Fluoroquinolone Update
Recent studies shed new light on penetration and wound healing.
BY CHRISTOPHER KENT, SENIOR ASSOCIATE EDITOR

When new pharmaceuticals are introduced to the marketplace -- whether they're new drugs or different formulations of familiar drugs -- it takes time for the advantages and disadvantages of each one to become clear. In that spirit, studies continue to analyze the newest ophthalmic fluoroquinolone products. Here, we summarize some of the most recent efforts in this area.

Reaching the Aqueous Humor

One of the most common uses of fluoroquinolones is surgical prophylaxis, technically an off-label use. Because it's possible for bacteria to enter the anterior chamber during surgery, the ability of these drugs to penetrate into the aqueous humor has become a significant issue.

	Moxifloxacin HCL

To compare the ability of moxifloxacin (Vigamox) and gatifloxacin (Zymar) to achieve significant levels inside the anterior chamber, several recent studies have measured aqueous humor concentration following different protocols, using both human and animal models.

Mimicking post-op dosing preoperatively in humans. Drs. J. P. McCulley, G. Surratt and W. Shine, of the Department of Ophthalmology at the University of Texas in Dallas, are conducting an ongoing double-masked study designed to compare the penetration of moxifloxacin 0.5% and gatifloxacin 0.3% into aqueous humor in human patients following topical application.

Dr. McCulley explains the rationale behind the study: "Many studies done in the past did pre-op loading of a fluoroquinolone and then determined the aqueous concentration at the time of surgery. Pre-op loading is important on the surface, and to a degree in the corneal stroma, but it's not important relative to post-op aqueous concentrations. As soon as the surgeon enters the anterior chamber, the antibiotic and its concentration are lost. Furthermore, we're not treating our patients with that protocol post-op. We're going to be treating q.i.d. following surgery, so the effect of the q.i.d. protocol on the aqueous humor is what we want to know.

"Unfortunately, we can't find out by going back a day or two after cataract surgery to tap the aqueous humor. The risks would outweigh the benefit. So, to determine the benefits of post-op dosing, we arranged to use it preoperatively."

In this ongoing study, patients with no detectable disease who are scheduled for standard phacoemulsification instill one of the drops q.i.d. the day before surgery and 1 hour prior to the operation. Samples of aqueous humor are later chromatographically analyzed.

To date, based on 29 subjects, mean resulting concentrations are:

• moxifloxacin 0.5%: 1.93 mcg/ml
• gatifloxacin 0.3%: 0.26 mcg/ml.

For comparison, published values for other fluoroquinolones after similar dosing were:

• ciprofloxacin 0.3%: 0.07 mcg/ml
• levofloxacin 0.5%: 0.28 mcg/ml.

(Ofloxacin levels after q.i.d. dosing were not available.)

"In our study, gatifloxacin is getting above the MIC₉₀s of many, but not all organisms," Dr. McCulley points out, "and it's not sufficiently above the MIC₉₀s to ensure kill. In contrast, the moxifloxacin preparation gets above the MIC₉₀s of virtually all the organisms we're concerned about and achieves a kill concentration." (Note: Dr. McCulley has a financial relationship with Alcon Laboratories.)

Comparing gatifloxacin 0.3%, moxifloxacin 0.5%, and ciprofloxacin 0.3% in human patients. Drs. R. Solomon, E. Donnenfeld, H. Perry, R. Snyder, C. Nedrud, J. Stein and A. Bloom, working out of Ophthalmic Consultants of Long Island in Rockville Centre, N.Y., and the University of Arizona in Tucson, conducted a prospective, double-masked clinical comparison of aqueous concentration resulting from topical gatifloxacin 0.3%, moxifloxacin 0.5%, and ciprofloxacin 0.3% in human patients. In this study, 52 cataract patients used one of the three drops q.i.d for 3 days before surgery, and every 15 minutes for 1 hour before the procedure. Fluoroquinolone concentrations in fluid aspirated during the procedure were analyzed using reversed phase high-pressure liquid chromatography assay technique.

Resulting mean concentrations were:

• moxifloxacin 0.5%: 1.31 mcg/ml (SD 0.46)
• gatifloxacin 0.3%: 0.63 mcg/ml (SD 0.30)
• ciprofloxacin 0.3%: 0.15 mcg/ml (SD 0.11).

The authors note that anterior chamber levels of moxifloxacin and gatifloxacin may be due to the difference in antibiotic concentration and may not reflect an intrinsic difference in solubility. (Two of the seven authors have commercial relationships with both Allergan and Alcon.)

