Cryopreservation technology for improving the stability of liposomes and its precise drug monitoring in clinical drug research

ABSTRACT

The instability of liposomes in blood samples during clinical drug research and drug monitoring results in the inability to accurately determine the actual drug concentrations in the body at the time of collection, mainly due to lipid deterioration, particle fusion or aggregation, and phase separation degradation, resulting in payload leakage. To improve drug monitoring accuracy, we developed a cryopreservation strategy in this study by innovatively combining cryoprotective agents (CPAs), such as L-proline, sucrose, and polyvinyl alcohol (PVA), to prevent liposomal leakage and maintain stability for reliable drug monitoring and clinical drug research applications. Doxorubicin liposomes were prepared, and the CPAs were tested at various concentrations and under different freeze‒thaw protocols in biological matrices, with the stability and leakage of the liposomes assessed. Each CPA contributes distinct stabilization mechanisms, with L-proline's osmoprotective ability, sucrose's hydrogen bonding, and PVA's steric hindrance to form a protective barrier. The optimized CPA combination demonstrated superior performance at 85% (v/v) by preserving liposomal integrity, offering the best cryoprotective effect for liposomes in plasma stored at −20 °C, achieving about 90% entrapment efficiency, compared to about 60% in the control group without CPAs. Mechanistic investigations confirmed that CPAs protect liposomes against mechanical stress, prevent membrane disruption, and reduce ice damage by inhibiting recrystallization and adjusting bilayer hydration. These findings offer practical solutions for accurate pharmacokinetic assessments and reliable personalized dosing, safer alternative for liposomal drug research, biobanking, and real-world therapeutic monitoring.

GRAPHICAL ABSTRACT

A multi-cryoprotective approach for preserving liposomal doxorubicin in biological matrices during drug monitoring in clinical drug research.

1. Introduction

Clinical drug research emphasizes the discovery, assessment, and enhancement of pharmaceutical products to verify their efficacy, safety, and pharmacokinetics prior to clinical use (Marques et al. 2024). This method relies on maintaining drug stability in biological samples to provide accurate drug monitoring (Tey and See 2021), especially for intricate delivery systems such as liposomes (Gomez-Lopez 2020; Lai et al. 2024). However, the structural complexity of liposomal drug delivery systems make them susceptible to physicochemical instabilities such as lipid deterioration, degradation (Mouritsen et al. 2016), fusion (Akbarzadeh et al. 2013), phase separation, drug crystallization, and leakage (Crommelin et al. 2020), especially in blood and plasma (Zhai et al. 2024; Wang et al. 2023b). These instabilities result in early drug leakage, complicating the correlation between measured drug levels and real in vivo concentrations (Palmer and Dasgupta 2021).

Cryopreservation, a technique for storing biological materials at ultra-low temperatures, offers a robust solution for preserving the structural and functional integrity of sensitive therapeutic formulations (Yu and Hubel 2019), including liposomal drug carriers (Roque et al. 2022). At ultra-low temperatures, molecular motion is slowed, and a reduction in enzymatic activity, oxidation, and other degradation processes that compromise liposomal integrity (Sydykov et al. 2018a, 2018b) during transport, storage, and freeze‒thaw cycles.

Cryopreservation application is promising, but careful optimization is required to mitigate risks associated with cryo-injuries, ice recrystallization, and osmotic injuries during the freeze‒thaw cycle (Yang et al. 2016, Zhao and Fu 2017), which can disrupt lipid bilayers and accelerate drug leakage (Saadeldin et al. 2020; Chang and Zhao 2021; Whaley et al. 2021). Although conventional cryoprotective agents (CPAs), such as dimethyl sulfoxide (DMSO) and glycerol, demonstrate high vitrification capabilities and have been used to prevent cryo-induced injuries (Susa et al. 2021), their inherent toxicity (Pal et al. 2012; Camp et al. 2020), time-consuming washing processes, and slow transmembrane diffusion rates (Liu et al. 2001; Aita et al. 2005) pose substantial practical barriers.

This study therefore explores the use of non-toxic CPAs such as osmoprotectants (L-proline (Pro)), disaccharides (sucrose (Suc)), and polymers (polyvinyl alcohol (PVA)). Each CPA employs unique mechanisms to provide cryoprotection with L-proline balancing osmosis on both sides of the membrane, thereby mitigating shear forces, preventing water flow, and constituent leakage in a hypertonic environment (Yang et al. 2017). Sucrose forms hydrogen bonds with the polar head groups of phospholipids to inhibit phase transition and membrane damage (Yu et al. 2021), while PVA inhibits ice crystal growth and recrystallization and forms glassy matrices around lipid membranes because of its polymeric network and water-retention ability (Deller et al. 2014), thereby increasing stability and reducing ice-induced stress.

Sucrose and PVA have been mostly utilized in the lyophilization of liposomes during preparation and formulation rather than for cryopreservation (Van Winden et al. 1997), while L-proline, to the best of our knowledge, has not been used in the cryopreservation of liposomes. Therefore, use of L-proline in combination with sucrose and PVA presents a novel approach to increase the stability of liposomes in biological matrices during drug monitoring (Figure 1). While individual components such as sucrose, amino acids, and polymers have been previously applied as cryoprotectants (Bailey et al. 2021), the specific ternary system of L-proline combined with sucrose and PVA has not been systematically evaluated for liposomal stabilization in blood and plasma matrices. The closest instance is a Chinese patent (CN111789109A) detailing cryopreservation formulations comprising an amino acid ice-controlling agent (such as L-proline), a disaccharide (sucrose, 0.1–1.0 M), and optionally PVA (Rong 2022).

Figure 1.

Illustration of the procedural steps and the multi-cryoprotective mechanisms for preserving DOX LIP integrity to enhance accurate drug monitoring. In the presence of CPAs, mechanisms such as hydrogen bonding for water replacement and osmotic protection preserve liposome stability, preventing constituent leakage. Conversely, the absence of CPAs leads to altered liposome integrity because of osmotic stress and ice-induced damage, resulting in leakage.

In this study, we developed a cryopreservation strategy by innovatively combining these non-toxic CPAs to preserve the structural integrity of liposomes and completely prevent leakage during storage and transportation for accurate liposomal drug monitoring. Doxorubicin liposomes (DOX LIP) were used as a model and were formulated using an advanced microfluidic technique, which enhances drug loading and stability of liposomes compared to conventional methods (Wang et al. 2023a; Boafo et al. 2022b). DOX LIP was selected because of its extensive clinical application and thorough characterization as a liposomal chemotherapeutic, with well-documented instability in plasma and whole blood, specifically concerning premature leakage and bilayer degradation (Aloss and Hamar 2023). This clinically relevant formulation allows for the direct assessment of the cryopreservation issues faced during drug monitoring, as they occur in practical liposomal drugs. This establishes a stringent and clinically significant standard for evaluating the necessity and efficacy of CPAs in drug monitoring processes.

Our study integrates formulation accuracy, stability preservation, and practical monitoring procedures within clinically pertinent contexts, unlike previous research that individually investigates liposomal design or cryopreservation. Through dynamic scanning calorimetry (DSC), attenuated total reflectance Fourier transform infrared spectrometry (ATR-FTIR), ice recrystallization inhibition (IRI), and transmission electron microscopy (TEM) investigations, we elucidated the cryoprotective mechanisms at the molecular and bilayer levels. These outcomes present a methodologically rigorous platform that facilitates precise pharmacologically relevant cryopreservation, ensuring accurate pharmacokinetic assays, individualized dosing paradigms, and expanded utility for biobanking, personalized therapeutics, and nanoparticle-based pharmacodynamic monitoring.

2. Materials and methods

2.1. Materials

Dipalmitoylphosphatidylcholine (DPPC) and dipalmitoylphosphatidylglycerol (DPPG) (Xian Ruixi Biotechnology Co. Ltd., China), cholesterol (Sinopharm Chemical Reagent Co. Ltd.), doxorubicin hydrochloride (DOX·HCl) (98% purity, Shanghai Aladdin Biochemical Technology Co. Ltd., China), Dialysis bag (MWCO 35 kDa, Solarbio Science & Technology Co. Ltd.), L-proline (99% purity, neoFroxx Gmbh), poly(vinyl alcohol) (PVA) (99% purity, Hefei BASF Biotechnology Co. Ltd., China), sucrose (99% purity, Tianjin Hengxing Chemical Preparation Co. Ltd., China), condensed sheep blood (Hongquan Biological®, China), methanol HPLC/Spectro (Anaqua Chemicals Supply), HPLC-grade formic acid (>88% purity, Tianjin Comeo Chemical Reagent Co. Ltd., China), SOLA HRP SPE cartridge 10 mg/1 mL (Thermo Scientific™), and purified water were used.

2.2. Preparation of DOX LIP using microfluidic chip-based coaxial electrostatic spray method

DOX LIP was formulated using a coaxial microfluidic electrostatic spray technique (Wang et al. 2023a). Phospholipids, including DPPC, DPPG, and cholesterol, were combined at a 2:2:1 mass ratio in methanol. The hydrophilic drug DOX was dissolved in 90% methanol at 0.5 mg/mL. A three-phase coaxial flow was established, with the outer, middle, and inner phases comprising 40% ethanol, phospholipid, and 90% methanol solutions, respectively. A polytetrafluoroethylene (PTFE) tube introduced the three-phase stream into the microfluidic chip, and droplets formed within a glass capillary. The external electric field generated a Taylor cone, drawing the organic solvent into tiny droplets.

Liposomes were collected and evaporated, and the formulation was dialyzed to remove unencapsulated DOX and any remaining solvent. The gas chromatographic analysis of residual solvents in the liposomes revealed that less than 6% of each solvent was present, as reported in our recently published article in the Chemical Engineering Journal, which employed similar advanced microfluidic techniques in liposome formulation (Shaw et al. 2025b). Moreover, the dialysis procedure, which serves as a purification step, almost completely removes all these residual solvents, which is in line with the ICH Q3C guidelines, and no organic solvents remain at biologically significant concentrations.

