Development of pharmaceutically scalable inhaled anti-cancer nanotherapy – Repurposing amodiaquine for non-small cell lung cancer (NSCLC)

Abstract

New drug and dosage form development faces significant challenges, especially in oncology, due to longer development cycle and associated scale-up complexities. Repurposing of existing drugs with potential anticancer activity into new therapeutic regimens provides a feasible alternative. In this project, amodiaquine (AQ), an anti-malarial drug, has been explored for its anti-cancer efficacy through formulating inhalable nanoparticulate systems using high-pressure homogenization (HPH) with scale-up feasibility and high reproducibility. A 3² multifactorial design was employed to better understand critical processes (probe homogenization speed while formulating coarse emulsion) and formulation parameters (concentration of cationic polymer in external aqueous phase) so as to ensure product quality with improved anticancer efficacy in non-small cell lung cancer (NSCLC). Optimized AQ loaded nanoparticles (AQ NP) were evaluated for physicochemical properties, stability profile, in-vitro aerosol deposition behavior, cytotoxic potential against NSCLC cells in-vitro and in 3D simulated tumor spheroid model. The highest probe homogenization speed (25,000 rpm) resulted in lower particle size. Incorporation of cationic polymer, polyethylenimine (0.5% w/v) resulted in high drug loading efficiencies at optimal drug quantity of 5 mg. Formulated nanoparticles (liquid state) exhibited an aerodynamic diameter of 4.7 ± 0.1 μm and fine particle fraction of 81.0 ± 9.1%, indicating drug deposition in the respirable airways. Cytotoxicity studies in different NSCLC cell lines revealed significant reduction in IC₅₀ values with AQ-loaded nanoparticles compared to plain drug, along with significant cell migration inhibition (scratch assay) and reduced % colony growth (clonogenic assay) in A549 cells with AQ NP. Moreover, 3D simulated spheroid studies revealed efficacy of nanoparticles in penetration to tumor core, and growth inhibition. AQ’s autophagy inhibition ability significantly increased (increased LC3B-II levels) with nanoparticle encapsulation, along with moderate improvement in apoptosis induction (Caspase-3 levels). No impact was observed on HUVEC angiogenesis suggesting alternative anticancer mechanisms. To conclude, amodiaquine can be a promising candidate for repurposing to treat NSCLC while delivering inhalable nanoparticles developed using a scalable HPH process. Despite the involvement of complex parameters, application of DoE has simplified the process of product and process optimization.

1. Introduction

Nanoparticles (NPs) have attained considerable attention in delivering therapeutics due to their unique characteristics related to size, surface, payload capacity and flexibility in surface modifications to exert targeting potential [1]. Available marketed products of NP systems include Cimzia^®, Eligard^®, Pegasys^®, Renagel^® etc. [2]. NPs have demonstrated reproducible results on small laboratory scale production batches, however, a major hindrance exists in the translation from laboratory scale to commercial production due to manufacturability challenges [3,4]. Manufacturing challenges such as feasibility of scale-up, product quality and process repeatability are to be considered during the development of NPs [5]. Moreover, lack of sufficient information and knowledge on the scale-up technologies makes it difficult to introduce NPs into the market at the same pace of their development [1,5]. Further, scale-up is known to affect the characteristics of NPs and necessitates further optimizations at industrial scale. A better strategy would be to develop NPs using commercially scalable approaches at the laboratory scale, and researchers should pay more attention to formulation scale-up in addition to the development of complex delivery systems [5]. Among several existing methods of NPs production, high pressure homogenization (HPH) has gained much importance due to its scale-up feasibility and efficiency in providing reproducible results [6]. HPH allows for seamless translation from academic/small scale research labs to commercial scale production in cGMP facilities, and facilitates manufacturing of various formulations such as emulsions, suspensions and NPs in larger quantities [4,6,7]. During the HPH process, formulations are subjected to both mechanical forces such as cavitation, impact, shear, turbulence and high pressure [8] which aid in accomplishing NPs with optimal characteristics such as minimal particle size and maximum drug loadings as reported by Hoyer et al [6]. Shi et al have reported that optimal homogenization pressure and number of homogenization cycles have resulted in small sized NPs with uniform size distribution [9]. In another study, high HPH was used to achieve NPs of minimal particle size and good stability [10].

While effective, there are multiple variable parameters in HPH process, which need to be optimized and fine-tuned to reproducibly develop a stable and effective delivery system. Due to the involvement of several critical processes and formulation parameters, it is imperative to utilize statistical design of experiment (DoE) approach to assist in optimizing formulation while improving the knowledge of several parameters [4]. For instance, Petersen and Steckel have investigated various parameters such as homogenization pressure, pre-emulsion quality and formulation variables using DoE to identify the critical parameters and developed an optimized emulsion [11]. The importance of DoE can be understood from several recent works where application of systematic DoE approach has assisted in developing solid lipid nanoparticles of Efavirenz, poorly soluble drug [4] and inhalable microparticles of highly lipophilic drug [12].

One therapeutic area where particle characteristics play a vital role in particle accumulation and performance, is localized delivery to the lungs for respiratory diseases. Accumulation into the deep lungs is imperative for treatment of majority of respiratory diseases including pulmonary hypertension, pulmonary fibrosis, and non-small cell lung cancer (NSCLC) [13,14]. Accumulation of drug-loaded nanocarriers in the lungs after pulmonary delivery has been studied by several researchers, and improved efficacy of inhalable carriers in NSCLC has been reported earlier [15]. For instance, gemcitabine-loaded gelatin nanocarriers [16], paclitaxel-loaded nanocomposite microparticles [17], quinacrine-loaded bovine serum albumin modified cationic nanoparticles [15], pirfenidone-loaded liposomes [18] and lactoferrin/chondroitin-functionalized monoolein nanocomposites [19] etc. have been reported for their capability as inhaled nanocarriers in providing localized drug delivery in NSCLC treatment. In addition, Abdelaziz et al [13], Mangal et al [20], Ahmed et al [21] and Anderson et al [22] have also highlighted the significance of nanocarriers in pulmonary delivery along with details of further investigations being carried out. Particle size, surface morphology and mass median aerodynamic diameter (MMAD) of inhaled therapeutics determines their fate and the site of deposition in deep lungs [23]. MMAD within the range of 2–5 μm is essential for deep lung accumulation of particulate therapeutics, which is otherwise difficult to maintain due to inherent differences in particles’ physicochemical properties. Variability in physicochemical properties brings immense variability in therapeutic performance, thus making scalability a major challenge [24,25].

In this study, we aim to investigate the utility of DoE based nanoparticle development using HPH and test the efficacy of optimized formulations in a relevant respiratory disease model. Here, we chose amodiaquine, 4-[(7-chloroquinolin-4-yl) amino]-2- [(diethylamino) methyl] phenol (AQ), an FDA approved anti-malarial compound (Basoquin), to formulate biodegradable polymer (PLGA)-based NPs using HPH and to evaluate their efficacy against NSCLC. While AQ has been approved as an anti-malarial therapeutic, it is also a potent autophagy inhibitor with an ability to induce apoptosis in-vitro [26,27]. Qiao et al for the first time reported AQ’s ability to cause autophagic-lysosomal and proliferative blockade in melanoma cells [26]. Another recent report has also highlighted AQ for its drug repurposing efficiency through a structure-based screening approach [28]. Non-small cell lung cancer (NSCLC) is also characterized by autophagy induction and deficits in apoptotic mechanisms for its pathogenesis [29], hence targeting these processes using AQ is of considerable interest. AQ makes an interesting model drug as (i) its long-term safety is well documented being a FDA approved molecule; (ii) there is no AQ nanoparticulate delivery system developed so far thus making it feasible to understand the effects of its physicochemical properties on formulation development and characterization. As discussed earlier, NSCLC is a disease of the deep lungs, and inhalation therapy is gaining importance in recent years due to successful local delivery of therapeutics to cancer cells in treating NSCLC, where NPs act as potential carriers [13].

In this project, a 3² multifactorial design has been applied for better understanding of critical processes and formulation factors to ensure product quality with accepted Quality Target Product Profile (QTPP) for improved anticancer efficacy. The objective of this project is to formulate scalable, inhalable AQ-loaded PLGA NPs using HPH approach and to explore their efficacy in NSCLC.

2. Materials and methods

2.1. Materials

Poly (lactic-co-glycolic) (PLGA 50:50, 25–35 kDa, acid terminated), polyethyleneimine (PEI), poly vinyl alcohol (PVA), and amodiaquine (AQ) were purchased from Sigma-Aldrich (St. Louis, MO, USA). A549, H4006, H358, H2122, H460 and H157 NSCLC cell lines were obtained from ATCC and maintained in RPMI medium (Corning) supplemented with 10% FBS (Atlanta Biologicals), sodium pyruvate, penicillin-streptomycin (Corning) at 5% CO₂/37 °C. Human embryonic kidney cell line (HEK-293) was obtained from ATCC and maintained in DMEM medium (Corning) supplemented with 10% FBS (Atlanta Biologicals) and penicillin-streptomycin (Corning, NY, USA) at 5% CO₂/37 °C. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), dichloromethane (DCM), dimethyl sulfoxide, coumarin-6, crystal violet dye, 16% paraformaldehyde (PFA) solution, HPLC grade methanol, acetonitrile (ACN) and water were purchased from Fisher Scientific (Hampton, NH, USA). Molecular biology kits, supplies, and antibodies were purchased from other commercial vendors which are listed at appropriate places throughout the manuscript.

