In the present study, GEF liposomes were fabricated as TD patches aiming to improve the PK and BA, and to overcome the side effects associated with its oral administration. The optimized GEF-LIP-TD was evaluated for in vitro, in vivo biodistribution (PK and BA), and skin irritation studies in female Wistar rats. Through liposomes, the solubility and BA issues of GEF were resolved, and GEF delivery (PK) was effectively managed by fabrication as a TD patch.

GEF-LIP-TD was of interest for our study owing to the restricted clinical use of GEF in BC due to low BA (< 60%). Also, it was considered due to the continuous emergence of novel mutations, the number of resistance mechanisms with GEF, and the invention of novel classes of EGFR-based therapeutics. It is observed that the treatment of BC can rely on numerous factors, and can comprise combinations of surgery, hormones, chemotherapy, radiation, and targeted therapy. The efficacy of GEF therapy in BC remains debated as numerous side effects were observed after oral administration during phase II Clinical trials. These include a higher rate of nausea, and vomiting, decreased incidence of hot flashes, and improved incidence of adverse events in patients [48].

GEF is proven to be an active inhibitor of BC proliferation, either unaided or in adjunct with other cytotoxic agents. Also, studies prove that GEF is effective in Asian patients with EGFR mutation-positive cancer [49]. To highlight, GEF is also used in combination with a variety of drugs. These include paclitaxel, trastuzumab, tamoxifen, and platinum compounds like cisplatin, carboplatin, and oxaliplatin [50]. The inconsistency and failure of GEF lies in the fact that GEF monotherapy yields high response rates in estrogen receptor-positive. But, is suppressed with combinative therapy, for example with tamoxifen. Thus, the molecular mechanisms of endocrine sensitivity (inhibition of EGFR signaling) of GEF remain unclear and therefore may hinder clinical benefits. Also, preceding exposure to adjuvant endocrine therapy in BC patients may decrease the effectiveness of GEF supplementation [51]. In such circumstances, the treatment may be prolonged and possibly make GEF therapy a failure. Thus, a need for an alternative route of administration of GEF is highly suggested. Understanding the fact, a novel liposomal TDDS was designed to deliver GEF to reduce the adverse reactions and complications arising in therapy including low BA and to deliver the drug at site.

A means to improve drug delivery in BC is by use of liposomal carriers which are more biocompatible and site-specific. These provide a controlled and sustained drug release action with the least drug resistance and side effects. However, the success of liposome drug delivery lies in the interaction with the cell membrane. This is either attained by receptor-mediated, or local fusion (adhesion), nonspecific endocytosis, or phagocytosis and absorption into the cell membrane. The interaction of liposomes within the cell is influenced by its composition (Lipid: Cholesterol), the diameter and surface charge of the vesicle, the targeting ligand on the surface of liposomes, and the bio-environment [52]. These factors need to be optimized during the formulation development of liposomes and were effectively studied by DOE.

A CCD design was used to develop and optimize the GEF-LIP-TDDs, and to analyze the optimum concentrations of the selected variables (phospholipid, cholesterol) in liposomal formation (UL, ML, OL vesicles). The actual fitting of the model was calculated using statistical software. Run order was randomized to confirm the effect of time-related variables, and also to satisfy the necessity of independence of observations. The ANOVA for quadratic model response (% EE) is given in Table 2. By polynomial equation (Eq. 5) it was observed the equation in terms of coded factors can be cast-off to make forecasts about the response for the given levels of each factor.

In this study, RSM was used within a CCD to explore the effects of independent variables on the % (%EE) of GEF in liposomes (LIP). Specifically, the study focused on the concentration of lipoid S100 and cholesterol as the two key independent factors influencing %EE. Tables 1 and 2 summarize the designed experiments, statistical parameters, and the lack of fit for various models.

The values in Table 2 suggest that the quadratic model is significant for predicting %EE, with a model F-value of 37.97 and a p-value of less than 0.0001, indicating strong statistical significance. A significant lack of fit (LOF) is undesirable in a model, so an insignificant LOF suggests the model is highly accurate in predicting results. Adequate precision, which measures the signal-to-noise ratio, is also crucial. A ratio greater than 4 is necessary, and the ratio of 17.09 in this case specifies an adequate signal. This confirms that the model can reliably explore the design space (Behnajady et al., 2018).

The correlation coefficient (R2) is another key measure of model significance. An R2 value close to 1 suggests that the experimental results align well with the model. In this study, the regression model produced an R2 of 0.9644 and an adjusted R2 of 0.9390. This specifies that 96.44% of the variation in %EE was explained by the independent variables. The predicted R2 was also in good agreement with the adjusted R2, as the difference between them was less than 0.1. According to ANOVA results, the parameters A, B, AB, A2, and B2 are significant, with a p-value (Prob > F) of less than 0.05 for each.