2.3. Characterization of DOX LIP

2.3.1. Particle size determination

The dynamic light scattering principle (Zetasizer Nano ZS 90, Malvern Co. Ltd., UK) was used to measure the particle size, polydispersity index (PDI), and Zeta potential. A total of 20  μL of liposome solution was diluted with 1  mL of ultrapure water into the particle size pool. The hydrated particle size of the liposomes was measured using a Malvern particle size-zeta potentiometer, and the average value was calculated at room temperature. The PDI was used as a size distribution measurement of the particle size. A pipette was used to aspirate the identical liposome sample solutions, which were subsequently transferred to the Zeta potential sample cell. After ensuring that the liquid level remained below the limit line, the Zeta potential of the liposomes was measured five times, and the average value was also computed. The Zeta potential, known as the surface charge, was obtained to measure the electrophoretic mobility at 25 °C. In contrast, the particle size was determined as an intensity-weighted Gaussian distribution, with a chi-square value < 3.

2.3.2. Transmission electron microscopy (TEM)

A TEM (Titan G2 60-300, FEI Co. Ltd., USA) operated at an accelerating voltage of 200  kV was used for the liposome imaging. The prepared liposomes were diluted 10-fold. A total of 10 µL of the sample was placed onto a copper mesh and dried at 25 °C for approximately 5 minutes (min). A filter paper was used to remove the excess sample. The sample was then stained with 2% phosphotungstic acid for approximately 30 seconds (s). The excess phosphotungstic acid was removed, and the sample was desiccated at 25 °C for approximately 5 min. Finally, the TEM image of the DOX LIP was scanned.

2.3.3. Drug release profile

The in vitro drug release assay was conducted using the dialysis method (Wallenwein et al. 2019; Boafo et al. 2022b) at three different pH levels (5.4, 6.4, and 7.4). The process involved adding 2 mL of liposome solution to a pre-treated dialysis membrane, then placed in phospate-buffered saline (PBS), followed by oscillating at 37 °C and 100  rpm under dark conditions in a shaker incubator (HZQ-X300C, Bluepard Instruments, China). At different times, 2 mL of the release media was withdrawn and replenished with fresh media. DOX was quantified using high-performance liquid chromatography (HPLC) at 490 nm. Free DOX across all pH levels were used as a control.

2.3.4. Drug loading and encapsulation efficacy

This study utilized HPLC to quantify DOX using reference standard solutions. A stock solution (1 mg/mL) of DOX in methanol was prepared and diluted to various concentrations of 1, 2, 4, 6, 8, and 10 µg/mL. The peak areas were determined using an HPLC-UV detector. A sample solution was prepared by adding methanol to the DOX LIP and sonicating for 30 min at 37 °C. It was ensured that the liposome sample used was pure by removing unencapsulated drugs using dialysis. A non-sonicated control sample was used to confirm that leakage did not occur during sample preparation prior to lysis. The standard calibration curve was constructed using linear regression. The encapsulation efficiency (E.E.) and drug loading (D.L.) of DOX were calculated using Equations (1) and (2).

$$\text{E.E}(\%)=\frac{\mathrm{Amount}\mathrm{of}\mathrm{drug}\mathrm{encapsulated}}{\mathrm{Amount}\mathrm{of}\mathrm{drug}\mathrm{added}}\times 100\%$$ (1)

$$\text{D.L}(\%)=\frac{\mathrm{Amount\; of\; drug\; encapsulated}}{\mathrm{Total\; amount\; of\; materials}+\mathrm{amount\; of\; drug\; encapsulated}}\times 100\%$$ (2)

2.3.5. In vitro DOX LIP stability

The in vitro stability of the DOX LIP was assessed by monitoring the particle size and PDI over 90 days when it was stored at 4 °C. Using dynamic light scattering (DLS), particle size distribution data were collected on days 1, 2, 5, 10, 15, 30, 40, 50, 60, and 90 to investigate changes in liposome size or PDI. Three replicate measurements were performed under identical conditions, and the data were analyzed.

2.4. Development of cryopreservation strategy for enhanced therapeutic drug monitoring

2.4.1. Preparation of CPAs

Three CPAs, PVA, Pro, and Suc, were used both individually and in combination. Specifically, the CPAs were dissolved in PBS to produce different concentrations of each CPA (0.1, 0.2, 0.5, 1, 2, and 5 M for Pro and Suc. 1, 2, and 5 mg/mL for PVA). CPA combinations were also prepared by adding the best concentration of the individual CPAs in the CPA volume ratio to liposomes (70% (v/v), 80% (v/v), 85% (v/v), and 90% (v/v)), as shown in Table 1. The concentration ranges of CPA to liposomes used in our investigation were adapted from previous studies, which showed that different types of cryoprotectants and their mixtures, within 1–6 M and 80% (v/v), inhibited membrane leakage in liposomes and provided storage stability (Sydykov et al. 2018a, 2018b). Additionally, these concentrations solubility and osmolarity limits are compatible with blood and plasma matrices.

Table 1.

The concentration of CPAs added to DOX LIP to obtain the CPA combinations.

CPA ratio (v/v)	PVA/Pro/Suc	PVA/Pro	PVA/Suc	Pro/Suc
70%	0.23 mg·mL⁻¹/1.17 M/0.47 M	0.35 mg·mL⁻¹/1.75 M	0.35 mg·mL⁻¹/0.70 M	1.75 M/0.70 M
80%	0.26 mg·mL⁻¹/1.33 M/0.53 M	0.40 mg·mL⁻¹/2.00 M	0.40 mg·mL⁻¹/0.8 M	2.00 M/0.8 M
85%	0.28 mg·mL⁻¹/1.42 M/0.57 M	---	---	---
90%	0.30 mg·mL⁻¹/1.50 M/0.60 M	0.45 mg·mL⁻¹/2.25 M	0.45 mg·mL⁻¹/0.9 M	2.25 M/0.9 M

2.4.2. Cryopreservation of DOX-LIP (CPA screening)

The effectiveness of cryoprotectants in enhancing liposome stability was done using a systematic approach to select the most effective CPAs (Date et al. 2010; Jangle and Thorat 2013). The particle size and PDI of the liposomes were measured after treatment with various cryoprotectants and freeze‒thaw protocols. The slow-freezing slow-thawing (SFST) and fast-freezing fast-thawing (FFFT) protocols were used. SFST was performed using a laboratory-scale freeze-dryer to cool samples from ambient temperature to −20 °C at a controlled rate of 1 °C min⁻¹ for 20 min, then thawed to 10 °C at 1 °C min⁻¹ for 60 min. The temperature profile was verified using a thermocouple probe inserted into a reference vial, confirming consistent cooling and thawing rates (±0.1 °C min⁻¹).

FFFT involved immersing samples in liquid nitrogen (−196 °C) for 5 min and thawing them at 25 °C for 20 min at ambient pressure. This protocol approximates a rapid cooling rate exceeding 100 °C min⁻¹, based on prior validation in cryobiological systems (Leibo and Pool 2011; Abdelhady et al. 2024). Both protocols were repeated seven times for each sample, with untreated liposomes used as a control.

2.4.3. Chromatographic measurements of DOX leakage from liposomes in buffer

The retention of DOX in liposomes after exposure to CPAs and cryogenic temperatures was analyzed using liquid chromatography, similar to the methods of Crowe and colleagues (Crowe and Crowe 1988). CPAs (80% (v/v) were added to the DOX LIP at a lipid concentration of 1.13  mg/mL, which was subsequently stored in liquid nitrogen and then analyzed after 24 h. The DOX LIP samples were diluted in 2 mL of 10 mM HEPES (pH 7.5), and the maximum peak area after complete leakage was determined by adding 60  μL of 1% (v/v) Triton X-100 to 2 mL of medium with 250  μL of liposome sample (50 μL liposome solution and 200 μL CPA). Initially trapped DOX levels were assessed directly after liposome preparation, and the percentage of DOX retention inside liposomes was calculated by comparing initial peak area values with those after adding CPAs and freezing using Equation (3).

$$\mathrm{DOX}\mathrm{retention}(\%)=\hspace{0.17em}\frac{{\mathrm{\text{P}}}_{\mathrm{B}}-{\mathrm{\text{P}}}_{\mathrm{A}}/{\mathrm{\text{P}}}_{\mathrm{B}}}{{\mathrm{\text{P}}}_{\mathrm{y}}-{\mathrm{\text{P}}}_{\mathrm{x}}/{\mathrm{\text{P}}}_{\mathrm{y}}}\times 100$$ (3)

P_A and P_B represent the peak area readings of treated samples before and after the addition of Triton X-100, respectively. P_x and P_y are the reference peak area values before treatment and after the addition of Triton X-100, respectively. The same procedure was repeated for the CPA combinations, and the liposome's DOX retention was assessed.

2.5. Differential scanning calorimetry (DSC) determination of CPA effect on phase transition temperatures

The phase transition behaviors of liposomes in the presence of CPAs were analyzed using differential scanning calorimetry (DSC, TA, DSC2A-02169). The effect of PVA, L-proline, sucrose, and their combinations on the freezing point depression and melting point of the DOX LIP were assessed. The optimal concentrations of CPAs were prepared in PBS (1  mg/mL PVA, 5 M Pro, and 2 M Suc) and added to the DOX LIP. A total of 10  μL of the samples were placed in a 25 μL aluminum pan and sealed hermetically. DSC was used to screen the ability of freezing point depression and melting point by cooling the samples to −60 °C at −10 °C min⁻¹, followed by heating to 30 °C at 10 °C min⁻¹ while monitoring the heat flow. The thermal events were then analyzed from the resulting thermograms.

2.6. Determination of CPA ice recrystallization inhibitory (IRI) properties

The standard splat assay was used to determine ice recrystallization inhibition (IRI) (Ampaw et al. 2021). In detail, a microscope slide was placed on dry ice to pre-cool to −78 °C. A total of 6  μL samples (PVA (1  mg/mL), L-proline (1, 20, 40, and 60  mg/mL) and sucrose (1, 20, and 40  mg/mL) were dropped from a height of approximately 1.5 m onto the glass slide and then instantly dispersed and frozen to form a thin layer. The microscope slide was then quickly moved into a cold stage maintained at −8 °C. The sample was annealed for 20 min to allow sufficient time for ice recrystallization. The ice crystals were observed with a polarizing microscope, and the mean largest grain size (MLGS) was obtained by calculating the mean size of the ten largest crystals in the field of view. MLGS is an index for the quantitative analysis of IRI activity.

