60/CAV-1 system, is actually transported via the denatured albumin transport system involving gp18/30.65 The pathway of nanoparticle internalization is critical, as it determines the intracellular fate of the cargo. While degradation before reaching the molecular target can reduce therapeutic efficacy, this is not the case for Abraxane. Transport via gp18/30 still allows for efficient delivery of paclitaxel, outperforming administration without a carrier. This mechanism also holds significance for overcoming treatment resistance, a topic explored further in the subsequent sections.

Figure 7 Gp18/30 recognizes damaged albumin and targets it for degradation. Damaged albumin is recognized by glycoprotein receptors on the cell surface, specifically by gp18/30 (I). Upon binding, albumin is internalized via clathrin-mediated endocytosis (II) and becomes entrapped in an endosome (III). The activity of H+ ATPase leads to a reduction in pH (IV). The endosome subsequently merges with the lysosome, and degradation enzymes are released (IV). Albumin is digested into amino acids, and its fatty acid cargo is salvaged which can be further used by cancer cells (V).

CD36

The interaction between albumin and the cluster of differentiation 36 (CD36) transmembrane receptor is well-documented in the literature, yet therapeutic approaches targeting this specific interaction remain largely unexplored. Cancer cells, notably, often overexpress CD36, a receptor essential for the uptake of long-chain fatty acids and lipoproteins, and which plays a pivotal role in immunological recognition, inflammation, cell adhesion, and apoptosis.100,101 Albumin, both in its native and modified forms, serves as one of its primary ligands.102 A series of experiments conducted by Raheel et al shed light on the functional relevance of this interaction, demonstrating that CD36 mediates albumin transcytosis across the dermis, and, to a lesser extent, the pulmonary microvascular endothelium. Interestingly, they found that this process of transcytosis could be saturable and exhibited variability depending on the tissue type, suggesting the potential for targeting this pathway in therapeutic applications.54

Transferrin Receptor Protein 1

TfR1 plays a key role in cancer cell survival through NF-κB activation,103 stimulation of neoangiogenesis104 and altered energy metabolism, leading to its overexpression in various cancers.103,105,106 Despite efforts to use anti-TfR1 antibodies for targeted drug delivery, safety concerns have hindered their success.107 TfR1 can also recognizes heme-albumin, facilitating the transport of nutrients or drugs into cells.108 HSA binds heme at two stites: one unaffected by fatty acids.109–111 The role of HSA in heme scavenging and delivery has been extensively reviewed.112 Brell et al demonstrated that TfR1 recognizes heme-albumin in Abraxane, promoting endocytosis and proliferation as efficiently as iron-loaded transferrin. Inhibition of TfR1 with the monoclonal antibody mAb 5–528, educed apoptotic cells by 25%, likely due to impaired endocytosis and reduced cytotoxicity.108 The potential of heme-albumin in targeting tumors compared to unmodified albumin remains to be explored. Additionally, HSA-based nanoparticles may spontaneously incorporate hemin, enabling TfR1-mediated uptake and improving drug delivery in cancer therapy.

Bae et al proposed another potential application of increased TfR expression on cancer cell surfaces. In their study, they developed doxorubicin-loaded nanoparticles functionalized with transferrin using sulfosuccinimidyl-4-(N-maleimidomethyl)cyclohexane-1-carboxylate (Sulfo-SMCC). The transferrin modification aimed to enhance receptor-mediated endocytosis and successfully overcame multidrug resistance in MCF-7/ADR cells, which is significant given that this resistance is typically mediated by MDR1. This highlights the critical importance of targeting TfR1 as a viable strategy for bypassing MDR1-associated resistance and improving therapeutic efficacy in drug-resistant cancers.113

Albumin and Multi Drug Resistance in Cancer

Cancers possess the ability to expel harmful substances from their cells by inducing multidrug resistance (MDR) systems, such as MDR1. When albumin delivers toxic drugs to cancer cells, it raises the question of whether tumors counteract this by upregulating MDR1. Insights into this question come from years of research on Abraxane, the first albumin-based drug approved by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA). A related and equally critical question is whether albumin can circumvent an already active MDR1 system, effectively delivering the drug despite the presence of functional efflux pumps. These issues will be addressed in this section.

