CONTINUOUS PRODUCTION OF RALOXIFENE HYDROCHLORIDE LOADED NANOSTRUCTURED LIPID CARRIERS USING HOT-MELT EXTRUSION TECHNOLOGY

Abstract

The aim of this study was to utilize a continuous process for the production of orally administered raloxifene hydrochloride (RX-HCl) loaded nanostructured lipid carrier (NLC) formulations for extended drug release using hot-melt extrusion (HME) technology coupled with probe sonication, and also to evaluate the in vitro characteristics of the prepared NLCs. Preparation of the NLCs using HME technology involved two main steps, first formation of a pre-emulsion after extrusion and then size reduction of the pre-emulsion using probe sonication to obtain the NLCs. A screw speed of 100 rpm and a barrel temperature of 85 °C, were used in the extrusion process. NLCs prepared by HME technology showed a lower particle size compared to those prepared by the conventional probe sonication method. The prepared NLCs had high entrapment efficiency values (>90 %). In vitro drug release was evaluated using dialysis bag diffusion technique and USP apparatus I. Overall, the RX-HCl loaded NLCs had a higher rate of drug release than the pure drug. The release profile for the F4-3 NLC formulations and pure drug at the beginning and end of the stability study were comparable. The particle size of the prepared NLCs remained stable over the storage period and all PDI and zeta potential values were ≤ 0.5 and in the range of −15 to −30 mV, respectively, indicating good physical stability of the formulations. In summary, HME technology and probe sonication were successfully used to prepare RX-HCl loaded NLC formulations with shorter processing times as compared to the conventional probe sonication method, which makes this technique a uniquely more industry-friendly method.

Graphical Abstract

INTRODUCTION

Nanostructured lipid carriers (NLCs) are the second generation of lipid nanoparticles which offer some advantages over the solid lipid nanoparticles (SLNs) [1]. These advantages include higher drug incorporation capacity, improved drug release properties, and less amount of stabilizers required compared to the SLNs [2]. NLCs are produced using a combination of solid lipids and liquid lipids which are organized in nano-compartments inside the solid lipid matrix [3]. Some of the solid lipids reported in literature include glyceryl monostearate, precirol ATO5, compritol 888ATO, glyceryl tristearate, dynasan 114, and dynasan 116, among others [4]–[6]. Liquid lipids reported in literature include capmul MCM C8 & C10, castor oil, oleic acid, labrafac lipophile WL 1349, labrafil M 1944 CS, and labrafac PG & CC, among others [4]–[6]. Surfactants and co-surfactants are important excipients used in the formulation of NLCs and are used to stabilize the prepared NLC system [4]–[6]. Surfactants and co-surfactants reported in literature include polysorbate 80, polysorbate 20, cremophor EL, lutrol F68, and poloxamer 188, among others [4]–[6]. The Active Pharmaceutical Ingredient (API) in the NLC is incorporated in the liquid lipid and encapsulated by the solid lipid [3]. This configuration gives the drug some degree of mobility and offers stability to some extent [3]. NLCs are safe and compatible colloidal drug carriers which are able to provide controlled delivery of the API and have various drug delivery applications via the oral, dermal, pulmonary, and ocular routes [7].

Various methods have been used to produce NLCs and most of these methods reported in literature include high-pressure homogenization, micro-emulsification, solvent displacement, supercritical fluid technology, high-shear homogenization, and ultrasonication [7]. Production of NLCs using high-pressure homogenization mostly involves the use of any of the two types of homogenizers currently available on the market, that is, jet-stream and piston-gap homogenizers [6]. Typically, the drug is dissolved or dispersed in the molten lipid, and then a pre-emulsion is formed by dispersing the drug-loaded hot lipid melt in a surfactant solution of equivalent temperature by high-speed stirring [1]. The obtained pre-emulsion is then passed through a high pressure homogenizer. The obtained hot nanoemulsion is then cooled in a controlled way, and the lipid mixture recrystallizes to form NLCs. Preparation of NLCs by high-pressure homogenization may involve hot or cold homogenization. The hot homogenization process is carried out at temperatures above the melting point of the lipids used in the formulation, while the cold homogenization process involves cooling the lipid melt and then grinding the solid lipid to lipid microparticles [6]. In the microemulsion process, a hot microemulsion is formed composed of melted lipid with surfactant/surfactant-cosurfactant and water, which is then poured into cold water [1]. The microemulsion breaks by this dilution with water forming nanoparticles, and at the same time the lipid recrystallizes in the cold water. These conventional production methods pose some disadvantages that include multistep processing, poor energy efficiency, potential for dilution of particle dispersion, and frequent failures due to batch-to-batch variations [3]. These disadvantages make the conventional methods used for NLC production less industry-friendly and hence the need for continuous processing methods.