2.7. Infrared spectroscopic analysis of Interaction between the CPAs and lipids

A Fourier transform infrared (FTIR) spectrometer with an attenuated total reflection (ATR) accessory was used to study CPA‒lipid interactions. The spectra were recorded using an automatic CO₂/H₂O vapor correction algorithm. The samples were incubated at room temperature for 30 min and then added to the ATR sample holder. The interaction was examined by monitoring the phospholipid head group and hydrocarbon tail, focusing on the position of PO₄-asymmetric and PO₄-symmetric stretching vibrations (νPO₄) and CH₂-stretching vibrations (νCH₂) within the 1260–1220 and 1110–1070 cm⁻¹, respectively. Shifts in the νPO4 and νCH₂ peak positions were analyzed.

2.8. Cryoprotective resistance of CPAs on liposome morphology

The ability of cryoprotectants to protect the morphology of liposomes from ice injury during freezing was assessed via TEM. Optimal CPA solutions (1 mg/mL PVA, 5 M Pro, and 2 M Suc) were added in various combinations and then to DOX LIP at 85% (v/v). The CPA cocktails were allowed to undergo distribution equilibrium with the liposomes for approximately 30 min to 1 hour (h), ensuring uniform dispersion. The samples were then stored in liquid nitrogen (−196 °C) for 24 h and thawed at 25 °C, based on the FFFT technique. The excess CPAs were then removed using dialysis, and TEM was performed similarly to the previous TEM method.

2.9. Assessment of liposome stability in biological matrices

Condensed sheep blood commercially purchased from Hongquan Biological® (China) was used for ex vivo cryoprotection analysis to simulate the drug monitoring process. The product was provided under certified veterinary inspection and did not require procedures conducted on live animals; therefore exempting it from institutional ethical review. The solid phase extraction (SPE) procedure was used to separate the free DOX from the encapsulated DOX in the biological samples. SPE was performed with a SOLA HRP SPE cartridge 10 mg/1 mL. The HPLC-UV was used to analyze the extent of DOX leakage following the SPE.

2.9.1. Solid phase extraction method

Using the SOLA HRP SPE cartridge, 0.5 mL of methanol was used to purge the system, followed by equilibration with water. The pre-treated sample was loaded and washed with PBS (pH 7.4). The effluent containing the liposomal doxorubicin fraction was further cleaned with 1 mL of water/methanol (90:10 v/v), followed by elution with 0.5  mL of methanol +0.1% formic acid. The resulting eluent was then diluted with an equal amount of water before HPLC analysis.

2.9.2. Baseline evaluation

The stability of the DOX LIP in biological matrices (plasma and whole blood) stored without CPAs was assessed as a baseline evaluation of DOX leakage. The DOX LIP was added to the plasma and whole blood, vortexed for approximately 30 s and allowed to reach distribution equilibrium within the biological matrix for approximately 30 min to 1 h at room temperature to ensure consistent initial contact between the liposomes and matrix components, and then stored at 4 °C. At various time points, the samples were thawed at 37 °C. The whole blood samples were thawed and then centrifuged at 1800 × g for 10 min, and the supernatant was collected, while the plasma samples were only thawed. The SPE method was then used to separate the released DOX from the encapsulated DOX in the samples, followed by the HPLC analysis as described by Dara et al. (Krishna Rao Dara and Liddicoat 2014). This was done to determine the leakage of DOX without the addition of any CPAs.

2.9.3. Cryoprotective effect on DOX LIP stability during drug monitoring

To examine the effect of CPAs on the stability of liposomes in biological samples, optimal CPA concentrations of 1 mg/mL PVA, 5 M Pro, and 2 M Suc were prepared. The DOX LIP was added to the plasma and blood, gently vortexed, and allowed to reach distribution equilibrium within the biological matrix at room temperature, ensuring uniform dispersion of the liposomes and initial lipid‒protein corona formation prior to storage.

At different concentrations, the CPA combination (PVA/Pro/Suc) was added to the blood and plasma samples containing DOX LIP, as shown in Table 2. The samples were stored at 4, −20, and −196 °C (liquid nitrogen) for 24 h. After that, SPE method was used to separate the released DOX and the encapsulated DOX. Then, the liposome's entrapment efficiency (Equation (1)) was subsequently assessed using the HPLC-UV.

Table 2.

Volume of blood or plasma, DOX LIP, and CPA concentrations.

CPA ratio (v/v)	Whole blood (μL)	Plasma (μL)	DOX LIP (μL)	PVA/Pro/Suc (μL)	Final CPA concentration (PVA/Pro/Suc)
70%	400	400	100	233.3	0.069 mg·mL⁻¹/0.34 M/0.14 M
80%	400	400	100	400	0.103 mg·mL⁻¹/0.51 M/0.21 M
85%	400	400	100	600	0.133 mg·mL⁻¹/0.67 M/0.27 M
90%	400	400	100	900	0.167 mg·mL⁻¹/0.83 M/0.33 M

Hemolysis was quantified spectrophotometrically at 540 nm and expressed as % hemolysis using Equation (4) for blood samples after freeze‒thaw cycles at all temperatures. Red blood cells (RBCs) in PBS were used as a negative control, whereas RBCs lysed in 1% Triton X-100 were used as a positive control.

$$\text{Hemolysis}(\%)=\frac{A_{\text{sample}}-A_{\text{negative control}}}{A_{\text{positive control}}-A_{\text{negative control}}}\times 100$$ (4)

where, A_sample represents the absorbance of the test sample, A_{negative control} represents the absorbance of RBCs in PBS, and A_{positive control} represents the absorbance of fully lysed RBCs.

The long-term ability of the optimal CPA cocktail to protect liposomes in the optimal biological samples was also examined by storing the samples at their optimal storage temperature for up to 90 days, after which their entrapment efficiency was analyzed.

2.10. Statistical analysis

The data were analyzed and expressed as the mean ±  standard deviation (S.D.) using GraphPad Prism 10 software. All the statistical analyses and comparisons were done using one-way analysis of variance (ANOVA) for the single-factor experiments and two-way ANOVA for the multiple-factor experiments. The Tukey's test was used to determine significant differences between groups for post-hoc comparisons. ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05 were considered significant. A minimum of five separate replicates were carried out concurrently under similar conditions and averaged to produce each reported value.

3. Results and discussion

3.1. Formulation and characterization of doxorubicin liposomes

A successful microfluidic chip combined with the coaxial electrospray technique was designed and used to formulate the DOX LIP, as shown in Figure 2A. By adjusting the flow rate ratios and total flow rates, the hydrodynamic diameter and PDI of the DOX LIP were found to be 136.87 ± 0.72 nm and 0.24 ± 0.01, respectively, with a negative zeta potential of −12.16 ± 0.65 mV (Figure 2B). The liposomes made of DPPC and DPPG loaded with DOX in their hydrophilic cores had an entrapment efficiency of 85.82 ± 1.37% and a drug loading rate of 12.00 ± 2.03% after HPLC-UV was used for detection (Figure 2D). TEM imaging confirmed the spherical morphology of the nanoparticles with a particle size of approximately 100  nm (Figure 2C).

Figure 2.

Characterization of DOX LIP. (A) Schematic representation of the advanced microfluidic design system made of a three-phase chip. Replicated from ref (Shaw et al. 2025a) with permission. (B) Size distribution of DOX LIP with Zeta potential. (C) TEM image showing the morphology of DOX LIP (this same image is also presented in Figure 11A to illustrate fresh DOX LIP before cryopreservation, since both panels represent the same experimental conditions). (D) The standard calibration curve of DOX derived from HPLC readings. (E) TEM image of DOX LIP after storage at 4 °C for 15 days. (F) Storage stability study of DOX LIP at 4 °C over a period of 90 days. ns represents no significant difference. (G) In vitro drug release profile of free DOX and DOX LIP in PBS at pH 5.4, pH 6.4, and pH 7.4. All data points represent the mean values ± S.D. (n = 5). Scale bar = 100 nm.

The microfluidic chip combined with the coaxial electrospray technique produced homogenous liposomal particles with relatively small particle diameters, as well as higher encapsulation efficiency and drug loading rates than other microfluidic methods (Hood et al. 2013; Maeki et al. 2022), where there is lower drug loading and entrapment efficiency since doxorubicin is loaded via an ammonium sulfate gradient or via passive diffusion. Additionally, in comparison with liposomes prepared using the thin film hydration method in our previous research (Boafo et al. 2022b), the advanced microfluidic electrospray technique produced better liposomes with a smaller particle size, higher encapsulation efficiency, and drug loading capacity.

This finding aligns with Wang et al.'s study, where the advanced microfluidic technique was also shown to produce better liposomes compared to the thin film hydration technique (Wang et al. 2023a). The success of this advanced microfluidic technique can be attributed to the unique design of the three-phase chip in which the solution was stretched into a Taylor cone by a high-pressure power supply in a stable laminar flow state, where the aqueous phase inside containing DOX was wrapped in the hollow core while forming phospholipid hydration to form liposomes.

The stability of the DOX LIP was evaluated during storage by monitoring changes in morphology, particle size, and PDI. No significant changes in morphology, particle size, or PDI were observed following storage at 4 °C for 90 days, as shown in Figure 2F, and the TEM image after 15 days is presented in Figure 2E, hence indicating the superb placement stability of the DOX LIP. The excellent stability of the DOX LIP affirms the ability of cholesterol to increase membrane rigidity and decrease permeability by reducing fluidity, which is consistent with findings of Farzaneh et al. (2018) but extends the verification over longer storage periods. In brief, this stability is crucial for its therapeutic efficacy in prolonged drug delivery and reduced clearance from the bloodstream.

