Ocular Pharmacology and Toxicology of TRPV1 Antagonist SAF312 (Libvatrep)

Abstract

Purpose: To evaluate the pharmacology and toxicology of SAF312, a transient receptor potential vanilloid 1 (TRPV1) antagonist.

Methods: TRPV1 expression in human ocular tissues was evaluated with immunohistochemistry. Inhibition of calcium influx in Chinese hamster ovary (CHO) cells expressing human TRPV1 (hTRPV1) and selectivity of SAF312 were assessed by a fluorescent imaging plate reader assay. Ocular tissue and plasma pharmacokinetics (PK) were assessed following a single topical ocular dose of SAF312 (0.5%, 1.0%, 1.5%, 2.5%) in rabbits. Safety and tolerability of SAF312 were evaluated in rabbits and dogs. Effects of SAF312 on corneal wound healing after photorefractive keratectomy (PRK) surgery were assessed in rabbits.

Results: TRPV1 expression was noted in human cornea and conjunctiva. SAF312 inhibited calcium influx in CHO-hTRPV1 cells induced by pH 5.5 (2-[N-morpholino] ethanesulfonic acid), N-arachidonoylethanolamine, capsaicin, and N-arachidonoyl dopamine, with IC₅₀ values of 5, 10, 12, and 27 nM, respectively, and inhibition appeared noncompetitive. SAF312 demonstrated high selectivity for TRPV1 (>149-fold) over other TRP channels. PK analysis showed highest concentrations of SAF312 in cornea and conjunctiva. SAF312 was found to be safe and well tolerated in rabbits and dogs up to the highest feasible concentration of 2.5%. No delay in wound healing after PRK was observed.

Conclusions: SAF312 is a potent, selective, and noncompetitive antagonist of hTRPV1 with an acceptable preclinical safety profile for use in future clinical trials.

Translational Relevance: SAF312, which was safe and well tolerated without causing delay in wound healing after PRK in rabbits, may be a potential therapeutic agent for ocular surface pain.

Introduction

The human ocular surface (cornea and conjunctiva) is densely innervated,^1,2 and corneal sensory nerves serve important sensory functions.³ Any injury or nerve abnormalities in these structures may contribute to pain or severe discomfort.⁴ Ocular surface pain (OSP) is a complex multifactorial condition, which has a debilitating impact on physical and social functioning, resulting in poor quality of life and increased healthcare burden.^5,6 

Despite several developments, there is no established standard of care or topical treatment for long-term management of OSP.^7,8 In addition, the use of topical nonsteroidal anti-inflammatory drugs (NSAIDs), a commonly used medication following surgery, is associated with serious ocular side effects such as burning, stinging, conjunctival hyperemia, superficial punctate keratitis, corneal melts, and delayed wound healing.^9–11 There is a need for targeted ocular medication for management of OSP without a risk of delayed wound healing and adverse effects. 

Transient receptor potential ion channel subfamily V member 1 (TRPV1) is a nonselective cation channel and a polymodal receptor expressed in cornea¹² and conjunctiva.¹³ It has been shown that TRPV1 has dual functions in corneal tissues; it detects and regulates pain and heat signals and plays an important role in innate inflammatory responses.^14–16 

In the past few years, several orally administered TRPV1 antagonists have been identified and evaluated for various pain conditions, but their clinical development was discontinued because of systemic side effects, including decreased noxious heat detection and hyperthermia.^17–21 However, there is a lack of studies evaluating the role of TRPV1 antagonist in ocular pain.¹⁷ Recently, phase 2 and 3 studies of SYL1001, a short interfering RNA targeting TRPV1, showed that topical ocular administration (1.125%) of SYL1001 was safe and effective in reducing hyperemia and ocular pain after 10 days of treatment in patients with dry eye disease (DED).^22–24 

SAF312 (libvatrep), a quinazolinone derivative,²⁵ is a novel TRPV1 antagonist being investigated as a possible treatment option for OSP. The purpose of this study was to evaluate the pharmacology and toxicology of SAF312 as part of preclinical studies for treating OSP. 

Methods

In vitro pharmacology studies assessed TRPV1 expression in human ocular tissues, as well as antagonistic activity, selectivity, and safety pharmacology of SAF312. In vivo pharmacokinetic (PK) studies assessed the tissue and plasma distribution of SAF312 following topical ocular administration. Safety and tolerability of SAF312 were evaluated by in vivo toxicokinetic and toxicologic studies. Also, the effect of SAF312 on corneal wound healing after photorefractive keratectomy (PRK) surgery was assessed in vivo. 