Comparing the effect of different pre-op and post-op dosing protocols using an animal model. Drs. J.M. Levine, R. Noecker, L. Herrygers and T. Clark at the University of Arizona in Tucson conducted an animal study to see what effect a change in pre- and post-operative dosing would have on aqueous levels of moxifloxacin and gatifloxacin.

The study was designed to determine aqueous levels of moxifloxacin and gatifloxacin following two different intensive topical dosing protocols that mimic different dosing regimens before and after cataract surgery. Both protocols included a 1-minute-long anterior chamber "washout" with balanced salt solution (BSS) intended to simulate anterior chamber irrigation during surgery.

"We looked at both the pre- and post-op concentration," says Dr. Levine. "Post-op dosing is performed to prevent endophthalmitis, but post-op penetration has never really been studied. It's likely affected by factors such as the temporal corneal wound and inflammation, which might in-crease penetration.

"Post-op penetration is hard to check if you can't bring the patient back and sample the aqueous again. So, we simulated cataract surgery in the rabbit by making a wound and washing out the eye; then we dosed again and sampled again."

In the study, 40 eyes of 20 New Zealand white rabbits received either topical gatifloxacin or moxifloxacin, following one of the two protocols:

► In the first protocol, the eyes were dosed q.i.d. for 4 days; the aqueous humor was sampled 30 to 60 minutes after the last dose. The eyes then received the washout, and the aqueous was sampled a second time 30 to 60 minutes later.

► The second protocol was adjusted to simulate additional intensive pre- and post-operative dosing. The eyes were dosed q.i.d. for 4 days; then each eye received four doses, 10 minutes apart. Ten minutes after the last dose, the aqueous was sampled. Following the washout with BSS, each eye received four more doses, 5 minutes apart. Five minutes after the last dose the aqueous was sampled again.

Mean aqueous fluoroquinolone concentration (assayed with HPLC) was:

► First protocol, before washout:

• gatifloxacin 0.3%: 0.269 mcg/ml
• moxifloxacin 0.5%: 2.111 mcg/ml.

After washout:

• gatifloxacin 0.3%: 0.229 mcg/ml
• moxifloxacin 0.5%: 0.619 mcg/ml.

► Second protocol, before washout:

• gatifloxacin 0.3%: 1.871 mcg/ml
• moxifloxacin 0.5%: 4.940 mcg/ml

After washout and 5-minute interval dosing:

• gatifloxacin 0.3%: 1.677 mcg/ml
• moxifloxacin 0.5%: 9.470 mcg/ml.

The authors noted that the dramatic differences in aqueous concentration following the two protocols shows that intensive dosing produces significantly higher pre- and postoperative antibiotic levels. Dr. Levine also notes that following the second protocol, the level of moxifloxacin was actually greater than the level at the time of surgery.

Although q.i.d. dosing of both fluoroquinolones achieved levels in excess of the reported MIC₉₀s for methacillin-sensitive S. aureus and Streptococcus sp., gatifloxacin did not achieve levels in excess of the MIC₉₀s for S. epidermidis (given by the authors as 2 mcg/ml) and methacillin-resistant S. aureus (2 to 4 mcg/ml) under either protocol.

"Both are excellent antibiotics, and both did achieve high levels of aqueous concentration post-op," says Dr. Levine. "Nevertheless, moxifloxacin achieved higher levels of both pre-op and post-op concentration, which probably relates to its higher formulated concentration.

"Both are good at killing organisms on the surface of the eye," he adds. "But if you're looking for post-op penetration in the eye to kill the organisms that do get in there, gatifloxacin did not reach the MIC₉₀ for S. epidermidis with typical post-op dosing. If you're going to use gatifloxacin following surgery, you may want to do more frequent dosing, perhaps every hour for the first day, to ensure a higher aqueous concentration." (Note: Dr. Noecker has a commercial relationship with Allergan, Inc.)

Gatifloxacin

Wound Healing and Epithelial Barrier Functioning

Several recent studies have noted differences in the effect that gatifloxacin and moxifloxacin have on protein synthesis in the epithelium, with resulting impact on wound healing and epithelial barrier functioning.

Gatifloxacin and moxifloxacin in the presence of dry eye. William Farley and Drs. Lihui Luo, Michael E. Stern and Stephen C. Pflugfelder conducted a study comparing the effect of topically applied gatifloxacin 0.3% and moxifloxacin 0.5% on corneal epithelial barrier function in the presence of experimentally induced dry eye.

