Liposomes as “Trojan Horses” in Cancer Treatment: Design, Development, and Clinical Applications

Abstract

Liposomes started to be studied for drug delivery in 1970s, taking advantage of their ability to encapsulate hydrophilic and hydrophobic drugs using biodegradable and biocompatible molecules. Nowadays, they remain one of the most promising strategies for drug delivery not only in cancer treatment but also in gene therapies and vaccines. The design and development of liposomal systems have evolved significantly over the past decades, moving from conventional formulations to advanced, stimulus-responsive, and multifunctional nanocarriers. Analogous to the myth of the Trojan Horse, liposomes must mislead the host immune system to reach the interior of cancer cells in order to deliver the therapeutic payload. There are many barriers that liposomes have to overcome to circulate through the bloodstream and specifically target cancer cells without damaging other tissues. Crucial parameters such as lipid composition, particle size, zeta potential, and PEGylation have been systematically optimized to enhance pharmacokinetics and biodistribution and to improve delivery efficiency. Furthermore, conjugation with antibodies, peptides, or small molecules has enabled active targeting, while stimuli such as pH, temperature, and enzymatic activity have been exploited for controlled drug release within the tumor microenvironment. Such innovations have laid the groundwork for translating liposomal formulations from the bench to clinical applications. In this paper, we evaluate the physicochemical features of liposomal design that underpin their suitability and efficacy for anticancer drug delivery. We aimed to focus on two main aspects: conducting an exhaustive review of the physicochemical parameters of liposomal drugs that have already been approved by regulatory agencies, while maintaining a pedagogical approach when explaining the key design parameters for the optimal design of liposomes in oncology in detail.

1. What Are Liposomes?

Liposomes are self-assembling colloidal particles in which a bilayer of lipids encapsulates a fraction of the surrounding aqueous medium. These lipids are amphiphilic in nature because they have a hydrophilic headgroup and hydrophobic tails. The polar heads orient themselves toward the aqueous medium and the hydrophobic tails constitute the inner region of the membrane and thus form a bilayer.

Since liposomes were discovered by Alec D. Bangham in the 1960s [1], they have been used to study the physical behavior of biological membranes like ion transport across bio membranes, physiochemical characterization of lipids, or protein interactions [2]. In the 1970s, Gregoriadis [3] began studying these liposomes as delivery systems for drugs or other bioactive molecules, and nowadays, they have become one of the most promising strategies for drug delivery not only in cancer treatment but also in gene therapies and vaccines; they are biodegradable, biocompatible, and stable in colloidal solutions and they enable specific functionalization to tune their physicochemical behavior in the human body.

Phosphatidylcholines are the main components of the eukaryotic lipidic bilayer (Figure 1a) [4]. Phosphatidylcholine contains a polar headgroup that consists of a choline moiety that is linked to a phosphate. The headgroup of phosphatidylcholines also contains a glycerol group connected to a pair of fatty acid tails. This hydrophilic head is zwitterionic, meaning that the P-N dipole has a balanced positive and negative charge, although a cooperative orientation of the dipoles in the bilayer can infer an overall electric potential to the membrane. Cationic and anionic lipids are also constituents of lipidic bilayers, but their related toxicity is usually higher. The head portion can be modified by attaching a functional group. 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine (DSPE) (Figure 1b,

Table 1. Main phospholipids and components commonly used in liposome bilayers, including their molecular formula, net charge, transition temperature (Tc), and molecular weight.

Component Name	Molecular Formula	Charge	Tc (°C)	Molecular Weight (g/mol)	References
Dimyristoylphosphatidylcholine (DMPC)	C36H72NO8P	0	24	678	[5]
Dipalmitoylphosphatidylcholine (DPPC)	C40H80NO8P	0	41	734	[6]
Distearoyl phosphatidylcholine (DSPC)	C44H88NO8P	0	55	790	[7]
Distearoyl phosphatidylglycerol (DSPG)	C42H83O10P	0	55	779	[8]
1,2-Dioleoyl-sn-Glycero-3-Phosphocholine (DOPC)	C44H84NO8P	0	−16.5	786	[6]
Dimyristoylphosphatidylethanolamine (DMPE)	C33H66NO8P	0	50	635.8	[8]
Distearoylphosphatidylethanolamine (DSPE)	C41H82NO8P	0	74	748	[8]
Hydrogenated soybean phosphatidylcholine (HSPC)	C42H84NO8P	0	54	762	[9]
1,2-dioleoyl-3-trimethylammoniumpropane (DOTAP)	C42H80NO4+	1	-	663	[10]
Cholesterol (Chol)	C27H46O	0	-	386.6	[11]
Cholesterol hemisuccinate (CHEMS)	C31H50O4	0	-	486.7	[12]
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)	C42H82NO8P	0	−7	760	[13,14]
1,2-Dioleoyl-sn-glycero-3-phospho-L-serine sodium salt (DOPS)	C42H78NO10P	0	−11	788	[9,15]
MPEG-2000-DSPE	C45H87NNaO11P	0	-	872	[16]
Sphingomyelins (SM)	C24H50N2O6P+	1	-	493.6	[17]

) is an example of a functional phospholipid used to conjugate other polymers like polyethylene glycol (PEG), which helps to increase the circulation time in blood.

The hydrophobic tails of phospholipids, composed of fatty acid chains, vary in length (number of methylene groups) and in their degree of saturation (presence or absence of double bonds). These characteristics critically determine many of the phospholipids’ biophysical properties within the membrane, such as the transition temperature (Tc), which marks the shift from an ordered gel phase, the gel phase (Lβ), to a liquid-crystalline or liquid-disordered phase (Lα or Ld), in some cases mediated by an intermediate rippled phase (Pβ’), depending on the lipid’s composition. Phospholipids with long, saturated tails generally exhibit higher Tc values and are therefore preferred in drug delivery systems, as they form tightly packed and more stable bilayers under physiological conditions—reducing the risk of drug leakage. A commonly used phospholipid in liposomal formulations is 2-distearoyl-sn-glycero-3-phosphocholine (DSPC) (Figure 1a), a cylindrical-shaped phosphatidylcholine with two saturated 18-carbon chains (18:0), leading to a total hydrophobic tail length of approximately 3.5 nm and a Tc of around 55 °C (Table 1).

2. Preparation Methods

Liposomes are relatively easy to prepare, both in laboratory settings and at the industrial scale. Among the various techniques available, thin-film hydration is the most commonly used and straightforward method [18]. In this approach, a mixture of lipids is first dissolved in an organic solvent, which is then evaporated, typically in a round-bottom flask, to form a thin lipid film on the flask’s inner surface. Upon rehydration of the film with an aqueous medium, followed by agitation or sonication, large multilamellar vesicles (MLVs) are formed. These can be subsequently downsized by extrusion through membranes with pore sizes smaller than 200 nm, yielding unilamellar liposomes with low polydispersity.

Other techniques, such as freeze–thaw cycling [19], ethanol injection [20], and the reverse-phase evaporation (REV) [21] method, can also be employed to produce liposomes with similar results. The freeze–thaw cycling method involves subjecting a lipid suspension to multiple cycles of freezing at very low temperatures followed by thawing at higher temperatures, which induces membrane permeabilization and liposome fusion. [22]. In the ethanol injection method, an ethanolic lipid solution is rapidly injected into a larger volume of aqueous buffer. The abrupt dilution of ethanol causes local supersaturation of lipids as the ethanolic solvent loses its ability to solubilize phospholipids. Consequently, the lipids precipitate and spontaneously self-assemble into bilayer fragments, which subsequently reorganize and close to form stable liposomal structures [23]. Finaly, the REV method entails dissolving lipids in an organic solvent to form a thin film, redissolving them in another solvent to generate inverted micelles, and mixing with an aqueous phase to create a water-in-oil (W/O) microemulsion. Removal of the organic solvent by rotary evaporation yields a viscous gel that collapses into an aqueous suspension of liposomes. This method provides high encapsulation efficiency and is suitable for loading large biomolecules, though post-processing steps such as extrusion are required to narrow the particle size distribution [24].

3. Drug Loading Methods

Regardless of the lipid composition, selecting the optimal drug loading method is another critical factor in enhancing encapsulation efficiency (EE%). This parameter is defined as the percentage of the drug successfully loaded into liposomes, typically calculated as the molar concentration of the encapsulated drug in the final formulation divided by the initial molar concentration used during preparation [25].

