ABSTRACT
PURPOSE: To compare, in vitro, the cytotoxicity profile of a new formulation of travoprost 0.004%/timolol 0.5% fixed combination ophthalmic solution preserved with polyquaternium-1 0.001% (travoprost/timolol PQ) instead of benzalkonium chloride (BAK) with (1) commercially available travoprost 0.004%/timolol 0.5% fixed combination ophthalmic solution (travoprost/timolol BAK), (2) commercially available latanoprost 0.005%/timolol 0.5% fixed combination ophthalmic solution (latanoprost/timolol BAK), and (3) their associated BAK concentrations. METHODS: Compounds tested on Wong-Kilbourne-derived human conjunctival epithelial cells: (1) phosphate-buffered saline, (2) polyquaternium-1 0.001% (Polyquad(®), PQ), (3) travoprost/timolol PQ, (4) travoprost/timolol BAK with 0.015% BAK (DuoTrav(®)), (5) BAK 0.015%, (6) latanoprost/timolol BAK with 0.020% BAK (Xalacom(®)), and (7) BAK 0.020%. Toxicological assays were used to assess cell viability [neutral red (NR), Alamar blue (AB)], apoptosis (YO-PRO-1, Hoechst 33342), and oxidative stress (H(2)DCF-DA, hydroethidine). The apoptosis and oxidative stress assays were each reported according to cell viability as observed with NR and AB (totaling 10 analyses per treatment). RESULTS: The NR and AB assays demonstrated that cells incubated with travoprost/timolol PQ had significantly better viability than cells incubated with latanoprost/timolol BAK, travoprost/timolol BAK, BAK 0.015%, and BAK 0.020% (P<0.0001 for all). As assessed with YO-PRO-1 and Hoechst 33342 relative to cell viability determined with NR or AB, travoprost/timolol PQ produced significantly less apoptosis than travoprost/timolol BAK and latanoprost/timolol BAK and their respective BAK concentrations alone (P<0.0001 for all). Also, travoprost/timolol BAK induced less apoptosis than latanoprost/timolol BAK (P<0.0001). As assessed with H(2)DCF-DA as a ratio to NR or AB, all of the compounds without BAK (phosphate-buffered saline, PQ 0.001%, and travoprost/timolol PQ) and travoprost/timolol BAK produced significantly less reactive oxygen species than latanoprost/timolol BAK (P<0.0001 for all). As assessed with hydroethidine as a ratio to NR or AB, travoprost/timolol PQ produced significantly fewer superoxide anions than latanoprost/timolol BAK (P<0.0001). In contrast, release of superoxide anions (hydroethidine method) after incubation with travoprost/timolol BAK was not significantly different from incubation with latanoprost/timolol BAK or travoprost/timolol PQ. CONCLUSION: Travoprost/timolol PQ may be better for ocular surface health than either BAK preserved formulations of latanoprost/timolol or travoprost/timolol.