Rabbits and dogs were used for in vivo studies. All animals were treated in accordance with the protocol approved by the Institutional Animal Care and Use Committee and the Novartis Animal Care and Use Committee. Also, all in vivo procedures were performed in accordance with the Association for Research in Vision and Ophthalmology Statement for the use of animals in ophthalmic and vision research. 

Pharmacology

Expression of TRPV1: Immunohistochemical Staining

TRPV1 expression in human ocular tissues was evaluated with immunohistochemistry. Anterior segments of human eye globes with no known corneal defects procured during postmortem (Lions Eye Institute, Miami, FL, USA) with consent of donors or donors’ next of kin, in accordance with the Eye Bank Association of America Medical Standards, US/Florida law for human tissue donation, the Declaration of Helsinki and Food and Drug Administration regulations, and Novartis human tissue registration working practice guidelines regarding research using human tissues, were used for immunohistochemical staining. The tissue was fixed in 4% paraformaldehyde for 2 hours and then transferred to 10% sucrose for 1 to 2 hours, 20% sucrose for 2 hours, and 30% sucrose overnight. Then the tissue was flash frozen in optimal cutting temperature compound. Cryosections, 10 µm thick, were fixed with ice-cold acetone for 15 minutes and incubated with TRPV1 antibody (1:100 dilution, ALX-210-417-C100; Enzo Life Sciences, Inc., NY, USA) and β-III tubulin antibody (1:800 dilution, ab78078; Abcam, MA, USA) overnight at 4°C. Secondary donkey anti-rabbit antibody conjugated to Alexa Fluor 594 (1:1000 dilution, A21207; Invitrogen, NY, USA) and goat anti-mouse antibody conjugated to Alexa Fluor 488 (1:1000 dilution, A11001; Invitrogen) were used. Images were captured using the AxioVision Program v4.8.2. 

Antagonistic Activity of SAF312: FLIPR Assay

Antagonistic activity of SAF312 against a wide variety of stimuli was assessed by a fluorescent imaging plate reader (FLIPR) assay. Chinese hamster ovary (CHO) cells expressing human TRPV1 receptor were grown in Ham's F-12 Nutrient Mixture (SH30026.01; HyClone, MA, USA), and cells were harvested at approximately 80% confluency and plated onto 384-well black cell culture plates (cat. 781091; Greiner Bio-One, NJ, USA). Following instructions of the Calcium 6 Assay Kit (#R8190; Molecular Devices, CA, USA), 20 µL/well of loading dye was added to the cells and incubated at 37°C for 1.5 hours. SAF312 (10 µL) was transferred to the cell plate using Vertical Pipetting Station 384ST (Agilent Technologies, CA, USA) and incubated for 10 minutes. Cells were stimulated by the application of 10 µL of agonist (low pH 2-[N-morpholino] ethanesulfonic acid [MES, pH 5.5] or N-arachidonoylethanolamine [AEA, cat. A0580; Sigma, St. Louis, MO, USA] or capsaicin [cat. 12084; Sigma] or N-arachidonoyl dopamine [NADA, cat. A8848, Sigma]) per well at its EC₈₀ concentration (agonist EC₈₀ determination data not shown) in the FLIPR instrument, and fluorescence was measured for 5 minutes. The half-maximal inhibitory concentration (IC₅₀) of SAF312 was calculated, and an average of at least three independent experiments was considered. 

The antagonistic activity of SAF312 was also evaluated against protein kinase C (PKC)–induced human TRPV1 activation (using Fluo-4 AM or Fura-2 AM) and heat-sensitized human TRPV1 activation (using Fura-2AM). Detailed methodology is provided in the supplementary file. 

Furthermore, to determine whether SAF312 is a competitive or noncompetitive antagonist, concentration–response curves to capsaicin and NADA were obtained in the presence of increasing concentrations of SAF312 (one to three titrations starting from 370 nM in the presence of six concentrations of either capsaicin [one to three dilutions from 0.5 µM] or NADA [one to three dilutions from 25 µM], respectively). Response curves were drawn with ligands (capsaicin or NADA) on the x-axis and FLIPR readings on the y-axis. 

