Improving the efficacy and targeting of letrozole for the control of breast cancer: in vitro and in vivo studies.

Introduction
Among cancer-related causes of death, breast cancer (BC) ranks second and is the most commonly diagnosed condition among women worldwide (DeSantis et al. 2019; Nasser and Yusof 2023; Trayes and Cokenakes 2021). Throughout their lifetimes, one in every eight women will receive a BC diagnosis. Current projections indicate that there are approximately 1,384,155 new cases and 459,000 fatalities associated with BC globally (DeSantis et al. 2019; Tao et al. 2015; Yassemi et al. 2020). Most BC cases are hormone-dependent, with estrogen playing a critical role in the progression to metastasis by activating estrogen receptors (Clusan et al. 2023; Yassemi et al. 2020). Estrogens stimulate cell differentiation through receptor-mediated hormonal pathways and enhance metabolic activity via cytochrome P450, which in turn increases mutation rates (Russo and Russo 2006). Aromatase is a key enzyme responsible for converting androgens into estrogens (Czajka-Oraniec and Simpson 2010). Thus, modulating aromatase activity has emerged as an effective therapeutic strategy for managing BC (Rachner et al. 2020; Ratre et al. 2020). Letrozole (LTZ), a potent third-generation aromatase inhibitor, is commonly used to treat postmenopausal patients with estrogen receptor-positive and metastatic BC (Ratre et al. 2020). LTZ regulates estrogen levels, leading to a reduction in tumor mass or inhibition of tumor growth (Bhatnagar 2007; Ratre et al. 2020). However, challenges such as water insolubility, poor bioavailability, unregulated release, and inadequate targeting raise concerns about the efficacy of oral LTZ treatment (Gomathi et al. 2017; Wempe et al. 2007). Consequently, the primary goal of this research is to develop a transdermal invasomal formulation of LTZ for use as a treatment for BC.

Several benefits make transdermal medication delivery a promising alternative to oral administration, including controlled drug release, avoidance of first-pass metabolism, fewer side effects, improved bioavailability, and targeted delivery (Puri et al. 2017; Ranade 1991; Tijani et al. 2021). However, the penetration of medications is significantly restricted by the stratum corneum (SC). To address this challenge, extensive research has focused on enhancing the effectiveness of nanocarriers for drug delivery across the SC (AlEbadi and Al-Lami 2022; Altameemi and Abd-Alhammid 2022; Regenthal et al. 2018; Salem et al. 2022; Xi et al. 2010). These nanocarriers improve permeability, which leads to increased bioavailability and efficacy while reducing adverse effects (AlEbadi and Al-Lami 2022; Altameemi and Abd-Alhammid 2022; Regenthal et al. 2018; Xi et al. 2010).

Invasomes are modified liposomes that incorporate permeation enhancers, such as terpenes and ethanol, to improve skin permeability (Nangare and Dugam 2020; Salem et al. 2022). They serve as vesicular systems capable of encapsulating a high load of both hydrophilic and hydrophobic therapeutics compared to conventional liposomes. Additionally, they exhibit exceptional deformability and a small size, enabling targeted delivery of therapeutics (Nangare and Dugam 2020; Salem et al. 2022; Vidya and Lakshmi 2019). Ethanol penetrates the SC, causing desaturation of the soft keratin architecture by loosening the fibril arrangement within corneocytes. It disrupts the short lamellar structures of the SC, resulting in an uneven configuration (Fouad et al. 2024; Salem et al. 2022). The permeation enhancement activity of terpenes is attributed to their interaction with the polar head and lipophilic tail groups of intercellular lipids; thus, terpenes weaken the skin barrier by removing or fluidizing SC lipids (Jain et al. 2021). Invasomes enhance drug selectivity, bioavailability, effectiveness, and patient adherence (Nangare and Dugam 2020; Salem et al. 2022; Vidya and Lakshmi 2019). Furthermore, invasomes promote the sustainability and localization of drugs within the targeted area (Bendas et al. 2013; Moghassemi and Hadjizadeh 2014).
Hydrogels, which are three-dimensional, hydrophilic polymer frameworks, can absorb and retain large volumes of water or biological fluids while maintaining their structural stability (Caló and Khutoryanskiy 2015). Due to their exceptional mechanical properties, high water content, and biocompatibility, hydrogels have been employed in various biomedical applications, including tissue engineering and drug delivery (Caló and Khutoryanskiy 2015; Singh and Bhardwaj 2021). Furthermore, they have the ability to protect sensitive pharmaceuticals, regulate medication release, and enhance a drug’s selectivity for target sites. Given these advantages, the incorporation of LLI formulations into hydrogels creates a beneficial synergism that enhances their properties and supports the primary objectives of these formulations.
No research has yet investigated the possibilities of transdermal LTZ-loaded invasomes (LLI) hydrogel for treating BC. Thus, the fundamental purpose of this project is to construct an LLI hydrogel formulation that not only sustains LTZ release but also enhances its effectiveness, permeability, and targeting for BC management. Multiple LLI formulations were optimized employing the Box-Behnken design. The LLI hydrogel formulation was developed by combining the selected LLI formulation with carbopol, followed by further in vitro analysis. The anti-cancer efficacy of the LLI hydrogel formulation was assessed in vivo with the aid of an experimental 7,12-dimethylbenz(a)anthracene (DMBA)-induced carcinoma rat model.