2.2. UPLC method development for amodiaquine (AQ)

A reverse-phase liquid chromatography method was developed for quantifying AQ using Waters series Acquity UPLC (Waters, USA). The column used was Xselect^™ HSS T3 (3.0 × 100 mm; 2.5 μm particles). The mobile phase consisted of aqueous phase of 0.1% orthophosphoric acid in water, and organic phase of ACN at 50:50 at a flow rate of 0.5 ml/min at 343 nm wavelength. Retention time was found to be 0.731 min with total run time of 1.5 min. Data were collected and analyzed using the Empower 3.0 software.

2.3. Optimization of process and formulation variables using design of experiment (DoE)

In this study, design of experiment (DoE) was used to develop nanoparticle formulations through which all potential factors were evaluated simultaneously, systematically, and quickly; using MINITAB statistical software (Minitab release 17, State College, PA, USA). Critical parameters such as probe homogenization speed and formulation variables involved in the formulation of nanoparticles were identified and were studied for their effect on drug loading which has been identified as a critical attribute while screening through DoE approach. The aim was to establish Quality Target Product Profile (QTPP) for drug-loaded nanoparticles with optimal characteristics of (i) the highest drug loading (> 3%), (ii) lower particle size (< 250 nm), along with (iii) uniform size distribution (polydispersity index of < 0.1). These specifications were established according to the intended route of administration i.e., inhalation route to ensure product safety and efficacy. A 3² randomized full factorial design, in which two factors (PEI concentration and homogenization speed) were studied at three levels (i) PEI concentrations – 0, 0.5 and 1% (w/v) (ii) homogenization speed – 6000, 12,500 and 25,000 rpm. A design-space was estimated from these two factors (probe homogenization speed and PEI concentration), with a risk of failure limit of ± 30 mV zeta potential and 600 nm particle size. The levels were chosen based on the preliminary studies performed to formulate nanoparticles using high pressure homogenization. Homogenization pressure of 30,000 psi, reverse flow pattern, and number of cycles of 7 to circulate the formulation was kept constant throughout the study. Reverse flow pattern refers to the configuration where inlet and outlet fluids flow in opposite direction resulting in maximum shear, cavitation and impact. This results in products of desired quality due to the efficient emulsification during each pass through the emulsifying cell [30]. Experimental trials were performed at all nine possible combinations; coded as A1–A9. From this set of experiments, an optimized formulation with the highest % drug loading (> 3%) and lower particle size (< 250 nm) will be identified to establish the optimal parameters for further studies. Experimental trials of factorial design containing different combinations are shown in Table 1. The effect of different variables on three different responses [i.e. particle size (PS), zeta potential (ZP) and drug loading (%DL)] was investigated.

Table 1.

Composition and characteristics of amodiaquine loaded nanoparticles in a 3² full factorial design.

Formulation	Coded value		Drug quantity	Polymer	EAP	Probe homogenization time	Particle size (nm)	PDI	Zeta potential (mV)	% Drug loading
	X1	X2
A1	−1	−1	2.5 mg	PLGA 502H	PVA 1% w/v	10 min	179.1	0.287	−19.9	1.06
A2	−1	0					198	0.224	−15	1.31
A3	−1	+1					192.4	0.222	−6.75	1.20
A4	0	−1					323.3	0.123	17	1.59
A5	0	0					478.8	0.333	10.9	4.40
A6	0	+1					218	0.31	8	3.35
A7	+1	−1					204.8	0.18	11.1	3.35
A8	+1	0					463.3	0.476	1.23	2.53
A9	+1	+1					579	0.253	13.4	3.09
Coded values	Actual values		X1- Polyethyleneimine concentration (%w/v) in EAP, X2-Probe Homogenization speed (RPM), EAP-External Aqueous Phase, PDI-Poly Dispersity Index
	X1	X2
−1	0	6000
0	0.5	12,500
+ 1	1	25,000

Optimized formulation, A6, with respect to predetermined particle size, zeta potential and entrapment efficiency, was subjected to further investigation to establish the optimal drug quantity to achieve QTPP as described earlier. Effect of different drug quantities (1, 2.5, 5 and 10 mg) on particle size, PDI, % entrapment efficiency (%EE) and % drug loading (%DL) is shown in Table 2. Formulations were coded as C1–C4. Other variables (including; formulation and operation parameters) were kept constant throughout the study. The association between dependent variables and responses was interpreted using response surface plots.

Table 2.

Characterization of different formulations with different drug quantities.

Formulation	Drug quantity	Polymer	EAP	Probe homogenization time	Particle size (nm)	PDI	Zeta potential (mV)	% entrapment efficiency	% drug loading
C1	1 mg	PLGA 502H	PVA 1% w/v	10 min	222.9	0.195	11.3	35.33449958	0.58
C2	2.5 mg				211.5	0.139	10.1	65.67726887	2.63
C3	5 mg				229.4	0.058	17.8	52.31175785	4.02
C4	10 mg				229.9	0.156	14.3	39.84788486	5.69

In brief, AQ NPs were formulated through high-pressure homogenization by modifying a method reported by Dong and Feng [31] with slight modifications. Briefly, coarse emulsion was prepared by homogenization of aqueous solution of AQ at 5 mg/ml (Set I) or varying concentrations such as 2,5,10 and 20 mg/ml (C1-C4) respectively and organic solution of PLGA RG-502H (20 mg/ml of DCM) in aqueous 1% w/v PVA solution in 1× phosphate buffered saline (PBS; pH 7.4) with or without PEI at respective concentrations. Coarse emulsion was then processed through high-pressure homogenizer (Nano DeBee; BEE International, South Easton, MA, USA) at a pressure of 30,000 psi for 7 cycles. The obtained emulsion was subjected for overnight stirring to remove organic solvent. Next day, nanoparticles were washed twice using 1× PBS by centrifugation at 21,000 RCF for 15 min, to remove excess PVA and unprocessed drug/polymer. Due to less solubility of AQ in PBS, it was optimized as the dilution medium for PVA to reduce drug leaching during washing process.

2.4. Characterization of AQ NPs

2.4.1. Particle size, poly dispersity index (PDI) and zeta potential

Particle size, PDI and zeta potential were measured by dynamic light scattering (DLS) using Zeta Sizer (Malvern Instruments, Malvern, UK). Formulations were diluted 100-fold with milli-Q water to obtain samples for measurement of size and zeta potential. The respective samples were loaded into the zeta cells (DTS 1070) and analyzed.

2.4.2. Drug content

To determine encapsulated drug amount in nanoparticle formulation, a direct vesicle lysis approach was carried out. To 20 μl of formulation, 1980 μl of ACN:Water:DCM – 98:1.5:0.5 was added followed by centrifugation for 45 min at 4 °C at 21,000 RCF to lyse nanoparticles and to obtain loaded drug into the analyzing solution. Clear supernatant was collected, analyzed for the drug content using UPLC as described earlier and % EE/%DL was calculated.

$$\mathit{\%Entrapment}\mathit{efficiency}(\mathrm{\%}EE)=\left[\frac{\mathit{Drug}\mathit{entrapped}\mathit{in}\mathit{nanoparticles}}{\mathit{Total}\mathit{drug}\mathit{added}\mathit{initially}}\right]\times 100$$

$$\mathrm{\%}\mathit{Drug}\mathit{loading}(\mathrm{\%}DL)=\left[\frac{\mathit{Drug}\mathit{entrapped}\mathit{in}\mathit{nanoparticles}}{\mathit{Total}\mathit{polymer}+\mathit{drug}\mathit{added}}\right]\times 100$$

2.4.3. In-vitro release studies

In-vitro release studies were performed for determining the drug release pattern from optimized AQ NPs using dialysis method as reported previously [32]. Briefly, 10,000 MWCO dialysis cassettes (Slide-A-Lyzer, 10,000 MWCO, 0.1–0.5 mL, Thermo-Scientific, Waltham, MA) were preconditioned in PBS buffer (pH 7.4) for 10 min to allow them to get hydrated. Then, using syringe with 19G1_1/2 TW BD filter needle, 500 μl of formulation was loaded into the cassettes membrane though one of the ports. The cassettes were immersed in a beaker with 150 ml of PBS buffer with 1% polysorbate 80 (pH 7.4; 37 °C) using a floater while stirring at a speed of 200 rpm. At regular time points of 0.5, 1, 2, 4, 8, 12 and 24 h, samples were withdrawn with replenishing the dissolution medium with fresh PBS with 1% polysorbate 80. The amount of drug released was estimated using UPLC as discussed above.

2.4.4. Morphological analysis using scanning transmission electron microscopy (STEM)

Morphology of nanoparticles was studied by Scanning Transmission Electron Microscope (STEM) using FEI Talos F200x Transmission/Scanning Transmission Electron Microscope (S/TEM) equipped with A-Twin objective lens operating in STEM mode. The samples were prepared by placing 10 μl of the sample solution carefully on a lacey carbon-coated copper grids (300 mesh, Ted Pella Inc., Redding, CA, USA). The sample was allowed to adhere to the grid by subjecting it to air drying. The grids were then imaged using Smartcam digital search and view camera.

2.4.5. Solid state characterization studies

2.4.5.1. Powder X-ray diffraction (PXRD) studies.

X-ray diffraction spectroscopy was carried out using XRD-6000 (Shimadzu, Kyoto, Japan). The diffractometry was performed by using a graphite monochromator consisting of copper-Kα1 radiation of wavelength 1.5418 Å operating at 40 kV, 30 mA. The samples were spread uniformly on a glass micro sample holder and were analyzed in the range of 5 to 60° at the scanning speed of 2° (2θ)/minute.