Resistance to Abraxane

Abraxane (ABI-007, nab-paclitaxel) is a 130 nm albumin-bound paclitaxel formulation used for various solid tumors.114 It is twice as effective as unbound paclitaxel, eliminates toxic solvents, and allows for higher dosing.114–116 Paclitaxel is a well-known substrate for MDR1.117 In a Phase 3 clinical trial (NCT00046514), Abraxane was administered to patients with metastatic breast cancer who had likely developed taxane resistance likely due to MDR1 induction. The study demonstrated that Abraxane effectively bypassed cross-resistance, even in patients heavily pretreated with solvent-based paclitaxel or docetaxel.118

Table 2 HSA-Based Nanoparticles in MDR1-Expressing Cancer Cell Lines and Their Outcomes

The potential of Abraxane to circumvent multidrug resistance (MDR) was further investigated in vitro (Table 2). Zhao et al established a non-small cell lung cancer cell line resistant to Abraxane (A549/Abr) and used RNA-Seq to compare its transcriptome with that of the parental line. They reported significant MDR1 overexpression in A549/Abr cells, a common mechanism of resistance to unbound paclitaxel, and confirmed that MDR1 contributes to resistance by reversing the phenotype with verapamil, an MDR1 inhibitor.119 In another study, Zhao et al analyzed global protein expression changes between treatment-naïve and treatment-resistant A549/Abr cells.120 These authors reported MDR1 overexpression and suggested that resistance arises from the release of paclitaxel from albumin, as MDR1 cannot export the large albumin‒paclitaxel complex. This study identified additional proteins related to lipid metabolism, the cell cycle, the cytoskeleton and apoptosis processes, including novel resistance markers such as the E3-ubiquitin ligase ring finger protein 139 (RNF139) and an enzyme involved in cholesterol synthesis, 3-hydroxy-3-methylglutaryl-CoA synthase 1 (HMGCS1).120 Similarly, Vallo et al reported that MDR1 overexpression in urothelial cell lines caused resistance to nab-paclitaxel.121 While Abraxane remains pioneering, analogous systems have been developed for the controlled delivery of various chemotherapeutics. However, the MDR1 substrate repertoire accommodates a variety of hydrophobic substances. Kong et al tested gemcitabine-loaded HSA nanoparticles in pancreatic cancer cells resistant to gemcitabine due to MDR1 and MRP1 overexpression. The nanoparticles inhibited proliferation and induced apoptosis in these resistant cells.122 The authors proposed that this effect was likely due to ATP consumption during albumin degradation, which limits ATP availability for ATP-dependent ABC transporters such as MDR1 or MRP1.

Onafuye et al developed HSA-based nanoparticles encapsulating doxorubicin and tested them in neuroblastoma cells resistant to vincristine and doxorubicin respectively. The doxorubicin-loaded nanoparticles were effective in MDR1-expressing cell lines, particularly in vincristine-resistant cells. In doxorubicin-resistant cells, the effect was present but less pronounced, possibly because the released drug faces a resistance mechanism similar to that of doxorubicin in solution.123 This highlights the possibility that multiple resistance mechanisms may be involved depending on the specific chemotherapeutic agent used.

Yuan et al investigated triple-negative breast cancer, which is characterized by a high percentage of cancer stem cells (CSCs), which are key drivers of tumor progression and metastasis.124 CSCs are chemoresistant due to mechanisms such as MDR1 overexpression and the activity of detoxifying enzymes, such as aldehyde hydroxygenase (ALDH). The authors reported that while Taxol and Abraxane had similar efficacy in reducing tumor size, Abraxane significantly reduced the frequency of CSCs by 3--9-fold and was more effectively taken up by ALDH+ cells than Taxol.125 Although multidrug resistance has not been specifically studied, the use of albumin as a carrier may enhance chemotherapy outcomes, even in cases of drug resistance.

Overall, despite conflicting evidence, albumin-based delivery systems show promise in facilitating drug transport, potentially overcoming multidrug resistance independent of active MDR1. There are possible specific modifications that would allow for active MDR1 targeting.

HSA-Based Drug Delivery in Presence of Active MDR1

Cancer cells often overexpress MDR1, reducing the efficacy of therapy by decreasing the bioavailability of therapeutics.126 Specific allosteric MDR1 inhibitors (reviewed in127), have been developed, but clinical trials have shown that they can cause significant toxic side effects due to MDR1’s role in regulating the distribution of various substances in the body.128 Therefore, enhancing drug delivery into cells may be more beneficial than inhibiting MDR1 directly (Table 2).