Materials and methods

Materials
LTZ was supplied by Hikma Pharmaceuticals (Egypt). Phospholipid, cholesterol, carbopol 940, triethanolamine, ethanol, Tween 80, 7,12-dimethylbenz(a)anthracene (DMBA), dialysis bags, and cineole were sourced from Sigma-Aldrich (Agitech, Cairo, Egypt). Methanol and chloroform were provided by Cornell Lab Company (Cairo, Egypt).

Experimental design
The Box-Behnken design was established using Design-Expert software (version 12, Stat-Ease Inc., USA) to optimize the LLI formulation. A total of 15 distinct LLI formulations were developed, as outlined in Table 1, in order to determine the effect of three separate variables: phospholipid concentration (ranging from 1 to 3%), ethanol concentration (ranging from 0 to 5%), and cineole concentration (ranging from 0.5% to 1.5%). The responses measured included entrapment efficiency (EE%) and particle size (PS) to determine the influence of the varying variables.

Fabrication of LLI formulation
With the help of the thin film hydration method, 15 different LLI formulations (Table 1) were prepared (Abdelrahman et al. 2015). LTZ (10 mg), phospholipid, cineole, and cholesterol (0.15%) were dissolved in a methanol-chloroform solution within a rotary evaporator flask. A thin, dry film was formed by evaporation of the organic phase using a rotary evaporator (Heidolph VV 2000, Burladingen, Germany) at 100 rpm and under vacuum set at 40 °C. The resulting thin film was subsequently rehydrated using a solution of phosphate buffer (pH 5.5) and ethanol while maintaining a speed of 60 rpm for 2 h under normal pressure. Following this process, all LLI formulations were stored at 4 °C for subsequent characterization.

Characterization of LLI formulations

Measurement of entrapment efficiency (EE%)
The LLI formulations were spun in a cooling centrifuge (Sigma 3–30 KS, Osterode, Germany) at 15,000 rpm for a duration of 2 h. The supernatant was separated from the invasomal pellets and recentrifuged for an additional hour to remove any particulate matter. Spectrophotometry (Shimadzu UV1650, Spectrophotometer, Kyoto, Japan) at 240 nm was performed in triplicate to ascertain the LTZ concentration in the supernatant. The EE% was subsequently determined, and the results were presented as the mean ± standard deviation (SD) of three separate assessments (Abdelrahman et al. 2015):

Measurement of particle size (PS)
The PS of LLI formulations was evaluated using dynamic light scattering technology (Salem et al. 2021). One milliliter of the invasomal suspension was diluted with 9 mL of distilled water, and the PS readings were taken using a Zetasizer Nano ZS (Malvern Instruments; Worcestershire, UK). The results were presented as the mean ± SD of three separate assessments.

Optimization of LLI formulations
To determine the best model for fitting the EE and PS data, the Design-Expert software conducts an ANOVA. The factors considered during the selection process include the p-value, lack of fit, F-value, correlation coefficient (R2), observed R2, expected R2, and adequate precision (Salem et al. 2018). Additionally, the Design-Expert software generates regression equations and 3D plots to illustrate the effects of each variable on EE% and PS. The optimization process begins by setting constraints that aim to maximize EE% while minimizing PS. To identify the optimal formulation that maximizes the desirability index, the Design-Expert software employs numerical optimization alongside the desirability index. A confirmation process is performed to validate the optimization by preparing the selected optimal formulation and comparing its PS and EE% results with the predicted data generated by the software.

Characterization of the optimized LLI formulation

Differential scanning calorimetry (DSC)
The thermal behavior of the optimized LLI formulation was assessed using DSC. Samples of the optimized LLI, phospholipid, and LTZ were heated at a rate of 10 °C per minute across a temperature range of 25 to 300 °C, utilizing a DSC (DSC-60 Shimadzu, Germany) (Salem et al. 2021).

Fourier transform infrared spectroscopy (FTIR)
FTIR was employed to assess possible chemical interactions among the components of the optimized LLI formulation (Tulbah and Gamal 2021). Samples of LTZ, phospholipid, cholesterol, and the optimized LLI were analyzed using an FTIR (Bruker FTIR spectrophotometer, Alpha, Germany). Before examination, each sample was carefully ground and mixed with KBr, with measurements taken between 4000 and 400 cm⁻1.

Transmission electron microscopy (TEM)
TEM was employed to examine the morphology of the optimized LLI formulation using a TEM microscope (Joel JEM 1230; Tokyo, Japan) (Fouad et al. 2024). The optimal LLI formulation was placed on a carbon-coated grid and dyed with 2% phosphotungstic acid.

