An ion‐paired moxifloxacin nanosuspension eye drop provides improved prevention and treatment of ocular infection

Abstract There are numerous barriers to achieving effective intraocular drug administration, including the mucus layer protecting the ocular surface. For this reason, antibiotic eye drops must be used multiple times per day to prevent and treat ocular infections. Frequent eye drop use is inconvenient for patients, and lack of adherence to prescribed dosing regimens limits treatment efficacy and contributes to antibiotic resistance. Here, we describe an ion‐pairing approach used to create an insoluble moxifloxacin–pamoate (MOX–PAM) complex for formulation into mucus‐penetrating nanosuspension eye drops (MOX–PAM NS). The MOX–PAM NS provided a significant increase in ocular drug absorption, as measured by the area under the curve in cornea tissue and aqueous humor, compared to Vigamox in healthy rats. Prophylactic and treatment efficacy were evaluated in a rat model of ocular Staphylococcus aureus infection. A single drop of MOX–PAM NS was more effective than Vigamox, and completely prevented infection. Once a day dosing with MOX–PAM NS was similar, if not more effective, than three times a day dosing with Vigamox for treating S. aureus infection. The MOX–PAM NS provided increased intraocular antibiotic absorption and improved prevention and treatment of ocular keratitis, and the formulation approach is highly translational and clinically relevant.

Tear production, reflexive blinking, and nasolacrimal drainage limit residence time, while formulation and drug properties can further limit the potential for the rapid intraocular drug absorption needed. [1][2][3] Many pharmaceutical drugs are water-soluble salts 4,5 and when dosed topically to the eye, water-soluble drugs often show increased systemic absorption and rapid drug elimination. 6 Low solubility drugs are often formulated as suspensions of particulates that must also traverse the ocular mucus barrier that protects the surface from allergens, pathogens, and debris. 7,8 Thus, regardless of the dosage form, intraocular drug absorption is low, necessitating frequent administration. 9,10 In the case of antibiotic eye drops, drops may be prescribed for use in treating bacterial keratitis and conjunctivitis every hour for the first 48 h and then once every 4 h until the infection is resolved. 11 As the required number of doses per day increases, patient compliance, and thus, treatment efficacy, decreases. [12][13][14] Antibiotic eye drop formulations that are more effective with less frequent dosing are needed.
Antibiotics are also used extensively for prevention of postsurgical ocular infection. 15 Approximately, 40% to 80% of all endophthalmitis cases occur after cataract surgery, with Staphylococcus species being the most common causative agent. 16,17 Both eye drops and intracameral injections have been studied for the prevention of postcataract surgery endophthalmitis, and there is evidence to suggest that using eye drops in addition to injections has the lowest risk of infection. 18 Fluoroquinolones show broad spectrum efficacy in treating bacterial conjunctivitis and keratitis, 19 and thus, are often used for preventing postsurgical infections. 18 Moxifloxacin hydrochloride (MOX) is often the drug of choice, as it has been shown to have higher intraocular bioavailability compared to other fluoroquinolones. 3,20 However, overall limitations in drug absorption with eye drops mean that prophylactic antibiotic drops still must be used three times or more per day. 21,22 Various strategies have been employed for increasing intraocular drug penetration with topical dosing. 23 One common approach is the addition of viscosity enhancing materials or the use of creams and ointments, which have issues with messiness and blurring vision. 2 Penetration enhancers may also be used, though frequent and chronic dosing can lead to ocular irritation and toxicity. 24 An alternative approach that has been demonstrated for a wide variety of mucosal surfaces, including the eye, female reproductive tract, gastrointestinal tract, and airways is to formulate nanoparticles with mucoinert surface coatings. 25,26 These so-called mucus-penetrating particles (MPP) are nonadhesive to the sticky mucin proteins and small enough in size to fit through the net-like pores, leading to enhanced delivery of small-molecule drugs and nucleic acids alike. 25 In the context of eye drops, MPP provide rapid and enhanced intraocular drug absorption, leading to the Food and Drug Administration (FDA) approval of two loteprednol etabonate (LE)-based products for treating ocular inflammation and pain. 7 Here, we describe an approach for forming an insoluble ion-paired complex of MOX with pamoic acid that can be formulated into a mucoinert nanosuspension (MOX-PAM NS). We demonstrate here that the MOX-PAM NS provides equivalent or better prevention and treatment of ocular Staphylococcus aureus infection with once a day dosing compared to three times a day dosing with the commercial formulation, Vigamox ® . We anticipate that such a reduction in dosing frequency while maintaining therapeutic effect would have a positive impact on patient care and quality of life.