Hot-melt extrusion (HME) processing was established in the early 1930s and during that time it rapidly became the most widely applied processing technology in the plastic, rubber, and food industries [5]. HME is a continuous pharmaceutical process that involves pumping polymeric materials with a rotating screw at temperatures above their glass transition temperature (T_g) and sometimes above the melting temperature (T_m) to achieve molecular level mixing of the APIs and thermoplastic binders, polymers, or other excipients [8]. This molecular level mixing converts the components into an amorphous product with a uniform shape and density, thereby increasing the dissolution profile of the poorly water soluble drug [5]. HME has also been utilized for the delivery of water-soluble drugs with several applications such as taste masking [8]. HME has gained increased interest in the pharmaceutical industry due to advantages that include providing a faster and more efficient time to achieve the final product, increased efficiency of drug delivery to the patient, formulations with high drug loading can be manufactured by HME with the desired release profile, good content uniformity can be achieved for very low drug loading due to intimate mixing of components, HME allows conversion of API to amorphous form which can result in enhanced bioavailability. Moreover, HME is a scalable continuous process, and its product quality can be monitored online, inline, or offline, and desired dosage forms can be easily manufactured by HME process [8], [9]. Because of these advantages over the conventional pharmaceutical manufacturing processes, HME has emerged as an alternative platform technology for pharmaceutical dosage forms, such as tablets, capsules, films, and implants for drug delivery via oral, transdermal, and transmucosal routes [10]. Hot-melt extrusion technology has been previously utilized to prepare nanoformulations [3], [10], [17]. Several research articles have been published describing the use of HME as the novel technique of choice in dealing with the day-to-day formulation challenges of new active pharmaceutical ingredients, with numerous aspects of the HME technology being extensively cited in the literature and the number of patents based on HME techniques rising steadily worldwide in recent decades [11]–[19]. HME technology gained a lot of attention and importance in the recent decade after the US Food and Drug Administration encouraged the use of continuous manufacturing processes [3].

Raloxifene hydrochloride (RX-HCl) is a selective estrogen receptor modulator that acts as an estrogen agonist on bone and the liver and therefore increases bone mineral density and decreases fracture incidence [7]. RX-HCl is used in the prevention of osteoporosis and invasive breast cancer in postmenopausal women, although its therapeutic efficacy is limited by its low solubility in physiological pH conditions and also undergoes extensive first pass metabolism [20].

In the current study, a continuous process was used for the production of orally administered RX-HCl loaded NLCs for extended drug release using HME technology coupled with probe sonication. Physicochemical characteristics of the prepared NLCs were evaluated by particle size, PDI, zeta potential, entrapment efficiency, drug loading, and in vitro drug release. Process parameters for the extrusion process like feeding rates for the volumetric feeder and peristaltic pumps, were optimized. The continuous production process used in this study will overcome most of the limitations presented by conventional methods.

MATERIALS AND METHODS

Materials

RX-HCl, liquid lipids (capmul MCM C8, castor oil, and labrafil M 1944 CS), and surfactants (polysorbate 80, polysorbate 20, and cremophor EL) were purchased from Fisher Scientific (Hanover Park, IL USA). Solid lipids, glyceryl monostearate, compritol 888ATO, precirol ATO5, and dynasan 118 were gifted from Gattefossé (Paramus, NJ USA). Poloxamer 188 (Lutrol^® F68) was purchased from Sigma-Aldrich (St. Louis, MO, USA). All other chemicals and solvents used in the study were of HPLC grade and de-ionized water was used throughout the study.