The in vitro cumulative drug release profile of the DOX LIP was also examined under various pH conditions to observe the release properties of DOX in the liposomes. As shown in Figure 2G, the release of DOX from the liposomes was observed under pathological (pH 5.4, pH 6.4) and physiological (pH 7.4) conditions relevant to drug delivery. Free DOX was factored into the quantification to correct for membrane-associated drug loss and ensure that the release profiles reflected true liposomal stability. Free DOX release rapidly reached 95.44 ± 0.87% at pH 5.4 and 87.87.70 ± 1.72% at pH 7.4 within 12 hours, and complete release was recorded within 24 hours at both pH values, compared with liposomal DOX at all pH values.

DOX LIP exhibited sustained and pH-sensitive release of DOX, which reached equilibrium within 24 hours, with the highest release of 74.55 ± 1.07% and the lowest release of 49.22 ± 0.07% observed at pH 5.4 and pH 7.4, respectively. The variation in the release could be associated with the different conditions mimicked. Dox is released from the liposomes in a pH-sensitive manner, with the fastest release occurring at a lower pH (pH 5.4), which aligns with prior observations of pH-sensitive liposomes (Aghdam et al. 2019). This low pH mimics endosomal conditions where the highly acidic intracellular environment causes disruption of the liposomal membranes, leading to significant drug release inside cells to exert its therapeutic effect.

3.2. Impact of cryoprotective agents on liposome stability over repeated freeze-thaw cycles

The stress of freezing and thawing has been shown to influence the stability of lipid membranes (Van Winden et al. 1997). Excipients can be incorporated to mitigate the mechanical stress caused by ice crystal formation, cryo-concentration caused by ice front propagation, membrane fusion, and dimensional expansion (Litschel et al. 2018). Hence, CPAs (PVA, L-proline, and sucrose) were added to the DOX-loaded liposomes, and several freeze‒thaw cycles were used to test the ability of the liposomes to withstand freezing stressors using two distinct techniques: SFST and FFFT.

While these freezing techniques do not replicate long-term pharmaceutical storage conditions (−20 °C or 4 °C), they act as accelerated stress models commonly employed to assess CPA efficacy and the stability of liposomal membranes under severe freeze‒thaw conditions. CPAs were added to liposomes individually and in combination at varying concentrations, and their particle size and PDI were analyzed using the DLS after the freezing techniques. The concentrations of Pro and Suc added were 0.1–5 M, whereas PVA was added at concentrations of 1–5 mg/mL.

3.2.1. Slow freezing and slow thawing

After seven cycles, the SFST method resulted in significant differences in the size distributions of the liposomes in the presence of PVA, L-proline, and sucrose compared to those of the pretreated liposomes (DOX LIP). The particle size of liposomes preserved without CPAs (Wt CPAs) was larger (192.17 ± 0.49 nm), with a significant decrease in PDI (0.11 ± 0.02) compared to the DOX LIP (136.87 ± 0.72 nm). As shown in Figure 3A, L-proline and sucrose resulted in significant increases in particle size at all concentrations except at 1 M and 5 M Pro and 2 M Suc, where liposomes did not significantly increase in size.

Figure 3.

Slow-freezing slow-thawing effect on liposomes showing (A) particle size of DOX LIP after cryopreservation without CPA (Wt CPA), and with L-proline and sucrose. (B) PDI of DOX LIP after cryopreservation of Wt CPA, and with L-proline and sucrose. (C) Particle size of DOX LIP after cryopreservation of Wt CPA, and with PVA. (D) PDI of DOX LIP after cryopreservation of Wt CPA, and with PVA. All data points represent the mean values ± S.D. (n = 5).

Figure 3B shows reduced PDI values after freeze‒thaw cycles with L-proline and sucrose at all concentrations, where 0.1 M, 0.2 M, and 0.5 M exhibit the lowest values. PVA at concentrations of 2  mg/mL and 5  mg/mL failed to prevent liposomes from being damaged by ice crystal formation, resulting in liposomes with the smallest size (Figure 3C), but their PDI increased tremendously, with a PDI greater than 0.6 (Figure 3D). However, liposomes preserved with 1  mg/mL PVA showed increased particle size (188.40 ± 12.12 nm, Figure 3C) with reduced PDI.

Research has shown that a slow freezing rate can damage the lipid bilayers by forming very large crystals (Yang et al. 2021). An increase in liposome size frequently signifies membrane destabilization, possibly arising from liposome fusion or aggregation during freeze‒thaw cycles, which aligns with prior findings (Crowe et al. 1984). This fusion may jeopardize the lipid bilayer, resulting in gaps or defects in the membrane, hence facilitating the leakage of encapsulated DOX.

Additionally, water within and surrounding the liposomes may induce osmotic stress throughout the gradual freezing and thawing process. This stress may cause the liposome to enlarge, resulting in an increased size and probable rupture of the membrane, leading to drug leakage. Furthermore, if these consistently larger liposomes emerge via fusion or aggregation, they may exhibit reduced stability and increased susceptibility to leakage, as the encapsulated drug has greater potential to escape through the compromised membrane.

3.2.2. Fast freezing, fast thawing

The FFFT technique employed liquid nitrogen (−196 °C), in which liposome samples are immersed to flash freeze and then quickly thawed at 25 °C. This technique is frequently employed in biomedical applications to guarantee that delicate molecules, such as liposomal drugs, maintain their efficacy and structural integrity post-cryopreservation. It differs from slow freeze‒thaw cycles which often facilitate the formation of larger ice crystals, hence heightening the danger of disrupting liposome membranes and resulting in drug leakage.

Figure 4A and 4B show that liposomes cryopreserved Wt CPA compared to the DOX LIP had a higher particle size and a slightly reduced PDI, respectively. The liposomes exhibited various effects based on the type and concentration of CPA employed following the FFFT cycles. L-proline preserved the liposomes best at the 5 M concentration, as shown in Figure 4A, with a particle size of 135.80 ± 8.04 nm and a PDI of 0.22 ± 0.04, whereas sucrose at 1 and 2 M maintained the liposome particle size (Figure 4A), with the best preservation occurring at 2 M since both the particle size (131.23 ± 4.56 nm) and PDI (0.22 ± 0.02, Figure 4B) were more stable compared to before freezing (DOX LIP).

Figure 4.

Fast-freezing fast-thawing effect on liposomes showing (A) particle size of DOX LIP after cryopreservation of Wt CPA, and with L-proline and sucrose. (B) PDI of DOX LIP after cryopreservation of Wt CPA, and with L-proline and sucrose. (C) particle size of DOX LIP after cryopreservation of Wt CPA, and with PVA. (D) PDI of DOX LIP after cryopreservation of Wt CPA, and with PVA. All data points represent the mean values ± S.D. (n = 5).

As shown in Figure 4C, the particle size of the liposomes with PVA increased regardless of the PVA concentration after freeze–thaw cycles. However, liposomes with 1  mg/mL PVA exhibited a stable PDI compared to the other concentrations, suggesting that the liposomes maintained their uniform size distribution, as presented in Figure 4D. Hence, the observed increase in hydrodynamic diameter upon the addition of PVA aligns with the adsorption or coating of PVA on the liposomal surface, resulting in the formation of a hydrated steric layer that increases the measured hydrodynamic radius (Cao et al. 2022).

On that basis, the optimal effect of PVA after freeze-thaw was observed at 1  mg/mL, with a liposomal particle size of 178.30 ± 2.25 nm and a PDI of 0.20 ± 0.03, which was closest to that of the DOX LIP. Rapid freezing in liquid nitrogen has been shown to reduce ice crystal size, preserving liposomal integrity (Fahy et al. 2020), but systematic optimization with non-toxic CPAs has been limited. This work demonstrates that the FFFT method combined with specific CPAs (5 M L-proline, 2 M sucrose, or 1 mg/mL PVA) significantly preserves the liposomal size and PDI after repeated freeze‒thaw cycles, outperforming the SFST method and aligning with practical needs for storage.

The variations in particle size noted throughout the SFST and FFFT protocols indicate modifications in lipid bilayer integrity caused by freeze‒thaw stress. Freezing induces extracellular ice crystallization, which excludes solutes and causes hyperosmotic dehydration in liposomes, leading to tighter phospholipid packing and increased vesicle‒vesicle interactions (Boafo et al. 2022a). Thawing, rehydration, and mechanical relaxation facilitate membrane fusion, aggregation, and defect formation, which are observed macroscopically as increases in hydrodynamic diameter and PDI (Susa et al. 2021). The effects are exacerbated by phase transitions between gel (Lβ) and liquid-crystalline (Lα) lipid states, resulting in packing discrepancies and temporary cracks that damage bilayer integrity (Boafo et al. 2022a).

In accordance with established cryobiology principles, cryoprotectants such as sucrose and proline reduce these effects by stabilizing the hydration layer and preserving headgroup spacing (Murray and Gibson 2022), while PVA, owing to its substantial polymeric nature, results in steric expansion of the hydrodynamic shell (Boafo et al. 2022a). The relatively smaller size perturbations noted under FFFT compared to SFST for certain CPA combinations suggest enhanced molecular stabilization of the bilayer, which is likely attributable to more uniform cooling and diminished osmotic shock. The trends in particle size during freeze‒thaw cycling act as sensitive indicators of molecular-level rupture or preservation of the liposomal membrane structure.

3.2.3. Comparison between fast freezing, fast thawing and slow freezing, slow thawing

The two freezing rates exhibited different effects at the same cryoprotectant concentrations, with liposomes generally having better outcomes under a fast freezing rate than under controlled slow freezing. One possible explanation for this phenomenon is the size of the ice crystals that developed during the freezing step, in which immersion in liquid nitrogen produced smaller ice crystals, which can induce less stress on the liposome membranes (Susa et al. 2021). The cryoprotectants were observed to protect the liposomes at higher concentrations (1–5 M) in both freeze‒thaw techniques, with a better protective effect in FFFT.

