Article Cite This: Mol. Pharmaceutics XXXX, XXX, XXX-XXX
pubs.acs.org/molecularpharmaceutics
Excipient-Free Pulmonary Delivery and Macrophage Targeting of Clofazimine via Air Jet Micronization Ashlee D. Brunaugh, Syed Umer Jan,† Silvia Ferrati, and Hugh D. C. Smyth* College of Pharmacy, The University of Texas at Austin, 2409 West University Avenue, PHR 4.214, Austin, Texas 78712, United States ABSTRACT: Clofazimine (CFZ) is highly active against mycobacterium, including resistant Mycobacterium tuberculosis, but its therapeutic efficacy via the oral route is limited by severe adverse effects, poor aqueous solubility, and slow onset of action. Pulmonary delivery of CFZ is an attractive alternative to target mycobacterium-harboring alveolar macrophages. This study explores the use of air jet milling to develop a respirable, cost-effective CFZ formulation. Jet milled CFZ was readily dispersed from an off-the-shelf dry powder inhaler without the need for additional excipients or carrier particles. Additionally, milled CFZ was internalized by J774.A1 alveolar macrophages within 8 h, with evidence of intracellular biotransformation of the CFZ crystals and macrophage sequestration by 24 h. Less macrophage toxicity was noted in comparison to solubilized drug. Compared to macrophage uptake rate, dissolution of milled CFZ was limited, thereby potentially reducing systemic absorption and subsequent side effects. These results suggest that jet milling is an effective manufacturing method in the development of a CFZ formulation for pulmonary delivery and alveolar macrophage targeting. KEYWORDS: jet milling, tuberculosis, clofazimine, pulmonary delivery, macrophage
1. INTRODUCTION
this mechanism leads to a stable population of intracellular mycobacterium.3 CFZ is a weakly basic iminophenazine antibiotic that exhibits activity against mycobacterium, such as Mycobacterium leprae,4 Mycobacterium avium complex (MAC),5−8 and M. tuberculosis,9 with a minimum inhibitory concentration (MIC) ranging from 0.125 to 2 μg/mL.10−12 Importantly, CFZ exhibits activity against drug-resistant TB12−14 and is now recommended as a second-line agent by the World Health Organization in treatment of MDR-TB.1 CFZ also exhibits numerous other properties that may be highly beneficial in the treatment of TB, including shorter duration of therapy;15 synergy with other antimicrobial agents such as pyrazinamide, rifampicin, fluoroquinolones, and amikacin, which results in enhanced bactericidal activity against stationary phase bacilli;16 and antiinflammatory activity.17 In particular, CFZ demonstrates a unique affinity for macrophage uptake and sequestration. Upon uptake of the drug, macrophages transform CFZ into liquid crystal structures bounded by a bilayer membrane.18,19 These unique intracellular CFZ structures may serve as a protective mechanism against cytotoxicity and allow for the mobilization and accumulation of drug at the site of infection in order to maximize therapeutic efficacy.18−21
There is a growing and urgent need for new drugs for use against tuberculosis (TB). 10.4 million new TB cases were reported worldwide in 2015, with 580,000 of these cases considered to be multidrug resistance tuberculosis (MDR-TB), defined as Mycobacterium tuberculosis resistant against rifampicin, or rifampicin and isoniazid.1 In addition, every region of the world has exhibited cases of extensively drug resistant TB (XDR-TB), defined as M. tuberculosis resistant against isoniazid and rifampicin, plus any fluoroquinolone and at least one of three injectable second-line drugs (amikacin, kanamycin, or capreomycin).1 With the advent of globalization and mass migration from high disease burden areas, these resistant strains are expected to spread. As treatment options dwindle, the reformulation of poorly tolerable, highly active anti-infective agents such as clofazimine (CFZ) is a potential method to overcome resistant TB, particularly if they can be targeted to the infection site. Several challenges exist in the development of such formulations. In order to be effectively implemented in the low resource countries in which TB predominates, any potential treatment must be cost-effective as well as easily transported and administered. Additionally, a potential treatment must exhibit a high specificity toward alveolar macrophages through which the M. tuberculosis infection is initiated and propagated.2 Infectious bacilli are inhaled as droplets and phagocytosed by alveolar macrophages and survive the hostile intracellular environment by restricting acidification of the macrophage and limiting lysosome fusion. In chronic infection, © XXXX American Chemical Society
Received: August 10, 2017 Revised: September 22, 2017 Accepted: October 2, 2017
A
DOI: 10.1021/acs.molpharmaceut.7b00690 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
high velocity streams of air and inducing particle−particle collisions, may be an ideal manufacturing technique to produce cost-effective, respirable CFZ particles. In contrast to the “bottom-up” micronization of poorly water-soluble drugs, air jet milling is a green manufacturing technique, requiring no organic solvents, and is mechanically simple and easily sterilized through application of sterile air. However, an inhalable CFZ formulation carries several development barriers, as summarized in Table 1. Micronization