Selective Action of SAF312: Selectivity Assay

Selectivity of SAF312 for 18 human TRP channels was evaluated by FLIPR assay using two cell types, CHO (expressing TRPM5 channel) and human embryonic kidney (HEK; expressing 17 channels). Cells were trypsinized, counted, seeded in black clear-bottomed 96-well plates at a density of 50,000 cells per well, and incubated overnight. Next day, media were removed from the cell plates, and 25 µL of assay buffer was added. TRP channels (A1, V1, V2, V3, V4, V5, V6, M2, M3, M4, M5, and M8) were tested using Calcium 5 dye solution (10 µL; SB Drug Discovery, Glasgow, UK), and other TRP channels (C1, C3, C4, C5, C6, and C7) were tested using red membrane potential dye (10 µL; SB Drug Discovery). 

SAF312 was added using a manual multichannel pipette and incubated for 10 minutes at room temperature. The plates were then placed in FLIPR, and fluorescence was monitored every 1.52 seconds. After 20 seconds, 10 µL of appropriate standard agonist was added, and fluorescence was monitored for 2 minutes at excitation (Ex)/emission (Em) of 488 nm/510 to 570 nm. For TRPV1, SAF312 was tested at a 10-µM starting concentration with a threefold serial dilution, seven points in triplicate, whereas for other channels, the starting concentration was 30 µM. The IC₅₀ values of SAF312 were determined. Detailed methodology is provided in the supplementary file. 

Safety Pharmacology: In Vitro Study

The in vitro safety pharmacology of SAF312 was assessed using a panel of 278 targets (Supplementary Table S1 and Supplementary Table S2). Also, as SAF312 has structural similarity to estrogen, it was screened for its agonist and antagonist activity at human estrogen receptors α or β (ERα or ERβ), using a HEK293T reporter cell line. Detailed methodology is provided in the supplementary file. 

PK and Tissue Distribution

The PK parameters (maximum concentration [C_max], time to C_max [T_max], and area under the concentration–time curve [AUC]) were assessed in ocular tissues (cornea, conjunctiva, aqueous humor, lens, vitreous humor, and retina) and plasma following a single topical ocular dose of SAF312 eye drops (0.5%, 1.0%, 1.5%, and 2.5% dosed at 0.175, 0.350, 0.525, and 0.875 mg/eye/35 µL, respectively) in both eyes of male New Zealand white rabbits. 

Toxicology

Toxicokinetics: In Vivo Animal Studies

Toxicokinetic plasma parameters (C_max, T_max, and AUC at different time intervals) were assessed in male and female Dutch belted rabbits (in 15-day and 13-week studies) and Beagle dogs (in a 13-week study). In a 15-day study, SAF312 was administered bilaterally eight times daily (75 minutes apart) for 15 days (0.5%, 1.5%, and 2.5% dosed at 0.185, 0.555, and 0.925 mg/eye/dose, respectively), and parameters were assessed after the eighth dose on days 1 and 15. In the 13-week studies, SAF312 was administered bilaterally four times daily for 13 weeks (0.5%, 1.5%, and 2.5% dosed at 0.7, 2.1, and 3.5 mg/eye/d, respectively), and parameters were assessed on days 1 and 91. 

Toxicology: In Vivo Animal Studies

Potential adverse toxicologic effects of topical ocular administration of SAF312 were assessed in repeated-dose toxicity studies (2-week studies and 13-week studies). In two different 2-week studies, SAF312 eye drops (0.5%, 1.5%, and 2.5%) were administered bilaterally either four times daily in eyes with PRK surgery (in New Zealand white rabbits [F1 strain]; non–good laboratory practice [non-GLP] study) or eight times daily (in Dutch belted rabbits; GLP study) for 14 days. In the 13-week GLP studies, SAF312 eye drops (0.5%, 1.5%, and 2.5%) were administered bilaterally four times daily for up to 13 weeks in two different species, Dutch belted rabbits and Beagle dogs. The clinical parameters and histopathologic changes in any tissue or organ were assessed. 

Corneal Wound Healing Following Photorefractive Keratectomy

The effect of topical ocular SAF312 on corneal wound healing following PRK was assessed in New Zealand red and white X and New Zealand white pigmented rabbits. The study comprised nine groups consisting of six rabbits per group (two males and four females, N = 54) that underwent a laser PRK procedure on the right eye (OD; wounded). Each treated animal received one drop to each eye (left eye [OS] and OD) four times daily for 14 days, according to the group assignment below. 