A comparison of the effects of the optimal CPA concentrations (5 M Pro, 2 M Suc, and 1  mg/mL PVA) on the DOX LIP, which is the control for pre-freezing the liposomes, they showed to effectively protect the liposomes and their constituents under fast freezing rates, as shown in Figure 5A and 5B. This could possibly be due to the unique properties of cryoprotectants, with L-proline and sucrose maintaining osmotic balance and dehydration during freezing, respectively. Whereas PVA provides a protective shield against mechanical stress and ice crystal formation (Mitchell et al. 2019).

Figure 5.

Comparison between slow-freezing slow-thawing and fast-freezing fast-thawing between their (A) particle sizes and (B) PDI. The data points represent the mean values ± S.D. (n = 5), two-way ANOVA followed by Tukey's test was used, ns represents no significant difference, * represents p < 0.05, ** represents p < 0.01, and **** represents p < 0.0001.

Although the current SFST and FFFT models offer important insights into CPA-mediated preservation under repeated freeze‒thaw stresses, they are laboratory stress simulations rather than clinical storage procedures. This study employed accelerated freeze‒thaw models to simulate high cryogenic stress, aiming to clarify CPA performance and identify formulations that are most resistant to ice crystal-induced damage. Clinical preparations of liposomal drugs are generally maintained at regulated temperatures (−20 or 4 °C) with specified cooling rates, which are subsequently explored via controlled-rate freezing and long-term storage in biological matrices.

While the FFFT protocol demonstrated significant cryoprotection, the clinical application of rapid plunge-freezing demands caution, as ultra-fast cooling may entrap liposomes in non-equilibrium bilayer states or produce minor defect subpopulations that are not consistently identifiable through DLS measurements (Baranova et al. 2023). Additionally, rapid freezing can produce uncommon micron-scale aggregates and alter surface chemistry, influencing protein corona formation, complement activation, and in vivo biodistribution, which are essential factors for intravenous drug delivery (González-garcía et al. 2022). For these reasons, even though FFFT is highly effective for screening and short-term stability assessment, its advancement to translational studies must be substantiated by complementary orthogonal structural analyses (cryo-TEM, SAXS), leakage-kinetics assays, hemocompatibility evaluations, and complement-activation profiling to guarantee comprehensive safety and translational feasibility.

3.3. Combinational effects of cryoprotectants on doxorubicin liposomes

To harness the varying effects of the cryoprotectants PVA, L-proline, and sucrose, their optimal concentrations that best preserve liposomes were assessed in various combinations. Since the fast-freezing rate was optimal, the effect of CPA combinations was evaluated using the FFFT method by analyzing the size distribution after seven cycles. Figure 6A shows that all the two combinations of CPAs added to liposomes at higher concentrations (90% (v/v)) caused the liposomes to aggregate, resulting in larger particle sizes, except for PVA/Pro, where the particle size did not increase significantly compared to DOX LIP. However, as shown in Figure 6B, the PDI of PVA/Pro decreased at 90% (v/v), indicating drug leakage. Overall, the combination of the 3 CPAs (PVA/Pro/Suc) preserved the liposomes best at 85% (v/v) compared to DOX LIP, as shown in Figure 6C.

Figure 6.

Combination effect of cryoprotectants. (A) Effect of PVA/Pro, PVA/Suc, and Pro/Suc on the particle size of DOX LIP after cryopreservation. (B) Effect of PVA/Pro, PVA/Suc, and Pro/Suc on the PDI of DOX LIP after cryopreservation. (C) Effect of PVA/Pro/Suc on the particle size distribution of DOX LIP after cryopreservation. The data points represent the mean values ± S.D. (n = 5).

While prior studies recognized the protective roles of individual CPAs (e.g. sucrose and glycerol as osmoprotectants) (Kim and Popova 2023) but lacked systematic combinatorial analysis, this work is the first to demonstrate that the combination of these particular CPAs (PVA/Pro/Suc) at 85% v/v optimally preserves liposomes during FFFTs, maintaining the particle size and PDI compared to pre-freeze (DOX LIP). This suggests that the three CPAs might work in a complementary and multimodal manner to maintain the liposome size as L-proline penetrates the aqueous core to minimize the osmotic gradient caused by cryoconcentration (Kannan et al. 2015), while PVA and sucrose remain in the external compartment of liposomes to stabilize the lipid bilayers, hence limiting modifications in physical properties and potential drug leakage (Franzé et al. 2018).

3.4. Cryoprotectant effects on liposomal membrane permeability and stability

The efficacy of the CPAs, that is, PVA, L-proline, and sucrose, in maintaining the stability of DOX-loaded liposomes was further determined by measuring DOX leakage following incubation with the cryoprotectants and subsequent optimal freeze‒thaw cycles, which is the FFFT. High DOX retention indicates excellent liposome stability. Liposomal samples stored without CPAs have a DOX retention of about 50%–60% (Figure 7A and 7B); approximately 40% of DOX leaks out of the liposomes when stored without any excipient.

Figure 7.

Retention of DOX in liposomes after the addition of (A) PVA only, (B) L-proline and sucrose, (C) PVA/Pro, PVA/Suc, and Pro/Suc, (D) PVA/Pro/Suc, and then cryopreserving with various CPA-to-liposome volume ratio concentration. The data points represent the mean values ± S.D. (n = 5).

Figure 7A shows that PVA at 1  mg/mL protected the liposomes from leakage, with a DOX retention of approximately 95%, which is consistent with the particle size distribution assessment. However, DOX retention decreased drastically as the concentration increased in the presence of PVA. The addition of L-proline and sucrose was ineffective at preventing DOX leakage at lower concentrations (0.1–0.5 M), whereas at higher concentrations, the liposomes showed good DOX retention, with sucrose and L-proline having a DOX retention of almost 100% at 1 and 5 M, respectively (Figure 7B). The liposomes were still stable in the presence of sucrose at 2 and 5 M, showing DOX retention of 90.48 ± 3.54% and 88.30 ± 2.96%, respectively.

For the CPA combinations, cryoprotectant concentrations were added at volume ratios to the liposomes, ranging from 70% (v/v) to 90% (v/v). Figure 7C and 7D show that at very high concentrations (90% (v/v)), all the CPA combinations did not significantly protect the liposomes with DOX retention between 40% and 60%. While cryopreservation employs excipients to enhance stability, higher concentrations of cryoprotectants can harm lipid membranes, leading to drug leakage (Bhattacharya 2018; Boafo et al. 2022a). Even though the addition of Pro/Suc to liposomes achieved a DOX retention of 80.46 ± 4.63% at 80% (v/v) (Figure 7C), PVA/Pro/Suc showed to best protect liposomes against DOX leakage at 85% (v/v), with a DOX retention of 91.63 ± 2.90%.

While high CPA concentrations have been reported to induce membrane damage (Bhattacharya 2018), our findings reveal that the addition of the optimal CPA concentrations (5 M L-proline, 2 M sucrose, and 1 mg/mL PVA) to liposomes help retain more than 90% DOX following freeze‒thaw cycles, indicating a significant improvement over conventional DMSO-based methods, which exhibit higher leakage rates. The comprehensive screening under repeated freeze‒thaw cycles underscores the importance of concentration-specific CPA use for effective cryoprotection in liposomal systems.

Incorporating excipients mitigates the effects of mechanical stress resulting from ice crystal formation, cryo-concentration caused by ice front propagation, membrane fusion, and increased dimensions (Cabane et al. 2006; Litschel et al. 2018). When excipients such as L-proline diffuse through the membrane into the aqueous core during the freezing stage, they preserve liposome size and reduce the osmotic gradient caused by cryo-concentration (Kannan et al. 2015). Additionally, sucrose tends to form hydrogen bonds with the hydrophobic methyl group and the carbonyl and phosphate groups of the polar heads (Boafo et al. 2022a). However, the presence of PVA in the liposomes' outer compartment stabilizes the lipid bilayers, stopping modifications in their physical characteristics and eventually preventing drug leakage (Franzé et al. 2018). Hence, combining the three CPAs (PVA/Pro/Suc) at the optimal concentrations successfully protected the DOX LIP against drug leakage.

3.5. Impact of cryoprotectants on phase transition temperature (freezing point depression and melting point)

To investigate the mechanism by which CPAs enhance liposome stability during cryopreservation, the phase transition temperatures, specifically the freezing and melting points, were analyzed using DSC. This demonstrates the ability of CPAs to inhibit ice formation by binding to water since ice injury is a major predicament during freezing (Liu et al. 2022). It is evident from the DSC thermograms in Figure 8A that the addition of all CPAs at their optimal concentrations increases the freezing point depression by decreasing the freezing temperature of liposomes, with the cryoprotectant combination (PVA/Pro/Suc) exhibiting the greatest effect, where the onset freezing point decreased to approximately −30 °C, compared to approximately −16 °C for the pure DOX liposomes. The incorporation of CPAs enhances the degree of freezing point depression, indicating their ability to disrupt ice nucleation and crystal growth.

Figure 8.

DSC thermograms showing the (A) exothermal freezing peaks and (B) endothermal melting peaks without (DOX LIP, black) and in the presence of 1 mg/mL PVA (red), 5 M L-proline (blue), 2 M sucrose (green), and their combination (purple). (C) Representation of the strong hydrogen bonding between sucrose and water to mitigate ice formation.

Since freezing point depression is mainly evaluated and achieved by the colligative effect (Johnson et al. 2022; Fatahi et al. 2022), these CPAs tend to increase it based on their colligative properties. As solutes, L-proline and sucrose reduce the chemical potential of water, consequently decreasing its freezing point. Sucrose exclusively reduces the freezing point through a colligative mechanism (Jayawardena et al. 2017; Deryabin and Trunova 2021), whereas L-proline contributes both chemical and kinetic effects to its freezing point depression capability (Zhang et al., 2025).

Nonetheless, sucrose performs better than the other CPAs do even at 2 M, possibly because of its added advantage of forming stronger hydrogen bonds with water molecules to reduce the amount of free water available for ice lattice formation, as depicted in Figure 8C. PVA, which is a relatively large molecule, preferentially adsorbs to ice crystal surfaces (Naullage et al. 2017; Dyubko et al. 2022), further delaying crystallization and increasing freezing point depression. Hence, when these cryoprotectants are combined, their effects on water structure and ice nucleation further increase the degree of freezing point depression to prevent damage to lipid membranes.