2.4.5.2. Differential scanning calorimetry (DSC) studies.

The thermograms for AQ, AQ NP, blank NP and physical mixture of AQ and blank NP were generated using a DSC 6000 (PerkinElmer, Inc.; CT, USA) equipped with an intra-cooler accessory. An accurately weighed sample (1–5 mg) was sealed in an aluminum pan and analyzed over a temperature range of 30 °C to 210 °C and compared to a sealed empty aluminum pan maintained as reference. The heating rate was maintained at 10 °C/min under a nitrogen purge having flow rate of 50 ml/min.

2.4.6. In-vitro aerosol performance lung deposition test

In-vitro lung deposition behavior of AQ NPs was evaluated using a M170 Next Generation Impactor^™ (NGI: MSP Corporation Shoreview, MN, USA) in accordance with earlier published studies [33]. The NGI was equipped with a stainless-steel induction port (USP throat adaptor) attachment and specialized stainless steel NGI^™ gravimetric insert cups (NGI Model 170, MSP Corporation, Shoreview, MN, USA). AQ NP formulation (2 ml) was placed into a PARI LC PLUS^® nebulizer cup of Pari FAST-NEB compressor system. (Boehringer Ingelheim Pharmaceuticals, Inc. Ridgefield, CT, USA), which was attached to a customized rubber mouth piece connected to the NGI^™. A Copley HCP5 vacuum pump (Copley Scientific, UK) was used to produce a flow rate of 15 L/min. Using a Copley DFM 2000 flow meter (Copley Scientific, UK), flow rate was adjusted before each experiment. In brief, 2 ml of formulation was nebulized with PARI LC PLUS^® nebulizer which passed through induction port into the NGI using pump at a flow rate of 15 L/min for 4 min. Prior to running the NGI, the plates were refrigerated at 4 °C for 90 min to cool the NGI plates. Samples were collected from each stage i.e. Stages 1–8 including throat and induction port as well which is important in determining emitted dose through rinsing with ACN:water (75:25) and analyzed with established UPLC method for drug content and effective deposition at each stage. All experiments were performed in triplicate (n = 3). Fine particle fraction (FPF, %) was determined as the fraction of emitted dose deposited in the NGI with d_ae < 5.39 μm. Mass median aerodynamic diameter (MMAD, d_ae < 5.00 μm) and geometric standard deviation (GSD) are the critical parameters for inhalation testing, and were calculated by quantifying the nanoparticle deposition at each stage in the NGI using log probability analysis (n = 3) [33,34].

2.5. Stability studies

Stability of the optimized AQ NP (C3) was evaluated while storing the samples at temperatures of 4 °C, 25 °C and 37 °C for four weeks as reported previously [32]. Samples were withdrawn after week 1, 2, 3 and 4; diluted with water (100-fold) and analyzed for particle size, PDI and zeta potential using Malvern Zeta Sizer. Entrapment efficiency was determined by lysing the samples as described above using UPLC. All experiments were performed in triplicate.

2.6. Cellular uptake studies

Cellular uptake studies were performed using a protocol reported earlier [35]. NPs for cellular uptake were formulated by replacing AQ with coumarin-6, a fluorescent dye, for easier tracking. In brief, A549 cells were plated in tissue culture treated cell imaging 8 chambered cover glass (Eppendorf, Hauppauge, NY, USA) at a seeding density of 10,000 cells/chamber followed by overnight incubation. Next day, cells were incubated with coumarin-6, coumarin-6 loaded nanoparticles at 1 μg/ml concentration for 1 and 3 h. After each interval, cells were washed with ice cold PBS twice and fixed with 4% PFA for 10 min. Fixed cells were washed again with ice cold PBS twice. Then, the chamber was removed and 20 μl of vectashield hardset mount with DAPI nuclear stain (H1500, Vector laboratories, Burlingame, CA, UAS) was placed on a glass slide dropwise followed by placing a cover glass. After hardening of mounting medium through overnight at 4 °C, slide was imaged using EVOS-FL microscope (Thermo Scientific, Waltham, MA, USA).

2.7. Cytotoxicity studies

AQ NPs along with plain AQ were evaluated for their cytotoxicity in six different non-small cell lung cancer (NSCLC) cell lines: A549, H157, H358, H2122, H460 and H4006 as reported earlier with slight modifications [35,36]. Briefly, cells were grown in FBS supplemented RPMI-1640 media as described in Materials section, and were seeded in TC treated 96-well plates (Eppendorf, Hauppauge, NY, USA) at a seeding density of 2500 cells/well (7500 cells/cm²), incubated overnight for adherence at 37 °C/5% CO₂, and treatments were added next day at different AQ concentrations ranging from 0.39 – 100 μM. Corresponding volumes of AQ NPs were calculated based on the drug entrapment efficiency. Blank culture media was added as control. After 72 h of incubation, % cell viability was determined by performing MTT assay as described earlier [36], by reading the absorbance of dissolved formazan crystals at 570 nm (Tecan Spark 10 M; Tecan, Männedorf, Switzerland). AQ and AQ NP were also tested for their cytotoxicity efficacy against A549 cells with different incubation periods of 6 and 12 h. The safety of AQ NP formulations was determined by evaluating the cytotoxicity of Blank-NPs on human embryonic kidney (HEK-293) cell line. Briefly, cells were seeded into 96-well plates as described earlier and were incubated overnight followed by treatment with equivalent amounts of Blank-NP corresponding to 10 μM of AQ NP for 72 h. % cell viabilities were determined through MTT assay as described earlier.

2.8. Scratch assay

In-vitro scratch assay was used to study the cell migration. Briefly, scratches were created on a confluent cell monolayer. The cells on the edge of the scratch will migrate toward the center to close the scratch, thus establishing new cell-cell contacts. The assay was performed as previously reported [37]. Briefly, A549 or H460 cells were plated in 24 well cell culture plates at a seeding density of 1 × 10⁵ cells/well followed by overnight incubation. Next morning, scratches were made along the center of all wells with the help of a sterile p200 (200 μl) pipette tip. All the wells were washed with PBS twice and reference markings were drawn near the scratch area from the bottom side of the plate with a fine tip marker. Then scratch images were captured within the marked area using inverted microscope (Laxco, Inc., Mill Creek, WA, USA) with 10× magnification. Later treatments of control, AQ, AQ NP (10 μM) were added to the respective wells and incubated followed by imaging after 24 and 48 h (A549); 24 and 120 h (H460). The images captured were analyzed quantitatively to assess the inhibitory effect of nanoparticles on cell migration. Scratch width was measured from all wells using ImageJ software and % scratch closure was calculated further.

2.9. Clonogenic assay

Clonogenic assay is an in-vitro cell survival assay which is based on the single cell’s capability to grow into a colony. Using this assay, the effectiveness of the AQ and AQ NPs toward colony inhibition was determined. The protocol reported previously [35,38] was briefly modified and followed in this study. A549 or H4006 cells were seeded into 6-well cell culture plates at seeding density of 500 cells per well for each cell line. Plates were kept for overnight incubation to allow time for cells to adhere. Next day, media was replaced, and cells were treated with AQ or AQ NPs (10 μM), or control for 48 h after which media was replaced with fresh culture medium on alternative days for 7 days. On 7th day, the colonies were stained with crystal violet. Briefly, the contents of the well were removed and washed with ice-cold PBS buffer twice followed by fixation with 4% PFA solution (in ice-cold PBS) for 10 min. Fixed cells were again washed with ice-cold PBS twice followed by staining with 0.01% (w/v) crystal violet solution with 1-hour incubation. After staining cells were washed with distilled water and images were captured using digital camera. Cell colonies were counted by colony counter software Open CFU [39].

2.10. 3D spheroid study

An effective NSCLC therapy is not only determined by enhanced cellular uptake or cytotoxic potential, but also improved penetrability of nanoparticles into solid tumors. 3D spheroid cell culture studies are capable of mimicking the in-vivo features of tumors as reported in our previous studies [36]. Briefly, 2000 cells per well were seeded in Nunclon Sphera 96-well U bottom plates, and were allowed to grow into solid tumor mass due to the shape of these ultra-low attachment cell culture plates. This in-vitro spheroid study was conducted through two kinds of models i.e. prophylactic and therapeutic models, differing in tumor growth and treatment strategies.

In prophylactic study, 2000 cells (A549) were seeded into each well of Nunclon Sphera 96 well ultra-low attachment treated spheroid microplates (Thermo Fisher Scientific, Waltham, MA, USA) and incubated overnight under standard conditions of 37 °C/5% CO₂. Next day, media was replaced with either fresh media (control) or respective treatment solutions of AQ or AQ NP to achieve 10 and 25 μM in the wells. On day 1, all the wells were observed for spheroid formation with a rigid margin. Images were captured using inverted microscope (Laxco, Mill Creek, WA, USA) on day 3, 5, 7 10 and 12. NIH ImageJ software was used to measure diameter of all the spheroids. Spheroids were treated on respective days by replacing half of the medium (100 μl) from every well with respective treatments in a gentle manner to avoid the bubble formation and the aspiration of spheroid itself.