SAF312 eye drops were administered either in the presence of a contact lens (0.5%, 1.5%, and 2.5%; daily doses of 1.4, 4.4, and 7.9 mg/d, respectively) or without a contact lens (only 2.5%). Two currently approved customary topical treatments after PRK surgery, ACULAR LS Ophthalmic Solution- Allergan, Madison, NJ, USA (0.5% ketorolac tromethamine, an NSAID) and MAXIDEX Ophthalmic suspension- Alcon Laboratories, Inc., Forth Worth, Texas, USA (0.1% dexamethasone, a corticosteroid), were used as reference treatments. A vehicle (placebo equivalent) with or without contact lens and an untreated control group (PRK wounded without topical ocular treatment and without contact lens receiving no eye drops) were also used to compare the duration of wound healing after PRK. 

Statistical Analysis

In the antagonistic activity assay, IC₅₀ of SAF312 was calculated using a nonlinear regression analysis of sigmoidal-logistic curves with the HELIOS (PROD 2) system, and an average of at least three independent experiments was considered. Data are expressed as mean ± standard error mean (SEM). For the selectivity assay, dose–response curves and IC₅₀ values were generated by Prism software (GraphPad Software, La Jolla, CA, USA). In the PK analysis, ocular tissue and plasma parameters (C_max, T_max, and AUC) of SAF312 were assessed using Phoenix WinNonlin (Version 8.0; Certara, Princeton, NJ, USA) with a sparse sampling linear trapezoidal linear/log interpolation noncompartmental analysis method. Values were rounded off to three significant digits, and data are presented as either mean or mean ± standard deviation (SD). In the PRK wound-healing study, quantitative analysis was conducted using ImageJ software (Version 1.46r; National Institutes of Health, Bethesda, MD, USA) to measure the wound area in each captured image. The average wound area per group was plotted per study day to assess the overall noteworthy difference in the rate of wound healing between the study groups. Data are expressed as mean ± SD. 

Results

Pharmacology

Expression of TRPV1

Immunohistochemical staining of corneal sections showed TRPV1 expression in human corneal epithelial cells and stroma (red signal in Fig. 1A). TRPV1 was also expressed in corneal nerves, including nerves deep in the corneal stroma, with colocalization of TRPV1 with nerve marker β-III tubulin (green signal; Fig. 1A). TRPV1 expression was also observed in conjunctival epithelial cells and the nerve localized in conjunctiva connective tissue (Fig. 1B). 

Figure 1.

 

TRPV1 was expressed in the human cornea and conjunctiva. (A) TRPV1 expression (red) colocalized with nerve marker β-III tubulin (green) in human cornea, including nerves deep in the corneal stroma (40×). (B) Higher magnification of immunofluorescent detection of TRPV1 expression in conjunctival epithelium and some nerves located in conjunctival connective tissue (20×). Cell nuclei were stained with DAPI and appear blue. DAPI, 4′,6-diamidino-2-phenylindole; H&E, hematoxylin and eosin.

Antagonistic Activity of SAF312

In the FLIPR-based Ca²⁺ assay, SAF312 appeared as a potent antagonist, which inhibited the calcium influx in CHO cells expressing human TRPV1 activated by different stimuli, such as low pH 5.5 MES, AEA, capsaicin, and NADA, with mean ± SEM IC₅₀ of 5 ± 2 nM, 10 ± 0 nM, 12 ± 1 nM, and 27 ± 6 nM, respectively. SAF312 also antagonized human TRPV1 activated by PKC (IC₅₀: 14 ± 2.4 nM) and noxious heat (44°C; IC₅₀: 57 ± 3 nM). 

Titrations of SAF312 with up to 25 µM NADA or 0.5 µM capsaicin resulted in the Ca²⁺ flux plateauing at different levels of activity with similar EC₅₀ values (Figs. 2A, 2B), which indicates the mechanism of SAF312 inhibition of TRPV1 is noncompetitive with respect to these ligands. 

Figure 2.

 

SAF312 inhibited (A) NADA-stimulated and (B) capsaicin-stimulated human TRPV1 receptor in a selective and noncompetitive manner. FLIPR, fluorescent imaging plate reader; NADA, N-arachidonoyl dopamine; SD, standard deviation; TRPV1, transient receptor potential cation channel subfamily V member 1.