Selection of Lipids and Surfactants

According to the FDA’s Inactive Ingredients Database (IID), all the solid lipids, liquid lipids, and surfactants that were chosen for screening in this study, have been used in approved drug products for oral administration. Selection of lipids and surfactants was further based on literature search of peer-reviewed journal articles for the commonly used solid and liquid lipids and surfactants in the formulation of colloidal drug carrier systems like NLCs and SLNs, that are used for controlled drug delivery. Lipids were screened for their potential to solubilize RX-HCl and assessment of their solubilization capacity for RX-HCl was done by means of visual observation with the naked eyes under normal light coupled with HPLC analysis. Briefly, an excess amount of drug was added into each lipid vehicle. The solid lipid was melted first before adding excess drug. This was followed by vortex mixing for 5 minutes. The mixtures were then shaken at 100 rpm for 24 hours in a thermostatically controlled shaking water bath (Thermo Scientific^™ Precision^™ Reciprocal Shaking Water Bath, Waltham, MA, USA). Glyceryl monostearate and precirol ATO5 were shaken at 65 °C while compritol 888ATO and dynasan 118 were shaken at 80 °C. The liquid lipids were shaken at ambient temperature. At the end of the study, mixtures were visually observed and then filtered through a 0.45 μ Millipore membrane filter (Burlington, MA, USA). Filtrates were suitably diluted with methanol and analyzed using the HPLC analysis method described in this study. Surfactant selection was based on reported data of compatibility of RX-HCl with various surfactant solutions, structural classification of the surfactants, and drug solubility using the method described for the lipids. After the screening studies, two solid lipids, two liquid lipids, and two surfactants were selected for further studies. The solid and liquid lipids used in this study were reported to be safe, biocompatible, and biodegradable. All experiments were conducted in triplicate.

Preparation of NLC Formulations

A. Preparation using the conventional probe sonication method for preliminary studies:

Two batches consisting of twenty four blank NLC formulations in each batch, were prepared by the conventional probe sonication method. The 1^st batch consisted of only one primary surfactant while the 2^nd batch consisted of a primary surfactant and a co-surfactant. The co-surfactant used was poloxamer 188 (Lutrol^® F68). Composition of the blank NLC formulations is as shown in Table 1. Based on stability results, formulations F4-2 and F4-3 from the 2^nd batch were selected for the preparation of RX-HCl loaded NLC formulations using conventional probe sonication and HME methods. RX-HCl concentration was kept constant at 1.2 % w/v in all the formulations.

Table 1.

Composition of blank NLC formulations

Formulation	Composition (%w/v)
	Solid Lipids		Liquid Lipids		Surfactants
					1st Batch		2nd Batch
	Glyceryl Monostearate	Compritol 888ATO	Capmul MCM C8	Castor Oil	Polysorbate 80	Cremophor EL	Polysorbate 80	Cremophor EL	Poloxamer 188
F1-1	5		1			1		0.75	0.25
F1-2	4.5		1.5			1		0.75
F1-3	4		2			1		0.75
F2-1	5		1		1		0.75
F2-2	4.5		1.5		1		0.75
F2-3	4		2		1		0.75
F3-1	5			1		1		0.75
F3-2	4.5			1.5		1		0.75
F3-3	4			2		1		0.75
F4-1	5			1	1		0.75
F4-2	4.5			1.5	1		0.75
F4-3	4			2	1		0.75
F5-1		5	1			1		0.75
F5-2		4.5	1.5			1		0.75
F5-3		4	2			1		0.75
F6-1		5	1		1		0.75
F6-2		4.5	1.5		1		0.75
F6-3		4	2		1		0.75
F7-1		5		1		1		0.75
F7-2		4.5		1.5		1		0.75
F7-3		4		2		1		0.75
F8-1		5		1	1		0.75
F8-2		4.5		1.5	1		0.75
F8-3		4		2	1		0.75