In therapeutic 3D model, A549 cells were seeded into Nunclon Sphera plate at density of 500 cells/well and were incubated at 37 °C/5% CO_2. Then, all wells were observed for spheroid growth and images were captured on day 1, 3, 5 and 7. On day 7, spheroids were subjected to two kinds of dosing treatments i.e. a single dose and a multiple dose treatment. Briefly, both single and multiple dosing spheroids were treated with 20 μM and 50 μM concentrations of AQ and AQ NP (to maintain original concentrations of 10 and 25 μM employed in the beginning), and images were captured. For single dosing, 100 μl of media was replaced with fresh media on further days of imaging. For multiple dosing, wells were replenished with 10 and 25 μM concentrations of AQ, AQ NP or fresh media (control). Images were captured on day 3, 7 and 10 and 15 days following treatment. NIH ImageJ software was used to measure diameter of all the spheroids.

2.11. Live-dead cell assay

Live-dead cell assay was performed on spheroids on day 15 of single and multiple dosing in therapeutic model according to manufacturer’s protocol. Briefly, 100 μl of 2 μM calcein AM/4 μM Ethidium homodimer III (EthD-III) staining solution was added to spheroids after complete removal of media from the respective wells. The plate was incubated for 45 min at room temperature. This provides green/red fluorescent staining of viable and dead cells, respectively. Images were captured using (EVOS-FL, Thermo Fisher Scientific); and green fluorescence intensity (GFP)/mm² of spheroid surface area was quantified using ImageJ software which signifies the presence of live cells in spheroid mass.

3. Mechanism of action

3.1. Autophagy inhibition microplate assay

Autophagy enables cancer cells to survive under micro-environmental stress conditions and promotes tumorigenesis [40]. While AQ is a known autophagy inhibitor, it is crucial to determine the capability of AQ NPs to inhibit autophagy. Autophagy inhibition microplate assay was performed using CYTO-ID^® Autophagy Detection Kit (Enzo Life Sciences, Farmingdale, NY, USA). In brief, A549 cells were plated at density of 2.5 × 10⁴ cells/well in 96 well plate (Fisher scientific, Hampton, NH, USA), and starved for 24 h in serum free media, and were later incubated with AQ 10 μM, AQ NP 10 μM or drug-free control for 18 h. Treatments were replaced with 100 μl of 1× assay buffer followed by addition of 100 μl of dual color detection reagent (CYTO-ID^® Green Detection Reagent + Hoechst 33342 nuclear stain in cell culture medium without phenol red indicator supplemented with 5% FBS). Plates were protected from light and incubated for 30 min at 37 °C. Later, cells were washed with 200 μl of 1× assay buffer to remove excess dye. Then 100 μl of 1× assay buffer was added to each well, and plates were read with a FITC filter (Ex ~480 nm, Em ~530; green detection reagent), and with a DAPI filter set (Ex ~340 nm, Em ~480 nm; Hoechst 33342 Nuclear Stain). Here, CYTO-ID^® Green Detection Reagent serves as a bright fluorescent probe in vesicles produced during autophagy [41].

3.2. Apoptosis: Caspase 3/7 assay

Apoptosis, or programmed cell death, plays a critical role in assessing the potency of drug delivery systems. Caspase-3 is one of the major markers of apoptotic cell death, which can be activated in apoptotic cells both by extrinsic and intrinsic pathways [42]. AQ and AQ NP were tested for their cytotoxicity efficacy against A549 cells with different incubation periods as well. Representative graph related to their cytotoxicity after 6 and 12 h can be found in Fig. S6. After 6 h of incubation, no significant difference was observed between AQ and AQ NP in their cytotoxicity potential. Hence, incubation period of 6 h and treatment concentration of 10 μM were chosen in performing the present assay. Caspase-3 activity was measured using EnzChek^™ Caspase-3 Assay Kit (Molecular Probes, Eugene, OR, USA) and assay was performed as per manufacturer’s specifications. Briefly, A549 cells were seeded at a density of 1 × 10⁶ cells per tissue culture dish (100 mm diameter) (Thermo Scientific, Rochester, NY, USA) and treated with AQ, AQ NP (10 μM) or control for 6 h followed by harvesting and washing of cell pellets. Cell lysis was carried out using 1 × cell lysis buffer while subjecting to freeze-thaw cycle, followed by centrifugation at 3000 RCF for 5 min. Supernatants obtained were transferred to 96 well plate to which 50 μl of 2× substrate working solution (10 mM Z-DEVD-AMC substrate +2× reaction buffer) was added and incubated for 20 min. Fluorescence was measured at excitation/emission 342/441 nm [43].

3.3. In-vitro angiogenesis study

Angiogenesis or formation of new blood vessels from preexisting vasculature is a key process in some physiological conditions including cancer pathogenesis. Angiogenesis have a pivotal role in pathogenesis of NSCLC among other cancer kinds in the form of over-proliferation of blood vessels [44]. To understand AQ loaded nanoparticles’ anti-angiogenic ability, angiogenesis assay was performed as per manufacturer’s instructions (3470–096-K; R&D Systems, Minneapolis, MN, USA). Briefly, human umbilical vein endothelial cells (HUVEC: LONZA, Basel, Switzerland) cells were seeded into a T-25 flask prior to performing angiogenesis assay. Next day, 50 μl of Cultrex^® RGF BME was aliquoted into each well of a 96 well plate and incubated at 37 °C for 60 min to allow the BME to gel. HUVEC cells were then labeled with 2 μM calcein AM; seeded at 1.3 × 10⁴ cells/well on BME coated 96-well plates after preparing dilutions of AQ and AQ NP (10 μM). Cells were incubated for 6 h at 5% CO₂/37 °C. Images were taken using fluorescence microscope (EVOS-FL, Thermo Fisher Scientific, Waltham, MA, USA) and tube formation was evaluated and quantified using NIH ImageJ software with the angiogenesis analyzer plug-in [45].

3.4. Statistical analysis

All data were addressed as mean ± SD or SEM, with n = 3 unless otherwise mentions. Three trials of cytotoxicity studies were performed for each control or treatment with n = 6 for each trial. Data Unpaired student’s t-test was used to compare two groups and one-way ANOVA followed by Tukey’s multiple comparisons test was used to compare more than two groups using GraphPad Prism software (Version 7.04 for Windows, GraphPad Software, San Diego, California USA). A p value of < 0.05 was considered statistically significant and was presented in data figures as a single asterisk (*). However, some studies have demonstrated a smaller p-value of 0.01 or less, which is included at respective places.

4. Results

4.1. UPLC method development for amodiaquine

A rapid, UPLC method was developed for determination of AQ. Retention time was found to be 0.731 min with run time of 1.5 min. Peak was eluted with good resolution. A representative chromatogram is presented in Fig. S1. The method provided excellent linearity between 0.05 and 6 μg/ml.

4.1.1. Optimization of process and formulation variables using design of experiments (DoE)

As presented in Table 1, adopted Minitab^® 18 software provided a numerical optimization technique to produce set of formulations with anticipated responses such as lower particle size, positive zeta potential with highest % drug loading. The developed model of experimental runs for three quality attributes supported the proposed design space. Accordingly, the acceptable region that fulfills those targeted features is achieved when 1% or < 1% (w/v) PEI is used in nanoparticle preparation. Results revealed that PEI concentration had great impact on particle size, zeta potential and % DL; each studied at three levels. As for particle size, it was found that particle size increased with increasing PEI concentration, at same homogenization speed, while positive zeta potential was observed only at PEI concentration of ≥0.5% w/v. % DL was found to be higher in case of 0.5% and 1% w/v PEI. Positive charge density of PEI enhances encapsulation of weakly basic drug molecules such as AQ [46]. Results showed that coarse homogenization speed could only affect particle size; while having negligible effect on zeta potential and drug entrapment. Coarse homogenization at 6000 rpm resulted in smaller particle size compared to 12,500 and 25,000 rpm; at all PEI concentrations. Largest particle size (579 nm) was observed in formula (A9); containing 1% PEI and homogenized at 25,000 rpm. At 0.5% PEI concentration, increase in speed from 6000 to 25,000 rpm rendered a concomitant increase in the breaking energy, resulting in smaller emulsion droplets [47]. Interactive influences of PEI concentrations of at all three levels (0, 0.5 and 1% w/v), and homogenization speed at the three levels (6000, 12,500 and 25,000 rpm) on particle size, zeta potential and encapsulation efficiency are represented in response surface plots presented in Fig. 1A–C (A. particle size, B. zeta potential and C. % drug loading). For future investigations on effect of initial drug amount on PS, ZP, %EE and %DL, an optimized formula was selected through relating constraints between independent variables and dependent responses. The criteria for optimized formulation was set to be demonstrating the highest % drug loading (> 3%) and lower particle size (< 250 nm). According to obtained results, formulation A6 (0.5% PEI, 25,000 rpm) was selected for further optimization. Even though, A7 had the highest entrapment and lower particle size, it was not chosen considering the presence of highest concentration of PEI, a cationic polymer, with potential of intracellular toxicity. While both A5 and A6 exhibited significantly higher % drug loading (> 3%), only A6 resulted in optimal particle size (218 nm). Hence, parameters pertaining to A6 were chosen for further experimental trials to optimize drug quantity.

Fig. 1.

3-Dimensional Surface response curves showing influence of homogenization speed and polyethyleneimine (PEI) concentration in EAP on (A) particle size (nm), (B) zeta potential (mV) and (C) % drug loading of amodiaquine-loaded nanoparticles. D. Effect of drug quantity on % drug loadings of AQ NP. Set II formulations were loaded with different quantities (1, 2.5, 5 and 10 mg) of amodiaquine.