Zeta potential and size distribution
An important measure for determining the stability of invasomes is zeta potential (ZP), which quantifies the attractive and repulsive forces between LLI vesicles (El-Ela et al. 2022). Additionally, the homogeneity of the formulation and the particle size distribution were evaluated using the polydispersity index (PDI) measurement (El-Ela et al. 2022). A 1-mL sample of the optimized LLI formulation was diluted, and PDI and ZP were measured as previously described in the PS measurement section. The results for PDI and ZP are presented as the mean ± SD of three independent assessments.

Preparation and characterization of the optimized LLI hydrogel

Preparation of the optimized LLI hydrogel
Carbopol 934 polymer was utilized for the preparation of both the free LTZ hydrogel and the optimized LLI hydrogel. The gel base was made by dissolving 2% carbopol 934 in distilled water under constant stirring (Princely and Dhanaraju 2018). For the purpose of neutralization, triethanolamine was added until a translucent gel formed. The synthesis of the free LTZ and optimized LLI hydrogels involved incorporating the free LTZ dispersion and the optimized LLI formulation into the carbopol gel matrix, respectively, with a mechanical stirrer at 800 rpm for 30 min.

pH measurement
The LLI hydrogel and free LTZ hydrogel samples were dissolved in distilled water, and their pH values were measured using a pH meter (Griffin, model 80, Britain) (Chaudhary et al. 2011). The results were presented as the mean ± SD of three separate assessments.

Viscosity evaluation
The effectiveness of hydrogel for transdermal application can be assessed by measuring its viscosity (Kushwaha et al. 2018, 2014). The optimized LLI, LLI hydrogel, and free LTZ hydrogel were subjected to shear rate measurements ranging from zero to one hundred seconds per second using a Brookfield viscometer (Brookfield DV-III ULTRA, USA) (Salem et al. 2020). The results were presented as the mean ± SD of three separate assessments.

In vitro release kinetic study
To determine the saturated solubility of LTZ in water, the equilibrium solubility test was obtained (Salem et al. 2022). A thermostatically controlled shaker was employed to mix glass bottles containing phosphate buffer with 0.1% Tween 80 (PBT, pH = 5.5) and an excess of LTZ. The concentration of LTZ was measured using a spectrophotometer calibrated to 240 nm. The results were presented as the mean ± SD of three separate assessments.
The dialysis bag method was utilized to perform the release study (Fouad et al. 2024). Dialysis bags, featuring a molecular weight cutoff of 12 kDa and a pore size of 2.4 nm, were filled with free LTZ suspension, the optimal LLI formulation, and LLI hydrogel, each containing 2 mg of LTZ. The bags were immersed in 50 mL of PBT as the release medium within a Hanson dissolution apparatus (Hilab, Düsseldorf, Germany) at 37 °C and 100 rpm. At specified time intervals up to 24 h, 3-mL samples were withdrawn and replaced with an equal volume of fresh solution to maintain sink conditions. The concentration of LTZ in each sample was measured at 240 nm using a spectrophotometer. The results were expressed as the mean ± SD of three independent assessments.
Utilizing DDsolver software, the release kinetics and mechanism of both the optimized LLI hydrogel and the free LTZ suspension were examined. To determine which release model is best, we examined the R2, model selection criterion (MSC), and Akaike information criterion (AIC) values (Fouad et al. 2024). Additionally, using the Korsmeyer-Peppas model, we determined the release exponent “n” to have a better understanding of the releasing process. Another measure used to verify the long-term viability of the LLI hydrogel was the mean dissolution time (MDT) (Hossain et al. 2013). The results were presented as the mean ± SD of three separate assessments.

Ex vivo permeation study
A 3-cm2 effective permeability region was highlighted in an ex vivo permeation experiment that utilized rats’ abdominal skin (Fouad et al. 2024). The abdominal skin was soaked in PBT for 24 h to be used as diffusion membranes. The permeation of LTZ was evaluated using three different formulations: an LLI hydrogel, the optimized LLI suspension, and a free LTZ suspension, each containing 3 mg of LTZ. After adding the appropriate amounts of each formulation to the diffusion membrane, it was immersed in a beaker that contained 50 mL of PBT to serve as the receptor medium. We used a Hanson dissolution apparatus that was maintained at 32 °C and 100 rpm for the investigation. To maintain sink conditions, 3-mL samples were collected at regular intervals over a 24-h period, and each sample was replaced with fresh dissolution medium. Using a 240-nm-calibrated UV–visible spectrophotometer, the LTZ concentration was measured. Data regarding the steady-state flux and cumulative percentage of LTZ permeation were recorded as the mean ± SD of three separate assessments.

In vivo evaluation of LLI hydrogel

Study design
The in vivo study adhered to all the requirements outlined in the ARRIVE guidelines and received approval from the Ethics Committee of Beni-Suef University (BSU-IACUC: 024–013).