B. Preparation using the hot-melt extrusion (HME) method:

Preparation of RX-HCL loaded NLCs using HME technology involved two main steps. The first step involved the formation of a pre-emulsion by extruding a combination of solid lipid, RX-HCl, liquid lipid, and the aqueous phase through the HME barrel, and the second step involved size reduction of the pre-emulsion utilizing probe sonication to obtain the NLC formulation. Extrusion was carried out on an 11-mm co-rotating twin screw extruder (11-mm Process 11^™, Thermo Fisher Scientific, Karlsruhe, Germany). Schematic representation of the preparation of the NLC formulations using HME and probe sonication is as shown in Figure 1. RX-HCl was uniformly mixed with the solid lipid, and introduced in the barrel using a volumetric feeder. Liquid lipid and the aqueous phase (solution of surfactants and water) were heated to 85 °C and then respectively, injected in zone 2 and zone 4 of the barrel using peristaltic pumps. A screw speed of 100 rpm, barrel temperature of 85 °C, and sonication time of 10 minutes, were used in this study. Volumetric feeder and peristaltic pump feeding rates were optimized in this study. Figure 2 shows the screw configuration that was used for the extrusion process [3]. The pre-emulsion obtained was subjected to probe sonication (SONICS Vibracell^™, Sonics and Material, Inc., Newton, CT) with an amplitude of 40 %, to obtain the RX-HCl loaded NLCs.

Figure 1.

Schematic representation of preparation of RX-HCl NLCs using HME technology coupled with probe sonication.

Figure 2.

Screw configuration used for the extrusion process

Characterization of NLC Formulations

Particle size (z-average diameter) and PDI were measured by dynamic light scattering technique using Malvern Zetasizer Nano ZS (Malvern Instruments, UK) at 25 °C [21]. Zeta potential was also measured using the same instrument. Prior to measurements, the formulations were suitably diluted with distilled water (100x dilution factor) and all measurements were carried out at a scattering angle of 90° at 25 °C. Percent entrapment efficiency and drug loading of the prepared NLC formulations were determined [22]. Entrapment efficiency was determined by calculating the entrapped drug after removal of the unentrapped drug using Amicon filters (100 kDa MWCO) centrifuged at 13,000 rpm for 15 minutes at 20 °C. The filtrate was diluted appropriately with methanol and analyzed using a suitable HPLC method as described in the analytical method section. The following formulas were used to calculate percent entrapment efficiency and drug loading [23];

$$\text{\% Entrapment efficiency =}\frac{\text{Amount of RX–HCl entrapped in NLCs}}{\text{Amount of RX–HCl added to the formulation}}\times 100$$

$$\text{\% Drug loading =}\frac{\mathit{\text{Amount of RX–HCl entrapped in NLCs}}}{\mathit{\text{Amount of RX–HCl and lipids added to the formulation}}}\times 100$$

HPLC Analysis Method for Raloxifene Hydrochloride

RX-HCl was analyzed using HPLC (Waters Corporation) equipped with a UV-vis detector operating at 288 nm. Samples were chromatographed on a stainless steel RP-C18 column with dimensions of 250 × 4.6 mm, packed with 5 μm particles. The mobile phase consisted of a mixture of acetonitrile (33 % v/v) and monobasic potassium phosphate buffer (67 % v/v). A flow rate of 1 mL/min and an injection volume of 10 μL were used for analysis [24].

In Vitro Drug Release Studies

In vitro release of RX-HCl from NLCs was evaluated using a dialysis bag diffusion technique with some modifications [25]. The receptor compartment (drug release media) consisted of 500 mL pH 6.8 phosphate buffer. To maintain sink conditions, 0.5% v/v polysorbate 80 was added to the dissolution media and a larger volume of the dissolution media (500mL) in respect to the saturation solubility of raloxifene hydrochloride was used. RX-HCl loaded NLC formulation (equivalent to 18 mg of RX-HCl) was placed in the dialysis bags (12–14 kDa MWCO) and then the bags were tightly sealed at both ends to avoid any leakages. The dialysis bags were immersed in the receptor compartment placed in USP apparatus I (basket) and maintained at 37 ± 1 °C and 100 rpm, and at predetermined time points, aliquots were withdrawn from the release media and filtered through 0.45 μm filter. The withdrawn aliquots were replaced with equal amounts of fresh release media. Samples were suitably analyzed using HPLC. All experiments were conducted in triplicate and the average values were recorded.