Results describing effect of varying initially loaded drug amount on particle size, zeta potential and entrapment efficiency are demonstrated in Table 2. From results, it can be observed that increasing initial drug loading led to subsequent increase in %EE and %DL. The highest %DL was observed in formula C3 and C4, loaded with (5 and 10 mg) amodiaquine; 4.0% for C3 and 5.7% for C4 respectively (Fig. 1D). Formula (C3; further denoted as AQ NP) containing 5 mg amodiaquine was chosen to be the best achieved formulation, as it showed higher %EE (52.3%) compared to C4 (39.8%), thus indicating significantly less wastage of starting raw material, as seen in Table 2. Moreover, C3 was found to exhibit a particle size of 229.4 nm with a PDI of 0.058 i.e. an almost monodispersed nanoparticle formulation. These studies revealed that C3 (AQ-NP) was able to meet the specification set as per QTPP. Hence, further in-vitro studies were performed using C3 formulation which is denoted as AQ NP. s

4.2. Characterization of AQ NPs

4.2.1. Particle size, polydispersity index (PDI), zeta potential

Nanoparticle formulation (C3) was found to have a particle size, PDI and zeta potential of 250.33 ± 43.0 nm, 0.18 ± 0.1 and 16.96 ± 1.0 mV respectively. Formulations were found to have a positive zeta potential due to the presence of PEI. A histogram representing particle size distribution of C3 can be found in Fig. 2A representing uniform particle size distribution with single peak.

Fig. 2.

Physicochemical characterization of nanoparticles. A. Size and Polydispersity index (PDI) of nanoparticles were measured by intensity-size distribution histogram for the finalized AQ NP. B. Scanning Transmission Electron Microscopy (STEM) image of AQ NP reveal the spherical morphology of nanoparticles. Magnification 20kX. The scale bar represents 500 nm. C. In-vitro drug release profile from AQ NP in PBS (pH 7.4). Data represents mean ± SD (n = 3). D. XRD patterns of amodiaquine (AQ), amodiaquine loaded nanoparticles (AQ NP), physical mixture of amodiaquine (AQ) and blank nanoparticles (Blank NP) and Blank NP. The characteristic sharp crystalline peaks of amodiaquine are evident in the AQ and physical mixture samples. Unlike the physical mixture and amodiaquine, AQ NP did not illustrate any sharp peaks, which can be attributed to AQ encapsulation in nanoparticle core. E. Thermograms of amodiaquine (AQ), amodiaquine loaded nanoparticles (AQ NP), physical mixture of AQ and blank nanoparticles (Blank NP) and blank nanoparticles. Amodiaquine thermogram displays a distinct sharp endothermic peak at 166.8 °C in relation to the melting point of the molecule. The physical blend thermogram observed an endothermic peak at 166.8 °C indicating the characteristic peak for amodiaquine. There is no considerable difference between the thermograms of unloaded and AQ loaded nanoparticles. Absence of endotherm representing melting point of AQ allows us to suggest that drug is entrapped in nanoparticle core, also supported by the XRD data.

4.2.2. Drug content

As shown in Table 1, A6 was found to be an optimized formulation among A1–9 with lower particle size and highest % DL. It was found that C3 had the highest % EE of 52.3% along with high % DL (4.02%) as seen in Table 2 and Fig. 1D. C3 was chosen for further studies and denoted as AQ NP.

4.2.3. In-vitro release studies

In-vitro release studies were performed to determine the drug release pattern from AQ NP (C3) in phosphate buffered saline (PBS; pH 7.4, 37 °C) to mimic physiological fluid environment. The AQ NP (C3) was found to exhibit a controlled release pattern with about 80% (85.3 ± 6.4%) drug being released after 4 h as shown in Fig. 2B. This reveals the ability of nanoparticles to release drug completely in 4–8 h after reaching physiological environment.

4.2.4. STEM studies

A representative STEM image presented in Fig. 2C indicated that the nanoparticles had uniform spherical shape with a smooth surface. As anticipated based on zeta potential measurement, no aggregation of nanoparticles was observed during STEM analysis. This reveals the uniform dispersion of nanoparticles in the formulation. These results agree with DLS measurements.

4.2.5. Solid state characterization

4.2.5.1. Powder x-ray diffraction (PXRD).

Due to its crystalline nature, AQ showed distinct peaks at 2θ values of 19.88 and 25.88 in XRD spectra whereas there were no AQ peaks present in AQ NP indicating encapsulation of drug inside the nanoparticles as shown in Fig. 2D. Results were found to be consistent with earlier studies of artesunate-amodiaquine microparticles [48]. Physical mixture of Blank NP and AQ also exhibited peaks inferring the crystalline nature of drug.

4.2.5.2. Differential scanning calorimetry (DSC).

The DSC studies were performed to understand melting and crystallization behavior of AQ when present in different forms such as NP encapsulated AQ, physical mixture with blank NPs, as compared to control. As can be seen in Fig. 2E, thermogram of AQ showed a sharp endothermic peak at 166.8 °C due to its melting transition. The absence of a sharp peak in AQ NP indicated complete drug encapsulation in the nanoparticle formulation core and suggested a consistent result with that of XRD as discussed above. Results found to be consistent with earlier studies [49].

4.2.6. In-vitro aerosol performance and lung deposition

Once nebulized, aerodynamic properties of particles govern their deposition profile in the airways and alveolar deep lung regions. Using a next generation cascade impactor (NGI), crucial parameters to determine respirability of the nanoparticles were determined. Mass median aerodynamic diameter (MMAD), which describes the aerodynamic particle size distribution of an aerosol by mass, and geometric standard deviation (GSD), which describes the spread of the aerodynamic particle distribution, were determined. The MMAD of AQ NP was 4.7 ± 0.1 μm suggests that majority of the emitted dose will be delivered to respirable region of the lungs, while the GSD was 2.0 ± 0.4 μm. The % fine particle fraction (FPF), also called the respirable fraction, was 81.0 ± 9.1% as shown in Table 3 which suggests good aerosolization performance. Aerosol dispersion profile at each stage of NGI, and a graph representing % cumulative deposition respective to each stage can be seen from Fig. 3A & B respectively. Particle size distribution is one of the most important parameters in inhalational delivery which determines efficiency of the delivery system to deliver the particles deep enough to the alveolar region [50]. These data suggest that the prepared formulations possess all the characteristics to render them inhalable with deep lung deposition.

Table 3.

In vitro aerosol deposition profile of amodiaquine nanoparticles.

Formulation	MMAD (μm)	Recovery (%)	FPF (%)	GSD (μm)
C3	4.7 ± 0.1	72.8 ± 25.1	81.0 ± 9.1	2.0 ± 0.4

Fig. 3.

In-vitro deposition profile of AQ NP. A. Aerosol dispersion performance as % deposited on each stage of the Next Generation Impactor^™ (NGI^™) for amodiaquine nanoparticles. For Q = 15 L/min for 4 min. The effective cutoff diameters (D₅₀) for each impaction stage are as follows: stage 1 (14.1 μm), stage 2 (8.61 μm), stage 3 (5.39 μm), stage 4 (3.3 μm), stage 5 (2.08 μm), stage 6 (1.36 μm), and stage 7 (0.98 μm). After nebulization of AQ NP of 2 ml volume, each stage was washed with ACN:water (75:25) and the washings were analyzed by HPLC to determine the drug deposition. B. Cumulative % deposition plot representing cumulative % of particles deposited at each stage. (n = 3, Data represent mean ± SD).

4.3. Stability studies

Stability is a major concern while dealing with nanosized drug delivery systems due to emulsion instability and particle aggregation [32]. As shown in Fig. S2, stability analysis data reveal that AQ NP formulation was stable at 4⁰ and 25 °C, with no significant changes in particle size or zeta potential (Fig. S2A & B). Here, the positive zeta potential of the AQ NP prevents the particles from coalescing and aggregating. There was no detrimental effect of temperature and storage time on entrapment efficiency either. Hence, the formulations were found to retain their physicochemical properties during their storage at different temperatures of 4 °C and 25 °C over a period of 4 weeks. Formulations stored at 37 °C to expose them for temperature excursions [51] and were found to exhibit increased particle size which could be due to coalescence. However, regular labeled storage conditions do not include 37 °C.

4.4. Cellular uptake studies

Fig. 4 represents intracellular uptake of nanoparticle formulations using A549 NSCLC cells. In this experiment, AQ was replaced with fluorescent coumarin to make it possible to visualize the uptake and accumulation. Fluorescent images taken following 1- and 3-hour incubation with coumarin loaded nanoparticles clearly demonstrated significant internalization compared to plain coumarin in A549 cancer cells even after 1-hour incubation period. It can also be observed that there was higher accumulation of the Coumarin-NP around the nucleus (DAPI stained), the most evident location for nanoparticle disruption and drug release inside the cells [52]. Presence of positive charge on the Coumarin-NP enables interaction of nanoparticles with cell surfaces (negatively charged), thus further resulting in efficient cellular internalization.

Fig. 4.

Cellular uptake studies: Representative invitro cellular uptake of coumarin-6 loaded nanoparticles by A549 cells at two different time points; 1 h and 3 h. Coumarin-6 plain solution is used as control. Nuclei are blue (DAPI) and Coumarin NP are green. The highest internalization of Coumarin loaded nanoparticles was observed at 1- and 3-h time points compared to plain coumarin. This indicated the importance of nanoparticulate encapsulation and cationic charge in an efficient cellular uptake. Scale bar 100 μm. Experiments were repeated in triplicates and representative images are shown.

