Inhibition of COX in ocular tissues: an in vitro model to identify selective COX-2 inhibitors

Ocular penetration and anti-inflammatory activity of ketorolac 0.45% and bromfenac 0.09% against lipopolysaccharide-induced inflammation

PURPOSE

Anti-inflammatory activity of topical nonsteroidal anti-inflammatory drugs is mediated by suppression of cyclooxygenase (COX) isoenzymes. This study compared ocular penetration and inflammation suppression of topical ketorolac 0.45% and bromfenac 0.09% ophthalmic solutions in a rabbit model.

METHODS

At hour 0, 36 rabbits received ketorolac 0.45%, bromfenac 0.09%, or an artificial tear 3 times once every 20 min. Half of the rabbits in each group then received intravenous injections of lipopolysaccharide (LPS) and fluorescein isothiocyanate (FITC)-dextran at hour 1, and the other half at hour 10. Aqueous and iris-ciliary body (ICB) samples were collected in the former group at hour 2 (peak) and in the latter group at hour 11 (trough) An additional group of 6 animals received only FITC-dextran, and samples were collected 1 h later. Peak and trough nonsteroidal anti-inflammatory drug concentrations were compared with previously determined half-maximal inhibitory concentrations (IC(50)) for COX isoenzymes.

RESULTS

Peak and trough aqueous and ICB concentrations of ketorolac were at least 7-fold or greater than those of bromfenac. At peak levels, both ketorolac 0.45% and bromfenac 0.09% significantly inhibited LPS-induced aqueous prostaglandin E(2) and FITC-dextran elevation (P < 0.01). At trough, both study drugs significantly inhibited LPS-induced aqueous prostaglandin E(2) elevation (P < 0.05), but only ketorolac 0.45% significantly reduced LPS-induced aqueous FITC-dextran elevation (P < 0.01). Aqueous and ICB ketorolac concentrations exceeded its IC(50) for COX-1 and COX-2 at peak and trough. Aqueous and ICB bromfenac levels exceeded its IC(50) for COX-2 at peak and trough, but not for COX-1 at trough aqueous levels and peak and trough ICB levels.

CONCLUSIONS

Both ketorolac 0.45% and bromfenac 0.09% effectively suppressed inflammation at peak. At trough, only ketorolac 0.45% effectively suppressed inflammation as measured by FITC-dextran leakage. The difference in inflammation suppression may be due to differences in tissue concentrations and/or greater COX-1 suppression by ketorolac 0.45%.

The expression of cyclooxygenase 2 in retinoblastoma: primary enucleated eyes and enucleation after conservative treatment

PURPOSE

To provide insight into the relationship between cyclooxygenase-2 (COX-2) expression and histopathologic features of retinoblastoma specimens treated either by primary or secondary enucleation.

DESIGN

Laboratory investigation.

METHODS

Twenty-five retinoblastoma specimens received between 1994 and 2003 were retrieved for this study from the Ocular Pathology Registry, Federal University of São Paulo, Brazil and the Henry C. Witelson Ocular Pathology Laboratory and Registry, McGill University, Montreal, Canada. The specimens retrieved were divided into two groups: Group I, enucleation was performed as a form of primary treatment (n = 15) and Group II, enucleation after failure of conservative treatment (n = 10). Patient information and formalin-fixed, paraffin-embedded specimens were obtained. New sections of these blocks were used for hematoxylin and eosin stain and for immunoassay using a monoclonal mouse anti-COX-2 antibody. Two independent ophthalmic pathologists reviewed all of the microslides.

RESULTS

Twenty-three specimens (92%) presented a high expression of COX-2 (15 in Group I; eight in Group II) and there was no statistical difference in COX-2 expression between the two groups (P = .07). However, all specimens expressed COX-2 to a different degree. The areas of tumor invasion were positive for COX-2 in 87.5% of the two groups.

CONCLUSION

Most retinoblastoma specimens revealed a high COX-2 expression. Future studies will be necessary to correlate the high expression of COX-2 in retinoblastoma and a possible applicability of anti-COX-2 medications in the treatment of these tumors.

Rapid determination of nimesulide in rabbit aqueous humor by liquid chromatography

A rapid method was developed for quantification of nimesulide (methanesulfonamide, N-[4-nitro-2-phenoxyphenil]) in rabbit aqueous humor. The analyses were performed by high-performance liquid chromatography using a C(18) reversed-phase column (Ultracarb ODS) with UV detection at 300 nm. The mobile phase consisted of acetonitrile-water containing 1% triethylamine (TEA) adjusted to pH 3.2 with orthophosphoric acid. The retention time was 4.5 min. A simple pre-treatment with acetonitrile was used to deproteinize aqueous humor samples. The limit of quantitation was 50 ng/ml. The recovery was over 90%. The relationship between peak areas and concentration was linear over the range between 0.05 and 2.5 microg/ml, with r(2) values over 0.99. The assay provided good reproducibility and accuracy and proved to be suitable for pharmacokinetic studies of nimesulide.