Yang et al developed HSA-based nanoparticles decorated with the EGFR-targeting antibody cetuximab to deliver short interfering RNA (siRNA) targeting MDR1 (siMDR1) and doxorubicin to breast cancer tissues. This nanoparticle successfully reversed multidrug resistance in vitro and in vivo with minimal off-target toxicity.129 Similarly, Liu et al coencapsulated paclitaxel and chemosensitizer-2 in folate-conjugated HSA nanoparticles. This approach aimed to reduce multidrug resistance and improve antitumor efficiency. The nanoparticles showed enhanced target specificity and reduced damage to noncancerous tissues, suggesting that folic acid receptor-mediated uptake limited the action of MDR1 (folic acid as a target for HSA modification is further described in Section 4.3).130

PEG, which is commonly used as an excipient in pharmaceuticals, can be efficiently conjugated to HSA and HSA-core nanoparticles through specific binding sites on albumin. This involves the use of functional groups, such as the -NHS ester of PEG, which bind to primary amines on the free N-termini of lysine residues within the albumin structure. This modification enables HSA nanoparticles to enhance drug delivery and inhibit MDR1 activity, improving drug bioavailability.131,132 Hodaei et al demonstrated that the effectiveness of PEG depends on its molecular weight.133 While the exact mechanism is not fully understood, it may be related to changes in membrane fluidity134,135 and direct inhibition of MDR1 through alkyl and unsaturated C‒C bonds in PEG.136

HSA Utilization in Receptor-Based Targeting Strategies

The intrinsic properties of albumin as a drug carrier offer a unique advantage in bypassing MDR1-mediated drug resistance by facilitating efficient cargo delivery to cancer cells. Yet, the challenge of multidrug resistance extends beyond simple transport—it demands strategies that simultaneously enhance membrane translocation and therapeutic specificity. By addressing these dual requirements, it becomes possible to disrupt the mechanisms that enable cancer cells to expel therapeutic agents. HSA-based nanotherapeutics employ passive targeting, mainly EPR effect, and active targeting strategies. Due to its large molecular size, passive diffusion of albumin is not feasible; instead, tissue accumulation occurs primarily through endocytosis.137 Active targeting involves functionalizing the surface of albumin nanoparticles with ligands that bind to receptors overexpressed on cancer cells (Figure 1). This approach enhances specific targeting of affected tissues, increasing the uptake of drug-loaded nanoparticles by cancer cells while reducing systemic exposure and toxicity. Thanks to albumin’s inherent properties and ease of modification, ligands can be conjugated to the protein, facilitating the internalization of nanoparticles and their cargo via endocytosis (Figure 8).

Figure 8 Utilization of human serum albumin as a versatile drug carrier.

In pharmaceutical applications, albumin serves as a crucial carrier for drugs through two primary mechanisms: albumin–drug conjugation and drug encapsulation within albumin nanoparticles. Albumin–drug conjugates include albumin-fusion proteins, drugs coupled to exogenous albumin, and prodrugs designed to bind efficiently to endogenous serum albumin. While modified albumin molecules can enhance the pharmacokinetic profiles of drugs, nanoparticle formulations improve their stability and solubility. Fan et al proposed a preclinical optimization strategy that correlates a drug’s binding affinity to albumin with its serum half-life, influencing dosing regimens and therapeutic effectiveness.138

Protein modification techniques can enhance drug delivery precision and reduce chemotherapy side effects. By chemically modifying the hydroxyl, amine, and carboxyl groups on HSA, selective binding to cancer cell receptors can be achieved. Strategies targeting tumor-specific markers, such as transferrin receptors, aptamers, targeting peptides, CD44, folate receptors, and others, have been explored in nanocarrier systems.139 Despite promising reports, only a few of these modifications have been widely implemented (Table 3). Advances in albumin modification, as described in recent studies, represent a critical step toward achieving this objective (Figure 8). In this section, we delve into the innovative modifications of albumin that not only improve drug delivery efficiency but also hold potential to overcome resistance mechanisms, paving the way for more effective cancer therapies.

Table 3 Examples of Cancer-Specific Modifications of Albumin

Aptamers

Aptamers are short single-stranded oligonucleotides capable of selectively binding to tumor tissue, offering advantages over monoclonal antibodies, such as lower cost, reduced immunogenicity, and easier high-throughput screening and chemical synthesis.152 However, their small size (approximately 20 kDa) leads to rapid renal clearance, limiting their clinical use. Conjugating aptamers with larger molecules, such as albumin, extends their circulation time in the bloodstream.153 Although aptamers have shown potential to improve cancer therapy (as reviewed in154–156), human serum albumin‒aptamer conjugates have yet to be explored in this context. There have been attempts to enhance immunotherapy with bovine serum albumin conjugated with PD-L1157,158 or CTLA-4 aptamers.159 In non-cancer applications, Zhang et al reported the double conjugation of a newly designed bone anabolic aptamer targeting sclerostin, which exhibits high binding affinity for human serum albumin, thereby extending the system’s half-life.153 These findings, along with results from aptamer-conjugated BSA nanoparticles, underscore the potential of albumin as an aptamer carrier. This research holds significant promise and is crucial for advancing translational oncology.