4.5. Cytotoxicity studies

Cell viability studies were performed using MTT assay to evaluate the cytotoxic potential of AQ-loaded NPs versus plain drug. From this study, it was revealed that AQ cytotoxicity against NSCLC cell lines was significantly enhanced by nanoparticle encapsulation. Similar trend was found in all six NSCLC cell lines: A549, H4006, H157, H358, H2122 and H460. Fig. 5 illustrates the cytotoxic effects of AQ and AQ NP in all six NSCLC cell lines. The IC₅₀ values for plain AQ and AQ NP were found to be 21.8 ± 8.3 μM, and 9.4 ± 2.1 μM in A549 cell line; 14.4 ± 1.6 μM, and 1.9 ± 0.5 μM in H2122 cell line; 22.5 ± 2.3 μM, and 16.3 ± 2.8 μM in H157 cell line; 28.8 ± 17.7 μM, and 10.3 ± 10.2 μM in H460 cell line; 28.8 ± 9.7 μM, and 13.5 ± 5.3 μM in H4006 cell line; 35.7 ± 9.7 μM, and 13.6 ± 8.0 μM in H358 cell line respectively (Table 4).

Fig. 5.

Cytotoxicity study: Inhibitory effects on different NSCLC cells after treatments with amodiaquine (AQ), and amodiaquine nanoparticle (AQ NP). AQ NP was found to exhibit enhanced cytotoxicity effect compared to AQ. Cells were incubated with different concentrations of AQ and AQ NP for 72 h and cell viability was determined using MTT assay. Cells without treatment were considered as control (100%). Data represent mean ± SD (n = 6) of at least 3 independent trials.

Table 4.

Comparison of IC₅₀ for AQ and AQ NP in Different NSCLC cell lines. Data represent mean ± SD (n = 6) for 3 independent trials.

Cell line	IC50 (μM)
	AQ	AQ NP
A549	21.8 ± 8.3	9.4 ± 2.1a
H2122	14.4 ± 1.6	1.9 ± 0.5b
H157	22.5 ± 2.3	16.3 ± 2.8c
H460	28.8 ± 17.7	10.3 ± 10.2d
H4006	28.8 ± 9.7	13.5 ± 5.3e
H358	35.7 ± 9.7	13.6 ± 8.0f

a, b, c, e, f: ****

^d:****.

^***p (α < 0.001)

^****p (α < 0.0001)

Smaller and consistent particle size and positive surface charge of AQ NP facilitates efficient internalization of nanoparticles resulting in more accumulation of nanoparticles inside the cells, thus leading to higher cytotoxic potential at the same dose. In-vitro studies to establish safety of optimized nanoformulation were performed on HEK-293 human embryonic kidney cell line. When HEK cell line was treated with blank drug-free AQ-NP equivalent nanoformulation equivalent to 10 μM AQ for 72 h, cell viability of > 80% (81.7 ± 12.4%) suggested that optimized AQ-NP was not toxic to HEK cells, in absence of the drug. The tested concentration of 10 μM was chosen based on the observed IC₅₀ values from cytotoxicity studies in NSCLC cell lines. These observations were expected since polymer used to prepare nanoparticles have long been known to produce little or no cytotoxicity. The safety profile of nanoparticles has been presented in Fig. S7.

4.6. Scratch assay

Scratch assay is a well-established method to access cell-cell interaction, and cellular migration [37]. For these experiments, a scratch was made as described earlier, cells were treated with AQ or AQ NP or fresh media (control), and images of scratched area were taken at respective time points.

Fig. 6A represents images at 0, 24 and 48 h following treatment in A549 cells. After 48 h, % scratch closure was found to be 97.2 ± 0.1, 91.7% ± 2.3 and 31.4 ± 6.8% for control, AQ 10 μM and NP 10 μM respectively in case of A549 (AQ 10 μM vs NP 10 μM; p < 0.0001). From Fig. 6A & B, it can be understood that scratches treated with control and AQ show migration of cells and those treated with AQ NP didn’t exhibit any significant cell migration or scratch closure. Similarly, AQ NP also inhibited scratch closure in case of H460 cell line as well, imaged at 0, 24 and 120 h (not shown); respective graph can be found in Fig. S3. This indicates the efficacy of AQ NP with enhanced anti-migratory effects, thus reducing the tumor metastasis probability.

Fig. 6.

Scratch assay: In-vitro scratch wound healing assay with A549 cells treated with AQ and AQ NP with no treatment as a control. A. Shows representative images for indicated treatments taken at 10× magnification using LAXCO microscope. AQ NP inhibited the migration of A549 cells significantly compared to control and AQ. B. The graph shows % area closure of scratch after 24 and 48 h. The uncovered area has been quantified by the ImageJ Software at each time point. Significance between the groups was analyzed by one-way ANOVA and Tukey’s multiple comparison test. Scale Bar 500 μm. Data represent mean ± SD (n = 3). Images were taken at 10× magnification.

4.7. Clonogenic assay

Clonogenic assay is an in-vitro cell survival assay used to determine the colony formation capability of single cancer cells [53]. AQ and AQ NP were evaluated for their long-term efficacy using clonogenic assay in two different NSCLC cell lines, A549 and H4006. From Fig. 7A, it can be illustrated that colony growth was significantly inhibited by AQ NP compared to AQ, in both A549 and H4006 cell lines. After 48-h treatment period and 7-day incubation, % of colonies survived after treatment with 10 μM AQ or AQ NP were 94.5 ± 15.5%, 27 ± 3.0% (A549; Fig. 7B, p < 0.001); and 49.3 ± 6.2%, 13.7 ± 3.0% (H4006; Fig. 7C, p < 0.0001) respectively, considering number of colonies to be 100% in drug free treatment control wells (Fig. 7B & C). The data suggest an approximately 4-fold higher efficacy against cancer colony formation with AQ-loaded NPs as compared to plain AQ. This data may be considered a representative of AQ NP’s efficacy in eliminating the possibility of tumor relapse, from single cancer cells left behind following chemotherapy and surgical intervention; and can well be linked to enhanced intracellular (and intratumoral) drug accumulation with nanoparticles formulation.

Fig. 7.

Clonogenic assay: A. A549 and H4006 cells were incubated with control, AQ and AQ NP at 10 μM concentration for 48 h in six-well plates. Then, treatments were replaced with fresh media and cultured for additional 7 days with media replacement on alternative days. Colonies were then washed with PBS twice, fixed with 4% paraformaldehyde followed by staining with crystal violet and photographed. Representative images are presented. B, C. Quantitative representation of clonogenic assay as % colony growth with AQ or AQ NP treatment as compared to control (B: A549, C. H4006). Significance between the groups was analyzed by oneway ANOVA and Dunnet’s multiple comparison test. Data represent mean ± SEM (n = 4).

4.8. 3D spheroid cell culture study

A549 cell line was able to form spheroid masses from day 1 during culture, which indicated strong cell-cell communications. While preventive or prophylactic treatment assists in determining the treatment’s potential in inhibiting tumor growth, and may mimic therapy following early diagnosis; therapeutic model represents traditional cancer therapy focused on reducing an already established tumor mass.

Preventive (prophylactic) model involves plating of cells followed by overnight incubation and treatment on next day as soon as the cells accumulate to take shape of a conglomerated mass. Representative images for A549 cell line can be seen in Fig. S4A (Day 0), where cells are seen to be loosely associated in the form of a mass. % Spheroid volumes 12 days after a single treatment with AQ & AQ NP (10 or 25 μM) are shown in Fig. 8A. As can be seen, on day 12 spheroid volumes for AQ NP treated groups were found to be significantly smaller than control volumes. Spheroid diameters (μm) and volumes (mm³) on day 12 after respective treatments were as follows respectively: Control: 1850.0 ± 77.9 μm, 3.3 ± 0.4 mm³; AQ 10 μM: 1818.8 ± 58.9 μm, 3.2 ± 0.3 mm³; AQ NP 10 μM: 1704.5 ± 226.2 μm, 2.7 ± 0.9 mm³. With time, spheroid volumes increased significantly in control groups. However, AQ-loaded NP inhibited spheroid growth significantly compared to plain AQ (p < 0.001). Results revealed that nanoparticle formulation was able to inhibit the spheroid growth in NSCLC cell line (A549).

Fig. 8.

3D spheroid study: A. A549 cells were treated with control, AQ and AQ NP (10 μM). % of spheroid volumes on 12th day of treatment were compared between the treatments. B. Effect of treatment on growth of tumor in 3D spheroids of A549 cells in therapeutic model. Images represent the spheroids on 0, 3 and 15 days of treatment. C, D. Spheroid volume vs time (days) plot for single (C) and multiple dose (D) study. Treatments were administered on day 7 of spheroid study. Spheroids of A549 treated with AQ NP have showed significantly reduced spheroid size in comparison to control after 15 days of treatment where no significant reduction in spheroid volume has been observed after treatment with AQ compared to control. Data represent mean ± SD (n = 6). Scale bar for the images represents 400 μM. Significance between the groups was analyzed by unpaired student’s t-test individually. *p(α < 0.05).

In therapeutic spheroid model, cells were allowed to form tight cellular masses for a period of 7 days, as would be the case with late diagnosis of a well-established tumor. After tumor formation, spheroids were treated by either a single dose (one dose in the beginning) or a multiple dose (dosing every 3rd day, mimicking chemotherapy treatments).