To further validate the model, normal probability plots and residuals versus run plots were analyzed. Residuals, representing the difference between experimental and predicted values, were calculated, and the linearity of the normal probability plot confirmed the model’s accuracy. The random distribution of residuals versus run numbers further supported this accuracy. The predicted versus actual values are shown in Fig. 1b.

Contour and 3D plots were used to examine the effect of the concentration of lipoid S100 on %EE. As seen in Fig. 3h, the %EE initially increases and later decreases at very high lipid concentrations. This suggests that lipid concentration significantly affects %EE. The contour plot in Fig. 3h indicates that a lipid concentration between 145 and 217.5 mg is ideal for achieving a %EE above 96%.

Similarly, the effect of cholesterol on %EE was also observed. The %EE was found to increase initially and later decreased at high concentrations. A combined increase in both lipid and cholesterol concentrations initially reduced the %EE, but was later improved. A further increase in the concentration of both lipids decreases the %EE. This indicates that the concentrations of lipids and cholesterol significantly influence the drug loading of GEF in liposomes. For final model validation, the predicted results were compared with experimental results in Fig. 1b. The R2 value of 0.9390 indicates a strong correlation between the experimental data and the RSM predictions. The optimal conditions for maximum %EE were found at 145 mg of lipid and 28.12 mg of cholesterol, resulting in a %EE of 97.79%. Figure 1i shows the combined effect of both variables, with a desirability factor of 1. Likewise, all the other formulations experienced a slight decrease in %EE in comparison with F3. The lower %EE at higher phospholipid and cholesterol concentrations suggests that the GEF may have saturated the lipid bilayer, leading to an increased amount of free drug. As phospholipid and cholesterol concentrations rise, more energy is used to distribute the phospholipids rather than forming smaller vesicles, which may increase liposome size and lead to the formation of multi-vesicular vesicles (MVV).

From Fig. 1h, it is evident that %EE is highest at 145 mg, and slightly decreases when the phospholipid concentration increases to 217.5 mg. This indicates that increasing the phospholipid concentration leads to larger vesicles thereby reducing the %EE. Similar findings were also reported by Jaafevr-Maalej et al. (2010), where higher lipid concentrations resulted in larger, less stable liposome structures, thus decreasing %EE.

The polynomial equation with the coded factors shows A, B, AB and A2 are all negative signifying a high negative influence on the formulation. Thus, by increasing their concentrations the %EE will be decreased. The polynomial equation in terms of coded factors is given as;

$$Y\left( x \right) = + . - 0.*A\left[ \right] - 0.*B\left[ \right] - 0.0*}\left[ \right] - 0.*A^} \left[ \right] - 0.*B^} \left[ \right]$$

(5)

The optimized formulation was selected based on the desirability graph, which indicates how well each variable meets the criteria for combined desirability. In this case, the desirability was close to 1, indicating a strong response. The optimized formulation chosen was F3. The %EE, avg. particle size, and ZP of F3 were found to be 97.79 ± 1.38, 96.07 ± 3.30, and − 46.0 ± 0.80, respectively. Overall, there were no significant differences (p < 0.05) between the experimental and predicted values for all responses according to the RSM, confirming the validity of the proposed final composition of phospholipid and cholesterol.

Lipoid S100 was chosen as it provided the highest %EE (> 40%) at a strength of 50% of the dose size of GEF and it produced stable vesicles of size between 100 and 1500 nm. S100 is L-α-phosphatidylcholine (contains phosphatidylcholine ≥ 94.0% and phosphoglycerides) and is widely used in NDDs.

Liposomes are spherical hollow structures formed from both natural and synthetic phospholipids. They are uniformly dispersed in an aqueous medium. The ultimate structure, organization, and physicochemical properties of the liposomes depend on the following factors. These include the types, vesicle size, surface morphology, concentration, and charge of the constituent lipids. In addition, properties such as ionic capacity, pH, and temperature can also influence the structure and organization of liposomes [53]. Also, lipid composition can impose a commendable effect on vesicle size, rigidity, fluidity, and stability of liposomes. The surface (electrical) charge may also be unfavorable at times. These factors were taken as utmost priority in our study. Thus, lipoid S100 (phospholipid) was selected, and the ideal concentration was established with cholesterol to entrap more drugs and to improve the drug loading.

The primary features liposomes should possess include high %EE (DL), low vesicle size, and stability designated to suit permeation to the specific tumor cell. The vesicle size of the liposome needs to be optimized to favor increased circulation and residence time in the blood. These liposomal vesicles can favor enhanced in vivo drug release and significant accumulation of drug into the tumor cell. The optimization process was effectively controlled and adopted during operations where the pH (5.5), temperature (40–50 °C), and stirring speed (30–60 rpm) of RFE were suitably adjusted. It was observed that a high pH can also result in variable size and unstable liposomal vesicles.