Experimental tumor induction and treatment
We obtained 24 male Wistar rats, each weighing between 200 and 250 g, from Nahda University and housed them under standard laboratory conditions at Beni-Suef University’s Faculty of Pharmacy. We utilized G*Power (version 3.1.9.2) to evaluate the statistical significance of the sample size for the animal experiment (Faul et al. 2007). After a week of consistent feeding to allow acclimatization to their new environment, the rats were randomly assigned to four groups, each consisting of six rats. The first group served as the negative control, while the second group acted as the positive control. The third group was designated as the free LTZ group, and the final group was the LLI hydrogel group. The negative control group (n = 6 rats) adhered to a normal diet and received no treatment. The positive control group was administered 7,12-dimethylbenz(a)anthracene (DMBA) to induce BC. The free LTZ group received DMBA and an oral suspension of LTZ at a dosage of 1 mg/kg (Alabiad et al. 2024), which was delivered using a syringe into their oral cavities. The LLI hydrogel group received DMBA and a transdermal LLI hydrogel, also at a dosage of 1 mg/kg.
To induce BC, the positive control, free LTZ, and LLI hydrogel groups (n = 18 rats) had their backs shaved 48 h before the experiment began. The method outlined by Dias et al. was utilized (Dias et al. 2018) by giving the tumor inducer DMBA as a single dose of 1 mg in 200 µL of acetone.

Anti-tumor activity measurement
Tracking the volume of a tumor is a common method for assessing the effectiveness of a treatment. Digital calipers were used to measure the tumor’s width and length twice per week to calculate its volume (Salem et al. 2018). The results were presented as the mean ± SD of six separate assessments.

Histopathology study
The trial concluded with an intraperitoneal injection of a 1:1 mixture of 90 mL/kg of ketamine and 5 mL/kg of xylazine for all rats. Euthanasia was performed via cervical dislocation. Tumor samples were collected and preserved in 10% buffered formaldehyde. Hematoxylin and eosin staining was conducted on thin tumor slices (4–6 µm) for histopathological examinations (Gamal et al. 2021).

Targeting study
The tumor and liver samples from the LLI hydrogel and free LTZ groups were sliced and preserved at −80 °C to evaluate the distribution of LTZ in each organ. The tumor and liver tissues were isolated and placed in separate tubes containing PBT, followed by homogenization for 30 min. The liquid was then centrifuged at 4500 rpm for 10 min to remove any unwanted contaminants (Praça et al. 2018). The clear supernatant that remained after passing through a filter with 0.45-µm holes was evaporated to dryness. A mixture of methanol and water was used to dissolve the residue, and the amount of LTZ present in each organ was measured using a spectrophotometer set to 240 nm. The results were presented as the mean ± SD of six separate assessments.

Statistical analysis
The data was statistically evaluated using SPSS (version 22.0, Chicago, USA). The results were evaluated using ANOVA, Tukey’s post hoc, and the Student t-test with a p-value less than 0.05.

Results

Experimental design
To develop and optimize the LLI formulations, the Box-Behnken design was employed. The evaluation of EE% and PS was carried out using a quadratic model. As shown in Table 2, this model was chosen because of its significant p-value, the highest F-value, an insignificant lack of fit, and the adjusted and predicted R2 values differing by less than 0.2. Additionally, the model demonstrated an adequate precision ratio that exceeded 4.

Characterization of LLI formulation

Entrapment efficiency measurement
According to Table 1, the EE% of LLI ranged from 68.67 to 93.06%. Figure 1A demonstrates that the three independent variables significantly (ANOVA, p-value < 0.0001) impacted the EE%. Phospholipid had a positive effect on EE%. Formulations F12 and F6, which contained 3% phospholipid, showed EE% values that exceeded those of formulations F7 and F11, which contained 1% phospholipid, by 20.86% and 20.53%, respectively. Conversely, ethanol has a negative impact on EE%. The EE% of the ethanol-free F1 and F8 formulations was 9.27% and 9.04% higher, respectively, compared to the 5% ethanol-containing F13 and F2 formulations. Additionally, cineole positively influenced EE%. The EE% of formulations F2 and F8, which contained 1.5% cineole, was 2.56% and 3.10% greater than that of formulations F4 and F9, which had 0.5% cineole, respectively.

Particle size measurement
Table 1 shows that the PS of LLI ranged from 114.27 to 363.70 nm. Figure 1B demonstrates that three independent factors significantly (ANOVA, p-value < 0.0001) influenced PS. Phospholipid had a positive effect on PS. Formulations F12 and F5, which contain 3% phospholipid, exhibited PS values that were 55.11% and 66.01% higher than those of formulations F7 and F13, which contain 1% phospholipid, respectively. Conversely, ethanol negatively affected PS. The PS of formulations F1 and F8, which contain 0% ethanol, was greater than that of formulations F13 and F2, which contain 5% ethanol, by 11.24% and 9.15%, respectively. Furthermore, cineole positively impacted PS. The PS of formulations F12 and F8, which contain 1.5% cineole, was 50.79% and 72.15% higher than that of formulations F6 and F9, which contain 0.5% cineole, respectively.

Optimization process
After applying the constraints for PS and EE%, the Design-Expert software recommended an optimal formulation based on the desirability index. This formulation, which includes 3% phospholipids, 5% ethanol, and 0.5% cineole, was selected because it achieved the highest desirability index of 0.858. An average EE% of 87.12 ± 0.41 and a PS of 115.44 ± 4.5 nm were revealed by in vitro experiments conducted on the optimized LLI formulation, confirming the optimization.