Stability Study

The physicochemical stability of the prepared NLC formulations was assessed based on particle size, PDI, zeta potential, percent entrapment efficiency, and percent drug loading. These parameters were assessed at the beginning and end of the stability period. The NLC formulations were stored at 25 °C/ 60 % RH and 40 °C/ 75 % RH for a period of one month. The stability results were statistically analyzed to check for any significant differences in particle size, PDI, and zeta potential between the NLCs prepared by the conventional method and those prepared by HME technique.

RESULTS AND DISCUSSION

Selection of Lipids and Surfactants

The first step in the selection of lipids for the preparation of NLC formulations, is the assessment of solubility of the API in the solid and liquid lipids. This is because the solubility of the API in the lipids is one of the most important factors for determining entrapment efficiency of the drug in NLC formulations [26]. With the aid of visual observation with naked eyes under normal light coupled with HPLC analysis of the filtrates, the solubility of RX-HCl in the lipids was based on the amount of precipitated drug at the bottom of the scintillation vials at the end of the solubility studies, and this amount of precipitated drug was inversely proportional to the solubility of RX-HCl in the lipids. Drug solubility results for the solid lipids were as follows: glyceryl monostearate (10.44 ± 1.01 μg/mL), compritol 888ATO (2.68 ± 0.37 μg/mL), dynasan 118 (2.04 ± 0.13 μg/mL), and precirol ATO5 (1.22 ± 0.09 μg/mL). Drug solubility results for liquid lipids were as follows: capmul MCM C8 (2.44 ± 0.10 μg/mL), castor oil (1.23 ± 0.27 μg/mL), and labrafil M1944CS (0.80 ± 0.02 μg/mL). Drug solubility results for surfactants were as follows: polysorbate 80 (9.57 ± 0.85 μg/mL), polysorbate 20 (9.95 ± 0.69 μg/mL), and cremophor EL (8.90 ± 0.41 μg/mL). Based on the results of the solubility studies, solid lipids, glyceryl monostearate and compritol 888ATO, and liquid lipids, capmul MCM C8 and castor oil, were selected for further studies since they had a lower amount of precipitated drug compared to the other lipids, and hence had a higher solubilization capacity for RX-HCl as confirmed with the HPLC analysis. The surfactants selected were reported to be compatible with RX-HCl [27], [28]. Surfactants with lower drug solubility were selected in order to avoid partitioning of drug in aqueous phase more than the oil phase. Based on structural classification of surfactants, non-ionic surfactants are reported to be less toxic compared to their ionic counterparts. Based on the above selection criteria for surfactants, polysorbate 80 and cremophor EL were selected for further studies.