3.1. Passive Loading Methods

Passive loading methods refer to those in which the drug is present during the liposome formation process. Hydrophobic or poorly soluble drugs could be mixed with the lipids in the initial inorganic solvent before the formation of the dry lipid film. Upon rehydration in an aqueous medium, these hydrophobic drugs tend to localize within the lipid bilayer of the resulting liposomes. In contrast, hydrophilic or soluble drugs may be dissolved in the buffer solution during hydration. In this case, the loaded drug is distributed equally between the interior and exterior of the liposomes.

To assess encapsulation efficiency, removal of the unencapsulated drug from the external aqueous phase is necessary. This can be achieved by using techniques such as centrifugation or size-exclusion chromatography (SEC). For a proper characterization of the loading method, it is convenient to analyze the amount of unencapsulated drug in the removed solvent to calculate the encapsulation molar rate. The choice of analytical method depends on the physicochemical nature of the drug. Some methods require a prior separation step between the free and encapsulated drug. For small-volume samples, this can be performed by centrifugation or ultrafiltration, although experimental conditions must be chosen carefully to avoid liposome rupture. SEC offers good performance, especially with large liposomes, but potential issues include sample dilution and material loss. For larger volumes of liposomal solutions, the dialysis method may be more convenient despite being time-consuming.

Once the loaded liposomes are separated from the solution containing the unencapsulated drug, quantification can be achieved using UV–vis spectroscopy [26] or high-performance liquid chromatography (HPLC) technique. To access and quantify the encapsulated drug, liposome disruption agents such as Titron X-100 or alcohol [27] can be used.

Other methods do not require previous separation of the unencapsulated drugs. Proton NMR (Nuclear Magnetic Resonance) takes advantage of the presence of a reagent that shows chemical shift perturbation with a pH dependency to quantify the amount of drug inside and outside of the liposome where a pH gradient was applied [28]. Fluorescent microscopy can also be used to determine the amount of encapsulated drug in a single vesicle. Liposomes are first optically trapped and then photolyzed with a short UV laser to release the encapsulated molecules. A confocal probe volume, focused a few micrometers from the liposomes, measures the number of released molecules crossing the probe volume [29].

The encapsulation efficiency strongly depends on liposome structure (size and composition). In general, for hydrophilic drugs, large vesicles exhibit a higher encapsulation efficiency than smaller ones, because this is proportional to the aqueous volume [30]. Lipid bilayers composed of long, highly saturated lipid chains tend to have higher encapsulation rates because they usually have elevated transition temperatures and remain in a rigid state under physiological temperatures [31]. Cholesterol is another biomolecule that changes the fluidity of the lipid bilayer: lower cholesterol concentrations in the membrane typically lead to higher encapsulation efficiency rates. However, the composition of the liposomes must be carefully selected because many of the design parameters that improve the loading of hydrophilic drugs will decrease the release rates. Excessively prolonged release times can decrease the antitumor efficiency of some drugs [32] and may decrease the reproducibility and robustness of the treatments. Although this is generally less critical than poor encapsulating rates, several methods of active drug release have been developed as thermosensitive [33] or pH-sensitive liposomes to controllably release the encapsulated drug at specific place and time [34,35].

The encapsulation efficiency of highly soluble drugs, which poorly interact with the lipid bilayer, is relatively low when they are loaded by a passive method. In contrast, drugs that interact more strongly with the lipid membrane generally exhibit higher encapsulation rates [36]. Covalently linking the drug molecules to lipids is an efficient strategy to load the drug within the liposomes to improve the encapsulation efficiency. For instance, muramyl dipeptide, a highly soluble therapeutic agent that enhances immune responses in cancer therapies [37], demonstrates poor encapsulation efficiency and leakage problems during storage. However, formulations in which muramyl dipeptide is conjugated to a synthetic ethanolamine via a peptide spacer significantly improve encapsulation efficiency while preserving its antitumor activity [38].

3.2. Active Loading Methods

Active methods are those where the drug is encapsulated after liposomes preparation. The most common active method for loading consists of creating pH or ionic gradients inside and outside the liposomes to allow for diffusion of the drug through the membrane only in one direction.

In the 1980s, Peter Cullis’ group [39] was the first to demonstrate the uptake of doxorubicin (DOX), a chemotherapeutic drug that inhibits gene expression in cancer cells, preventing it from reproducing into unilamellar liposomes using a pH gradient. Liposomes of pure eggPC (EPC) and egg PC/cholesterol at a molar ratio of 1:1 were prepared and extruded at low pH buffer, resulting in large unilamellar liposomes with an average diameter of 103 nm. Exchange of the external (untrapped) buffer for an alkaline buffer was performed by employing a gel filtration column to create a pH gradient between the inside and outside of the liposome. The DOX added to this liposome solution rapidly permeates through the liposome’s bilayer because, in an alkaline buffer, it stays neutral in its deprotonate form. Once the DOX reaches the acidic environment of the interior of the liposomes, it protonates, becomes charged, and its permeability to diffuse back to the exterior drastically decreases [40]. The encapsulation efficiency of DOX increased by up to 98%, with internal drug concentrations as high as 100 mM. Since this initial work, some other methods based on similar concepts have been developed to optimize the pH gradient or to load drugs of a different nature.

The ammonium sulfate method takes advantage of the low permeability of the counterion, SO₄⁻² (Figure 2). In this case, liposomes are formed in a highly concentrated ammonium sulfate salt solution. After the exchange of the extraliposomal solution for an alkaline buffer, neutral ammonia molecules diffuse out the liposome to compensate for its concentration. For each neutral ammonia out, one sulfate counterion and one proton are left inside, making the internal solution more acidic. Weak amphipathic basic drugs dissolved in this liposome solution will have a neutral charge in a high pH buffer and they will also diffuse into the interior with high permeability. In the interior of the liposome, the drug will interact with the highly concentrated protons and sulfates and, therefore, they will be untrapped in the liposome [41]. Furthermore, accumulation of the protonated base inside the liposome leads to an elevation of the internal pH, which increases the level of NH₃ and, therefore, once again reduces the pH, enabling more of the drug to enter [42]. For loading weakly acidic salt drugs, calcium acetate can be employed instead of ammonium sulfate to generate the pH gradient [43].

3.3. Microfluidic Methods

The development of microfluidic production methods, in which the formation of liposome formulations occurs within a confined microenvironment has been an active topic in recent research.

Generally, a stream of alcohol solution containing the proper composition of lipids is pumped to flow in the central microchannel. The lipid stream is intersected by two lateral streams of a buffer phase (Figure 3a). Because the alcohol and buffer solutions are miscible, the alcohol starts to diffuse to the aqueous phase when both streams are in contact until the alcohol concentration around the lipid decreases below the solubility of the lipid, enforcing their self-assembly in the lamellar phase to form liposome structures [44,45] (Figure 3b). The size of the focused stream, the volumetric flow rate ratio between the lipid and water phase streams, and the total flow rate can be tuned to control the final size and polydispersity of the liposome’s production [46] without the use of extra steps like extrusion or sonication [47]. Massive and rapid fabrication of size-predictable liposomes and loading of hydrophobic drugs have made great progress in the search for industrial-scale production methods [48]. However, more sophisticated devices recently developed to incorporate active loading of amphiphilic drugs [49,50] could mean a disruptive milestone for the commercial-scale fabrication of liposome drug delivery systems. Hood and co-workers [49] built a microfluidic device that combines liposome formation with in-line channels for liposome purification and remote loading of doxorubicin hydrochloride in one single step. This 12 cm system enables the formation of transmembrane pH (ΔpH = 3 and 5) and ion gradients, followed by immediate introduction of amphipathic drugs for real-time active loading into the liposomes. The total on-chip residence time is approximately 3 min, which is a significant improvement compared with bulk methods that require the combination of several steps that could take days to be completed. More recently, Gkionis and co-workers [50] used a similar device to evaluate the impact of lipid composition of PEGylated liposomes loaded with DOX on the encapsulation efficiency, in vivo drug release profiles or physicochemical properties of the drug-loaded liposome’s size, polydispersity, zeta potential, or temperature transition. Compared with in-bulk active-loading methods, microfluidic devices provide lower encapsulation efficiency (~EE = 60–80%) but it is still larger than passive loading [51]. Another drawback of the use of microfluidic systems is related to the difficulty of removing all trends of alcohol in the resulting lipidic membranes of the liposomes, which compromises the stability of liposomes for long-term circulation as well as changes the release profile of drugs [52].