| Nanosuspension formulation and characterization
Pharmaceutical anionic ion-pairing agents, sodium laurate, sodium oleate, and disodium pamoate, were able to convert hydrophilic and cationic MOX into hydrophobic ion pairs after simple mixing. The disodium pamoate showed highest ion-pairing efficiency (>95%) among the tested ion-pairing agents at the concentrations tested ( Figure 1(a)). The Fourier-transform infrared (FTIR) spectrum of MOX-PAM showed suppression of the N─H stretching band and shifting of O─H stretch of carboxylic acid in MOX (Figure 1(b)), indicating the interaction between MOX and disodium pamoate to form MOX-PAM ion pair via electrostatic interaction. A schematic for the formation of MOX-PAM and formulation of the nanosuspension is shown in Figure 1(c). Using our previously described wet-milling method, we successfully developed MOX-PAM NS with particle size of 233 ± 26 nm, average polydispersity index (PDI) of 0.15, and zeta potential value of À10.4 ± 0.37 mV (n = 3). Transmission electron microscopy (TEM) imaging of MOX-PAM NS showed near spherical morphology ( Figure 1(d)), where the average particle size measured using TEM images was in good agreement with the Zetasizer measurements (198 ± 54 nm). Short-term stability characterization over 2 weeks showed stable particle diameter when stored at room temperature, with a slight increase in size when stored at room temperature ( Figure S1).

| Topically administered MOX-PAM NS provides increased intraocular drug absorption compared to Vigamox
The ocular pharmacokinetics study in healthy rats showed that MOX-PAM NS provided rapid and increased delivery of MOX compared to Vigamox in the aqueous humor, with~1.6-fold greater C max (     Figure S4).
Once a day topical treatment with the moxifloxacin-pamoate nanosuspension (MOX-PAM NS) was more effective in treating ocular Staphylococcus aureus infection. (a) To assess therapeutic efficacy, rats were infected topically with S. aureus 24 h prior to initiating treatment. Rats received either no treatment (Infection control), once a day Vigamox (0.5% w/v), three times a day Vigamox, or once a day MOX-PAM NS (equivalent to 0.5% w/v MOX). (b) On the third day of treatment, corneal swabs were taken with a cotton-tipped applicator for determination of bacterial load (n ≥ 7 per group). (c) Twenty-four hours after the end of treatment (Day 4), the eyes were enucleated and were prepared for the determination of bacterial burden (n ≥ 7 per group). Once a day MOX-PAM NS had similar treatment efficacy as three times a day Vigamox. Data expressed as mean ± SEM, *p < 0.05; **p < 0.01; ***p < 0.001