Preparation of NLC Formulations

Preparation of RX-HCL loaded NLCs involved the formation of a pre-emulsion by extruding a combination of solid lipid, RX-HCl, liquid lipid, and the aqueous phase through the HME barrel, and finally size reduction of the pre-emulsion utilizing probe sonication to obtain NLC formulations. The extruder’s screw consisted of a conveying system and mixing zones which helped in transport and efficient mixing, a die at the end of the barrel, and downstream equipment that was used to collect the pre-emulsion for further processing [3]. The pre-emulsion collected downstream was simultaneously subjected to sonication without having to transfer or move the pre-emulsion from one processing equipment to another and this eliminated the possibility of batch-to-batch variabilities that would occur during the transfer process to the probe sonicator as in the conventional method. The screw configuration is as shown in Figure 2. The first mixing zone served the purpose of mixing the melted solid lipid and RX-HCl with the liquid lipid and the second mixing zone served the purpose of efficiently mixing the oil and aqueous phase to form a pre-emulsion that further underwent size reduction by probe sonication to form the NLCs. RX-HCl and the solid lipid were introduced in the extruder barrel using a volumetric feeder at an optimized feeding rate of 2.369 g/minute. The liquid lipid phase was heated to 85°C and then injected in zone 2 of the barrel using a 500 series (520S) Watson-Marlow peristaltic pump at a feeding rate of 0.7 rpm. The aqueous phase (solution of surfactants and water) was heated to 85°C and then injected in zone 4 of the barrel using a 300 series (323E) Watson-Marlow peristaltic pump at a feeding rate of 30 rpm. The conventional method involved preparation of the NLCs by standalone probe sonication that took a longer preparation time and produced NLCs of slightly higher particle size than those produced by the HME technique. Physically, NLCs produced by both methods appeared to be fine, but showed differences with further physicochemical evaluation, with the NLCs produced by the HME technique showing better and desirable physicochemical properties. For both the conventional and HME methods, an amplitude of 40 %, a pulse of 20 seconds on and 10 seconds off, and a sonication time of 10 minutes were used. The sonication batch sizes were 10 mL and 400 mL (of the pre-emulsion) for the conventional and HME methods, respectively. A 500 Watt 20 kHz SONICS Vibracell^™ ultrasonic processor was used for probe sonication, 2 – 3 % of the maximum power (500 Watts) was supplied for both the conventional and HME sonications. The temperature of the NLC formulations from both the conventional and HME methods was between 48.7 – 56.5 °C but started dropping immediately at the end of the sonication when the NLC formulations were taken out of the probe sonicator housing. It took about 2 hours to prepare 10 mL of the NLC formulation using the conventional method, while with the HME technique, 400 mL of the NLC formulation was produced in less than 1 hour. The conventional method took about 2 hours because of the multi-step processing involved i.e., heating and melting of the solid lipids and liquid lipids first, adding drug to the lipids and then stirring for some time, manually adding the aqueous phase to the lipid phase drop by drop until the aqueous phase is depleted and then continuing with the stirring for some time, and finally transferring the coarse emulsion for size reduction. In the HME method, a continuous processing method was used, and the processes involved were automated with just switching on/ off of the peristaltic pumps and the volumetric feeder and melting the solid lipids within the extruder barrel. In the HME method, the total running time from when the volumetric feeder and the peristaltic pumps were switched on to the depletion of all the API/solid lipid mixture, liquid lipid, and aqueous phase, was 6.75 minutes at the optimized process parameters. This makes the HME technique used in this study a faster and more industry-friendly method for the preparation of the NLC formulations. Other inline sonication methods include the use of flow cells that offer inline and continuous sonication of large volumes. In this type of inline sonication, the pre-emulsion would be continuously fed through an ultrasonic flow cell reactor, and can either be run as single-pass or be recirculated for further size reduction.

Characterization of NLC Formulations

Assessment of formulation variables of the blank NLC formulations from the 1^st and 2^nd batches was based on particle size, PDI, and zeta potential measurements after a 14-day storage stability period. Particle size, PDI, and zeta potential are the key factors for evaluating the stability of colloidal dispersions [17], [18]. Selection of formulations for further studies was based on stability results and the physical properties (e.g., gelling) of the formulations. Stability results are shown in Figure 3. Blank NLC formulations that showed some extent of gelation on day 0 were not subjected to the 14-day storage stability period of the preliminary studies and subsequent further studies because it was very important that we have final NLC formulations of desirable lower particle size and good physical stability. Based on the stability results, the 2^nd batch showed a relatively more stable particle size compared to the 1^st batch. On day 14, particle size was within a range of 242.9 ± 6.8 to 721.2 ± 188.2 for the 1^st batch and 192.9 ± 3.3 to 416.9 ± 7.9 for the 2^nd batch. This could be explained by the tendency of the primary surfactant/co-surfactant combination of the 2^nd batch to stabilize the system [31]. NLC formulations for further studies were selected based on overall stability results (particle size, PDI, and zeta potential) from day 1 through day 14. Blank NLC formulations from the 2^nd batch showed relatively stable particle size, PDI, and zeta potential than those from the 1^st batch. NLC formulations for further studies were also selected based on the range and stability of particle size. And hence F4-2 and F4-3 blank NLC formulations from the 2^nd batch were selected for further studies because they showed relatively stable particle size from day 1 through day 14 and also had the lowest particle size. On day 14, F4-2 and F4-3 blank NLC formulations had a particle size of 192.9 ± 3.28 and 224.5 ± 4.70, respectively. RX-HCl loaded NLC formulations prepared by the HME technique generally showed a lower particle size compared to the NLCs prepared by the conventional method. On Day 0, the particle size of the NLCs prepared by HME was 229.6 ± 6.4 and 199.5 ± 1.2 for F4-2 and F4-3 compared to 327.4 ± 9.8 and 284.2 ± 45.1 for F4-2 and F4-3 NLCs prepared by the conventional method. The reduced particle size can be explained by the additional shearing effect of the screws in the HME extrusion process that could cause a reduction in particle size of the extruded pre-emulsion material. This can be further explained by the fact that the probe sonication parameters (amplitude, pulse, and sonication time) were kept constant for both the conventional and HME methods. So, the difference in particle size was due to the HME extrusion step that provided a shearing effect of the screws that caused additional particle size reduction and ultimately led to a lower particle size of the NLCs.