Though highly active against mycobacterium, the therapeutic efficacy of the existing commercial CFZ oral formulation (Lamprene, Novartis) is limited by its poor water solubility (10 mg/L),22 slow onset of action, and significant adverse effect profile. Oral bioavailability ranges from 45 to 62%23 and exhibits a high degree of interpatient variability and food effect.24,25 Additionally, CFZ exhibits pH dependent solubility, with pKa values of 2.31 and 9.29.26 The change in pH that accompanies the transition from the stomach to intestinal environment can potentially lead to recrystallization and precipitation of CFZ and reduce systemic absorption. At least 30 days of administration is necessary to reach steady-state concentrations, necessitating the use of large loading doses,25 and a delay in bactericidal activity occurs for up to 2 weeks after oral dosing, regardless of the dose administered.27 The necessary high systemic doses are associated with adverse effects that include reddish-brown skin and conjunctiva discoloration (75−100%) and GI distress (40−50%) including abdominal pain, nausea, diarrhea, vomiting, and severe complications such as splenic infarction, bowel obstruction, and fatal bleeding secondary to accumulation of crystalline deposits.28 Additionally, availability of the oral formulation is limited. In the US, Lamprene is only available for treatment of MDR-TB through single-patient Investigation New Drug applications (INDs) administered by the US Food and Drug Administration (FDA).29,30 To reduce the undesirable systemic adverse effects and improve therapeutic efficacy, it is obvious that there is a need for a more targeted formulation of CFZ. Much of the previous formulation work has been focused on the alteration of the physicochemical properties of CFZ, through complexation,31 salt formation,22 amorphous solid dispersions,32 or liposomal encapsulation.33 However, we propose that these physicochemical properties of CFZ, while rendering it a poor candidate for oral or intravenous administration, make it ideally suited for targeted pulmonary administration via dry powder inhaler. The deep lung has limited fluid capacity,34 which typically presents a challenge for the dissolution and absorption of poorly watersoluble drugs. However, this may be advantageous for the targeted pulmonary delivery of CFZ powder. The drug will in fact not be expected to undergo dissolution and subsequent systemic absorption from the lung, but will concentrate at the site of action potentially leading to reduced adverse effects. In addition, insoluble particles delivered to the alveolar regions of the lung are primarily cleared through alveolar macrophage phagocytosis.34 Inhaled CFZ powder could take advantage of this natural uptake mechanism and target mycobacterium residing in the macrophages. To be effectively delivered to the lungs, coarse drug material must be micronized to a smaller particle size, typically less than 5 μm. Currently, the formulation approaches for inhalable antibiotic powder have been focused on the development of advanced spray drying techniques to achieve dispersible particles, for example, the Pulmosphere technology used in the TOBI Podhaler (Novartis) tobramycin delivery system.35 However, such costly technologies limit the commercialization potential in resource-poor countries where deadly pulmonary infections like tuberculosis predominate. Additionally, recent studies by Zhu et al.36 suggest that, depending on the device used, particles produced using conventional micronization methods like milling may offer similar aerosol performance to particles produced using more advanced engineering techniques. In light of this, air jet milling, in which particle size is reduced through entraining particles in
Table 1. Target Profile of CFZ DPI Formulation for Pulmonary Delivery barrier
method to address barrier
high dose needed for efficacy must target infection site in peripheral lung TB is accompanied by lung remodeling and altered lung function and affects pediatric patients TB predominates in low-resource countries
develop excipient-free formulation ensure MMAD < 3 μm achieve aerosolization performance independent of patient inspiratory flow rate utilize cost-effective manufacturing and delivery system
results in highly cohesive particles that are generally resistant to dispersion. This is typically overcome through formulation as interactive mixtures with larger lactose carrier particles (e.g., 45−75 μm) or other dispersion enhancing excipients.37 The recommended oral dose of CFZ for treatment of MDR-TB is 100 mg.38 Though it is possible that this dose may be lower with the higher local concentrations achieved with an inhaled formulation,39 the magnitude of the dose renders the inclusion of carrier particles impractical, as such a lactose blend formulation would drastically increase the amount of powder that the patient must inhale and increase the overall treatment burden. Thus, an excipient-free antibiotic formulation would be a better approach if it could overcome poor dispersion performance. Additionally, simple DPI devices typically exhibit an inspiratory flow rate dependency on powder dispersion, which may be an issue for the treatment of TB due to disease related changes in lung architecture 40 and pulmonary function.41 Lastly, to achieve targeted delivery of the drug to alveolar macrophages, inhaled CFZ must reach the peripheral lungs, which generally requires a mass median aerodynamic diameter (MMAD) under 3 μm.42 Previous work43 has demonstrated in vivo the improved efficacy of inhaled spray-dried CFZ versus orally administered CFZ in a TB mouse model; however, limited information is available on the aerodynamic performance of a CFZ formulation, and to our knowledge, the use of jet milling as a simple, cost-effective, and scalable manufacturing alternative for the formulation of an excipient-free respirable CFZ powder has not been explored. By coupling an inexpensive formulation method with a simple DPI device, inhalable CFZ could be more readily distributed to the low-resource countries where it is required the most. Thus, we propose that air jet milling can be used to develop an excipient-free formulation that is readily dispersed from an off-the-shelf DPI, that exhibits limited dissolution increasing local accumulation, and that is readily phagocytized by alveolar macrophages.