Figure 8B illustrates the enhanced cryoprotection against thawing stresses achieved by adding cryoprotectants to the liposomal formulation, as evidenced by the melting thermogram. PVA, L-proline, and sucrose reduce the onset melting temperature to −3.4, −9.50, and −9.64 °C, respectively, compared to −2.75 °C for DOX LIP, which is attributable to their distinct interactions with water molecules. Nonetheless, their combination offers the most effective cryoprotection, with an onset melting temperature of approximately −17 °C. The CPA combination results in a broad and shallow melting endotherm, indicating reduced ice recrystallization during thawing. Cryoprotectants lower the onset of melting temperature and restrict erratic ice melting transitions, which suggests reduced ice formation and enhanced preservation of liposome integrity during freeze‒thaw cycles. This finding supports the idea that a multi-CPA strategy offers superior stabilization for DOX liposomes during freezing and thawing.

Using DSC, we demonstrated that CPAs effectively lowered the freezing and melting points of liposomal formulations, which is consistent with colligative property theories (Franks 1985), but specific impacts on DOX liposomes with non-toxic CPAs have been inadequately characterized. This study provides a precise quantification of the magnitudes of depression (−30 °C onset with CPA combination vs. −16 °C without CPA), with the data confirming that CPA combinations produce broad, shallow melting endotherms, indicative of reduced ice recrystallization, enhancing liposome integrity during thawing. This detailed profiling clarifies how these non-toxic CPA mixtures enhance cryoprotection mechanistically.

3.6. Ice-recrystallization inhibition (IRI) activity of cryoprotectants

Any lipid present in a normal frozen formulation would be phase separated to the surface of ice crystals, producing a freeze-concentrated liquid and an ice phase. As the surface area of ice decreases, the lipids at the surface should become more concentrated, and hence, the likelihood of aggregation also increases (Roessl et al. 2015; Mitchell et al. 2019). In cryopreservation, ice recrystallization is one of the leading causes of damage (Lin et al. 2023), as it disrupts the liposomal bilayer, leading to increased particle size and drug leakage (Chang and Zhao 2021). Therefore, effective IRI by CPAs is essential for maintaining the stability and preventing DOX leakage from liposomes.

We examined the ability of cryoprotectants to prevent these atrocities by inhibiting ice recrystallization at varying concentrations. This study used a modified splat assay to measure the mean crystal growth length at subzero temperatures (−8 °C). Figure 9 shows that all the CPAs exhibited some level of IRI properties compared to the control PBS. At initial concentrations of 1  mg/mL, L-proline and sucrose did not exhibit significant IRI activity until the concentrations were increased, as seen in the ice crystal wafers in Figure 9A.

Figure 9.

Splat assay investigation of the ice recrystallization inhibition activity of the cryoprotectants. (A) Ice crystal wafers of L-proline and sucrose at 1  mg/mL, 20  mg/mL, 40  mg/mL, and 60  mg/mL; PVA at 1  mg/mL; and PBS buffer as a control. (B) Quantitative analysis of the mean largest grain size (MLGS) at significant IRI CPA concentrations (PVA at 1  mg/mL; L-proline and sucrose at 40  mg/mL). Scale bar = 20 μm. The data points represent the mean values ± S.D. (n = 10), one-way ANOVA followed by Tukey's test was used; ns represents no significant difference, and **** represents p < 0.0001.

In contrast, PVA has extremely high IRI activity even at the lowest concentration of 1 mg/mL, thereby inhibiting ice crystal growth. Figure 9B compares MLGS at average concentrations where each cryoprotectant inhibits ice crystal growth. Even though the primary functions of L-proline and sucrose are to act as osmoprotectants and stabilizing matrix, respectively, they have exhibited some ability to inhibit ice crystal growth at higher concentrations.

PVA at 1 mg/mL was the most IRI-active cryoprotectant after MLGS was examined. The potency of PVA as an IRI-active agent is due to its ability to increase the actual surface area of ice crystals (resulting in more small crystals) by decelerating the ice formation rate, preventing liposome aggregation (Mitchell et al. 2015). L-proline creates a high osmotic pressure that reduces water mobility and limits ice crystal formation (Aquino et al., 2025). This mechanism is beneficial because it stabilizes the liposomes without directly interacting with the lipid bilayer. Sucrose provides IRI through hydrogen bonding with the lipid head groups of the liposomes, creating a stabilizing matrix around them, as well as strong hydrogen bonding between water molecules to mitigate ice formation (Boafo et al. 2022a). This prevents the reorganization of water molecules, significantly reducing ice recrystallization (Marcantonini et al. 2022).

PVA's IRI activity is known (Chang and Zhao 2021), but comparative IRI evaluations with L-proline and sucrose in liposomal systems are lacking. This study provides the first detailed MLGS comparisons for these CPAs in liposomal systems, confirming the high IRI activity of PVA even at low concentrations, while L-proline and sucrose require relatively high concentrations for effective IRI. This insight supports strategic CPA selection for liposome stabilization during freeze‒thaw cycles. Therefore, using the CPA combination (PVA/Pro/Suc) would provide a multifaceted protective effect: L-proline's osmoprotective ability, sucrose's hydrogen bonding, and PVA's steric hindrance all work in tandem to provide the most potent IRI effect, resulting in minimal ice crystal formation and minimal leakage after freeze‒thaw cycles. The combined multimodal protective effect of the CPAs results in a cryopreserved liposome solution that is more stable and maintains drug encapsulation, which is useful in liposome drug monitoring.

3.7. Cryoprotectant interaction with lipids

The shifts in the peak positions of νPO₄ and νCH₂ were examined to understand the interactions of CPA with lipids to provide liposomal stability. DPPC lipids were utilized to study interactions since most lipids contain phosphatidylcholine. CPA concentrations were specifically increased to ensure distinct and accurate peak shifts following interactions. Figure 10A shows the infrared spectra of the lipid DPPC, which exhibit various characteristic peaks. The fingerprint region (1500–1000 cm⁻¹) has PO₄-asymmetric and PO₄-symmetric stretching vibrations at 1241.99 and 1084.98 cm⁻¹, respectively. The CH₂-symmetric and CH₂-asymmetric stretching vibrations originating from lipid acyl chains are expressed by the peaks at around 2851 and 2920 cm⁻¹, respectively, with the peaks between 3600 and 3000  cm⁻¹ associated with OH-stretching vibrations.

Figure 10.

Infrared absorption spectra of (A) DPPC (20 mg/mL) exhibiting various characteristic peaks of its stretching vibrations, (B) effect of PVA on the lipid, (C) effect of L-proline on the lipid, and (D) effect of sucrose on the lipid headgroup hydration level and membrane lipid acyl chain rotational freedom.

Figure 10B–10D clearly show that all the CPAs affect the peak positions of PO₄-stretching vibrations. PVA and L-proline (Figure 10B and 10C) shift the PO₄-symmetric stretching vibrations to higher wavenumbers of 1093.56 and 1089.97  cm⁻¹, respectively. Sucrose (Figure 10D) also showed that the νPO₄ peak shifted to a higher wavenumber (1268.60  cm⁻¹), which was observed in the asymmetric stretching vibrations. This suggests stronger hydrogen bonding and dehydration of the phosphate group, which enhances the rigidity of the phospholipid bilayer headgroup to provide stability (Pohle et al. 2001). In contrast, CH₂ stretching vibrations had the fewest interactions with the CPAs only in the presence of L-proline (Figure 10C), where the νCH₂ shifted to a lower wavenumber of 2774.08  cm⁻¹.

The shift to a lower wavenumber indicates improved hydrocarbon tail ordering or decreased motional flexibility, suggesting tighter packing of the tails due to the CPA inducing more organized lipid arrangements (Selle and Pohle 1998). The high νPO₄ peak shift of PVA and sucrose indicates their stronger interactions and ability to form a protective barrier and stabilizing matrix, whereas L-proline's additional νCH₂ peak shift might affirm its ability to permeate the lipid membrane and hence interact with the hydrocarbon tail. While prior studies noted hydrogen bonding by sugars (Crowe et al., 2016), this study uniquely demonstrated the strong interactions of PVA and sucrose with the lipid phosphate group and the interactions of L-proline with its hydrocarbon tail, confirming the mechanistic basis for CPA-mediated liposomal stabilization during freezing.

3.8. Effect of freezing on liposome morphology

Studies were conducted to examine the effects of CPA combinations on the morphology of liposomes and to further understand how these CPAs provide stability. During cryopreservation, shape alterations of liposomes are most likely to occur, resulting in liposome deformation, aggregation, or fusion (Yu et al. 2021; Bernal-Chávez et al. 2023). For instance, the mechanical stress exerted by ice crystal formation can deform liposomes (Rajankar et al. 2024), changing them from their typical spherical shape (Qin et al. 2020). As shown in Figure 11B, the shape of liposomes cryopreserved Wt CPAs was not observed even when the image was captured at 50 nm, indicating extreme ice damage to the lipid membrane compared to liposome morphology before freezing in Figure 11A.

Figure 11.

TEM images of (A) fresh DOX LIP before cryopreservation (same image as Figure 2C, since this panel serves as the baseline reference for the cryopreservation study), (B) after cryopreservation without any CPA, (C) after cryopreservation with PVA/Pro/Suc, (D) after cryopreservation with PVA/Pro, (E) after cryopreservation with PVA/Suc, (F) after cryopreservation with Pro/Suc, by freezing the liposomal samples at −196 °C for 24 h. Scale bar = 100 nm and 50 nm.

However, freezing liposomes with CPA combinations maintained the spherical morphology, with PVA/Pro/Suc having the best protective effect, similar to that before freezing, as shown in Figure 11C. Figure 11D and 11E present some morphological deformations in the liposomes preserved with PVA/Pro and PVA/Suc, respectively, with the former exhibiting peanut-shaped liposomes compared to DOX LIP. Furthermore, in liposomes cryopreserved with Pro/Suc (Figure 11F), where the shape was still spherical similar to that of the DOX LIP, the image shows some blank patches within the liposomes, which might be linked to drug leakage due to the probable absence of PVA to act as a protective barrier against mechanical stress in the liposomes' outer compartment.