Selective Action of SAF312

Selectivity assays using a panel of 18 human transient receptor potential (TRP) channels demonstrated that SAF312 is highly selective for TRPV1, with at least 319-fold selectivity achieved against most family members, with the closest activity shown against TRPM8 still displaying 149-fold selectivity versus TRPV1 (Table 1). 

Table 1.

 

IC₅₀ (µM) Values of SAF312 Against TRP Panel Showed High Selectivity for TRPV1

 

Safety Pharmacology

In the in vitro safety pharmacology profile assay, SAF312 did not show any activity of >50% inhibition at 10 µM or 30 µM for any of the targets (including G-protein–coupled receptors, ion channels, nuclear receptors, transporters, and enzymes) tested (Supplementary Table S1 and Supplementary Table S2). In addition, SAF312 did not show any potent effects on ERα or ERβ (Supplementary Fig. S1). Details are provided in the supplementary file. 

PK and Tissue Distribution

PK analysis following a single dose of SAF312 eye drop in both eyes of rabbits showed the highest exposure of SAF312 in the cornea and conjunctiva, followed by the aqueous humor, lens, retina, plasma, and vitreous humor (Table 2). The range of mean C_max was from 8710 to 25,000 nM in the cornea and from 9110 to 27,500 nM in conjunctiva. The corresponding AUC_last (AUC to the last quantifiable time point) ranges were 27,800 to 66,500 nM*h and 19,700–62,900 nM*h, respectively. The plasma concentrations were low, with C_max in the range of 29.0 to 90.4 nM. 

Table 2.

 

PK Analysis of Bilateral Single Dosing of SAF312 Eye Drops in the Rabbit Ocular Tissues and Plasma Showed Highest Exposure in the Cornea and Conjunctiva and Low Plasma Concentrations

 

Following a single topical ocular bilateral administration of SAF312 0.5%, the mean corneal and conjunctival concentrations of SAF312 were >30 times and >70 times higher than the IC₅₀ of SAF312 against NADA (27 nM) and against capsaicin (12 nM), respectively, for at least 12 hours (Fig. 3). The mean concentrations of SAF312 at 12 hours in these target tissues were 1290 nM and 852 nM, respectively, at a dose of 0.5%. 

Figure 3.

 

The mean concentration of a single bilateral topical ocular administration of SAF312 0.5% was >30× higher than the IC₅₀ of SAF312 against NADA and capsaicin in both (A) cornea and (B) conjunctiva through 12 hours. N = 2 rabbits and 4 eyes per time point.

Toxicology

Toxicokinetics

Toxicokinetic plasma parameters of the 15-day study period showed rapid absorption of SAF312 with a T_max of 0.5 hours and no significant accumulation of SAF312. The C_max and AUC increased in an approximately dose-proportional manner. Additionally, there was no consistent difference in plasma exposure between the male and female rabbits (Supplementary Table S3). 

During the 13-week study in rabbits, exposure to SAF312 rose with an increase in dose level from 1.4 to 7 mg/d in a roughly dose-proportional manner. Gender differences in SAF312 PK were not obvious, and only minimal accumulation in SAF312 C_max and AUC (average of respective values less than twofold) was observed after multiple four times daily doses (approximately 2 hours apart) in rabbits (Supplementary Table S4). 

In the 13-week study in Beagle dogs, exposure to SAF312 increased with an increase in dose level from 1.4 to 7 mg/d in a generally dose-proportional manner. Gender differences in the SAF312 mean C_max and AUC_0–12 values were generally less than twofold. No accumulation of SAF312 was observed after multiple four times daily doses (approximately 2 hours apart) in dogs (Supplementary Table S5). 

Toxicology

In the 2-week studies, bilateral topical ocular administration of SAF312 (four times daily [in PRK surgery-treated eyes] and eight times daily [in normal eyes]) showed no adverse effects in rabbits at concentrations of up to 2.5%. In addition, in the 13-week studies, bilateral topical ocular administration of SAF312 four times daily for 13 weeks was well tolerated in both rabbits and dogs. Increased ocular discharge (at dose of ≥1.5%) and multifocal punctate corneal fluorescein stain retention (at dose of 2.5%) in both rabbits and dogs and conjunctival hyperemia (at dose of ≥1.5%) in dogs were some SAF312-related ocular findings observed. Adrenal gland cortical necrosis and vacuolar degeneration observed in the 13-week rabbit study are of doubtful association with SAF312 due to a lack of a clear dose-response relationship, presence of necrosis in males only, unilateral distribution with only one adrenal gland being affected in individual animals, acute morphologic nature of the necrosis, lack of mechanistic connection between TRPV1 antagonist and adrenal effects in the literature, and lack of findings in oral rat and dog systemic toxicology studies conducted at doses up to 1000 mg/kg with exposures up to 350,000 ng*h/mL (rats). All SAF-312–related findings were considered mild in nature (generally grade 1) and hence viewed as nonadverse events. Based on these findings, the maximum feasible concentration of 2.5% was also found to be the no observed adverse effect level (NOAEL) for SAF312 administered eight times daily for 2 weeks in rabbits and four times daily for 13 weeks in rabbits and dogs (Table 3). 