Hyaluronic Acid (HA)

HA is an unbranched glycosaminoglycan that is an important component of the extracellular matrix. Its synthesis increases in proliferating cells, which indicates its role in the metabolism of cancer cells.160 The HA receptor CD44 is present to varying degrees on multiple cells in the body. It plays a role in cell adhesion and migration, inhibits apoptotic pathways, promotes cell proliferation, and may also enhance early angiogenesis in tumors (detailed in161,162). While present in normal tissue, CD44 is overexpressed in multiple tumors and has been demonstrated to be correlated with increased aggressiveness and a greater risk of metastasis.163 Spadea et al reported that cancer cell lines that express standard CD44 (the shortest CD44 isoform164), which is present in high quantities on the membrane of cancer cells, can efficiently take up HA. Furthermore, owing to heightened receptor activity, the uptake of HA by malignant cells surpasses that of noncancerous cells, as evidenced by various studies.164–166 This phenomenon may enhance drug accumulation within cancerous tissues relative to healthy tissues. Importantly, CD44 is also a molecular marker specific for cancer stem cells (CSCs). Moreover, hyaluronic acid is a nontoxic, biocompatible, and biodegradable polymer, which makes it an excellent candidate for use as a targeting ligand in delivering drugs to tumors, including HSA-based delivery systems (Table 3). Shen Y et al synthesized spherical albumin nanoparticles loaded with erlotinib and coated them with hyaluronic acid via the electrostatic deposition technique. This approach enhanced drug uptake and anticancer effects in the lung cancer cell line A549, which was further confirmed in experiments using a mouse model.140 Brindisi M et al designed a similar system in which they synthesized a hyaluronic acid-human serum albumin conjugate that was further organized into redox-crosslinked nanoparticles (HAlbNPs). The HA-HSA NPs were loaded with doxorubicin (DOX). The authors tested several breast cancer cell lines with various CD44 receptor expressions and evaluated cancer cell viability and the main molecular drug resistance and metastasis pathways. The results demonstrated a selective, 2- to 8-fold increase in the efficacy of DOX against cancer cells, without an increased toxic effect on healthy mammal cells.141 HA demonstrates excellent targeting capabilities for CSCs, as demonstrated by Li et al, through the utilization of all-trans retinoic acid encapsulated within a BSA-based nanoparticle functionalized with HA (HA-eNPs). These HA-eNPs exhibited heightened binding affinity to the CD44-enriched B16F10 cell population and effectively accumulated in the lungs of tumor-bearing mice, leading to significant inhibition of metastasis.142

Folic Acids

Folates play crucial roles in cell growth, survival and proliferation.167 Because of their hydrophilic properties, folates rely on folate receptors (FRs) for the cellular internalization through receptor-mediated transcytosis.168,169 Tumor cells require folates for essential processes, a vulnerability exploited by widely used anticancer drugs such as folate antagonists.170 FOLR1 and FOLR2 folate receptor isoforms are overexpressed in many types of carcinomas, including ovarian, endometrial, kidney, lung, and brain carcinomas; mesothelioma; and myeloid leukemia.171,172 This heightened expression could also be utilized in nanoparticle design (Table 3). Ulbrich et al demonstrated that, in general, cancer cells exhibit greater binding and uptake of HSA nanoparticles conjugated with folic acid than noncancer cells do.173 Sun et al demonstrated that FA-functionalized HSA nanoparticles loaded with cabazitaxel (FA-NPs-CTX), compared with CTX-loaded human serum albumin alone, induced greater cytotoxic effects against HeLa cells overexpressing the folate receptor (FR). Moreover, in a HeLa xenograft mouse model, researchers reported a significant increase in the accumulation of FA-NPs-CTX at the tumor site, leading to inhibited tumor growth and reduced toxicity. Importantly, FA conjugation enhanced the safety and tolerability of cabazitaxel.143 In another study, Ahlawat et al reported that the functionalization of HSA nanoparticles with folic acid enhanced the targeting of hydroxycaptothecin (HCPT) in lung adenocarcinoma and increased the cytotoxic effect of the drug (actinonin) encapsulated within the