Previous reports highlighted liposome deformation due to ice stress (Stark et al. 2010; Susa et al. 2021), but our TEM imaging uniquely shows that CPA combinations preserve spherical morphology post-freeze-thaw, with PVA/Pro/Suc showing superior protection. This finding validates the combined complementary and multimodal protective effects hypothesized but not empirically demonstrated in earlier works. Hence, combining the three CPAs effectively protects the liposome morphology against ice injury and mechanical stress.

Conventional CPAs such as sucrose and trehalose are used to vitrify liposomal systems since they are susceptible to damage and fusion during freezing, mainly due to mechanical stress induced by ice crystals, destabilization of the bilayer caused by dehydration, and phase separation of membrane lipids. However, the individual use of these CPAs fails to inhibit vesicle‒vesicle contact or polymer-induced bridging. Similarly, slow-freezing methods usually aggravate fusion due to the growth of large ice crystals and significant bilayer deformation. The cryopreservation method presented here combines three mechanistically complementary and multimodal protective effects of PVA, L-proline, and sucrose, resulting in a protective profile unattainable with any single agent CPA.

This study elucidates the mechanistic pathway of CPA-mediated cryoprotection, demonstrating that L-proline serves primarily as an osmoprotectant, equilibrating osmotic gradients during freezing to stabilize the hydration shell and mitigate phase separation-driven defects. Sucrose interacts with water and the phosphate head groups of phospholipids through hydrogen bonding, creating a vitrified matrix that stabilizes the lipid bilayer and reduces the availability of free water, thereby inhibiting ice nucleation. PVA provides steric hindrance and demonstrates high surface adsorption onto developing ice crystals, thus inhibiting ice recrystallization and vesicle‒vesicle fusion during thawing.

Collectively, these CPAs reinforce liposomal membranes by minimizing mechanical stress from ice, preserving membrane integrity, and limiting drug leakage. This mechanistic clarity corresponds with molecular-level evidence demonstrating νPO4 and νCH2 vibrational shifts in FTIR analyses, confirming that the combined use of CPA targets multiple pathways, such as osmotic stress mitigation, hydrogen bonding stabilization, and steric protection against ice recrystallization, thereby preserving the structure and function of liposomes during freeze‒thaw cycles. Integrating this mechanistic insight can inform the systematic design of CPAs for enhanced cryopreservation techniques in liposomal drug monitoring processes.

3.9. Cryoprotective assessment in biological matrices

3.9.1. Stability of liposomal doxorubicin in drug monitoring workflow

The stability of the DOX LIP in a simulated drug monitoring workflow was analyzed in sheep blood by examining the extent of DOX leakage. Sheep blood is a commonly used ex vivo model for evaluating hemocompatibility and cryopreservation effects in drug delivery research because it has a similar erythrocyte membrane composition, osmotic fragility profile, and hemolytic response to cryoprotective agents to human blood (Pasciu et al. 2021; Murray et al. 2022; Hu et al. 2023). This assessment was performed separately in whole blood and plasma to observe the difference in leakage over 24 h at 4 °C.

Examining the stability of the DOX LIP separately in plasma and whole blood is essential because of differences in their biochemical environments, physiological relevance, and regulatory and therapeutic implications (Wang et al. 2021). Continuous DOX leakage was observed in both biological samples without incorporating CPAs with increasing storage duration. Figure 12A shows that approximately 34.89 ± 0.61% of the DOX was released from the liposomes in the plasma samples after 24 h, with approximately 8.81 ± 0.52% leakage within the first 30 min of storage. Moreover, in whole blood samples (Figure 12B), the degree of DOX leakage within 30 min was 1.65 ± 0.24%, with 18.06 ± 0.92% leakage over 24 h. Comparatively, the release of DOX in plasma was always higher than that in whole blood at any given time, as Figure 12C shows.

Figure 12.

Drug leakage assessment from DOX LIP in (A) plasma, (B) whole blood, and (C) comparison between plasma and whole blood samples stored at 4 °C for 24 hours. The data points represent the mean values ± S.D. (n = 5). (D) Optical images of freeze-thawed whole-blood samples exhibiting hemolysis after centrifuging and even collecting the supernatant from whole-blood samples. (E) Hemolysis quantification of erythrocyte integrity at 540  nm in blood and then stored at 4, −20, and −196 °C for 24 h. The data points represent the mean values ± S.D. (n = 5), one-way ANOVA followed by Tukey's test was used, and **** represents p < 0.0001.

The observed differences in DOX leakage between plasma and whole blood highlight the distinct biochemical environments influencing liposomal stability. Plasma, which is devoid of cellular components, might enable greater interactions between the DOX LIP and plasma proteins, leading to destabilization and leakage, despite concerns that enzymatic and oxidative stress in whole blood may accelerate lipid degradation. This aligns with studies that show the destabilizing effect of albumin and other plasma proteins on liposomes, potentially by facilitating drug diffusion or altering the integrity of the lipid bilayer (Thakur et al. 2014; Tretiakova et al. 2022).

In contrast, whole blood might stabilize liposomes due to its complex cellular matrix, potentially shielding them from direct interactions with plasma proteins, which is supported by hemolysis-free quantification (1.07 ± 0.60%), confirming that erythrocytes remained unscathed. In hemolysis-free whole blood, intact erythrocytes contribute significantly to antioxidant capacity via intracellular catalase, superoxide dismutase, and glutathione peroxidase, therefore limiting oxidative stress to lipid membranes.

Moreover, red blood cells (RBCs) provide steric and rheological buffering that restricts liposome‒protein interactions. Therefore, the minimal leakage observed even in whole blood might suggest ongoing destabilization mechanisms, possibly enzymatic degradation or interactions with blood components such as fibrinogen (Sercombe et al. 2015; Yang et al. 2022), or undamaged erythrocytes in whole blood serve as a stabilizing reservoir for free DOX, which contributes to transient sequestration effects that reduce measurable leakage from the liposomal compartment.

Evaluating both matrices provides a fair idea of how to accurately characterize and stabilize liposomal doxorubicin for precise drug monitoring. Plasma provides a more precise measure of free drug levels, while whole blood may better reflect drug stability under physiological conditions, enabling a comprehensive understanding of liposome behavior.

3.9.2. Freeze–thaw effects on blood samples

This study examined the ability of the optimal cryoprotectant cocktail (PVA/Pro/Suc) to prevent liposome drug leakage in biological samples (plasma and whole blood) stored at 4, −20, and −196 °C at varying concentrations for 24 h. These temperatures represent standard clinical and laboratory storage conditions used for short-term, intermediate-term, and long-term preservation (Jyothi et al. 2022; Lehman et al. 2023; Dcunha et al. 2024). Specifically, 4 °C is commonly utilized for the temporary storage of blood orplasma in therapeutic drug monitoring protocols, −20 °C is extensively employed in clinical laboratories for the medium-term freezing of drug-containing samples when immediate analysis is unfeasible, and −196 °C is the designated temperature for long-term cryogenic biobanking.

Importantly, research on liposomes stored in whole blood at these temperatures is limited, and this current ex vivo model seeks to present preliminary evidence that requires validation in clinical-like storage protocols. After the blood samples were thawed, the RBCs were observed to undergo hemolysis, mainly for the samples stored at −20 and −196 °C, making it impossible to achieve clear plasma separation, as shown in Figure 12D and 12E. A comparison of the blood samples after storage at all temperatures to the hemolysis-free control sample showed the hemolysis was minimal at 4 °C (26.85 ± 1.82%) but increased substantially at −20 °C (60.91 ± 1.54%) and −196 °C (88.79 ± 2.07%), which is consistent with the literature on RBC fragility under deep cryogenic conditions, as presented in Figure 12E. Hemolysis was assessed independently of liposomal leakage, allowing clear separation of the effects of CPA on liposomes from temperature-induced erythrocyte damage. This makes SPE challenging because the samples cannot be easily filtered through the pores of the filter.

The challenges of hemolysis and subsequent plasma separation after freeze‒thaw significantly impact the reliability of SPE and subsequent HPLC analyses. Hemolysis introduces cellular debris and hemoglobin, which clog SPE filters and interfer with drug quantification (Farouk et al. 2023; SÆBØ et al. 2023). This introduces inconsistencies in sample composition, leading to variable outcomes in cryoprotective efficiency, as accurate separation of already released DOX and encapsulated DOX becomes difficult. The practical challenge of hemolysis during freeze‒thaw cycles complicates analysis, which aligns with prior reports (Tsimbaliuk 2021), and our findings reinforce the advantages of plasma for CPA-based cryoprotection due to reduced complexity and analytical reliability.

3.9.3. Cryoprotective effect on DOX LIP stability during drug monitoring

Liposomes exhibit different responses when stored at various temperatures, whether in plasma or whole blood. Storage at various temperatures is essential to simulate different clinical and research conditions. At 4 °C, liposomal formulations experience minimal stress, providing a baseline for comparison. At −20 °C, ice crystal formation and osmotic shifts present challenges that can disrupt liposomal integrity. Storage at −196 °C in liquid nitrogen introduces extreme vitrification conditions, where even minor cryoprotectant inefficacy becomes apparent. Evaluating the stability across these conditions ensures robust storage guidelines for both clinical and laboratory use.

By examining the E.E., liposomes in plasma stored without CPAs in liquid nitrogen were relatively stable, with 78.74 ± 0.38% E.E. compared to liposome plasma samples stored at 4 °C (65.11 ± 0.61%) and −20 °C (62.59 ± 0.53%) (Figure 13A). However, the addition of PVA/Pro/Suc to the plasma samples improved the stability of the liposomes at all temperatures, with the best cryoprotective effect observed at a CPA concentration of 85% (v/v) stored at −20 °C (90.08 ± 0.72% E.E.). Figure 13B exhibits the stability of liposomes in whole blood, where samples cryopreserved without any CPA (0%) showed a similar effect as the plasma samples, with those stored at −196 °C (91.89 ± 0.22% E.E.) being better than samples stored at 4 and −20 °C, having E.E of 81.94 ± 0.92% and 88.32 ± 0.14%, respectively. Moreover, the best cryoprotective ability of PVA/Pro/Suc was observed at a CPA concentration of 85% (v/v), similar to that of plasma, however not at −20 °C but rather at −196 °C, with an E.E. of 97.18 ± 0.26%.