4. Circulation and Stability

4.1. Stability

One of the main challenges for liposomes to become a referential drug delivery system in cancer therapies is the instability during the production and storage of liposome formulations. Instability issues in long-scale manufacturing cause poor batch-to-batch reproducibility, shorten circulating times in blood and drug leakage issues, and they are probably the main reason behind only a few liposome formulations entering the market [53].

Chemical and physical instability are the main reasons why liposomes fail in their purpose. Chemical instability mainly arises from the hydrolysis and oxidation of the phospholipids that compose the liposomal bilayer. Lipidic hydrolysis is a natural process that involves the cleavage of the ester bonds that link the fatty acid chains to the glycerol backbone of the phospholipids (Figure 4a). This reaction occurs under aqueous conditions, and is accelerated by temperature, pH, or the presence of catalytic ions. As a result of hydrolysis, the phospholipids break down into lysophospholipids (molecules that retain only one acyl chain) and free fatty acids [54]. More problematic for the long-term chemical stability of aqueous liposome dispersions is the chemical hydrolysis of phosphatidylcholines, which forms lysophosphatidylcholines (LysoPCs) [55]. Since they contain only one hydrophobic tail and a hydrophilic headgroup, LysoPCs have a conical molecular shape that disturbs the regular packing of the bilayer and introduces positive curvature stress. These molecules can flip between the inner and outer layer of the membrane and destabilize bilayers, resulting in leakage of encapsulated aqueous contents. LysoPCs can also stabilize non-bilayer forming lipids and induce aggregation and/or fusion in liposome formulations [54]. When their concentration surpasses a critical threshold, LysoPCs can even drive phase transitions from lamellar to micellar structures, compromising the bilayer’s integrity [56]. The formation of substantial amounts of these hydrolysis products may lead to an increase in particle size, an increase in the permeability of liposome bilayers, and also to various toxic effects [57]. Hydrolysis reactions increase with temperature, which is why it is strongly recommended to store liposome solutions at 4 °C [54].

Lipid oxidation represents one of the major degradation pathways compromising the integrity and life of liposomal formulations. This process primarily involves the peroxidation of polyunsaturated phospholipids containing bis-allylic hydrogen atoms, which are highly susceptible to radical attacks [58]. Oxidation occurs when dissolved molecular oxygen reacts with these sensitive phospholipids, leading to spontaneous lipid peroxidation. Although this reaction predominantly affects unsaturated fatty acyl chains, saturated fatty acids may also undergo oxidation under elevated temperatures [59]. Lipid peroxidation follows a free-radical chain mechanism consisting of three stages: initiation, propagation, and termination (Figure 4b). In the initiation step, reactive oxygen species abstract a hydrogen atom from a bis-allylic methylene group, generating a lipid radical (L·) and triggering the oxidative chain [58]. During propagation, these radicals react with oxygen to form lipid peroxyl radicals (LOO·), which oxidize neighboring lipids to form lipid hydroperoxides (LOOHs), sustaining the process. Termination occurs when antioxidants such as vitamin E donate hydrogen atoms to neutralize LOO· species, yielding nonradical products and halting the chain reaction [60].

The breakdown of lipid hydroperoxides produces numerous secondary oxidation products, including reactive aldehydes such as malondialdehyde (MDA) and 4-hydroxy-2-nonenal (4-HNE). These compounds are capable of diffusing across membranes, forming covalent adducts with proteins, and further amplifying oxidative damage [60]. As a consequence, the accumulation of oxidized phospholipids increases membrane polarity, disrupts bilayer packing, and can lead to pore formation or micellar structures within the membrane. In particular, oxidized phospholipids become more polar and tend to relocate toward the lipid–water interface. This redistribution disturbs the organization of the hydrophobic core, weakens lipid packing, and increases membrane permeability [58].

Oxidation can be accelerated by metal ions, elevated temperatures, and light exposure. To mitigate these effects, liposomal dispersions should be stored at low temperatures (around 4 °C), in the dark, and under inert atmospheres (nitrogen or argon) using degassed Milli-Q water or buffers to minimize oxygen concentration and reduce oxidation rates.

Physical stability relates to the capacity of liposomes to stay dissolved in a solution without aggregation. Physical stability is mainly governed by the balance of electrostatic forces, van der Waals, and hydration forces [61], and can be predicted by the classical DLVO (Derjaguin–Landau–Verwey–Overbeek) theory through a physicochemical characterization of size, zeta potential, and ionic strength [62]. The main factor affecting physical stability is the composition of the lipids in the bilayer, particularly the charge of their headgroups. Liposomes made up of neutral or zwitterionic lipids can last up to several weeks without degradation and they undergo irreversible aggregation after several days. The addition of cationic and anionic lipids into the liposomes significantly increases the zeta potential of the liposomes and avoids aggregation for long periods of time. Choosing the proper composition and size of the liposomes, it is possible to combine the electrostatic and van der Waals forces to promote aggregation in a secondary minimum of the DLVO potential. Aggregation in such conditions leads to the formation of controlled-size stable clusters of liposomes [63]. These small aggregates of several liposomes could be of potential interest because they combine high packing and encapsulation efficiency with a desirable and tunable size of clusters [61,64].

4.2. Cholesterol

Cholesterol is a 27-carbon organic molecule with a hydroxyl group, a steroid ring, and a carbohydrate tail (Figure 5a). Due to its string amphiphilic nature, cholesterol integrates into lipid membranes by embedding its hydrophobic sterol ring system among the fatty acyl chains of the bilayer, while its hydroxyl group forms a hydrogen bond with the hydrophilic headgroups of adjacent phospholipids. The short hydrocarbon tail of the sterol further intercalates within the hydrophobic region of the membrane. The addition of cholesterol to lipid bilayers modifies the gel-to-liquid crystalline phase transition of the bilayer. At low cholesterol concentrations, the transition temperature and the enthalpy significantly decrease. For concentrations larger than 25%, the gel-to liquid crystalline transition disappears, resulting in a more ordered liquid phase.

Molecular simulation studies indicate that cholesterol incorporation increases the ordering of phosphatidylcholine acyl chains, aligning them more closely with the bilayer normal [65], and reduces the thickness of the membrane. Consequently, the cohesiveness and packing of the bilayer increases, leading to a reduction in the permeability to non-electrolyte and electrolyte solutes [66]. The presence of cholesterol also affects the orientation of the lipidic headgroups, exposing more of the P-N dipole to the water phase, increasing the zeta potential values, and therefore enhancing the physical stability against aggregation.

Cholesterol reduces the interaction of liposomes with several plasma proteins present in blood, which inhibits macrophage digestion and increases the circulating time in the body [67]. The addition of cholesterol not only affects the stability of the liposomes but also increases the encapsulated rates and decreases the drug’s release. So, the optimal amount of cholesterol to be incorporated into liposomes is still unclear and it is very dependent on the other constituents of the bilayer and the nature of the encapsulated drugs. It is known that the maximum cholesterol that could be added to stable liposomes is 50%. The most frequently used proportion is a 2:1 ratio but this is usually defined by trial-and-fail in vitro experiments.

4.3. Size

The reticuloendothelial system (RES) is a part of the immune system made up of a heterogeneous population of phagocytic white cells that play an important role in the clearance of particles and soluble substances in the circulation and in tissues. Monocytes and macrophages are the main two types of white blood cells involved in liposome removal. Monocytes are irregularly shaped cells (10–15 μm in diameter) that mainly circulate in the bloodstream. Once monocytes reach an organ or a tissue, they differentiate into macrophages, which are larger in size and have more specific receptors in their surface for detection, phagocytosis, and the destruction of harmful organisms.

The detention of liposomes by macrophages is preceded by the interaction of the serum protein, which binds preferentially at the hydrophobic defects of the external layer of the lipidic membrane, creating a protein corona that “labels” the liposomes to be detected and removed by the RES. Liposomes smaller than 100 nm are rapidly removed from the bloodstream by these white cells [68]. The spleen and liver are the primary organs that also play an important role in the reticuloendothelial system by filtering the blood through pores that vary in size from 100 to 600 nm [69]. Most cancer therapies use liposomal formulations with sizes in the 80–150 nm range to avoid the reticuloendothelial system. One of the exceptions are therapies that target macrophages or some therapies that aim to release drugs in the liver and spleen.