Table 3.

 

Toxicity of Topical Ocular Bilateral SAF312 in 2-Week and 13-Week Repeated-Dose Toxicity Studies

 

Corneal Wound Healing Following Photorefractive Keratectomy

Qualitative and quantitative assessments of corneal wound healing showed that the groups treated topically with 0.5%, 1.5%, or 2.5% of SAF312, either with or without bandage contact lens, did not heal at a different rate than the untreated control group (PRK wound without topical ocular drug and no-bandage contact lens). Groups treated with Acular LS (NSAID) or Maxidex (steroid), reference treatments, showed a trend toward a slightly slower healing rate (no formal statistical analysis was performed), but this minimal difference was not considered clinically or biologically relevant (Fig. 4). 

Figure 4.

 

Topical ocular bilateral administration of SAF312 did not delay corneal wound healing after PRK surgery in rabbits, with or without the concurrent use of a bandage contact lens (CL). *Reference treatments.

Discussion

To our knowledge, this is the first study to evaluate the pharmacology and toxicology of SAF312 administered topically to the eye. We report that SAF312 is a potent, selective, and noncompetitive antagonist of human TRPV1. Also, SAF312 has an acceptable preclinical safety and tolerability profile following topical ocular administration, even with the highest feasible concentration of 2.5%. 

TRPV1 is a promising target to develop analgesics for a variety of pain conditions. TRPV1 antagonists with selective and potent action are expected to prevent TRPV1 activation by a variety of endogenous stimuli.²⁶ Our selectivity assay showed that SAF312 is highly selective (>149-fold) for TRPV1 over other TRP family receptors. SAF312 also strongly inhibited calcium influx in CHO-hTRPV1 cells, suggesting that it has the potential to inhibit TRPV1 activity induced by all known activation mechanisms. 

In addition, the noncompetitive mode of inhibition that was observed, suggests that SAF312 might interact with an allosteric binding site (other than ligand or agonist binding sites) on the TRPV1 receptor structure and prevent channel opening by agonists or block its aqueous pore^27–29 or potentially induce a conformational change within TRPV1 that perturbs the ligand binding site. As open channel blockers, noncompetitive antagonists are considered therapeutically more attractive as they preferably identify and block pathologically overactivated TRPV1 channels, thus reducing potential unwanted side effects.²⁸ The noncompetitive mechanism of SAF312 also suggests that functional target engagement can occur despite high concentrations of TRPV1 agonists like low pH that may be present in the disease setting. 

We observed TRPV1 expression in human cornea and conjunctiva, in agreement with previous reports.^12,13,30,31 In rabbit cornea and conjunctiva, the concentration–time profiles over 12 hours suggested that SAF312 reaches the target tissues at relevant concentrations that may assist with mitigation of ocular surface pain. After topical ocular bilateral administration of 0.5%, SAF312 concentration in the cornea and conjunctiva remained >30-fold higher than in vitro IC₅₀ of SAF312 for NADA- and capsaicin-mediated TRPV1 activation for at least 12 hours, suggesting sufficient drug levels to manage ocular pain over a clinically relevant period. The low exposure of SAF312 in plasma following topical ocular administration, even at the highest concentration of 2.5%, indicates that SAF312 is highly likely to avoid off-target systemic/ocular effects such as impaired heat pain perception and hyperthermia, which are commonly observed with orally administered TRPV1 antagonists.^17–19,32 

In vitro safety pharmacology profiling is considered an essential tool to predict and flag possible clinical adverse effects of any new chemical entity during early phases of drug discovery.³³ The in vitro safety pharmacology profiling assay showed that SAF312 has little or no effect on 278 tested targets, including COX2 and CGRP1, suggesting minimal or no clinical adverse effects with SAF312 likely owing to the low concentration of the plasma SAF312 following topical administration. 