Figure 13.

Entrapment efficacy of DOX LIP in (A) plasma and (B) whole blood stored at 4, −20, and −196 °C, after the addition of CPA combinations, PVA/Pro/Suc (1 mg/mL PVA, 5 M Pro, and 2 M Suc) was used. The data points represent the mean values ± S.D. (n = 5), one-way ANOVA followed by Tukey's test was used, ns represents no significant difference, and **** represents p < 0.0001.

Comparatively, the blood samples had a higher E.E. than plasma samples in the presence or absence of excipients. Although whole blood samples exhibited slightly higher E.E. values than plasma samples, freeze‒thaw cycles in whole blood led to significant hemolysis, resulting in the release of cellular debris and hemoglobin. These contaminants interfered with SPE and probably with HPLC, likely resulting in inaccurate separation of free and encapsulated DOX and compromised analytical reliability, respectively. This may have compromised the E.E. calculations and affected data reliability in the blood analysis. Hence, plasma is the ideal biological matrix for adding CPAs (PVA/Pro/Suc) since it avoids hemolysis-related complications and improves data accuracy. Furthermore, plasma's widespread use in drug monitoring (Lu et al. 2022; Hesham et al. 2024; Liang et al. 2024) aligns CPA applications with clinical relevance.

The study shows that the CPA combination PVA/Pro/Suc provides the best stability to liposomal doxorubicin in a plasma at 85% (v/v) at a storage temperature of −20 °C with approximately 90% E.E. compared to liposomes stored Wt CPAs with about 60% E.E., offering an optimized condition for cryopreserving liposomal formulations for clinical drug monitoring. These findings suggest that CPAs should be added to plasma due to their reduced complexity, clinical relevance, and increased baseline leakage risk before storage or preservation to obtain the best cryoprotective effect.

3.9.4. Long-term cryoprotective effect

The evaluation of PVA/Pro/Suc on the long-term stability of liposomes was performed in plasma and at −20 °C based on the initial short-term findings. Figure 13C shows that the stability of liposomes in plasma in the presence of PVA/Pro/Suc at 85% (v/v) was relatively stable across 90 days, with the E.E. still greater than 90%, whereas the E.E. Wt CPAs was approximately 60% and 35% within 24 h and 90 days, respectively. In contrast, the DOX continued to leak after 90 days of storage in the absence of any cryoprotectant. This suggests that −20 °C is favorable for long-term storage of plasma-containing liposomes using PVA/Pro/Suc at 85% (v/v). This type of stability is important in clinical drug research studies in which traceback is performed from the point of sample collection to the period of analysis and demands accurate concentration of the drug throughout the process.

These results demonstrate that the combined CPAs preserve liposomal integrity more effectively over extended storage conditions (90 days) at −20 °C compared with untreated controls, as seen in Figure 13C. This long-term evaluation complements the immediate freeze‒thaw protection data, establishing that the multimodal actions of the ternary system, that is, vitrification (sucrose), osmotic buffering (L-proline), and steric membrane shielding (PVA), collectively mitigate lipid bilayer degradation, ice-induced fusion, and leakage over time. This combined mechanism provides advantages over single-component cryoprotectants, which typically address only one mode of cryodamage.

Additionally, our developed ternary cryopreservation system poses no toxicity, unlike other studies that use multi-component cryopreservation fluids combining an amino acid (e.g. L-proline or polyamino acids), a disaccharide such as sucrose, and polyvinyl alcohol containing some amount of DMSO for cell and tissue storage in liquid nitrogen (Rong 2022). However, these DMSO-containing formulations are not optimized for liposomal systems and do not report nanoparticle size, PDI, or drug-leakage metrics over time. To our knowledge, no peer-reviewed experimental studies have evaluated a DMSO-free PVA/Pro/Suc ternary system for long-term stabilization of liposomal chemotherapeutics in blood or plasma matrices. Our work therefore represents the first systematic assessment of this triad in a nanomedicine context, with long-term cryostorage readouts directly relevant to pharmacokinetic sampling.

The minimal leakage in plasma and whole blood offers a significant ex vivo indicator of the potential clinical performance of liposomal doxorubicin and indicates that the developed CPA cocktail can preserve bilayer integrity in biologically active settings, preventing premature drug leakage. Importantly, the reduced leakage after CPA-assisted freezing indicates that the liposomes withstand clinically relevant storage and handling stresses. Collectively, these observations indicate that the cryopreservation technique has translational potential, while underscoring the need for confirmatory in vivo pharmacokinetic, biodistribution, and safety evaluations.

Therefore, the developed cryopreservation protocol should be investigated and expanded on translational and clinical research, where in industrial-scale lyophilized liposomal formulations, the combined effect of CPAs will provide desirable protection since heterogeneous cooling profiles and batch-dependent ice formation often compromise liposomal integrity (Tchessalov et al. 2022; Kazarin et al. 2023; Deck et al. 2024). All three components are biocompatible and commonly found in parenteral goods that have been approved; therefore, the CPA system complies with current regulatory standards and may facilitate enhanced lyophilization cycles, quicker drying cycles, and better storage stability by preventing liposome fusion and aggregation. The challenges of increased viscosity at higher concentrations of PVA, which can complicate high-throughput filling, and engineering controls to achieve consistent cooling and thawing rates must also be considered for future clinical-scale deployment.

This extended stability has direct implications for drug monitoring workflows, where repeated thawing, interim storage, and delayed analytical handling can compromise liposomal DOX quantification. The CPA cocktail needs to be practically tested through comprehensive clinical trials across relevant patient groups to validate the pharmacokinetic efficacy, safety profile, and procedural specifications of the cryopreservation method. While acknowledging that multi-month stability and in vivo pharmacokinetic confirmation require future work, the added long-term data now substantiate the superior effectiveness of the ternary CPA system beyond immediate cryopreservation effects.

The improvements in liposomal stability are not confined to phospholipid-based vesicles but can be translated to other nanoparticle-based drug delivery platforms, including those currently employed in immunotherapy to address universal biophysical failure points common to lipid nanoparticles, protein-nanocapsules, virus-like particles, and mRNA-LNP systems used in cancer immunotherapy and vaccine delivery. Additionally, the long-term stability supports the feasibility of applying this cryoprotective strategy to immunotherapeutic nanoparticles, where long-term storage, shipping, and thermal robustness remain critical bottlenecks.

PVA/Pro/Suc effectively prevents ice crystal formation, osmotic stress, fusion, lipid bilayer disruption and constituent leakage under these conditions. In cases where studies rarely combine CPA optimization with practical monitoring workflows, this finding directly addresses the gap in practical protocols for liposomal drug monitoring in real-world workflows, representing a significant advancement over studies that only assess cryoprotection in buffer systems.

4. Conclusion

This research effectively developed a complementary multimodal cryoprotectant system combining L-proline, sucrose, and PVA that improves the stability of liposomal doxorubicin during cryopreservation to enhance drug monitoring. We established structure‒function connections between chemical composition and cryoprotective efficacy by logically designing a ternary CPA cocktail (PVA/Pro/Suc). Key findings demonstrate that this cryoprotective cocktail addresses liposome degradation and fusion, by preserving over 90% drug encapsulation through complementary multimodal mechanisms in which L-proline acts as an osmolyte and cryoprotective kosmotrope, sucrose vitrifies the extracellular environment and substitutes bound water via hydrogen bonding at lipid headgroups, and PVA inhibits ice recrystallization and forms a steric barrier.

The unavailability of real-time cryopreservation instrumentation, such as real-time cryomicroscopy, cryoelectron microscopy, and cryo X-ray diffraction, limited the potential to precisely estimate the phase behavior of water within a lipid membrane cavity, assess the degree of ice recrystallization, and evaluate the mechanical strength of liposomal membranes at cryogenic temperatures in real time. Despite these limitations, the study relied on post-thaw examinations to provide a crucial foundation for advancing cryopreservation reliability and generalizability, as well as ex vivo experiments showing the stability and preservation of the liposomal bilayer under physiologically relevant conditions (plasma), which suggests initial translational potential, especially for formulations necessitating cold-chain transport, prolonged storage, decentralized clinical management, or drug monitoring.

The developed cryopreservation technique was effective on liposomes in biological matrices, exhibiting significant protection against membrane fusion, aggregation, and premature drug release by mitigating critical obstacles that hinder consistency and accuracy in liposomal drug monitoring. These findings provide a solid foundation for the preservation of nanocarriers, tackling a significant obstacle in drug monitoring and the storage of nanomedicine.

Future research should concentrate on computational modeling to enhance CPA‒lipid interactions, as well as broader application of the developed cryopreservation techniques across different liposomal and nanoparticulate drug delivery systems, and in vivo validation of the cryopreservation techniques. While the present FTIR, DSC, IRI, and leakage studies demonstrate robust lipid-CPA compatibility for DOX LIP, extension to other drug types and carriers will require empirical confirmation tailored to their molecular architecture and drug–excipient interactions.

The cryoprotective principles established in this research which minimize ice-induced stress, prevent osmotic shock, maintain bilayer packing, and reduce plasma protein-induced leakage, represent broadly transferable mechanisms since liposomal instability during storage, thawing, and sample processing is a major obstacle across all liposomal medicines and nanocarrier-based therapeutics. This study connects essential cryopreservation chemistry with practical applications, providing a framework for the stabilization of next-generation nanotherapeutics.

Funding Statement

This work was supported by the Hunan Provincial Natural Science Foundation of China (2024JJ8116, 2025JJ80163 and 2025JJ80076) Natural Science Foundation of Hunan Province.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All experimental data within the article are available from the corresponding author upon reasonable request.

Ethics statement

The authors have nothing to report since the study did not involve animal or human experiments.