Characterization of the optimized LLI formulation

DSC
As shown in Fig. 2A, the LTZ thermogram displays an endothermic peak at 208 °C, which corresponds to the drug’s melting point. A distinct endothermic peak appears on the phospholipid thermogram at 129 °C, indicating its melting temperature. Additionally, the cholesterol thermogram shows an endothermic peak at 146.4 °C, reflecting its melting temperature. The LLI thermogram reveals that these peaks are absent.

FTIR
The FTIR spectra for LTZ, phospholipid, cholesterol, and the optimized LLI formulation are presented in Fig. 2B. In the phospholipid FTIR spectrum, peaks indicative of the CH₂ stretching frequency appeared at 2924 and 2857 cm⁻1. The symmetrical C = O stretching frequency was observed at 1739 cm⁻1, and the PO₄ stretching frequency was noted at 1240 cm⁻1. For cholesterol, the FTIR spectra revealed the OH stretching group at 3407 cm⁻1, the CH₂ frequency at 2942 cm⁻1, and an asymmetric CH₂ vibration at 1460 cm⁻1. The LTZ peaks corresponding to the aromatic C-H group stretching frequency were found at 3100 cm⁻1, along with a nitrile group vibration at 2250 cm⁻1 and a C-N group stretching vibration at 1270 cm⁻1. The FTIR spectrum of the optimized LLI formulation exhibited the same peaks as the original components, indicating that there were no interactions among the formulation’s components.

TEM
Figure 2C shows that the TEM micrograph revealed the formation of circular vesicles devoid of aggregates, with an easily discernible center and periphery.

Zeta potential and size distribution
The optimized LLI formulation produced homogeneous vesicles, evidenced by a low PDI value of 0.283 ± 0.022. The zeta potential of −49.2 ± 0.49 mV suggests enhanced vesicle stability and a lower likelihood of aggregation in the optimized LLI formulation.

Characterization of the optimized LLI hydrogel

pH measurement
The pH values were 5.6 ± 0.15 for the free LTZ hydrogel and 5.5 ± 0.19 for the LLI hydrogel, confirming the lack of irritation.

Viscosity
The viscosity coefficients measured were 540.67 ± 42.34 cP for the LLI hydrogel, 365.28 ± 23.18 cP for the free LTZ hydrogel, and 72.55 ± 1.62 cP for the optimized LLI formulation. The viscosity coefficient of the LLI hydrogel was found to be (ANOVA, p-value < 0.0001) 7.45 times greater than that of the optimized LLI formulation and 1.52 times greater than that of the free LTZ hydrogel.

In vitro release studies
Figure 3A shows the performance of the LTZ-free suspension and the optimized LLI formulation regarding in vitro drug release when compared to the LLI hydrogel. Within 8 h, approximately 98.73 ± 0.98% of the LTZ was released from the LTZ-free suspension. After 24 h, the LLI hydrogel released only 37.93 ± 0.74% of its total LTZ, whereas the optimized LLI formulation released 58.42 ± 0.82%. The total amount of LTZ released from the LLI hydrogel was (ANOVA, p-value < 0.0001) 61.58% less than that of the LTZ-free dispersion and 35.07% less than that of the optimized LLI formulation.
We assessed the kinetics and mechanism of LTZ release from the LLI hydrogel formulation utilizing the DDSolver program. The Weibull model and the Korsmeyer-Peppas model were determined to be the best-fit models for the release data of the LLI hydrogel formulation and the free LTZ dispersion, respectively, since they exhibited the minimal AIC values, the highest R2 values, and the highest MSC values, as presented in Table 3. The computed “n” values were 0.517 and 0.556, suggesting that non-Fickian diffusion was the process by which LTZ was released from the LLI hydrogel formulation. The MDT (t50%) for the LLI hydrogel was 35.62 ± 2.18 h, compared to 1.88 ± 0.12 h for the free LTZ dispersion. The LLI hydrogel increased the MDT of the free LTZ dispersion by a factor of 18.95.

Ex vivo permeation study
Figure 3B illustrates the permeation profile of the LLI hydrogel formulation compared to the free LTZ suspension. The total amount of LTZ that permeated from the LLI hydrogel was 690.02 ± 15.11μg/cm2, while the free LTZ suspension yielded 194.11 ± 7.45μg/cm2. The steady-state flux (Jss) for the LLI hydrogel was recorded at 11.44 ± 1.02μg/cm2/h, in contrast to the Jss of the free LTZ suspension, which was 3.09 ± 0.34μg/cm2/h. The amount of LTZ permeating from the LLI hydrogel was significantly greater than that from the free LTZ suspension, showing a difference of 3.55 times. Furthermore, the Jss for the LLI hydrogel was greater than that of the free LTZ suspension by a factor of 3.69.