| DISCUSSION
About 70% of ocular infections are caused by bacteria. 19,27 In the United States, it was reported that there were 4 million cases of bacterial conjunctivitis and nearly 1 million clinic visits due to microbial keratitis per year. 28,29 Staphylococcus species are the most common causative agent for both conjunctivitis and keratitis, and topical fluoroquinolones are typically used for treatment. 19 Staphylococcus species are also the most common pathogens that cause endophthalmitis, and thus, fluoroquinolone eye drops are routinely prescribed off-label for preoperative and postoperative administration to prevent postsurgical endophthalmitis. 17 Despite the overall low risk of postsurgical infection, the more than 2 million procedures per year in the United States make cataract surgery the most common cause of endophthalmitis. 17,27 Antibiotic eye drops are prescribed for use at least three times a day, and up to once every hour for severe infections. As the required num- Similar to other exposed epithelial surfaces, the ocular surface is protected by mucus. 30 Membrane spanning mucins form a dense glycocalyx that protects the conjunctival and corneal epithelium, while soluble mucins secreted by the goblet cells in the conjunctival epithelium are released into the tear film. 31  2.2-fold and 3.5-fold higher than that of rat cornea. 37,38 Furthermore, the rats used here were albino, whereas it has been described that fluoroquinolones bind to melanin in the pigmented tissues in the eye. 39 One additional factor to consider is that the rats used for PK experiments here had intact corneas, whereas it has been described that higher amounts of drug may penetrate into the eye when there is a surgical incision. 40 Pamoate is a counter ion for several pharmaceutical drugs taken by the oral route and given as a sustained release intramuscular depot. 41,42 Furthermore, we recently described using pamoic acid as an ion-pairing agent to formulate sustained-release microcrystals that were injected in the subconjunctival space to provide protection of retinal ganglion cells. 43 Pamoic acid did not negatively impact retinal ganglion cell survival in vitro at concentrations as high as 10 mg/ml. 43 While encouraging, the safety of pamoic acid dosed topically would need to be demonstrated before use in an eye drop. Furthermore, while the short-term stability data are encouraging, longer-term characterization and, if required, additional optimization to ensure long-term shelf stability would be an important next step in development.
Increasing drug absorption efficiency can lead to equivalent efficacy with fewer doses. Adherence to eye drop regimens is an often cited problem, and eye drops can only be effective when they are used as prescribed. 44 Adherence is particularly a concern for elderly and pediatric patients, where reducing the required number of drops per day could significantly improve quality of life. 44,45 Better adherence in addition to increased and more effective drug concentrations can also circumvent issues with the development of bacterial resistance. 46 In the case of LE-MPP, the formulation was developed as a 1% suspension and approved in 2018 as the first twice-daily ocular corticosteroid for post-surgical inflammation and pain, where other prior products had to be dosed four times per day. 8 The development of an antibiotic eye drop that has similar efficacy with once a day dosing compared to three times per day dosing would also be clinically impactful. Reformulation of a drug at the same concentration or lower than the approved concentration is a relatively straightforward path to clinical development. 47   MOX was monitored at 296 nm, and the retention time was 2.75 min.

| HPLC analysis of MOX
Data were analyzed using Lab solution integrator software 5.87 SP1.

| Ion pairing
Hydrophobic ion-pair complexes of MOX with sodium oleate, sodium laurate, and disodium pamoate were prepared using previously reported methods. 50

| Development and characterization of MOX-PAM NS
The MOX-PAM NS was formulated based on thorough formulation development using a wet bead-milling method previously described. 54,55 The wet-bead milling was carried out using a lab-scale tissue homogenizer (TissueLyser LT, Qiagen Inc, Germantown, MD).

| Ocular infection prevention and treatment studies
Prevention and treatment of ocular infection in rats was assessed using a previously described model with minor adjustments. 56 S. aureus bacteria were propagated on nutrient agar plates followed

| Histological evaluation
For histological evaluation, the rats were euthanized, eyes enucleated, and fixed in formalin for 24 h. The eyes fixed in formalin were embedded in paraffin, cross-sectioned, and stained with H&E by the Johns Hopkins Reference Histology Laboratory for histological evaluation.

| Statistical analysis
The bacteria concentration in different groups was analyzed by oneway analysis of variance followed by Turkey's multiple comparison test. The differences were considered significant at p < 0.05.

CONFLICT OF INTERESTS
The mucus-penetrating particle technology is licensed and in clinical

DATA AVAILABILITY STATEMENT
The main data supporting the findings of this study are available within the paper and its Supplementary information. The associated raw data are available from the corresponding author on reasonable request.