Figure 3.

Particle size, PDI, and zeta potential for the blank NLC formulations

Entrapment Efficiency and Drug Loading

Percent entrapment efficiency and drug loading results for F4-2 and F4-3 RX-HCl loaded NLCs, are as shown in Table 2. The high entrapment efficiency values (> 90 %) can be explained by the high solubilization capacity of the lipids and a relatively high surfactant concentration that leads to increased solubility of the drug in the lipid [32]. F4-3 NLCs generally had relatively higher entrapment efficiency values (97.7 % ± 0.1 for the conventional method) compared to F4-2 (96.1 % ± 0.1 for the conventional method) and this can be explained by the effect of increasing the liquid lipid concentration from 1.5 % w/v in F4-2 to 2 % w/v in F4-3. The high proportion of liquid lipid may help increase drug solubility in the lipid matrix which leads to high entrapment efficiency [33]. NLCs prepared by HME showed a relatively higher entrapment efficiency and this could be explained by the high shear generated by the screws inside the barrel which may cause more interaction of drug, lipids, and surfactant resulting in a homogeneous emulsion and an increase in entrapment efficiency [10]. However, based on statistical analysis, there was no significant difference (t = −2.922, p > 0.05) in percent entrapment efficiency values between NLC formulations prepared by the conventional method and those prepared by the HME method. There was also no significant difference (t = −4.250, p > 0.05) in percent drug loading values between NLC formulations prepared by the conventional method and those prepared by the HME method.

Table 2.

Percent entrapment efficiency and drug loading for the prepared NLC formulations

	Conventional Method		HME
	% EE	% Drug Loading	% EE	% Drug Loading
F4-2	96.10 ± 0.07	16.02 ± 0.01	99.10 ± 0.01	16.65 ± 0.01
F4-3	97.65 ± 0.03	16.28 ± 0.01	99.12 ± 0.01	16.67 ± 0.01

In Vitro Drug Release Studies

The release profile of RX-HCl from the RX-HCl loaded NLCs is as shown in Figures 4 and 5. It can be observed that the pure drug initially showed a faster rate of drug release in the first 8 hours compared to the RX-HCl loaded NLCs which showed a slower and more sustained release of RX-HCl. From the 8^th to the 30^th hour, the pure drug showed a much slower and almost constant rate of drug release, while the RX-HCl loaded NLCs continued to show an increased and sustained release of RX-HCl. Overall, the RX-HCl loaded NLCs had a higher rate of drug release than the pure drug. These results indicate that lipid carriers play a predominant role in sustaining the drug release. The sustained release of the drug from the NLCs may be explained by the increased diffusional distance and hindering effects of the surrounding solid lipid shell. Lipid nanoparticles (e.g., SLNs and NLCs) with a particle size in the submicron range (50 to 1000nm) have been reported to offer a prolonged release of the encapsulated drug and rapid uptake by cells due to their smaller particle size and larger surface area. These lipid nanoparticles have also been reported to have the potential to overcome the solubility and bioavailability constraints and promote oral absorption of poorly water-soluble lipophilic drugs like raloxifene hydrochloride [6], [7], [34]. NLCs prepared by HME showed a higher dissolution rate than those prepared by the conventional method, and this could have resulted from the further reduction of particle size and increase in surface area of the HME prepared NLCs. F4-3 generally showed a higher dissolution rate than the F4-2 and this may be explained by the decrease of the solid lipid amount and increase in liquid lipid amount in F4-3 that could cause a reduction of formulation viscosity and hence lead to an increase in dissolution rate. Figure 5 shows the release profile of F4-3 NLCs at the end of the stability study. The release profile for F4-3 NLCs and pure drug at the beginning and end of the stability study were comparable, although the dissolution rate was slightly lower at the end of the stability study. These results indicate good physical stability of the prepared NLC formulations. The two common release kinetics models i.e., zero order model and Higuchi model, were fitted to determine the release pattern of the optimized NLC formulations at the beginning of the stability studies and F4-3 NLC formulation at the end of the stability studies. The release kinetics of the optimized formulations calculated by regression analysis (R² value) had higher linearity for zero order and Higuchi model. The R² values from the zero order modeling at the beginning of the stability studies were as follows; F4-2 (0.9734) and F4-3 (0.9798) for the conventional method, and F4-2 (0.9743) and F4-3 (0.9696) for the HME method. The R² values from Higuchi modeling at the beginning of the stability studies were as follows; F4-2 (0.9689) and F4-3 (0.9416) for the conventional method, and F4-2 (0.9719) and F4-3 (0.9624) for the HME method. The R² values from the zero order modeling for F4-3 NLC formulation at the end of the stability studies were 0.9749 and 0.9702 for the conventional and HME methods, respectively. The R² values from Higuchi modeling for F4-3 NLC formulation at the end of the stability studies were 0.9393 and 0.9602 for the conventional and HME methods, respectively. Therefore, it can be concluded that the optimized formulations follow zero order kinetics with diffusion controlled release mechanism.

