Glaucoma: Bimatoprost tolerability appears to be enhanced with switching strategy

Key Points

San Diego-Patients whose therapy is switched from latanoprost 0.005% (Xalatan, Pfizer) to bimatoprost 0.03% (Lumigan, Allergan) appear to tolerate bimatoprost better than if their prostaglandin analogue treatment was initiated with bimatoprost, said Jason Bacharach, MD, at the annual meeting of the American Glaucoma Society.

Dr. Bacharach's observation was based on outcomes from a multicenter, investigator-masked, randomized, parallel-group "switch" study that was designed to evaluate the efficacy and tolerability of bimatoprost compared with latanoprost in patients treated initially with latanoprost, along with data from other bimatoprost trials. He is associate clinical professor of ophthalmology, California Pacific Medical Center, San Francisco, and medical director, North Bay Eye Associates, a multispecialty private practice in Sonoma County, CA.

The switch trial enrolled 586 patients with glaucoma or ocular hypertension and an IOP expected to be controlled with prostaglandin analogue monotherapy. After washout of any existing IOP-lowering medication and a 6-week run-in on once-nightly latanoprost monotherapy, 543 patients were randomly assigned to 12 weeks of once-nightly treatment with latanoprost or bimatoprost.

At 6 weeks after randomization (week 12 of the study), significant lowering of IOP was achieved in both treatment groups compared with the untreated baseline, and the benefit was maintained at week 18. At both postrandomization visits, the IOP-lowering effect was significantly greater in the bimatoprost group, although the absolute difference compared with the latanoprost group was small (~1 mm Hg).

Both bimatoprost and latanoprost groups demonstrated a high degree of safety. Although a slightly larger percentage of patients in the bimatoprost arm developed moderate to severe hyperemia compared with the patients who continued to take latanoprost, 3.3% versus 1.8%, respectively, the difference was not statistically significant.

In addition, the mean hyperemia score for the bimatoprost group remained below "trace," and no patients in either treatment group withdrew from study participation due to hyperemia.

Also, no statistically significant differences were observed between study groups in overall adverse event rates or discontinuation rates.

Consistent findings

"The results of this trial provide information that is important for practitioners, considering that many patients must be switched off latanoprost due to managed care constraints and formulary restriction," he said. "Its findings are consistent with previous studies showing that the risk of hyperemia in bimatoprost-treated patients may be reduced if they initiate prostaglandin analogue treatment with latanoprost.

"Considering that in the pivotal trial leading to FDA approval of bimatoprost, 3.4% of patients discontinued treatment due to hyperemia compared with no patients in this study, it appears that patients may do even better when treatment is switched to bimatoprost from latanoprost than when it is started with bimatoprost as first-line therapy," he added.

About 97% of patients in each treatment group completed the study. The two study groups were comparable at randomization in their demographic characteristics, diagnoses, history of use of previous ocular hypotensive medication, and IOP measurements.

IOP was recorded at each visit at 8 a.m., noon, and 4 p.m. For efficacy determination, a diurnal IOP was calculated for each eye (average of the three time points), and the mean diurnal IOP for both eyes of each patient was used for the statistical analysis of IOP response.

Careful design

Dr. Bacharach said that the study was carefully designed, taking into consideration potential flaws/biases in prior "switch" studies. Because all patients who started the latanoprost run-in were eligible for randomization regardless of IOP response, patients were stratified for randomization based on both baseline IOP (≤25 mm Hg versus >25 mm Hg) and response to latanoprost (<20% IOP versus ≥20%). The baseline (untreated) mean diurnal IOP was about 22 mm Hg in both treatment groups.

In addition, by having two parallel study cohorts-latanoprost as the active control for the group switched to bimatoprost-the potential for confounding of IOP results from regression to the mean was addressed.

"With this design, any IOP differences between study groups are more likely to be related to differences in medication effects and not regression," Dr. Bacharach concluded.