In vivo evaluation of LLI hydrogel

Anti-tumor activity measurement
When compared to the positive control group, Fig. 4A shows that oral LTZ and LLI hydrogel were more effective against cancer. The initial tumor volumes for each group were as follows: 193.60 ± 18.22 mm3 for the positive control, 91.72 ± 5.64 mm3 for the oral LTZ group, and 166.74 ± 43.56 mm3 for the LLI hydrogel group. The positive control showed a significant 80.95% increase in tumor volume (ANOVA, p-value < 0.0001) from the beginning to the end of the trial. In contrast, the tumor volumes in the oral LTZ and LLI hydrogel groups significantly (ANOVA, p-value < 0.0001) decreased by 65.62% and 99.11%, respectively, during the same period. On the final day of the trial, the tumor growth inhibition percentage for oral free LTZ and LLI hydrogel was found to be 84.21% and 99.69%, respectively. Additionally, in comparison to the oral LTZ group, the LLI hydrogel group showed a significant (ANOVA, p-value < 0.0001) decrease in tumor volume of 51.04%.

Targeting studies
The buildup of LTZ in tumor and liver tissues is shown in Fig. 4B for both the oral LTZ and LLI hydrogel groups. The concentration of LTZ in the tumor from the LLI hydrogel was significantly (Student t-test, p-value < 0.0001) higher than that in the liver, with a 7.91-fold increase. Additionally, the LTZ concentration in the tumor from the LLI hydrogel was 9.36 times greater than that in the tumor from the free LTZ dispersion.

Histopathological examination
The histological analysis of the negative control group (Fig. 5A) revealed normal epidermis, subcutis with intact muscle structure, and hair follicles. In contrast, Fig. 5B illustrates the positive control group, which displayed an ulcerated epidermis and a significant hypercellular tumor characterized by rounded cells and notably pleomorphic spindles. Additionally, this group showed scattered giant cells embedded in a fibrotic stroma that extensively infiltrated the muscles. The oral LTZ group, represented in Fig. 5C, shows intact skin and a small hypocellular tumor composed of mildly pleomorphic rounded cells, with indications of apoptosis in a fibrotic stroma. Similarly, Fig. 5D, depicting the LLI hydrogel, also shows intact skin and a tiny hypocellular tumor made up of moderately pleomorphic spherical cells, exhibiting signs of cell death within a fibrotic stroma, with no evidence of muscle invasion.

Toxicological studies
Figure 6 presents the histological examination of the skin treated with LLI hydrogel in comparison to the control negative group. The LLI hydrogel group exhibited no signs of inflammatory reaction, reddening, or swelling in the dermal, epidermal, or subcutaneous layers of the skin.