As can be seen Fig. 8B, spheroids were found to achieve a diameter and spheroid volume of 1835.4 ± 97.4 μm and 3.3 ± 0.5 mm³ on an average after 7 days of culture i.e. day 0 for the treatments. After 15 days of treatment, A549 spheroids of single dose study had spheroid volumes of 16.8 ± 5.1 mm³ (Control), 13.9 ± 2.5 mm³ (AQ 10 μM), 7.7 ± 0.67 mm³ (AQ NP 10 μM), 11.9 ± 5.2 mm³ (AQ 25 μM), and 8.7 ± 1.2 mm³ (AQ NP 25 μM), thus demonstrating significant growth inhibition as compared to both drug free treatment and plain AQ (Control vs AQ NP 10 μM: p < 0.01; AQ 10 μM vs AQ NP 10 μM: p < 0.01; Control vs AQ NP 25 μM: p < 0.01) (Fig. 8C; Single dose). Multiple dosing (every 3rd day treatment) also resulted in significant spheroid growth inhibition after AQ NP treatment compared to AQ (10 μM: p < 0.001; 25 μM: p < 0.05) (Fig. 8D; Multiple dose). Interestingly, no significant reduction in tumor volume was observed with plain AQ treatment compared to AQ NP treatment groups for both single and multiple dose studies. Line graphs representing spheroid volume comparison after single and multiple dosing can be found in Fig. 8C & D respectively. Bar graphs representing spheroid volume (normalized to control) comparison can be seen in Fig. S4B & C confirming superior tumor growth inhibition by AQ NP compared to control and plain AQ treatments.

From the observed efficacy in both spheroid models, it was confirmed that AQ NP could inhibit tumor growth in earlier as well as later stages.

4.9. Live-dead cell assay

This study assists in determining viable and dead cells within a solid spheroid mass. One of the main objectives of developing a drug-loaded nanocarriers is to facilitate drug’s entry into the deep core of the tumor, otherwise inaccessible to the plain drug. However, a 2D tumor diameter or volume measurement may not reveal a nanoformulation’s efficacy in killing tumor cells within the tumor core, as the spheroid masses may consist of a dead cell core surrounded by a well-established structural periphery of live cells. 3D spheroid masses were reported to comprise of different regions where cells undergo proliferation, quiescence and necrosis [54]. Therefore, it is necessary to quantify live-dead cell portions out of a spheroid mass. Fig. 9A (single dose) & Fig. 9B (multiple dose) are representing terminal spheroid images on day 15th after treatment.

Fig. 9.

Live dead cell assay: A, B. A549 spheroids of therapeutic model (Single and multiple doing respectively) were stained using the viability/cytotoxicity assay kit to determine % of live and dead cells in respective spheroids. Then images were captured using EVOS-FL fluorescence microscope. Regions showing green indicate stained live cells and red as stained dead cells. C, D. Graphs represent comparison of GFP intensity/mm² (compared to control) (live cells) of A549 spheroids treated with single (C) or multiple dosing (D) through analyzing the fluorescent images by ImageJ software. Data represent mean value observed from single spheroid per each treatment (n = 1).

Like regular xenograft tumor in mice, 3D spheroids are also grown over an extended period, with intratumoral core regions going through lack of O₂, and nutrients, resulting in presence of necrotic and quiescent cells in the core regions. Due to this, even control spheroids are likely to exhibit red fluorescence due to the presence of necrotic cells. While AQ-NP treatments demonstrated significant reduction in tumor volume and cellular viability, a reduced red fluorescence (dead cells) were seen with all AQ/AQ-NP treatments in Fig. 9A & B. This observation can be explained on the basis of loss of tumor integrity with AQ/AQ-NP treatment, resulting in removal of dead cells during media changes. However, green fluorescence (live cells)/mm² of tumor surface area has been quantified, which represented distinct reduction in presence of live cells with AQ-NP treatment at both doses, in both single- and multiple-dose treatment (Fig. 9C & D). Green fluorescence intensity comparison on day 15th day of single dose (AQ NP) post treatment period revealed presence of live cells, 1.4–3.2 folds lower as compared to control and also there observed a clear reduction in live cell population compared to plain AQ treated groups (Fig. 9C).

For multiple dose study, AQ NP of 10 μM and 25 μM were found to exhibit 4.5–9-fold reduction in live cell population compared to control group. In addition, live cell population observed for AQ NP treated groups was found to be less compared to that of plain AQ groups (Fig. 9D). For both single and multiple dose studies, live cell proportion was found to be lowered in case of nanoparticle group compared to plain AQ or control treated groups, indicating ability of AQ NP in inhibiting tumor cell proliferation.

5. Mechanism of action

5.1. Autophagy inhibition microplate assay

Nanoparticles of AQ were found to cause autophagy inhibition significantly more compared to plain drug. This was confirmed from the results of fluorescence measurements. AQ NP (10 μM) treated cells exhibit accumulation of green fluorescent vesicles with higher intensity compared to cells treated with plain amodiaquine or starved media, as seen in Fig. 10A (blue fluorescence represents Hoechst 33342 nuclear stain). Following normalization of data to control as 100%, relative fluorescence intensities, representing co-localization of LC3 protein and thus accumulation of autophagosomes, were found to be 126.2 ± 1.4% (AQ), and 173.9 ± 28% (AQ NP) after treating A549 cells for 18 h (Fig. 10B). Increased green signal indicated accumulation of autophagic vesicles further confirming autophagy inhibition which leads to tumor growth due to blockade of this crucial pathway required for the survival and proliferation of cancer cells [55].

Fig. 10.

A. Autophagy Inhibition Microplate Assay: A549 cells were cultured and starved for 24 h followed by incubation with respective treatments of AQ or AQ NP or control 18 h. Inhibition of autophagy was determined using CYTO-ID^® autophagy kit. Green staining was due to co-localization of CYTO-ID green dye with LC-3 protein. In presence of AQ NP, very bright green fluorescent signals were observed compared to treatment with AQ or control. EVOS-FL fluorescence microscope was used for imaging. B. Graph represents the comparison of relative fluorescence units (RFU) after incubating with respective treatments. Higher the RFU value, higher the accumulation of autophagosomes which indicates the inhibition of autophagy. Significance between the groups was analyzed by one-way ANOVA and Dunnet’s multiple comparison test. C. Detection of caspase-3 levels in A549 cells using the EnzChek^® Caspase-3 Assay: A549 cells were either treated with AQ or AQ NP (10 μM and 25 μM) and incubated for 6 h at 37 °C/5% CO₂. Cells were harvested, lysed and assayed. Reactions were carried out at room temperature, and fluorescence was measured in a fluorescence microplate reader using excitation at 360 nm with emission detection at 460 nm after 20 min. Results indicate the presence of apoptosis from the fluorescence measurements. Data represent the mean ± SEM (n = 3).

5.2. Caspase 3 assay

To evaluate induced active caspase-3 enzyme, cleavage product of a fluorogenic caspase substrate, DEVD-AMC was measured. The EnzChek^® Caspase-3 Assay Kit allows the detection of apoptosis by assaying for increases in caspase-3 and other DEVD-specific protease activities (e.g., caspase-7). As seen in Fig. 10C, increased caspase-3 activity was observed in A549 cells with nanoparticle treatment compared to plain drug and control (fluorescence intensity measurements – Control: 5063.3 ± 1093.8, AQ 10 μM: 5018.0 ± 1715.1; AQ NP 10 μM: 7408.7 ± 2095.9; AQ 25 μM: 5664.7 ± 1161.6; AQ NP 25 μM: 7096.7 ± 1694.0) as seen in Fig. 10C. However, the results were not found to be statistically significant (p > 0.05). This experiment may signal toward presence of an alternate apoptotic pathway responsible for AQ’s tumor inhibiting properties.

5.3. In-vitro angiogenesis study

Angiogenesis i.e. formation of new blood vessels is a paramount mechanism for tumor growth, with its inhibition being a major antitumor mechanism. Endothelial tube formation and inhibition can easily be determined and quantified by analyzing microscopy images. While significant tube formation was observed in HUVEC endothelial cells mediated tube formation assay with drug free treatment (negative control), neither plain AQ nor AQ NP appeared to inhibit the tube formation in-vitro (Fig. S5). It was found that there was no potential angiogenesis inhibition activity suggesting the requirement of altered doses to exert their anti-angiogenesis activity or further investigation.