2. EXPERIMENTAL SECTION 2.1. Micronization and Characterization of CFZ. A labscale Aljet air jet mill (also known as a model 00 Jet-O-Mizer, Fluid Energy, Telford, PA) was used to micronize CFZ (Sigma; B
DOI: 10.1021/acs.molpharmaceut.7b00690 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
RCS40 (TA Instruments-Waters LLC, New Castle, DE, USA) refrigerated cooling system with nitrogen purge of 50 mL/min. Approximately 4 mg of each sample was loaded in standard DSC pans (DSC Consumables Inc., Austin, MN, USA), which were crimped using a Tzero sample press (TA InstrumentsWaters LLC, New Castle, DE, USA). Samples were heated at a rate of 5 °C/min from 30 to 300 °C. 2.2. Determination of Aerosolization Performance of Excipient-Free Milled CFZ. In vitro aerodynamic performance testing was performed utilizing a model 7 low-resistance Monodose RS01 DPI, a gift from Plastiape S.p.a (Osnago, Italy). Size 3 hydroxypropyl methylcelllulose (HPMC) capsules were provided by Capsugel Inc. (Morristown, New Jersey, USA). The resistance of the RS01 Monodose DPI used in the cascade impaction studies was determined using a dosage sampling unit according to apparatus B of USP Chapter 601,48 and was calculated to be 0.021 kPa0.5 min/L. Cascade impaction studies for milled CFZ were performed on the Next Generation Impactor (NGI) (MSP Corporation, MN, USA). Stage 1−7 cutoff diameters were determined using eq 1, and MOC cutoff diameters were determined using eq 2:
Lot: SLBL8945 V) to a particle size distribution within the respirable range of 0.5−5 μm. Based upon previously reported methods44−46 for air jet milling for DPI formulations in our lab, the air jet mill was set at 75 psi grind pressure, 65 psi feed pressure, and 1 g/min feed rate. Approximately 3−4.5 g of CFZ was milled per batch. In total, 3 batches were milled, with one batch remilled to further reduce the particle size distribution to the desired range. Based upon the intrinsic size classification present in jet mills,47 different fractions of powder were collected from different areas of the Aljet mill (Figure 1) for further analysis.
⎛ Q ⎞X D50, Q = D50, Q n⎜ n ⎟ ⎝Q ⎠
(1)
⎛ Q n ⎞1.36 = 0.14⎜ ⎟ ⎝Q ⎠
(2)
D80, Q
where D50,Q is the cutoff diameter at the flow rate, Q, and the subscript, n, refers to the archival reference value for Qn = 60 L/ min, and the values for the exponent, X, refer to the archival NGI stage cut size−flow rate calculations as determined by Marple et al.49 To reduce particle bounce and re-entrainment, the NGI plates were coated with 1% (v/v) silicon oil in hexane and allowed to dry. To determine the influence of particle size on aerodynamic performance of milled CFZ, analysis was performed on milled particles obtained specifically from the second milled batch. Two different populations of particles were tested: those with a median geometric diameter (D50) of 2.69 μm (CFZ2.69μm), derived from the cyclone unit of the jet mill (Figure 1), and milled particles with a D50 of 1.81 μm (CFZ1.81μm), derived from the collection vessel unit of the jet mill (Figure 1). Cascade impaction was performed on these samples at a 4 kPa pressure drop (equivalent to 93 L/min on the low-resistance RS01 device) at a duration of time sufficient to draw 4 L of air through the apparatus (2.6 s), according to USP specifications.48 To determine the flow rate dependency of milled CFZ dispersion, cascade impaction was also performed on CFZ1.81μm particles at a 1 kPa pressure drop through the device (equivalent to 47 L/min) for a duration of 5.1 s. All cascade impaction experiments were performed in triplicate. The resultant dispersed powder was collected from the capsule, the inhaler, the adapter, the induction port, stages 1−7, and the micro-orifice collector (MOC) by washing with ethanol. The drug mass in each sample was quantified by measuring the UV absorbance at a wavelength of 480 nm using a Tecan Infinite1 200 PRO multimode microplate reader (Tecan Systems, Inc., San Jose, CA, USA). The emitted fraction (EF) was calculated as the total drug emitted from the device as a percentage of the total mass of drug collected. The fine particle (