Discussion
One kind of elastic vesicle that has been developed to enhance the bioavailability, effectiveness, sustainability, and penetration of pharmaceuticals is the invasome (Abou-Taleb et al. 2024; Ahmed and Badr-Eldin 2019; El-Bakry et al. 2024; Sahu et al. 2024). The components ofinvasomes consist of terpenes, phospholipids, cholesterol, and ethanol. Phospholipids function as lipid carriers for the drug (Abou-Taleb et al. 2024; Ahmed and Badr-Eldin 2019; El-Bakry et al. 2024; Sahu et al. 2024). Cholesterol improves the integrity of vesicular membranes (Abou-Taleb et al. 2024). By making the vesicle’s lipids more fluid, ethanol improves the medicines’ permeability (Patil et al. 2023). By changing the SC’s lipid structure, the water molecules released by terpenes make penetration easier (Patil et al. 2023). The development of LLI formulations was led by a literature study and pre-formulation experiments, which helped determine the independent variables and their levels (Abou-Taleb et al. 2024; Ahmed and Badr-Eldin 2019; El-Bakry et al. 2024; Sahu et al. 2024). Pre-formulation studies indicated that cholesterol concentration positively influenced EE and PS. However, cholesterol levels above 0.15% compete for bilayer packing space, making LTZ entrapment into invasomes more difficult. Therefore, the cholesterol concentration was maintained at a constant level of 0.15% w/v for all LLI formulations. Results that were found by Salem et al. were also comparable (Salem et al. 2021). Additionally, research conducted prior to formulation revealed a noteworthy correlation between the invasomes’ PS and EE% and the contents of phospholipids, ethanol, and cineole. Accordingly, the selected independent variables were phospholipid concentration (ranging from 1 to 3%), ethanol concentration (ranging from 0 to 5%), and cineole concentration (ranging from 0.5% to 1.5%).
The Box-Behnken design was employed to develop various LLI formulations, which were subsequently analyzed for EE% and PS. The Design-Expert software utilizes the ANOVA test to evaluate the model in relation to the EE% and PS data (Abdelbary and AbouGhaly 2015). In order to find the best fit between the data and the model, the program computes various metrics, including p-value, lack-of-fit, F-value, adjusted R2, predicted R2, and adequate precision. A p-value below 0.05 is required for a model to be deemed significant. Since the degree to which a model’s predictions differ from the actual data is indicated by its lack of fit, it is expected that an insignificant lack of fit will be achieved (Awan et al. 2022). The model with the highest F-value should be chosen since it represents the relationship between the source’s mean square and the residuals. If the predicted R2 and adjusted R2 of the model are both among the highest values and differ by less than 0.2, the model is deemed reliable (Abdelbary and AbouGhaly 2015). Additionally, in order for the model to be optimized, the adequate precision must be greater than 4 (Abdelbary and AbouGhaly 2015). After reviewing the data, the quadratic model was selected for both EE% and PS.
The Design-Expert software generates regression equations and 3D plots to evaluate the effects of phospholipid, ethanol, and cineole on EE% and PS (Awan et al. 2022; Salem et al. 2020). The ILL formulations demonstrated effective encapsulation of LTZ, indicating that invasomes may serve as a promising delivery vehicle for LTZ. The observed positive correlation between EE% and phospholipid concentration can be attributed to the increased number of lipid particles resulting from higher phospholipid concentrations, which enhance the capacity for LTZ entrapment. This observation is consistent with previous findings reported by Teaima et al. (Teaima et al. 2022). In contrast, a negative correlation was noted between EE% and ethanol concentration, as higher ethanol levels can solubilize phospholipids, leading to increased leakage from the lipid bilayer. This trend is supported by earlier research conducted by Jain et al. (Jain et al. 2021). Furthermore, elevated levels of cineole can dissolve lipophilic drugs such as LTZ and integrate into the lipid bilayer, which explains the positive correlation observed between EE% and cineole concentration. This finding is corroborated by the research of Ahmed et al. (El-Tokhy et al. 2021).
PS is a critical component to consider during the optimal formulation assessment. Smaller invasome vesicles can permeate the SC, facilitating deeper skin penetration (Babaie et al. 2020; Samir et al. 2024). Invasome PS is augmented by thick bilayers formed by higher quantities of phospholipids; thus, there is a positive correlation between PS and phospholipid concentration. This confirms what Teaima et al. found in their earlier studies (Teaima et al. 2022). Ethanol, on the other hand, reduces membrane thickness and density as its concentration increases, leading to a negative correlation with PS. There is agreement between this finding and previous research by Jain et al. (). Additionally, PS is directly correlated with cineole concentration, as higher levels of cineole disrupt lipid bilayer packing, resulting in larger invasomes that require a greater membrane surface area for packing. This conclusion is supported by earlier findings from El-Tokhy et al. ().
Achieving minimum PS and maximum EE values was the criterion for selecting the optimized LLI formulation. Not only did this formulation achieve the maximum desirability index value, but it also showed a strong match between the expected values and actual EE% and PS of the optimized formulation. DSC was used to assess the invasomes’ thermal behavior and crystallinity (Abou-Taleb et al. 2024). The absence of endothermic peaks in the DSC analysis of the optimized LLI formulation suggests that LTZ was completely integrated into the invasomes in an amorphous form, as previously observed with other components. Further evidence for this finding comes from FTIR, which confirmed the chemical-free incorporation of LTZ into the vesicles by showing that the optimized LLI had peaks consistent with all components. The presence of non-aggregated nano-sized vesicles was confirmed by TEM examination of the optimized LLI formulation’s shape. The optimized LLI formulation had low PDI, which meant that the vesicles were evenly dispersed. Additionally, a negative zeta potential for the optimized LLI formulation was observed, indicating its physical stability and absence of aggregation.
Transdermal administration of the invasomal formulation is enhanced by utilizing a suitable hydro-gelling medium that offers favorable mucoadhesive and rheological properties (Madan et al. 2019). The choice of gelling agent type and concentration was informed by a thorough literature review and preliminary studies conducted previously (Madan et al. 2019). Carbopol 934 is a widely used anionic hydrogelling agent due to its excellent buffering capacity, high viscosity in aqueous solutions, acceptable rheological characteristics, and minimal skin irritation (Madan et al. 2019). During the preparation process, the thickness of the hydrogel was assessed using triethanolamine, which alters the basicity of the formulation and serves as a neutralizer (Madan et al. 2019). A review of the literature and pre-formulation studies concluded that a 2% concentration of carbopol 934 was optimal, as it produces a hydrogel that is more stable and elastic compared to other concentrations, despite findings reported by Madan et al. (). Consequently, the optimized LLI formulation was incorporated into the hydrogel using 2% carbopol 934. The consistency of the hydrogel is critical when applying it to the skin, with viscosity being a significant factor. The LLI hydrogel is suitable for transdermal drug delivery due to its viscosity coefficient. In comparison to the optimal LLI formulation, the LLI hydrogel formulation exhibits considerably higher viscosity coefficient, attributed to the cross-linking and elevated viscosity of the carbopol polymer.