Liposome size can be tuned with the lipidic composition of the bilayer by incorporating compounds with a particular geometry which forces the bilayer to adopt specific curvatures [70,71]. Selecting different flows in microfluidic devices for large-scale liposome production can also set a desired size distribution. However, for research purposes, filtering liposomes by extrusion is the more common method to fix the size of liposomes.

4.4. PEG-Ylation

Polyethylene glycol (PEG) polymers are hydrophilic molecules (C_2nH_4n+2O_n+1) which show high biocompatibility, low toxicity, and high solubility in aqueous and organic media, for which they are widely used in many biomedical fields and in food products.

PEG-coated liposomes, also known as “stealth” liposomes, can be prepared by adding PEG-grafted lipids into the preparation formulation. Such liposomes have a highly flexible and hydrophilic PEG layer that prevents aggregation [72] and avoids the interaction of serum proteins, which makes their detection by the macrophages more difficult and, consequently, prolongs the circulating time in the bloodstream. PEGylated liposomes have shown a significant increase in blood circulation time from 30 min to up to 5 h [73].

In poorly dense PEG-coated liposomes, the PEG polymers anchored at the liposome surface adopt a “mushroom” configuration. By increasing the concentration, PEG polymers stretch and adopt a “brush” configuration [74]. It is well established that brush configurations are needed to increase circulation times, and this occurs when the average distance between PEG polymers at the surface is smaller than the Flory radius [74] of the polymers (defined by the random walk law).

PEG has been considered to be nonimmunogenic. However, there is growing evidence that PEG might trigger a more immunogenic response than previously recognized. This is supported by the detection of anti-PEG antibodies in healthy humans who are increasingly exposed to PEG additives [75]. Since PEG polymers were approved by the Food and Drug Administration (FDA) in 1990, they have been commonly used as additives to improve the stability of many drug formulations and foods. Dams and co-workers were one of the first to report that “empty” PEGylated liposomes are rapidly removed from the body when they are administered twice in the same animal at certain intervals [76,77]. This unexpected phenomenon, usually called the accelerated blood clearance (or ABC) phenomenon, has been a topic of investigation in recent work, showing nonsystematic, and sometimes contradictory, results. Different animal models treated with repeated intravenous injections of Doxil showed very different ABC responses, which evidences that ABC phenomenon differs depending on the different immune system of the animal species. ABC phenomena also seem to depend on the lipid composition and time interval between injections [75].

The regular use of personal care products containing PEG additives (shampoo, toothpastes, and lotions, etc.) and the ingestion of foods containing PEG-modified molecules as antioxidants or emulsifier, results in a stimulation of the production of anti-PEG antibodies [78]. The presence of pre-existing anti-PEG antibodies might trigger further immunogenic responses to PEG when humans are treated with PEGylated liposomes, decreasing their circulating time and provoking failure of the final goal of cancer therapies [75]. This would also explain the different ABC responses in different individuals treated with PEGylated drug delivery systems.

5. Targeting

5.1. Passive Targeting: Enhanced Permeation and Retention (EPR)

Cancer cells are anomalous cells that divide themselves uncontrollably. The rapid growth of tumors demands a high supply of oxygen and constituents, and new blood vessels need to quickly be developed to cover all regions of the tumor. To stimulate the creation of blood vessels, tumor cells secrete outstanding amounts of vascular endothelial growth factor (VEGF). Under such a stimulus, vessels are created chaotically with abnormal shapes and architecture. Blood vessels surrounding tumor cells are usually made up of endothelial walls with defects, holes, and large fenestrations with diameters that range from 300 to 4700 nm [79]. Another characteristic of solid tumors is the lack of lymphatic drainage, which is responsible for eliminating toxins and extracellular harmful substances. The combination of these two phenomena leads to an enhanced permeability and retention (EPR) effect, which plays a critical role in the success of liposomes as cancer drug carriers (Figure 6b).

High-molecular-size molecules with a capacity for greatly prolonged circulation will preferentially leak out from those abnormal tumoral vessels and accumulate in the interstitial space of tumor. Since the EPR effect was initially described by Hiroshi Maeda in the late 1980’s [80], it has been well accepted as one of the universal pathophysiological characteristics of solid tumors and acts as a fundamental principle for designing and developing tumor-targeting delivery strategies for anticancer drugs [81]. The EPR effect in combination with the long circulation times of stealth liposomes were key features in the development of Doxil, the first FDA-approved drug that uses liposomes as a drug carrier, which achieves about a 10–15-fold higher concentration in tumor tissues compared with surrounding normal tissues [81,82].

5.2. Active Targeting

Once the liposomes accumulate at the interstitial space of the tumor by passive targeting, the drug slowly leaks through the membrane and diffuses in the interstitial region. A fraction of the drug reaches the desirable cells for treatment, but most of them are lost or reach healthy cells. In some cases, when the efficiency of the drug is strong and its toxicity is low, this could be enough to result in efficient treatment. But, in other cases, the amount of drug reaching the tumor cells needs to be optimized by active targeting.

Active targeting involves the attachment of a targeting ligand to the surface of liposomes for enhanced delivery of liposomal systems. The targeting ligands commonly used in liposomes are antibodies, aptamers, or proteins, which bind specifically to tumor cells and reduce their interaction with healthy tissue to minimize the adverse effects and avoid potential drug resistance.

The most common methods to add specificity into liposomes consist of adding modified lipids with anchored specific ligands prior the mixing with other lipid components during liposome preparation [83]. The major drawback of this method is that ligands are incorporated not only in the outer layer of the liposome but also in their internal side, reducing their stability and loading features (Figure 2). In addition, the exposure of ligands as aptamers or antibodies to organic solvents in some preparation methods could lead to the loss of their structure and binding properties. Modified lipids with ligands can also be incorporated into pre-prepared and pre-loaded liposomes through a micelle phase if the ligand–lipid molecules present amphiphilic behavior. Another possibility is based on the covalent coupling of the ligand with the PEG polymer through the amine, carboxylic acid, or thiol group in stealth liposomes. In this case, the targeting of liposomes with a ligand is performed after the liposome’s preparation [84]. The main disadvantage of this method is that the orientation of the ligand on the liposome’s surface cannot be always controlled, and this could lead to a loss of efficiency when binding with the targets and higher detection by the RES because more hydrophobic residues are exposed.

5.2.1. Antibody–Liposome Bioconjugates (Immunoliposomes)

Most studies on active targeting liposomes have focused on antibodies or antibody fragment conjugates, since procedures for producing highly specific monoclonal antibodies are well established. This liposomal formulation allows for the efficient targeting of the antibody–liposome bioconjugate to its matching antigen in the tumor cells (Figure 6c). It is well-accepted that antibody–antigen binding often triggers cell signaling cascades that culminate in the internalization of liposomes by endocytosis. Then, the liposome may either fuse with the cell membrane and release the load directly into the cytoplasm or internalize it via endosome for degradation and expulsion [85]. During the internal path of the endosome, hydrolytic enzymes influx protons into the endosome to decrease their internal pH and promote degradation, which provokes the release of drugs into the cytoplasm.

As described previously, the success of targeted binding depends on many different factors, not only on the antibody’s coupling, which makes comparisons of in vitro and in vivo studies with coherent conclusions particularly difficult. Many studies have focused on the efficiency of different strategies to conjugate antibodies with liposomes, without keeping in mind that any modification of the liposome’s surface can affect the drug’s encapsulation, the circulation time, or their stability. For instance, Gerben and co-workers [86] tested two different configurations of stealth liposomes with antibodies that specifically bind to cancer colon cells in rats in terms of targeting and macrophage uptake. When compared with antibodies coupled directly at the PEGylated lipidic bilayer, formulations with antibodies coupled at the distal end of PEG polymer showed enhanced target cell binding, probably because the antibodies were more exposed, but they were removed quickly because the randomly oriented antibodies in PEG exposed more the Fc region, which is easily detected by macrophages.