To generate dry eye in C57BL6 mice, researchers injected 0.25 mg of scopolamine subcutaneously q.i.d.; they exposed the mice to an air draft in a low humidity environment for up to 5 days. Both eyes of each mouse were treated with 1 microliter of either 0.3% gatifloxacin or 0.5% moxifloxacin q.i.d.

At the end of the 5 days, researchers used a Fluorotron Master to measure corneal epithelial permeability to 0.5% sodium fluorescein in six eyes treated with each antibiotic, as well as six untreated eyes. They also used scanning laser confocal microscopy to evaluate expression of the tight-junction-associated protein ZO-1 in the corneal epithelium.

Two significant differences were found:

► Eyes treated with moxifloxacin showed a significant and progressive increase in corneal fluorescein permeability at days two and five. In contrast, eyes treated with gatifloxacin showed only a slight, non-significant increase in corneal fluorescein permeability at day two, and returned to baseline by day five. As a result, at day five, moxifloxacin-treated eyes showed significantly greater permeability than eyes treated with gatifloxacin.

► Eyes treated with gatifloxacin showed a normal honeycomb pattern of ZO-1 protein immunoreactivity on days 2 and 5. In contrast, moxifloxacin-treated eyes showed focal loss of membrane-associated ZO-1 staining at days 2 and 5.

Dr. Pflugfelder comments: "This study is significant because dry eye affects up to 10% of the population, some of them as a result of surgery such as LASIK. Many of these patients end up using these drops.

"Of the two drugs, gatifloxacin didn't have any significant effect in the presence of dry eye, but moxifloxacin left corneal barrier function worse than pretreatment levels. It appears that Zymar might be a little bit friendlier to the corneal epithelium than Vigamox for a patient with dry eye."

Dr. Pflugfelder notes that there could be a correlation between the reduced corneal barrier function associated with moxifloxacin and the high levels of the drug other studies find in the aqueous humor. "We found cellular changes in the tight junction of the epithelium, which would allow more molecules to diffuse into the eye. The higher concentration of drug inside the eye is partly due to the higher concentration in the formula, but it could also reflect the drug's breaking down the permeability barrier. We don't know that for sure, but it makes sense."

Dr. Pflugfelder also noted some parallels between his study and the next two regarding protein synthesis. (See comments below.)

Effect on corneal epithelial healing. Drs. J. Gao, K.F. Siemasko, C. Vu, H. Etemadi, P.J. McDonnell and M.E. Stern, working out of the Biological Sciences division of Allergan Inc. in Irvine, Calif., and the Johns Hopkins University School of Medicine in Baltimore, conducted a study designed to assess the impact of gatifloxacin and moxifloxacin on corneal wound healing in rabbits following anterior keratectomy (AK).

Eighteen female New Zealand white rabbits were randomly divided into three groups of six. Following 8-mm central corneal AK on both eyes, researchers used digital slit lamp microscopy to establish a base value of epithelial wounding. The rabbits were then topically treated t.i.d. with 0.3% gatifloxacin, 0.5% moxifloxacin, or the vehicle of gatifloxacin, until day five. During this period microscopy was performed twice a day at 9 a.m. and 4 p.m.

Corneas from half of the rabbits were collected at 48 hours; the rest were collected on day 5 (at 96 hours). Imaging software was used to assess the rate of corneal re-epithelialization by determining the ratio of epithelial defect area to baseline wound value at each time point. Researchers used immunohistochemistry and Western blot to evaluate the distribution and quantity of type IV collagen expression, to document the quality of epithelial healing.

Data showed:

► Compared with the vehicle, topical gatifloxacin caused no significant delay or acceleration in the rate of epithelial healing. In contrast, moxifloxacin delayed wound closure during the linear phase of epithelial healing (6 hours post-AK).

► Corneas treated with moxifloxacin showed healing delays of 13.25%, 15.76% and 16.49% at 72, 78 and 96 hours post-AK, respectively, compared with the vehicle.

► Moxifloxacin-treated corneas showed markedly diminished collagen IV expression at 48 and 96 hours, especially at the leading edge and base of the wound. Quantitative analysis found that moxifloxacin-treated corneas had 30.5% fewer collagen IV proteins from newly migrating corneal epithelial cells than corneas treated with gatifloxacin.

The authors conclude that although topical gatifloxacin treatment did not delay corneal epithelial healing or decrease collagen IV expression following AK, topical moxifloxacin treatment caused delayed epithelial healing. The authors believe that this is probably due to the suppression of collagen IV expression in migrating corneal epithelial cells. They add that this data could be a cause for concern in terms of epithelial cell adhesion and the healing of corneal lesions.