Similarly, the vesicle size and drug loading will have a direct impact on drug delivery. These can impact BA, and are influenced by the nature of the drug and lipid, phospholipid and cholesterol composition. These parameters need to be efficiently controlled to develop stable liposomes. The temperature and the solvency conditions may also have a direct effect and need to be effectively controlled [54]. In our study, DOE was used to establish the formulation structure and to control the parameters crucial for obtaining a high %EE. A total of 13 formulations were fabricated and optimized with the least experimental time and composition. In the preparation of liposomes, the lipid and cholesterol concentrations were set between 42.46–247.53 mg, and 8.23–48.03 mg, respectively. These resulted in controlled vesicle size and good reproducibility.

DLS measurements (Table 4) show a significant reduction in vesicle size. It was observed that all the batches attained a particle dimension less than 373 (d.nm) indicating the method (TFH) chosen was ideal for the fabrication of liposomes. A high negative charge (ZP) was also detected in all batches, and the effect may be related to the polarity of ethanol [55]. Likewise, a high %EE (> 96%) was recognized in all the batches (F1–F13). F3 demonstrated a high %EE of 97.79 ± 1.38 and the least was observed in F6 (96.30 ± 3.89). It was observed that F3-GEF-LIP produced an extreme %EE (97.79 ± 1.38) and was further considered in the development of the TD patch. The %EE of all the formulations is illustrated in Table 3.

It is detected that liposomes of size 100–150 nm favor improved cellular uptake. These can also discharge from the blood capillaries within the infected diseased tissues (including kidney, heart, and lung) and enter through the (fenestrated) vessels into the tumor environment [56]. In our observation, F3-GEF-LIP may favor better cellular uptake due to small vesicle size (96.07 ± 03.30), high ZP (− 46.0 ± 6.80), and increased %EE (98.21 ± 0.24).

One of our prime interests was to understand the suitability of lipoid S100 in preparing liposomes, its vesicular formation, and its permeability and suitability as TDDS. A variety of lipids have been explored to enhance the solubility and targetability of drugs. Major lipids explored in the development of liposomes are L-α-dimyristoyl phosphatidylglycerol (DMPG), -α-dimyristoyl phosphatidylcholine (DMPC), L-α-distearoyl phosphatidylcholine (DSPC) and egg phosphatidylcholine (EPC) [57]. Yet, lipoid S100 was unexplored for fabricating liposomal—TDDS loaded with GEF. Lipoid S100 produced MVV of smaller size (< 100 d.nm) with high drug loading, and its drug entrapment was observed by TEM analysis (Fig. 3b). These types of vesicles will feature high payload and lipid content enabling better cellular adherence and drug delivery [58].

Observing the ex-vivo skin permeation of F3-GEF-LIP-TD, initially, a slow and controlled release of drug (< 9%) was observed at 30 min in comparison with regular liposomes (Fig. 4a). This may be due to the restricted barriers in the epidermal layer of the skin. A high drug release (%CDR, 83.32%) in F3-GEF-LIP-TD was observed at 48 h and may be attributed to a higher %EE (97.79 ± 1.35%) and a decrease in size of MVV (< 100 d.nm). A low %CDR (12.5 ± 0.17%) was observed in F11-GEF-LIP. This might be due to the higher proportion of lipid (217.5 mg) incorporated in the liposome leading to a decrease in drug diffusion.

Also, pure GEF exhibited a low drug release (38.1%) owing to its solubility characteristics (BCS class II, BA less than 60%) at pH 5.5. A slow (24 h) and extended drug release (48 h) in TD patches may be due to the effect of HPMC as it produces adequate swelling, gelling, and thickening resulting in prolonged drug release [59]. Also, HPMC (2.5% w/v) was found to produce a stable matrix for transdermal patches. The kinetic study reveals that the release behavior of F3-GEF-TD from the TD matrix resembled the Higuchi model (R2 = 0.9868), and follows diffusion [60]. It was noticed that the ideal proportion of lipid and cholesterol for liposomal formation was 145.0 and 28.12 mg, making it most suitable for effective and stable vesicle formation. The %CDR of F3-GEF-LIP-TD was significantly higher (83.32% ± 1.14, p > 0.05) at 48 h and was further chosen for skin irritation, biodistribution, and BA studies.

The safety and compatibility of TD patches were assessed by a skin irritation study using female Wistar rats. The Group II animals treated with F3-GEF-LIPs-TDP show no signs of rashes and edema in comparison with Group III treated with standard 0.8%v/v aqueous formalin (0 day, and after 72 h, Fig. 5). The irritation score index for F3-GEF-LIP-TD was between 0.0 and 0.5 for 72 h signifying non-irritancy (> 1.2 indicate-severe irritation). Skin irritation studies prove F3-GEF-LIP-TD to be safe and suitable for topical application.