Toxicokinetic analysis showed that topical ocular administration of SAF312 is well tolerated, as evidenced by low plasma exposure, rapid absorption (T_max of 0.5 hours), and generally minimal accumulation. Also, our toxicologic studies showed that SAF312 was a safe in rabbits and dogs, and the NOAEL was found to be the maximum feasible concentration of 2.5% administered eight times daily for 2 weeks in rabbits and four times daily for 13 weeks in rabbits and dogs. These findings were predictive and consistent with a recent first-in-human phase 1 clinical trial of SAF312 in healthy volunteers, which reported that SAF312 was well tolerated with no ocular or systemic safety concerns at the maximum feasible concentration of 2.5% dosed up to one eye drop eight times daily for 7 days.³⁴ The surgical rabbit model of PRK used in this study showed that topical ocular administration of SAF312 did not delay wound healing, indicating the potential of SAF312 to control ocular pain without prolonging wound healing, which has been a concern with NSAIDs that are frequently used to reduce inflammation and pain after ocular surgery.^35–38 

A key strength of this study is the in-depth assessment to understand the effect of SAF312 on TRPV1 activation and its PK and safety profile, aiming to assess its pharmacology and toxicology in the management of ocular surface pain. However, there are certain limitations. First, there is no established preclinical model for ocular pain after PRK; therefore, evaluating the effect of SAF312 on pain in a clinical setting is required to evaluate the potential effect on ocular surface pain. Second, although SAF312 had a low systemic exposure after topical ocular administration, potential consequences of systemic TRPV1 inhibition, such as loss of heat perception and hyperthermia, need to be assessed following topical ocular administration of SAF312 in humans. Finally, although rabbits are widely used for ocular PK studies, these data may not translate directly to humans. The most notable ocular differences between rabbits and humans are the lower tear turnover rate (approximately 50%) and spontaneous blink rate in rabbits (4–5 times/h vs. 6–15 times/min) compared with humans and the presence of a nictitating membrane in rabbits but not in humans. Consequently, translation from rabbits to humans is not necessarily absolute, but assessment of relative differences in formulation (e.g., drug concentration) remains valid.^39–41 However, despite species differences, the relatively large corneal and conjunctiva SAF312 concentrations relative to in vitro IC₅₀, as observed in the current rabbit study, suggest that SAF312 levels in corresponding human eye tissues will likely exceed IC₅₀. 

In conclusion, the findings of this study demonstrate that SAF312 is a potent, selective, and noncompetitive TRPV1 antagonist, and topical ocular administration of SAF312 has favorable preclinical PK, safety, and tolerability profiles. These encouraging preclinical findings may pave the way toward further research and clinical development. Clinical trials with well-defined endpoints are needed to gain better insights into the efficacy, safety, and tolerability of topical ocular SAF312 and to establish its potential role in ocular surface pain management. 

Acknowledgments

The authors thank Christopher Brain, Scott Womble, Mark Bock, Margaret McLaughlin, Leslie Lemke, and Margaret Weaver for their support in the acquisition, analysis, and interpretation of data. Medical writing support was provided by Nilesh Kumar Jain, PhD (Novartis Healthcare Pvt. Ltd., Hyderabad, India), in accordance with Good Publication Practice (GPP3) guidelines. 

Supported by Novartis. The sponsor or funding organization participated in the design of the study; management, analysis, and interpretation of the data; and preparation, review, and approval of the manuscript. 

Disclosure: M. Mogi, Novartis Institutes for BioMedical Research (E); A.E. Mendonza, Novartis Institutes for BioMedical Research (E); J. Chastain, Novartis Institutes for BioMedical Research (E); J.T. Demirs, Novartis Institutes for BioMedical Research (E); Q.G. Medley, Novartis Institutes for BioMedical Research (E), The Janssen Pharmaceutical Companies of Johnson & Johnson (E); Q. Zhang, Novartis Institutes for BioMedical Research (E); J.P.N. Papillon, Novartis Institutes for BioMedical Research (E); J. Yang, Novartis Institutes for BioMedical Research (E); Y. Gao, Novartis Institutes for BioMedical Research (E); Y.Y. Xu, Novartis Institutes for BioMedical Research (E); K. Stasi, Novartis Institutes for BioMedical Research (E), Adverum Biotechnologies (E)