On the other hand, selecting a proper target for immunoliposomes is the most crucial decision in the development of new strategies in active targeting and it requires quite a vast interdisciplinary background in cancer developments. The “ideal target” should be a stable antigen with highly specific overexpression in tumor cells [87] and poor expression in healthy cells, it should be exposed on the external side of the cellular membrane and either facilitate the internalization of external agents or play any vital biological role for the cancer cells. The human epidermal growth factor receptor (HER) family are transmembrane tyrosine kinases that regulate diverse cellular functions in response to extracellular ligands. Overexpression of HER proteins is found in approximately 20–25% of cases of breast and lung cancer [88]. Despite that the full signaling system that controls tumoral cell growth and differentiation is far to be fully understood, it is known that HER proteins play an important role in the signaling between the internal and external cells. Blocking HER signaling has become a promising strategy to inhibit tumor growth [89]. Anti-HER2 immunoliposomes have been developed for DOX delivery to breast cancer. Loaded liposomes conjugated with Fab portions against the HER2 receptors showed higher accumulation in tumor tissue and increased antitumor cytotoxicity with lower toxicity than free DOX [90]. Paclitaxel, a chemotherapy drug that inhibits microtubule stabilization during mitosis, has been loaded in immunoliposomes conjugated with anti-HER2 antibodies in the clinic, and showed longer circulation times and higher cellular uptake than commercial formulation of paclitaxel (Taxol^®) [91]. Paclitaxel has also been encapsulated alone in liposomal carriers, such as Lipusu, a formulation approved in China that avoids the use of Cremophor EL (responsible for severe hypersensitivity reactions) by incorporating the drug into a liposomal bilayer [92].

In some cases, antibodies not only work as detectors of cancer cells but also as therapy treatments themselves. This is the case in immunoliposome conjugates with anti-EGFR antibodies that are usually applied in combination with other loaded chemotherapy drugs to obtain a double effect in cancer growth. EGFR is a transmembrane protein of the HER family involved in the pathogenesis and progression of different types of cancers [93]. EGFR antibodies bind to EGFR transmembrane proteins, blocking the interaction with growth factors and, eventually, causing apoptosis while facilitating the internalization of chemotherapy drugs for glioma and carcinoma tumors.

A more sophisticated example of immunoliposomes is those targeting matrix metalloproteinases (MMPs), a family of zinc-dependent endopeptidases. Zhu and coworkers [94] designed responsive liposomes that change their surface in the presence of MMPS proteins. Pegylated liposomes were conjugated with anti-MMPS antibodies and a cell-penetrating peptide (short cationic peptides with 5–30 amino acids, which facilitates their penetration across the cell membrane). When the antibodies bind to MMPs undergoing cleavage in the tumor by the highly expressed extracellular MMP2, it causes the removal of PEG chains and exposes the penetration peptide, resulting in TAT peptide-mediated endocytosis, causing increased cell uptake [94].

5.2.2. Aptamers

Aptamers are short, single-stranded DNA or RNA molecules that selectively bind to a specific target (proteins, peptides, carbohydrates, or small molecules). Aptamers assume a variety of three-dimensional structures due to their tendency to form helices and single-stranded loops. They can be used for active targeting, and some studies have shown that aptamers have privileged characteristics over antibodies, especially in cancer treatment. They are shorter, smaller, and more stable than antibodies and easier to synthesize at a large scale. They show higher affinity and selectivity, and they do not have an Fc region that could be detected by macrophages.

An example of the use of aptamers in active targeting was designed by Kin and Co-workers [95], who conjugated anti-EGFR aptamers in PEGylated liposomes targeting human breast metastatic carcinoma cells loaded with therapeutic RNA and quantum dots as labels. Fluorescence microscopy showed higher signals in tumors in mice four hours after the injection of liposomes with aptamers compared to liposomes without aptamers [95]. Aptamer AS1411 is one of the most studied aptamers because it binds with high affinity and specificity to nucleolin, a key protein in the synthesis of ribosomes that is overexpressed in tumor cells [96]. The aptamer AS1411 was used to functionalize cationic liposomes in gene therapy to simultaneously deliver paclitaxel and a therapeutic siRNA in breast cancer cells. In vitro and in vivo experiments show a significant 62% inhibition of tumor growth [97].

5.2.3. Sensitive Liposomes

Sensitive liposomes refer to liposomal formulations that release their load as a response to a change in the environmental conditions. The drug’s release can be triggered by local changes in the pH, temperature, magnetic fields, light radiation, or after a particular interaction (as in the mentioned example of aptamer-targeted activation).

Thermosensitive liposomes (Figure 6d). The first and most simple thermosensitive liposomes are based on DPPC lipids, which have a transition temperature of 42 °C. In local hyperthermia conditions, DPPC liposomes release their drug in tumor regions at 44 °C, 100 times more than at 37 °C, not only because the permeability of the thermosensitive liposomes drastically increases but also due to the enhancement of the ERC effect.

The next-generation thermosensitive liposomes were based on lysolipids, derivatives of lipids in which one acyl derivative has been removed by hydrolysis. They stabilize pores in lipid bilayers undergoing phase transitions from gel to liquid and increase the drug release rate of DPPC liposomes. The incorporation of lysolipids such as monopalmitoylphosphatidylcholine into DPPC bilayers further accelerates the rapid release of the drug cargo under mild hyperthermia (40–42 °C).

ThermoDox^®, a lysolipid and thermally sensitive liposome product, has been tested in a phase III clinical trial to encapsulate DOX for the treatment of solid breast cancer, facilitating a 25-times-greater concentration of the drug around the targeted tumor compared with direct intravenous injections of DOX [98]. Other more sophisticated thermosensitive components have been proposed for use in cancer treatments as surfactants [99], polymers [100], or collagen-like polypeptides [101] in the search of higher thermal sensibility, more stable formulations, or larger circulating times.

One of the critical decisions in the application of thermosensitive liposomes is the selection of the heating method to provoke local hyperthermia, because achieving this temperature uniformity in large tissue volumes is challenging, especially in deep cancer tissues with difficult direct access [102]. Radiofrequency ablation (RFA) is already a clinically approved anticancer therapy that kills cancer cells with heat. RFA consists of the application of a radiofrequency electric current to tissue via electrodes inserted into the tumor under image guidance. It is also the most common heat method to use in combination with thermosensitive liposomes. High-intensity focused ultrasound (HIFU) is another option that employs focused ultrasound beams emitted from external ultrasound transducers into deep tissue regions, resulting in tissue heating with a precision in the range of mm. It has the advantage that it can be used in combination with magnetic resonance image techniques for real-time monitoring of the tissue’s temperature.

pH-sensitive liposomes. pH-sensitive liposomal formulation is another strategy to enhance drug delivery in the interior of cancerous cells and has seen increasing development in recent years (Figure 6e). The biophysical basis of pH-sensitive liposomes usually relies on the physicochemical features of phosphatidylethanolamine (PE). PE has a small and poorly hydrated headgroup that occupies a lower volume compared with its hydrocarbon chains. In contrast to the majority of phospholipids which display an overall cylinder shape, PE exhibits a cone shape that tends to adopt an inverted hexagonal structure in the absence of other molecules. Combining PE with amphiphilic molecules with protonable headgroups, (also known as stabilizers) such as cholesteryl hemisuccinate (CHEMS, Figure 6b) [103], is possible to force the lamellar phase and create stable liposomes in physiological conditions. However, in acidic environments, the carboxylic groups of the stabilizers protonate, losing their negative charge and, therefore, reducing the repulsion between headgroups. In such conditions, the liposomes become unstable, and the PE lipids revert into their inverted hexagonal phase.

pH-sensitive liposomes bind to cancer cells via specific target biomolecules (usually antibodies), and they are internalized through the endocytic pathway. Liposomes are retained in early endosomes, which mature into late endosomes for their degradation. During this process, protons are pumped into the endosome, which triggers the destabilization of pH-sensitive liposomes, releasing the cargo in the cytosol. The exact mechanism of endosomal escape still remains unclear: either the pH-sensitive liposome also promotes the fusion or destabilization of the endosome membrane by pore formation or the released drug inside the endosome diffuses through the endosome membrane into the cytoplasm [104]. In either case, pH-sensitive liposome formulations with different target conjugations have demonstrated a great efficiency in delivery drugs and nucleic acid into the cytoplasm of cancerous cells [105,106].

