**1. Introduction**

Alzheimer's disease is an irreversible neurodegenerative disorder characterized by neuronal deterioration that leads to the loss of cognitive functions [1–3]. Genetic, environmental, and aging factors; the presence of neurofibrillary tangles; and senile plaques in the brain caused by the agglomeration of poorly folded proteins have been highlighted as the main factors involved in Alzheimer's pathogenesis [1,2,4]. Drugs used in clinical practice can attenuate the disease symptoms, inhibiting acetylcholinesterase activity and avoiding acetylcholine hydrolysis in the synaptic cleft [5,6]. Examples of these drugs include galantamine, donepezil, and rivastigmine [1,7]. Among these, rivastigmine hydrogen tartrate, chemically known as (S)-N-ethyl-N-methyl-3-[(1-dimethylamino) ethyl]-phenyl carbamate hydrogen tartrate, is the most used as a reversible non-competitive dual inhibitor of acetylcholinesterase and butyrylcholinesterase, improving central cholinergic function through the increase of acetylcholine levels [7–9]. Nonetheless, it was reported that rivastigmine hydrogen tartrate undergoes an extensive first-pass effect in the liver, which decreases bioavailability [10,11]. This molecule also has a short half-life and a hydrophilic nature, which makes it difficult for it to pass through the blood brain barrier (BBB) and cerebrospinal fluid (CSF) [5,9,12]. In addition, the tight junctions between the BBB capillary endothelial cells restrict the passage, absorption, and permeation of drugs to the brain [13,14]. Therefore, high drug concentration and frequent dose administration are required to reach therapeutic levels, causing unpleasant cholinergic side effects, such as nausea, dyspepsia, bradycardia, and hallucinations [8,10,11].

More effective ways of delivering rivastigmine to the brain are required, such as the use of nanosystems administered through alternative administration routes [11,13,15–18]. Herein, the intranasal route has been considered for delivering drugs from the nose directly to the brain, avoiding the need to overcome the BBB [13,19,20]. The nasal cavity directly contacts the central nervous system (CNS) through the olfactory and trigeminal nerves that connect to the brain and the CSF, allowing direct drug transport [13,20–22]. The nasal route offers other advantages to improve drug delivery, including the avoidance of the first-pass effect and gastric degradation, high drug absorption, and reduction of adverse effects. However, this route shows some limitations, such as fast drug elimination by the mucociliary clearance mechanism, among others [11,19,23–25]. Notwithstanding, the composition of the nasal formulation is crucial to obtain high therapeutic efficiency, being influenced by excipients, the physical state of the dosage form, and the applied volume [23,24].

Regarding nanosystems, several studies have showed that they promote nasal delivery, providing sustained drug release while avoiding molecules degradation due to the protective shell [11,23]. In this area, lipid nanoparticles (solid lipid nanoparticles—SLN; nanostructured lipid carriers—NLC) have shown high potential as carriers for nose-to-brain drug delivery [13,17,18,26–28]. SLN and NLC seem more advantageous than other nanosystems for brain delivery, as they are made of physiological lipids and are generally recognized as safe (GRAS) excipients that are biocompatible and biodegradable [13,26,29,30]. Furthermore, they provide drug protection against enzymatic degradation and increase the residence

time in the nasal cavity, improving drug bioavailability [14,17,25]. Other advantages include the ease of production on a large scale without the use of organic solvents, the high encapsulation efficiency for lipophilic molecules, and having a controlled release profile [13,15,22,30–32]. Besides, it is possible to produce SLN and NLC with diameters below 200 nm and a polydispersity index (PDI) of around 0.3, which are recommended for nose-to-brain delivery [13,31,33,34].

Although the clinical use of nanosystems has been intensively studied, some specific regulatory requirements are lacking [35]. In this sense, the use of the quality by design (QbD) approach to optimize lipid nanoparticles is essential to design formulations with low risk of failure and to achieve the desired clinical attributes. Thereby, carrying out preliminary studies to ensure the quality of the final product is required to achieve high efficiency, stability, and reproducibility. Some of these studies have explored the definition of the desired administration route and drug release profile, followed by the evaluation of the formulation properties and the control of the variables of the production method [36,37].

The Food and Drug Administration (FDA) and European Medicines Agency (EMA) authorities have encouraged the use of the QbD approach as a continuous process that should be applied to the development of a new pharmaceutical product, defining the quality target product profile (QTPP) to obtain a final product with high quality, safety, and efficiency [38,39]. QbD starts with the selection of the critical process parameters (CPPs) and critical material attributes (CMAs) that interfere with the critical quality attributes (CQAs), which are based on risk management [38]. For the implementation and continuous improvement of the QbD approach, several quality tools described in the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Q8, Q9, and Q10 guidelines are used, namely the Ishikawa diagram, Pareto chart, response surface methodology, and design of experiment (DoE) tools [40,41]. These tools are fundamental in the optimization of formulations, reducing the number of required experiments and consequently saving time and costs [11,32].

The aim of this work was to use the QbD approach to optimize a rivastigmine-loaded NLC formulation for nose-to-brain delivery with the predefined QTPP for the particle size (<200 nm), PDI (<0.3), zeta potential (ZP) (close to ±30 mV), and encapsulation efficiency (EE) (>80%) [31,33,34,42]. To carry out a complete and accurate optimization of the formulation, the study was divided into two parts. First, the most suitable CMAs, which corresponded to the concentrations of the different formulation components (lipids and surfactants), were defined. Afterwards, the CPPs were selected, which corresponded to the production method (high-pressure homogenization—HPH; or ultrasound technique) to produce rivastigmine-loaded NLC formulations with the desired QTPP. A central composite design was used to optimize the CMAs to achieve high quality predictions for various factors at extreme levels [32,43,44], and a Box–Behnken design was used to optimize the CPPs, analyzing the effects of three variables and requiring less experiments [45–47]. Finally, the pH and osmolarity of the optimized NLC formulations were adjusted to the physiological values and in vitro release studies were performed. Rivastigmine quantification was assessed by a high-performance liquid chromatography (HPLC) method validated according to the European Pharmacopeia (Ph. Eur.) and ICH guidelines [48,49]. In addition, the long-term stability of the optimized rivastigmine-loaded NLC formulations was assessed by measuring the particle size, PDI, ZP, and EE values over 90 days of storage at 20.0 ± 0.5 ◦C and 4.0 ± 0.5 ◦C.