Figure 4.

In vitro drug release profile of prepared RX-HCl loaded NLC formulations

Figure 5.

Post stability in vitro drug release profile of F4-3 NLC formulations

Stability Study

Stability results are as shown in Table 3. Particle size, PDI, and zeta potential are key factors for evaluating the stability of colloidal dispersions [29], [30]. There was a statistically significant difference (p < 0.05) in particle size for NLCs prepared by both conventional and HME methods after one month of storage at 25 °C and 40 °C. Despite these differences, the particle size of NLCs were kept within a desirable submicron range (less than 300 nm), except for F4-3 (483.9 nm) that was prepared by the conventional probe sonication method. This could be explained by the low negative zeta potential value (−17.9 ± 2.6) of F4-3 on day 30 that could cause flocculation of the nanoparticles. All PDI and zeta potential values were within an acceptable range of ≤ 0.5 and −15 to −30 mV, respectively, and thus not considered to be significant in the present study. These results further indicated good physical stability of the prepared NLC formulations.

Table 3.

Particle size, PDI, and zeta potential of RX-HCl loaded NLCs

			Conventional Method		HME Technique
			F4-2	F4-3	F4-2	F4-3
Particle Size (nm)	Day 0		327.4 ± 9.79	284.2 ± 45.08	229.6 ± 6.34	199.5 ± 1.23
	Day 30	25 °C	336.9 ± 3.26	292.8 ± 24.70	259.4 ± 11.25	219.2 ± 10.25
		40 °C	483.9 ± 8.85	293.7 ± 11.84	276.6 ± 8.53	247.3 ± 3.80
PDI	Day 0		0.426 ± 0.03	0.432 ± 0.03	0.347 ± 0.08	0.376 ± 0.04
	Day 30	25 °C	0.513 ± 0.04	0.417 ± 0.02	0.434 ± 0.03	0.412 ± 0.06
		40 °C	0.523 ± 0.05	0.389 ± 0.01	0.464 ± 0.03	0.448 ± 0.04
Zeta Potential (mV)	Day 0		−18.1 ± 1.88	−25.2 ± 0.23	−17.9 ± 0.55	−22.7 ± 0.66
	Day 30	25 °C	−19.1 ± 0.21	−24.1 ± 1.37	−17.5 ± 1.91	−25.3 ± 0.70
		40 °C	−17.9 ± 2.63	−20.5 ± 1.21	−26.6 ± 1.29	−20.5 ± 1.21

CONCLUSION

HME technology coupled with probe sonication was successfully utilized to prepare RX-HCl loaded NLC formulations as a continuous manufacturing process with shorter processing times as compared to the conventional probe sonication method. HME coupled with probe sonication is a process that is able to overcome challenges in the variability of end-product quality attributes that are common with the conventional batch process. Results from this study not only suggest an efficient method that can be used to prepare RX-HCl loaded NLCs, but also leads to a product with a better drug release profile and good physical stability, and hence could provide a paradigm shift and better regimen in the prevention and treatment of osteoporosis and invasive breast cancer in postmenopausal women.