6. Approved Liposomal Formulations for Cancer Treatment

In 1990, the European Medicine Agency (EMA) approved the first liposomal formulation, AmBisome^®, an antifungal preparation of amphotericin B encapsulated in 60–80 nm liposomes composed of cholesterol and phosphatidylcholines. This formulation does not employ PEGylation, as rapid uptake by the reticuloendothelial system in the liver and spleen was considered desirable for targeting fungal infections [107].

In 1995, the U.S. Food and Drug Administration (FDA) approved the first liposomal formulation for anticancer therapy, Doxil^® (subsequently authorized by the EMA as Caelyx^®). Doxil comprises ~100 nm liposomes composed of cholesterol and phosphatidylcholine, which encapsulate doxorubicin hydrochloride, an anthracycline chemotherapeutic agent. Polyethylene glycol coating protects the liposomes from recognition by the mononuclear phagocyte system, markedly prolonging systemic circulation, with reported blood residence times of up to ~200 h. The introduction of Doxil had a profound impact on nanomedicine, establishing a new paradigm for the application of nanotechnology in cancer treatment. Although it has been reported to cost approximately 145 times more than conventional doxorubicin therapy [108], Doxil substantially reduces the risk of cardiotoxicity associated with the non-liposomal drug and can enhance antitumor efficacy by enabling longer treatment courses at higher cumulative doses [109].

Following the approval of Doxil^®, a liposomal formulation of daunorubicin citrate (DaunoXome^®), an anthracycline chemotherapeutic, was approved for the treatment of Kaposi’s sarcoma. This formulation does not employ PEGylation to prolong circulation; rather, it utilizes small liposomes (≈48–80 nm) to reduce uptake by the reticuloendothelial system and thereby modestly extends systemic exposure [110].

Since these two initial oncology approvals, several additional liposomal formulations for cancer therapy have been approved. Table 2 summarizes the principal liposome-based anticancer products authorized by major regulatory authorities, described according to the biophysical parameters outlined above.

Similarly, there are several promising formulations currently in clinical trial phases which are worth monitoring. FF-10832 is a liposomal formulation encapsulating gemcitabine, which is already in phase II trials for urothelial carcinoma and non-small-cell lung cancer (NSCLC), in combination with pembrolizumab [111]. New formulations of liposomal paclitaxel are also being tested in a randomized phase II trial; in this case, a cationic liposomal paclitaxel is designed to target tumor endothelial cells in advanced pancreatic cancer [112]. TLD-1 is a novel liposomal formulation developed to encapsulate doxorubicin for the treatment of solid tumors [113].

There are also numerous ongoing clinical trials investigating previously approved systems for new cancer indications. One of the most notable examples is Doxil, which is currently being evaluated as a treatment for triple-negative breast cancer in combination with atezolizumab and/or bevacizumab (phase II) [114], for recurrent ovarian cancer in combination with the IN10018 inhibitor (phase II) [115], and for ovarian cancer in combination with carboplatin (phase III) [116], among others.

Table 2. Comparison of clinically approved liposomal anticancer therapeutics. The table details, for each formulation, the manufacturer, year of approval, therapeutic application, active pharmaceutical ingredient (API), composition, drug loading, method of preparation, physicochemical stability (particle size, zeta potential, and Tc), circulation profile, targeting strategy, drug release characteristics, specific observations, and the literature references.

Commercial Name	Doxil/Caelyx	DaunoXome	DepoCyte	Myocet
Manufacturer	Baxter Healthcare Corporation (Deerfield, US) Janssen Pharmaceutica (Beerse, Belgium)	NeXstar Pharmaceutical (Boulder, US)	Pacira Ltd. (Watford, UK)	GP-Pharm (L’Hospitalet de Llobregat, Spain) Sun pharmaceutical (Mumbai, India)
Approval date	1995 (US, Doxil) 1996 (EU, Caelix)	1996 (US)	1999 (US) 2001 (EU)	2000 (US and EU)
Illness	Breast, Ovarian, Kaposi’s sarcoma (KS), and Multiple Myeloma	HIV-associated KS	Lymphomatous meningitis	Breast cancer
API	DOX Hydrochloride	Daunorubicin citrate	Cytarabine	DOX hydrochloride
Composition	MPEG-2000-DSPE:HSPC:Chol (5:55:40)	DSPC:Chol (2:1)	DOPC:DPPC:triolein:Chol	EPC:Chol (45:55)
Drug loading	Active (pH gradient of ammonium sulfate)	Passive	Passive	Active (pH-gradient with citrate buffer)
Preparation Method	Thin-film hydration + extrusion	Thin-film hydration + extrusion	DepoFoam technique	Thin-film hydration + extrusion
Stability (Particle Size, Zeta Potential, and Tc)	100 nm (Unilamellar); 55 °C	48–80 nm (Unilamellar); −5 mV	20 µm (Multilamellar)	150–250 nm (Unilamellar); −10 to −20 mV
Circulation profile	The half-life up to 231 h, with a mean of 73.9 h due to PEGylation, avoiding RES	Prolonged circulation due to small size, avoiding RES and phagocytosis Non-PEGylated	Cytotoxic drug concentrations in the cerebro-spinal fluid (CSF) are maintained up to 14 days	Larger size leads to faster recognition by RES, yet in vivo assays show extended circulation time Non-PEGylated
Targeting	Passive (EPR)	Passive (EPR)	Direct injection into the CSF compartment	Passive (EPR)
Drug release	Controlled by cholesterol; mechanism not fully understood	Sustained intracellular release over ≥36 h, maintaining cytotoxic levels within tumor cells	Liposome degradation	Passive release
Observations	Avoiding heart damage risk of free DOX	Suitable for tumors with high vascular permeability No longer marketed	No longer marketed	Low systemic toxicity. Reduced incidence of cardiac events and congestive heart failure compared to free DOX
References	[109,117,118,119]	[110,120,121,122,123]	[118,124,125]	[126,127]
Commercial name	Lipusu	Mepact	Marqibo	Onivyde
Manufacturer	Nanjing Luye Pharmaceutical (Shanghai, China)	Takeda France (Courbevoie, France)	Talon Therapeutics (San Francisco, US)	Merrimack Pharmaceuticals (Cambridge, US)
Approval date	2006 (China)	2009 (EU)	2012 (US)	2015 (US) 2016 (EU)
Illness	Breast, ovarian, and lung cancer	Osteosarcoma	Leukemia	Pancreatic cancer
API	Paclitaxel	Mifamurtide	Vincristine Sulfate (VCR)	Irinotecan HCL trihydrate
Composition	Lecithin–Chol	DOPS:POPC (3:7)	SM:Chol (60:40)	DSPC:Chol:MPEG-2000-DSPE (3:2:0.015)
Drug loading	Passive	Passive	Active (pH gradient)	Active (with triethylammonium sucrose octasulfate)
Preparation Method	Thin-film hydration + extrusion	In situ, mixed with 0.9% saline solution.	Ethanol injection + extrusion	Ethanol injection + extrusion
Stability (Particle Size, Zeta Potential and Tc)	<200 nm (Unilamellar)	2–3.5 µm (Multilamellar) 5 °C	130–150 nm (Unilamellar)	~110 nm (Unilamellar); –18 mV; 55 °C
Circulation profile		Rapidly cleared from serum with half-life of 2 h	Low protein binding, which results in a prolonged circulation time for the liposome Non-PEGylated	Long circulation time due to PEGylation, avoiding RES
Targeting	Passive (EPR)	Targeting immune system for immunotherapy	Passive (EPR)	Passive (EPR)
Drug release		Liposome degradation	Long release half-time, up to 117 h	Prolonged release; half-life of drug release up to 56.8 h
Observations	First paclitaxel liposome commercial in China	Used in children and young adults after resection surgery	No longer marketed	Improved tumor accumulation
References	[92]	[118,128,129,130,131]	[32,132]	[133,134,135]
Commercial name	Vyxeos	Celdoxome	Zolsketil
Manufacturer	Jazz Pharmaceuticals (Dublin, Ireland)	Baxter Holding (Utrecht, Netherlands)	Accord Healthcare (Barcelona, Spain)
Approval date	2017 (US) 2018 (EU)	2022 (EU)	2022 (EU)
Illness	Myeloid leukemia	Breast, ovarian, Kaposi’s sarcoma, and Multiple Myeloma	Breast, ovarian neoplasms, and Kaposi’s sarcoma
API	Daunorubicin, cytarabine (1:5)	DOX hydrochloride	DOX hydrochloride
Composition	DSPC:DSPG:Chol (7:2:1)	MPEG-2000-DSPE:HSPC:Chol	MPEG-2000-DSPE:HSPC:Chol
Drug loading	Passive + active (with copper gluconate buffer)	Active (pH gradient of ammonium sulfate)	Active (pH gradient of ammonium sulfate)
Preparation Method	Thin-film hydration + extrusion	Thin-film hydration + extrusion	Thin-film hydration + extrusion
Stability (Particle Size, Zeta Potential and Tc)	107 nm (bilamellar) –33 mV 55.3 °C	75–100 nm (Unilamellar) 55 °C	75–100 nm (Unilamellar) 55 °C
Circulation profile	~50× longer circulation time than free drug Uses anionic phosphtildylgrlycerol as alternative to PEG to avoid ABC phenomenon	The half-life up to 231 h, with a mean of 73.9 h due to PEGylation, avoiding RES	Long circulation times. With an average half-time of 73.9 h
Targeting	Preferential internalization of CPX-351 liposomes	Passive (EPR)	Passive (EPR)
Drug release	Low cholesterol optimized for tumor-controlled release		Passive release
Observations	Dual-drug formulation with a fixed synergistic ratio; in vivo efficacy is drug ratio-dependent Potentially avoids P-gp-mediated efflux, reducing treatment resistance	Authorized as generic	Bioequivalent to Caelyx
References	[136]	[137]	[138,139]

