Purpose: To determine the proportion of patients with primary open-angle glaucoma (POAG) or ocular hypertension (OH) who do not respond to latanoprost therapy. Methods: Three hundred forty consecutive patients with a new diagnosis of POAG or OH, or previously treated only with a beta-blocker and after an appropriate washout period, were treated with latanoprost for 1 month and then divided into three groups on the basis of the reduction in intraocular pressure (IOP): nonresponders (< 15 %), responders (&rt;= 15% but < 30%), and high-responders (&rt;= 30%). To give a wide picture of the drug effect, eight different cut-off points were used to present data on distribution of mean IOP reductions. Only the nonresponders entered a randomized cross-over study investigating the efficacy of timolol, brimonidine, and pilocarpine. Results: IOP at baseline and after 1 month's latanoprost therapy was respectively 24.1 +/- 1.4 and 16.9 +/- 2.4 mm Hg, with a mean reduction in IOP of 29.9 +/- 4.2%. Nonresponders accounted for 4.1% of the patients and high-responders for 41.2%. The nonresponders showed a statistically significant reduction in IOP after brimonidine treatment (P= 0.05), whereas the reduction after timolol and pilocarpine treatment was clinically relevant but not statistically significant. Conclusions: This multicenter prospective study found only 14 of 340 nonresponders to latanoprost. In the cross-over trial on nonresponders, IOP reduction reached statistical significance only after brimonidine, but their small number reduced its statistical power.