In a study performed to evaluate and compare preserved versus unpreserved fluoroquinolone antibiotics, cell culture assays demonstrated some degree of toxicity and elevated induction of apoptosis in each of three commercially available topical ocular fluoroquinolone antibiotics.
Fort Lauderdale, FL-In a study performed to evaluate and compare preserved versus unpreserved fluoroquinolone antibiotics, cell culture assays demonstrated some degree of toxicity and elevated induction of apoptosis in each of three commercially available topical ocular fluoroquinolone antibiotics, according to Seth P. Epstein, MD, who presented a poster on his work with colleagues at the annual meeting of the Association for Research in Vision and Ophthalmology.
Of the three, nonpreserved levofloxacin ophthalmic solution 1.5% (Iquix, Santen/Vistakon Pharmaceuticals) induced the least toxicity and apoptosis, followed closely by nonpreserved moxifloxacin ophthalmic solution 0.5% (Vigamox, Alcon Laboratories), while gatifloxacin ophthalmic solution 0.3% (Zymar, Allergan), which is preserved with 0.005% benzalkonium chloride (BAK), had the greatest levels of both toxicity and apoptosis, said Dr. Epstein, assistant professor of ophthalmology, Mount Sinai School of Medicine, New York, NY.
"There's been a great deal in the literature about the toxicity of BAK," Dr. Epstein said. "Previous work in our lab has determined that of the five classes of preservatives that are in use, BAK is one of the most toxic. It also is one of the most commonly used. Of course, this is somewhat understandable since preservatives are added to prevent contamination; preservatives need to be toxic, at least to single-cell microorganisms."
The averaged results of the cell lines showed that nonpreserved levofloxacin induced the least toxicity (40%), followed by moxifloxacin (48%), and preserved gatifloxacin (96%). The three antibiotics also were compared with raw levofloxacin 1.5% and BAK at the same concentration (0.005%) found in the gatifloxacin formulation, both in appropriate medium, and several controls: medium alone, formalin, and DNAse.
"It turned out that the toxicity of the raw levofloxacin was approximately 37%, only slightly below the formulated Iquix, which is surprising because the formulation would be expected to make the product less toxic," Dr. Epstein said. Greater toxicity would have been expected in the raw levofloxacin because it was not buffered or pH balanced.
Raw moxifloxacin and gatifloxacin could not be obtained for the study; the team instead tested ciprofloxacin and ofloxacin to determine whether their results were an artifact. They found that the toxicity level of ciprofloxacin was about 40%, similar to that of nonpreserved levofloxacin, while the ofloxacin 0.3% was more toxic.
As found with the MTT cytotoxicity assay, the results of the apoptosis assay demonstrated that nonpreserved levofloxacin induced the lowest apoptosis, followed closely by nonpreserved moxifloxacin. Figures resulting from the averaged cell lines were 12% for nonpreserved levofloxacin, 30% for moxifloxacin, and 58% for gatifloxacin.
The researchers also found that about 40% of the toxicity observed in the MTT assay could be accounted for by the apoptosis rather than by the true toxicity of the pharmaceutical agents.
In the future, the researchers hope to study why the levofloxacin in the raw form was approximately equal in toxicity to the formulation.
"Is this because the formulation isn't really the ideal and it can be made less toxic, or is the levofloxacin in its raw form really not all that toxic in that it does not severely alter the pH or the osmolality of the solution, which of course would affect the eyes?" Dr. Epstein said.
He also observed that while the tissue culture model was a rapid and cost-effective method of screening for toxicity, the results need to be correlated with in vivo wound healing and clinical findings.
Seth P. Epstein, MD
Phone: 212/241-6500 E-mail: firstname.lastname@example.org
Dr. Epstein has a commercial relationship with Vistakon Pharmaceuticals.