Effect on corneal wound integrity. Drs. L. Pedroza-Schmidt and R. Beuerman at the LSU Eye Center in New Orleans, La., conducted a study to evaluate the effects of gatifloxacin and moxifloxacin on corneal wound integrity in a rabbit model. Researchers made a 6-mm linear incision, similar to a clear cornea incision, in the eyes of 12 New Zealand white rabbits; the wounds were closed with two nylon 10-0 sutures. One eye of each rabbit was treated with either gatifloxacin 0.3% or moxifloxacin 0.5%, administered as recommended in the respective package inserts. Contra-lateral eyes served as the control; they were dosed with BSS at a frequency matching that of the fluoroquinolone.

Corneas were checked using white-light confocal microscopy at 1, 4 and 7 days after wounding. Rabbits were euthanized 7 days after wounding and all corneas were collected and prepared for light and transmission electron microscopy (TEM). The structural analysis was carried out in a blinded fashion.

Confocal microscopy of the wounds suggested that moxifloxacin treatment resulted in less stromal healing at 7 days; this was verified by histology. Other findings included:

► Both antibiotics slowed the process of stromal closure when compared to control treated wounds. However, the epithelial plug was expressed more quickly with stromal closure following gatifloxacin treatment and there was renewal of mature epithelial organization.

► TEM revealed that moxifloxacin-treated corneas had protein precipitation just below the basal lamina, which appeared to be thin, with few anchoring filaments. This protein precipitation was seen consistently outside the wound within 2 to 3 mm of the incision. In contrast, the gatifloxacin-treated wounds appeared similar to the contra-lateral controls, which displayed essentially normal structure along the basal lamina.

► Gatifloxacin produced significantly less keratocyte response and stromal disorganization than moxifloxacin.

The study's authors concluded that gatifloxacin caused significantly less disruption in corneal wound integrity than moxifloxacin. (Note: This study was supported by an unrestricted grant from Allergan, Inc.)

Dr. Pflugfelder notes that this study could have implications for humans following corneal transplant, corneal cataract incisions, corneal laceration or other corneal surgeries. "Treatment with moxifloxacin could delay wound healing, which could cause an irregular surface or haze, or eventually result in a larger scar," he says.

Dr. Pflugfelder sees some interesting correlations between his study and the last two. "All of these studies look at different proteins, but the results run along the same lines. In our study moxifloxacin appeared to be interfering with the production of the tight junction protein; in Dr. Gao's study it clearly interfered with the production of collagen type IV. In Dr. Pedroza-Schmidt's study, eyes treated with moxifloxacin showed clumping of proteins underneath the basement membrane with no normal basement membrane structure.

"It's possible that moxifloxacin has a toxic effect on protein synthesis in the corneal epithelium," he concludes. "This might be a function of the higher concentration or intrinsic differences in the molecule.

"If there is a slightly toxic effect, the epithelium should recover once the medication is stopped, and most people aren't treated with these drugs for prolonged periods. Nevertheless, the possibility of toxicity is a cause for concern."

	Levofloxacin hemihydrate

A New Alternative for Corneal Ulcer Treatment

The FDA recently approved a new formulation of levofloxacin with a concentration of 1.5% -- the highest concentration of any ophthalmic fluoroquinolone currently available. The new formulation, which will be marketed under the name Iquix, has been approved for the treatment of corneal ulcer.

Like gatifloxacin and moxifloxacin, levofloxacin attacks bacteria at two locations to prevent replication: DNA gyrase and topoisomerase IV. Levofloxacin has been shown to be effective against a wide range of organisms, including Staphylococcus aureus, Streptococcus pneumoniae and Pseudomonas aeruginosa, and clinical studies have demonstrated that tear film concentrations far exceed the MIC₉₀ values of key pathogens.

In two randomized, double-masked multi-center controlled clinical trials, Iquix achieved a clinical cure rate of 73% to 87% in patients with bacterial corneal ulcer. (The most frequently reported adverse events in the overall study population were headache and a taste disturbance following instillation, occurring in 8% to 10% of patients.)

Iquix contains no preservatives, and a series of four animal studies conducted by Clark et al. found no evidence of toxicity.

Although it is currently only approved for corneal ulcer, animal studies have shown that the new formulation penetrates the cornea well, creating high levels in the aqueous. How effective the new formulation will be at surpassing the MIC₉₀s for key bacteria inside the eyes of human patients remains to be established.

In the meantime, Iquix appears to be an effective new option for treating corneal ulcers.