Table 2. Comparison of clinically approved liposomal anticancer therapeutics. The table details, for each formulation, the manufacturer, year of approval, therapeutic application, active pharmaceutical ingredient (API), composition, drug loading, method of preparation, physicochemical stability (particle size, zeta potential, and Tc), circulation profile, targeting strategy, drug release characteristics, specific observations, and the literature references.

Commercial Name	Doxil/Caelyx	DaunoXome	DepoCyte	Myocet
Manufacturer	Baxter Healthcare Corporation (Deerfield, US) Janssen Pharmaceutica (Beerse, Belgium)	NeXstar Pharmaceutical (Boulder, US)	Pacira Ltd. (Watford, UK)	GP-Pharm (L’Hospitalet de Llobregat, Spain) Sun pharmaceutical (Mumbai, India)
Approval date	1995 (US, Doxil) 1996 (EU, Caelix)	1996 (US)	1999 (US) 2001 (EU)	2000 (US and EU)
Illness	Breast, Ovarian, Kaposi’s sarcoma (KS), and Multiple Myeloma	HIV-associated KS	Lymphomatous meningitis	Breast cancer
API	DOX Hydrochloride	Daunorubicin citrate	Cytarabine	DOX hydrochloride
Composition	MPEG-2000-DSPE:HSPC:Chol (5:55:40)	DSPC:Chol (2:1)	DOPC:DPPC:triolein:Chol	EPC:Chol (45:55)
Drug loading	Active (pH gradient of ammonium sulfate)	Passive	Passive	Active (pH-gradient with citrate buffer)
Preparation Method	Thin-film hydration + extrusion	Thin-film hydration + extrusion	DepoFoam technique	Thin-film hydration + extrusion
Stability (Particle Size, Zeta Potential, and Tc)	100 nm (Unilamellar); 55 °C	48–80 nm (Unilamellar); −5 mV	20 µm (Multilamellar)	150–250 nm (Unilamellar); −10 to −20 mV
Circulation profile	The half-life up to 231 h, with a mean of 73.9 h due to PEGylation, avoiding RES	Prolonged circulation due to small size, avoiding RES and phagocytosis Non-PEGylated	Cytotoxic drug concentrations in the cerebro-spinal fluid (CSF) are maintained up to 14 days	Larger size leads to faster recognition by RES, yet in vivo assays show extended circulation time Non-PEGylated
Targeting	Passive (EPR)	Passive (EPR)	Direct injection into the CSF compartment	Passive (EPR)
Drug release	Controlled by cholesterol; mechanism not fully understood	Sustained intracellular release over ≥36 h, maintaining cytotoxic levels within tumor cells	Liposome degradation	Passive release
Observations	Avoiding heart damage risk of free DOX	Suitable for tumors with high vascular permeability No longer marketed	No longer marketed	Low systemic toxicity. Reduced incidence of cardiac events and congestive heart failure compared to free DOX
References	[109,117,118,119]	[110,120,121,122,123]	[118,124,125]	[126,127]
Commercial name	Lipusu	Mepact	Marqibo	Onivyde
Manufacturer	Nanjing Luye Pharmaceutical (Shanghai, China)	Takeda France (Courbevoie, France)	Talon Therapeutics (San Francisco, US)	Merrimack Pharmaceuticals (Cambridge, US)
Approval date	2006 (China)	2009 (EU)	2012 (US)	2015 (US) 2016 (EU)
Illness	Breast, ovarian, and lung cancer	Osteosarcoma	Leukemia	Pancreatic cancer
API	Paclitaxel	Mifamurtide	Vincristine Sulfate (VCR)	Irinotecan HCL trihydrate
Composition	Lecithin–Chol	DOPS:POPC (3:7)	SM:Chol (60:40)	DSPC:Chol:MPEG-2000-DSPE (3:2:0.015)
Drug loading	Passive	Passive	Active (pH gradient)	Active (with triethylammonium sucrose octasulfate)
Preparation Method	Thin-film hydration + extrusion	In situ, mixed with 0.9% saline solution.	Ethanol injection + extrusion	Ethanol injection + extrusion
Stability (Particle Size, Zeta Potential and Tc)	<200 nm (Unilamellar)	2–3.5 µm (Multilamellar) 5 °C	130–150 nm (Unilamellar)	~110 nm (Unilamellar); –18 mV; 55 °C
Circulation profile		Rapidly cleared from serum with half-life of 2 h	Low protein binding, which results in a prolonged circulation time for the liposome Non-PEGylated	Long circulation time due to PEGylation, avoiding RES
Targeting	Passive (EPR)	Targeting immune system for immunotherapy	Passive (EPR)	Passive (EPR)
Drug release		Liposome degradation	Long release half-time, up to 117 h	Prolonged release; half-life of drug release up to 56.8 h
Observations	First paclitaxel liposome commercial in China	Used in children and young adults after resection surgery	No longer marketed	Improved tumor accumulation
References	[92]	[118,128,129,130,131]	[32,132]	[133,134,135]
Commercial name	Vyxeos	Celdoxome	Zolsketil
Manufacturer	Jazz Pharmaceuticals (Dublin, Ireland)	Baxter Holding (Utrecht, Netherlands)	Accord Healthcare (Barcelona, Spain)
Approval date	2017 (US) 2018 (EU)	2022 (EU)	2022 (EU)
Illness	Myeloid leukemia	Breast, ovarian, Kaposi’s sarcoma, and Multiple Myeloma	Breast, ovarian neoplasms, and Kaposi’s sarcoma
API	Daunorubicin, cytarabine (1:5)	DOX hydrochloride	DOX hydrochloride
Composition	DSPC:DSPG:Chol (7:2:1)	MPEG-2000-DSPE:HSPC:Chol	MPEG-2000-DSPE:HSPC:Chol
Drug loading	Passive + active (with copper gluconate buffer)	Active (pH gradient of ammonium sulfate)	Active (pH gradient of ammonium sulfate)
Preparation Method	Thin-film hydration + extrusion	Thin-film hydration + extrusion	Thin-film hydration + extrusion
Stability (Particle Size, Zeta Potential and Tc)	107 nm (bilamellar) –33 mV 55.3 °C	75–100 nm (Unilamellar) 55 °C	75–100 nm (Unilamellar) 55 °C
Circulation profile	~50× longer circulation time than free drug Uses anionic phosphtildylgrlycerol as alternative to PEG to avoid ABC phenomenon	The half-life up to 231 h, with a mean of 73.9 h due to PEGylation, avoiding RES	Long circulation times. With an average half-time of 73.9 h
Targeting	Preferential internalization of CPX-351 liposomes	Passive (EPR)	Passive (EPR)
Drug release	Low cholesterol optimized for tumor-controlled release		Passive release
Observations	Dual-drug formulation with a fixed synergistic ratio; in vivo efficacy is drug ratio-dependent Potentially avoids P-gp-mediated efflux, reducing treatment resistance	Authorized as generic	Bioequivalent to Caelyx
References	[136]	[137]	[138,139]

7. Future Trends

Liposomal nanomedicine, after decades of clinical validation with formulations such as Doxil^® and Onivyde^®, is now entering a convergence phase with precision oncology, synthetic biology, and AI-driven design [140,141]. Current therapies face major limitations, including suboptimal tumor penetration, a restricted range of payloads, and heterogeneous patient responses [142]. Overcoming these barriers is a key driver of research, motivating the development of advanced design principles, multifunctional architectures, and patient-tailored approaches.