The biodistribution and BA studies were performed on optimized F3-GEF-LIP TD using female Wistar rats. This was to analyze the localization of GEF-LIP in various organs and to recognize the importance of S100 in drug release and targeting. In the current study, a higher rate of GEF liposomal delivery was revealed in the liver (36,589 ± 238 ng) (Fig. 6a). In such a situation, liposomes can prominently reduce nephrotoxicity and cardiotoxicity. This type of distribution may be ascribed to the higher phagocytic uptake of liposomes in the liver than other organ cells [61]. As mentioned, GEF is highly lipophilic and can cross the blood–brain barrier (BBB). This was reflected in our study where a minimal drug accumulation was observed in the brain (10,254 ± 641 ng). The above data were statistically analyzed by two-way ANOVA and the results proved significant (p < 0.05).

BA was assessed for GEF (oral), GEF-LIP (oral) and F3-GEF-LIP-TD. A higher AUC (323.60) was observed for F3-GEF-LIP-TD in comparison with GEF (oral) and GEF-LIP (oral) after 24 h demonstrating improved drug permeation and absorption through the skin. In the BA study, the HPLC chromatogram of GEF exhibited a sharp peak at RT-15.41 min (Fig. 6c). The plasma drug conc. of oral GEF solution was found to be 0.0218 µg/ml, while that of GEF LIPs was 0.0286 µg/ml. A significant difference in plasma drug conc. was observed in F3-GEF-LIP-TD (0.0645 µg/ml) at 48 h (Table 5). The peak area for F3-GEF-TD was observed to be high (466.41 ± 06.18) at 18 h followed by a decrease at 48 h. The BA of TD patch was significantly increased after topical administration (74.05 ± 0.11%, p < 0.05).

From our results, GEF liposomes exhibited a controlled release and uptake of the drug with improved biodistribution. This could be vital in BC therapy as GEF-LIP-TD may be positively employed at various phases of breast cancer. Also, it was seen that F3-GEF-LIP exhibited low distribution (decreased AUC) and uptake in all the organs. In this situation, a more predictable and higher uptake of the drug at the site or by other tissues or diseased cells can be assured. These favor improved BA and enhanced therapeutic effects in BC [62]. The effective tissue distribution is influenced by interactions of liposomes with biological barriers and tunable nanoparticle characteristics. These comprise core composition and properties, vesicle dimension, surface alterations, and ligand functionalization [63]. These may impact the in vivo biodistribution and blood circulation half-life of circulating nanovesicles. In such circumstances, the liposomes may delay opsonization, reduce the level of nonspecific uptake, and increase tissue-specific accumulation [64]. In our observation, F3-GEF-LIP-TD had the best proportion of lipid and cholesterol to produce stable liposomes of appropriate size, and %EE, and was also reflected in biodistribution studies.

The BA study result specifies GEF-TD as a substitute dosage form for oral GEF. Also, topical administration of GEF can avoid unwanted GI complications and can deliver better patient compliance. A high AUC was observed in F3-GEF-TD which may be due to improved absorption and permeation of GEF through the skin. Permeation of GEF occurs mainly through the intercellular and intracellular pathways [65]. The enhanced absorption may be the result of the monodisperse nature of vesicles, nano size, and the influence of a high negative surface charge with liposomes. These can help in better permeation and were detected to be faster through the stratum corneum leading to enhanced drug absorption [66]. Similarly, the high uptake of GEF liposomes could be due to the formation of small MVV (< 100 nm) resulting in efficient drug loading (97.79 ± 1.38%). The BA of GEF-LIP-TD was observed to be 74.05 ± 0.11% in comparison with oral GEF-LIP (65.25 ± 0.08%) and pure GEF (58.10 ± 0.17%). A substantial increase in BA (74.05 ± 0.11) was observed in GEF-LIP-TD.

Stability studies on F3-GEF-LIP-TD indicate no change in drug content (96.05 ± 0.16%) for 3 months (p < 0.05). The results prove a combination of HPMC, PEG 600, and oleic acid was found to produce a stable and compatible TD film.

The present study emphasizes on the development of GEF-LIP-TD patch, and its examination of PK. However, the study has limitations including analyzing the exact mechanism in the biodistribution of GEF liposomes. Also, the targeting efficacy of GEF needs further examination, to prove its efficacy as tumor-free rats were used. Similarly, examining the kinetics of GEF liposomes under various pH conditions would probably draw a better assumption in the drug release behavior. To complete, GEF-LIP-TD patch was successfully developed and evaluated for biodistribution and BA studies. Our report suggests lipoid S100 to be unique and compatible in consideration in the design of an efficient liposomal TDDS.