6. Discussion

Nanoparticle drug delivery systems are promising carriers for both hydrophilic and hydrophobic anticancer drugs. Tseng et al developed gelatin nanoparticles with biotinylated epidermal growth factor conjugation for targeting lung cancer [56]. Through standard manufacturing approaches such as ultra-probe sonication, there exists a difficulty in scaling up to a large commercial scale. Moreover, payload capacities for hydrophilic drugs are low in those cases. Hence, there is great need to explore scalable approaches to develop nanoparticles at a laboratory scale. High pressure homogenization (HPH) has been utilized to formulate nanoparticles due to its scale-up feasibility and reproducible product quality [6]. HPH has been widely used in industries for large scale production of nanoemulsions and solid lipid nanoparticles [6]. However, formulation of nanoparticles using HPH involves multiple complex parameters to consider while obtaining an optimized system. Both process and formulation parameters can be studied to develop an optimized system. Application of design of experiments (DoE) assists in identifying the critical parameters and their respective levels [4]. Critical formulation parameter such as PEI concentration in external aqueous medium and critical process parameter such as probe homogenization speed were studied in the present study. For this, we have utilized multifactorial design to identify optimal levels related to pre-emulsion quality and cationic polymer concentration in the formulation while keeping all other optimized variables constant. Cationic polymer such as PEI is capable of imparting positive surface charge to the nanoparticle formulations. As drug repurposing provides an excellent opportunity to find novel anti-cancer uses for previously FDA approved drugs, a vital approach to explore them where toxicity and safety data is already available in the public domain [28,57], AQ (FDA approved anti-malarial) has been chosen to formulate NPs using scalable HPH approach. Qiao et al have reported the anti-cancer efficacy of AQ in melanoma. But, till now there were no studies reporting the efficacy of AQ or AQ nanoparticles in NSCLC. Moreover, development of efficacious NPs against NSCLC using a scalable approach is of significant translational value. Optimized nanoparticle formulation was identified through DoE and used further to assess the efficacy in treating NSCLC. As shown in response plots in Fig. 1A, and Table 1, effect of PEI concentration on NP particle size prepared at lower homogenization speeds of 6000 rpm, formulations A1, A4, and A7 is not distinctive. Increasing trend of particle size/polydispersity was observed at medium homogenization speed of 12,500 rpm with an increase in PEI concentration. (A2: 0% PEI: 198.0 nm; A5: 0.5% PEI: 478.8 nm and A8: 1% PEI: 463.3.0 nm of 0.476 PDI value). Preparation of pre-emulsion at high speed of 25,000 rpm resulted in the reduction of coarse emulsion globule size and allowed formation of smaller uniform sized nanoparticles. As seen in Fig. 1C, encapsulation efficiency increased from 30.1% to 83.7% with increase in PEI concentration from 0% (A3) to 0.5% (A6). The increase in encapsulation efficiency at higher speed of homogenization was found to be significant at 0.5% of PEI concentration. This may be due to the fact that a unidirectional and less turbulent flow in case of lower speed may have resulted in loss of drug from the organic phase as reported earlier by Sharma et al [58]. No significant effect of the drug quantity was observed on the particles size. However, the PDI, % entrapment efficiency (%EE) and overall drug loading was greatly affected by the drug quantities. With an increase of drug quantity from 1 to 10 mg, %EE declined drastically from 47.5% to 22.6%. It seems that at higher drug quantity, %EE was less, resulting in lower actual:theoretical drug loading ratio. A possible explanation is that the higher drug loading resulted in increased drug concentration gradient between the polymer matrix and outer aqueous phase, which in turn led to more drug loss in the fabrication process. The polymer itself may have a limited capacity to encapsulate the specific drug. Beyond its maximum capacity, more drug may be wasted during fabrication process and the EE thus decreased correspondingly [59]. Moreover, C3 formulation was found to exhibit particle size of < 250 nm, and PDI of < 0.1 along with highest % drug loading of 4.02. Hence, C3 formulation was identified as the monodispersed nanoparticulate system with optimal characteristics with and used for further studies.

STEM images represented in Fig. 2C revealed that the nanoparticles were of spherical shape with smooth surface with little or no aggregations indicating that the vesicles were uniformly dispersed. XRD and DSC studies as seen in Fig. 2D & E were found to be consistent with earlier reports [48,49], and also revealed AQ encapsulation inside nanoparticles while being amorphous.

Presence of positive charge on AQ NP surface refers to the stability of nanoparticles through inhibiting aggregation. STEM studies revealed that the nanoparticles were spherical in shape. Cationic surface charge of nanoparticles associates with higher cellular uptake and greater cytotoxicity in cancer cells [60]. Intracellular uptake shown in Fig. 4 reveals the same where coumarin nanoparticles were internalized by A549 cells efficiently within an hour compared to plain coumarin solution. Due to smaller particle size, there is a possibility of NP aggregation. Hence, stability studies were performed to determine stability of AQ nanoparticles at different temperatures. Results revealed that there was no significant difference in particle size (Fig. S2A), zeta potential (Fig. S2B) and drug entrapment over 4 weeks indicating formulation integrity.

Pulmonary route of administration is well known for its ability to deliver therapeutics locally to site of action but is challenged by major obstacles like poor control over deposition rates and site of inhaled molecules for efficacious delivery at the same time. NPs exhibiting optimal in-vitro aerosol lung deposition can be administered via inhalation, facilitating local delivery to the deep lungs, thus resulting in reduced exposure to other organs and reduced adverse events. Nebulizers can deliver the formulations as small droplets which will be deposited in the lung airways based on their MMAD [24]. PLGA polymer was chosen to formulate NPs due to its ability to produce small size systems, excellent stability, high cytocompatibility and rapid cellular uptake [61]. Moreover, it has been well documented about the distribution of PLGA NPs within alveolar cells [62]. In current study, MMAD value was found to be 4.7 ± 0.1 μm with 72.8 ± 25.1% FPF, confirming the respirability of formulated AQ nanoparticles and efficient lung deposition. Developed amodiaquine nanoparticles are capable of providing a promising strategy to overcome the physiological barriers involved in respiratory tract, and ensuring their capability to reach the respirable regions of the lung [24].

Extensive in-vitro cell culture studies were carried out to investigate the efficacy of AQ NP against different NSCLC cell lines. Even though each of the NSCLC cell line used in cytotoxicity study has its own characteristics, it was found that a significant reduction in IC₅₀ values was observed for AQ after encapsulating into nanoparticles in all 6 NSCLC cell lines as seen in Table 4 and Fig. 5. This reveals the significant anti-cancer efficacy of AQ NPs among wide range of NSCLC types. This could be a result of efficient cellular internalization of nanoparticles. Generally, cell migration is an important characteristic of any cancer which involves the movement of cells from one location to another, associated with cancer metastasis [63]. In this study, results from scratch assays seen in Figs. 6B and S3 revealed that amodiaquine nanoparticles had greater ability to inhibit cell migration significantly compared to amodiaquine in NSCLC cell lines. Clonogenic assay study revealed that AQ NPs can inhibit the ability of single cells to form colonies. Hence, it ensures the effectiveness of AQ NP on long term basis while preventing tumor recurrence. Altogether, AQ NP could be further evaluated for their establishment in NSCLC treatment explicitly.

Development of 3D in-vitro models is of utmost necessity today to mimic in-vivo solid tumor conditions which could be achieved by using specific low attachment spheroid plates as reported by Vaidya et al [35]. From results seen in Fig. 8A–D, it can be understood that nanoparticles were able to deliver the drug to spheroids’ core efficiently in both prophylactic and therapeutic model ensuring their clinical use. Drug-loaded nanoparticles facilitated the entry of drug into deep tumor core, which would otherwise be inaccessible for plain drug. Thus, the physicochemical properties of nanoparticles such as small particle size, charge or surface modifications are crucial in imparting good penetration behavior to them [64]. It has also been reported in the literature that smaller the nanoparticle size, more efficient will be their transport within the tumor [65]. In addition, live dead cell assay results confirmed ability of AQ NP to inhibit tumor cell proliferation in solid mass of spheroids leading to the diminished live cell proportion. In accordance with earlier report by Qiao et al [26], AQ exhibited autophagy inhibition with AQ NP demonstrating significantly higher efficacy in autophagy inhibition, potentially due to higher cellular internalization of the treatment [28]. Here, distinct accumulation of autophagosmes was found indicating the inhibition of autophagy process. AQ causes an increase in lysosomal pH, which inhibits lysosome function and blocks fusion of autophagosomes with the lysosomes [26], thus conserving LC-3 protein and producing a positive signal in the CYTO-ID^® Autophagy detection assay as seen in Fig. 10A & B.

Apoptosis is referred to a programmed cell death. Caspases are cysteine proteases, which play a key role in apoptosis. They will be activated in presence of death inducing stimuli [29]. From the results, it was observed that AQ NP treatment induced caspase-3 levels, demonstrating their ability to induce cell death through apoptosis as well. However, the results weren’t significant indicating toward presence of an alternate cell death pathway. The tube formation assay is a fast, reproducible method for in-vitro measurement of angiogenesis, a major event in tumorigenesis [66]. No significant difference was observed between AQ and AQ NP in their ability to inhibit tube formation compared to control indicating toward an alternate mechanism.

As the major challenge being scale-up issue with development of nanoparticulate systems, exploring a scalable approach of HPH is of special interest for successful translation of nanomedicines. Moreover, optimal respirable properties of the developed NPs enable a local delivery into the lungs. Repurposing of anti-malarial drugs for cancer treatment could be a promising approach to target predominant processes such as autophagy and proliferation involved in NSCLC progress. By utilizing the advantage of drug repurposing, AQ was found to be an efficacious molecule against NSCLC with enhanced efficacy in an optimized NP identified by DoE. Taken together, results from the current study including in-vitro cell culture studies as well as 3D spheroid models illustrating their pronounced efficiency in NSCLC treatment. Within this study, the potential of scalable HPH approach in formulating amodiaquine nanoparticles and the repurposing feasibility of an anti-malarial drug to treat NSCLC has been evaluated through a systematic approach, and optimal in vitro cell culture studies. Altogether, no research work has been reported until today, which accomplishes a comprehensive task of bridging the gap between laboratory scale and industrial scale manufacturing of nanoparticles while reducing development costs due to repurposing feasibility of amodiaquine, achieved in the present project.

7. Conclusion

With the present investigations, it can be concluded that inhalable nanoparticles of amodiaquine were successfully formulated and optimized using systematic approach of design of experiments by a scalable high-pressure homogenization (HPH) approach. As HPH approach is feasible for scale-up while providing reproducible results, developed nanoparticles can be easily scaled-up with ensured product quality. Despite involvement of multiple factors, by using a 3²-factorial design, it was possible to develop an optimized formulation with ease. In addition to cellular uptake studies, cytotoxicity studies and 3D spheroid studies, assays exhibiting the autophagy inhibition or apoptosis inhibition capability of amodiaquine nanoparticles are promising them as a potential treatment strategy for NSCLC. While the preliminary results are promising, large animal preclinical studies should be carried out to emphasize the drug repurposing of amodiaquine to treat NSCLC.