Hierarchical targeting strategies, addressing tissue accumulation, cellular internalization, and subcellular localization, alongside ligand engineering, aptamer or peptide conjugation, and multivalent display, have demonstrated improved efficacy in preclinical and some early clinical models [143,144,145]. Similarly, liposomal delivery of mRNA, siRNA, and CRISPR/Cas cargoes has transitioned from concept to clinical application, exemplified by LNP-based COVID-19 vaccines [146,147,148,149]. These approaches provide a robust foundation, yet further innovation is required to fully realize precision, adaptive, and personalized cancer therapies.

Biomimetic liposomes are emerging as a particularly promising frontier. Coating liposomes with natural membranes derived from cancer cells, leukocytes, platelets, or NK cells confers immune evasion, homotypic targeting, and extended circulation, enabling access to difficult-to-reach tissues such as the brain or metastatic niches. Recent studies show that the presence of specific proteins in the protein corona is essential for the stealth effect of nanoparticles. For example, leukosomes, which combine leukocyte membranes with liposomes, exhibit longer circulation times and greater accumulation in inflamed tissues [150]. In cardiovascular medicine, platelet-mimetic vesicles have achieved near six-fold enhanced accumulation in atherosclerotic plaques and effective rapamycin delivery [151]. In oncology, dual-membrane designs combining macrophage and glioma membranes achieved sequential blood–brain barrier (BBB) crossing and tumor targeting, significantly enhancing paclitaxel efficacy [152], while transferrin-conjugated glioma–membrane hybrids further improved survival outcomes [153].

Beyond cancer, biomimetic strategies have shown promise in infectious and inflammatory diseases. HA-P3-Lipo vesicles enhanced MRSA (Methicillin-resistant Staphylococcus aureus) clearance and modulated systemic inflammation in sepsis models [154], while biomimetic vaccine platforms (including bacterial outer membrane vesicles, exosomes, and membrane-decorated liposomes) enable stable, targeted, and immunogenic delivery for cancer and infectious disease applications [155,156,157]. In neurodegeneration, NK cell membrane-coated liposomes (BLIPO-CUR) delivered curcumin via meningeal lymphatic vessels, achieving ~20-fold improved brain targeting and mitigating Parkinsonian pathology [158]. Collectively, these studies illustrate how biomimicry transforms liposomes into adaptive, multi-layered nanoplatforms capable of immune modulation, vascular homing, BBB penetration, and neuroprotection.

In parallel, theranostic liposomes have emerged as powerful tools for integrated cancer diagnosis and therapy. Engineered liposomal nanoparticles combine therapeutic and diagnostic functionalities within a single platform, enabling the co-delivery of drugs and imaging agents to tumor sites for real-time monitoring of biodistribution, treatment efficacy, and tumor response. Several systems have reached preclinical and even clinical stages, underscoring their translational potential in image-guided and combination cancer therapy. However, premature drug leakage, limited in vivo stability, and suboptimal tumor accumulation remain major hurdles to clinical adoption [159].

Recent advances have sought to overcome these limitations through innovative design strategies. Gold-coated liposomes, for example, have demonstrated excellent photothermal conversion efficiency in photothermal therapy (PTT); the addition of a secondary liposomal layer and PEGylation enhanced in vivo stability and enabled PET imaging via ⁶⁴Cu radiolabeling, leading to improved tumor targeting and PTT efficacy in breast cancer models [160]. Likewise, thermosensitive ultra-magnetic liposomes loaded with antivascular drugs have been employed for MRI-guided therapy, achieving magnetic accumulation and HIFU-triggered drug release with pronounced tumor regression [161]. Multifunctional nanohybrids integrating liposomes with gold nanoparticles and graphene quantum dots offer dual imaging and therapeutic capabilities, enabling phototriggered chemotherapy and synergistic ROS-mediated tumor ablation [162].

Building upon these advances, hybrid theranostic liposomes are gaining attention for their improved biocompatibility and tumor specificity. Cell membrane-derived or enzyme-activated hybrid liposomes provide tumor microenvironment (TME) responsiveness, charge-reversal properties, and homologous targeting, achieving precise drug and gene delivery with minimal off-target effects [163]. These “smart” hybrid vesicles represent a major step toward integrating diagnosis, targeted therapy, and controlled release within a single nanoplatform. Despite these advances, clinical translation still demands improvements in reproducibility, long-term stability, and optimized biodistribution. Integrating imaging and therapeutic agents in the same vesicle also poses design trade-offs between diagnostic sensitivity and therapeutic potency. Future developments are expected to focus on modular, stimuli-responsive, and personalized liposomal systems capable of precise tumor targeting and real-time treatment monitoring.

Liposomal platforms are also central to nucleic acid delivery and genome editing. Non-viral, brain-targeted mRNA delivery via cationic liposomes has achieved robust expression in the cortex, striatum, and midbrain with minimal systemic exposure, while liposomal formulations optimized for CRISPR/Cas9 cargo demonstrate up to ~80% gene-editing efficiency in preclinical tumor models [164,165]. These developments position liposomes as cornerstone technologies for in vivo genome editing and personalized immunomodulation in oncology.

As liposomal systems become increasingly complex, computational approaches are now essential to guide rational design and optimization. Looking to the future, AI/ML (machine learning)-driven designs are emerging as key enablers of next-generation liposomal nanomedicine. By integrating predictive models with nanotechnology and gene editing, these tools can optimize formulation parameters, such as PEGylation, ligand selection, or hybrid lipid–polymer architectures, improving liposome stability, target specificity, and therapeutic efficacy [166,167]. Machine learning also streamlines nanoparticle synthesis and performance prediction, reducing experimental costs and accelerating translation. When coupled with biomimetic engineering, co-delivery systems, and theranostic approaches, AI-guided pipelines position liposomes to evolve from passive carriers into intelligent, adaptive nanoplatforms capable of delivering personalized and combinatorial therapies across oncology and beyond [168].

Dual-drug and co-delivery systems are also emerging, enabling synergistic combinations such as chemotherapy with immunotherapy or gene silencing. These strategies address tumor heterogeneity and drug resistance while integrating theranostic capabilities for the real-time monitoring of biodistribution and therapeutic responses [169,170,171].

Clinical translation will rely on robust biomanufacturing and predictive pipelines, including large-scale production, patient-derived organoids, and microfluidic tumor-on-chip models that better mimic human tumors [144]. Strategies to improve tumor penetration, including size reduced liposomes or matrix-degrading functionalities, further optimize intratumoral distribution.

In summary, while foundational liposomal platforms have already reshaped clinical practice, the integration of biomimetic engineering, nucleic acid therapeutics, gene editing, and AI-driven design is poised to establish the next generation of liposomes as intelligent, multi-functional agents, capable of personalized, combinatorial, and adaptive therapy in oncology.

8. Conclusions

This review consolidates key physicochemical principles underlying the design and optimization of liposomes for anticancer drug delivery. Cancer was selected as a paradigmatic target, not only due to its clinical relevance but also because it illustrates the full potential of liposomal systems—from chemotherapeutic drug delivery to the transport of biomolecules used in vaccination and immunotherapy. Liposomes are ideal candidates, owing to their remarkable biocompatibility, derived from their structural mimicry of biological membranes and their thermodynamic stability under diverse environmental conditions. Furthermore, their capacity for controlled or thermodynamically triggered release—through surfactant-mediated degradation or membrane poration—makes them versatile carriers for molecules with a broad range of solubilities and therapeutic functions. A detailed understanding of the physicochemical aspects involved in all stages—design, preparation, drug loading, blood circulation, targeting, and release—is essential for the optimization of current and future liposomal formulations in anticancer therapies.