1. The Role of Benzalkonium Chloride in the Occurrence of Punctate Keratitis: A Meta- Analysis of Randomized, Controlled Clinical Trials S Trocme, MD, 1 L-J Hwang, PhD, 2 G W Bean, BS, 3 M B Sultan, MD, MBA 2,4 1 Department of Ophthalmology and Visual Sciences, Case Western Reserve University and University Hospitals Eye Institute, Cleveland, Ohio; 2 Pfizer Inc, New York, New York; 3 Independent Consultant, Burkett, Texas; 4 New York Eye & Ear Infirmary, New York, New York Dr. Trocme declares no financial interest; Drs. Hwang and Sultan are employees of Pfizer Inc; Mr. Bean is a paid consultant to Pfizer Inc. This study was supported by Pfizer Inc.

2. Introduction / Background  Although benzalkonium chloride (BAK) is the most commonly used preservative, it has undergone considerable criticism based on in vitro and in vivo studies 1-5  Evidence of dose-dependent BAK-induced epithelial cell damage has been confirmed in cultured corneal 6 and conjunctival cells 7 and in cat and rabbit studies 3,4,8,9  At common clinical concentrations, however, not all preclinical studies demonstrate toxicity 10,11 and their relevance has not been confirmed in large clinical trials  Clinically, BAK has been associated with corneal epithelial cell injury manifesting as punctate keratitis 12-19  If punctate keratitis is caused by BAK at common concentrations in IOP-lowering drugs then increasing the amount of BAK exposure daily should cause more pathology References on notes page

3. BAK Concentrations in IOP-Lowering Ophthalmic Solutions by Daily Dose Drug BAK/ mL (%) Drops/ mL BAK µg /drop Dosing per PI Daily BAK µg Apraclonidine0.0132.93.03-6/day9.0-18.0 Betaxolol0.0128.53.51-4/day3.5-14.0 Bimatoprost0.00535.21.41/day1.4 Brimonidine0.00522.22.33/day6.9 Brinzolamide0.0125.14.03/day12.0 Dorzolamide0.007527.12.83/day8.4 Latanoprost0.0228.07.11/day7.1 Pilocarpine0.0123.04.32-8/day8.6-34.4 Timolol0.0132.53.12/day6.2 Travoprost0.01534.54.31/day4.3

4. Purpose To determine the incidence of punctate keratitis reported in double-masked trials comparing latanoprost and timolol ophthalmic solutions in which the amount of BAK was twice as much in the latanoprost arm as in the timolol arm

5.  A PubMed search of prospective, randomized, double-masked clinical trials comparing the efficacy and safety of latanoprost and timolol ophthalmic solutions in glaucoma or ocular hypertension for ≥6 months duration  The number of reports of punctate keratitis either as a finding or an adverse event were identified in each trial and were meta-analyzed  Incidence was calculated and forest plots generated to summarize risk difference and odds ratios Methods

6. Results 7 Studies Identified in PubMed Search No. of Patients and Dosing Frequency StudyDurationLatanoprost/VehicleTimolol Alm 6 months 183 QD/QD 84 BID Camras 6 months 128 QD/QD 140 BID Watson 6 months 149 QD/QD 145 BID Lass 12 months 127 QD/none 126 QD Pfeiffer 6 months 147 QD/QD 149 BID Higginbotham 6 months 140 QD/QD 140 BID Mastropasqua 12 months 18 QD/QD 18 BID Alm A, et al. Ophthalmology 1995;102:1743-52; Camras CB. Ophthalmology 1996;103:138-47; Watson PG, et al. Ophthalmology 1996;103:126-37; Lass JH, et al. 2001;108:264-71; Pfeiffer N, et al. Graefes Arch Clin Exp Ophthalmol 2002;240:893-9; Higginbotham EJ, et al. Arch Ophthalmol 2002;120:915-22; Mastropasqua L, et al. Ophthalmology 1999;106:550-5.

7. Results (cont’d)  Timolol maleate (0.01% BAK) was given twice daily and latanoprost (0.02% BAK) was given once daily with vehicle (0.01% [in 2 studies] or 0.02% BAK) given once daily in 6 of the 7 trials; 1 study dosed both drugs once daily  Of the 1694 patients enrolled in the double-masked portion of the trials (latanoprost, N=892; timolol, N=802) the overall incidence of punctate keratitis was 6.3% (106/1694)  The incidence of punctate keratitis was 6.5% and 6.0% in latanoprost- and timolol-treated patients, respectively, similar to rates reported in glaucoma and ocular hypertension patients treated with preservative-free eye drops

8. Incidence of Punctate Keratitis (PK) and Amount of BAK Study Latanoprost: Incidence (PK/Total) Timolol: Incidence (PK/Total) Amount of BAK ( µg/day): Latanoprost/Vehicle vs Timolol Twice Daily Alm 6.6% (12/183) 2.4% (2/84) 14.2 vs 6.2 Camras 13.3% (17/128) 17.9% (25/140) 14.2 vs 6.2 Watson 12.8% (19/149) 6.9% (10/145) 14.2 vs 6.2 Lass 3.9% (5/127) 5.6% (7/126) 7.1 vs 3.1* Pfeiffer 0.7% (1/147) 1.3% (2/149) 10.2 vs 6.2 Higginbotham 2.9% (4/140) 1.4% (2/140) 10.2 vs 6.2 Mastropasqua 0% (0/18) 14.2 vs 6.2 Total 6.5% (58/892) 6.0% (48/802) ~2 vs 1 * Latanoprost and timolol were dosed once daily

9. Combined Risk Difference for Punctate Keratitis: Latanoprost – Timolol

10. Combined Odds Ratio for Punctate Keratitis: Latanoprost Versus Timolol

11. Conclusions  To our knowledge, this meta-analysis is the first to evaluate the corneal effects of variable BAK concentrations in ophthalmic solutions  Although containing a higher concentration of BAK, latanoprost ophthalmic solution was not associated with a significantly increased risk of punctate keratitis vs timolol ophthalmic solution  The incidence of punctate keratitis found in this study (6%) was similar to that (8.9%) reported in patients treated with preservative-free eye drops 1 1 1 Jaenen N, et al. Eur J Ophthalmol 2007:17:341-9.

12. Conclusions (cont’d)  Study limitations  evaluation of punctate keratitis was not pre-specified in any of the trials  standardized grading for lesion severity was not used  patients with severe corneal pathology were not enrolled in the included studies  These results indicate that BAK does not produce significant corneal toxicity in the vast majority of patients with glaucoma or ocular hypertension at the concentrations used in these studies