To determine the required amount of LTZ dissolution medium and its saturation solubility, an equilibrium solubility investigation was conducted. In the release and permeation investigations, the sink condition was maintained using a solution of 50 mL of PBT. This solution was chosen since it exceeded the saturation solubility of LTZ. The release of LTZ from the LLI hydrogel was studied and compared with the optimized LLI formulation and free LTZ suspension. The presence of cholesterol and phospholipid in the optimized LLI and LLI hydrogel formulations significantly reduced LTZ release compared to free LTZ. In contrast to the optimized LLI formulation, because the carbopol polymer is cross-linked and has enhanced viscosity, the LLI hydrogel releases LTZ gradually. The in vitro release kinetics and mechanism of LTZ from the LLI hydrogel formulation were assessed using the DDSolver program (El-Bakry et al. 2024). This program uses criteria that include the minimum AIC, maximum R2, and MSC values, along with data entry, to identify the optimal model fit. Consequently, for the LLI hydrogel formulation, the Weibull model proved to be the most appropriate. The release exponent “n” was calculated to establish the release method. The release of drugs is thought to be caused by non-Fickian diffusion if “n” is larger than 0.45 and by Fickian diffusion if “n” is smaller than 0.45 (Hornsby et al. 2023). The MDT verified that the LLI hydrogel formulation had an extended-release pattern. The capacity of the formulation to retain medicines becomes more apparent when the MDT value rises (Hossain et al. 2013). The high MDT of the LLI hydrogel formulation supports its delayed release pattern compared to free LTZ.
Ex vivo permeation studies are highly valued for their ability to enhance our understanding of the behavior of nanovesicles in vivo (Nangare and Dugam 2020). When compared to the free LTZ suspension, the LLI hydrogel formulation had much higher LTZ permeation. This difference can be attributed to the breakdown of invasomes, which release terpenes, ethanol, and phospholipids to facilitate penetration. The increased fluidity and smaller particle size of the invasomes, driven by cineole and ethanol, contributed to their enhanced penetration. These results are in agreement with what Dragicevic-Curic et al. have already published (Dragicevic-Curic et al. 2008).
Among malignancies affecting females, BC ranks high. Tumors progress in a sequential fashion, with activation of RAS gene family members playing a role in this process. Animal models of chemically produced cancers, such as DMBA, mimic the histological characteristics of human tumors. Young rats are particularly susceptible to BC induction when administered DMBA due to the rapid proliferation of their cells and the ongoing development of their mammary glands (Gul et al. 2021). Consequently, rats were treated with DMBA to induce experimental BC. By monitoring factors such as tumor volume and histological features, it was confirmed that BC had progressed successfully (Gul et al. 2021).
The tumor volume was significantly higher in the positive control group when contrasted with the negative control group. These results were confirmed by histopathological examination. These results confirmed that the DMBA BC model was successfully established in the rats. Experimental results in an in vivo DMBA BC rat model corroborated the in vitro results. Both the oral free LTZ and the transdermal LLI hydrogel showed better anti-cancer effects than the positive control group, thanks to a notable decrease in tumor volume. These findings were further validated by histopathological examination. Additionally, compared to rats given an oral LTZ suspension, those given LLI hydrogel showed a more significant decrease in tumor volume as well as improved histology samples. This can be attributed to the penetration-enhancing properties of cineole and ethanol within the SC, as well as the smaller size of invasomal formulations. The levels of LTZ in hepatic and cancerous tissues were measured for both the oral LTZ and transdermal LLI hydrogel to highlight the selectivity and targeting effects of the latter. The findings illustrated that the LLI hydrogel effectively increased LTZ accumulation in cancerous tissues, thereby enhancing its anti-tumor efficacy. There was no sign of skin irritation or inflammation when the LLI hydrogel group was examined histologically. Based on these findings, transdermal LLI hydrogel appears to be a safe and effective alternative to current BC treatments.
The current study emphasizes the potentially transformative effects of ILL hydrogel formulation design on BC management. This innovation could lead to new treatment options aimed at preventing BC, which would benefit both patients and clinical research. Consequently, the likelihood of side effects may be reduced by lowering the dosage required. However, clinical trials and extensive preclinical studies in animals are still essential to secure approval for human use. Additionally, the method used to induce breast cancer presents a limitation of this research. The comparison of DMBA-induced tumors in rats to human BC indicates significant behavioral, genetic, and physiological differences. Moreover, DMBA may also impact normal cells and adjacent tissues, including the immune system, in addition to targeting tumors. To address these limitations, we will utilize supplementary models, such as xenograft mouse models, in our forthcoming article. We will conduct a comparative assessment of intratumoral ILL formulation versus transdermally administered ILL hydrogel for the treatment of BC.

Conclusion
BC is the most often diagnosed medical illness among women globally and ranks second in cancer-related death. LTZ, a potent third-generation aromatase inhibitor, is frequently used to treat metastatic BCs. This study’s primary objective is to develop a hydrogel formulation of LLI that sustains LTZ release while enhancing its effectiveness, permeability, and targeting. The total amount of LTZ permeated from the LLI hydrogel exceeded that of the free LTZ suspension. Histological analysis of the LLI hydrogel group confirmed its anti-cancer activity. Our findings indicate that the LLI hydrogel can increase LTZ accumulation in cancerous tissues, thereby enhancing its anti-tumor efficacy. Additionally, the histopathological examination of the LLI hydrogel group showed no signs of skin irritation or tissue inflammation. Consequently, the transdermal LLI hydrogel represents a promising and safe treatment option for BC. In our forthcoming article, we will utilize a xenograft mouse model to conduct a comparative assessment of intertumoral ILL formulation versus transdermal ILL hydrogel for the treatment of BC.