Neovascular age-related macular degeneration treated with long-term use of intravitreal ranibizumab (Lucentis, Genentech) was safe and well tolerated in a group of patients participating in an extension study. Between the original studies and the extension trial, patients were followed for approximately 4 years, during which no safety signals were observed. Although problems linked to vascular endothelial growth factor inhibition were suggested, the lack of a control group makes proving a correlation difficult.

Key Points

Los Angeles-Repeated intravitreal injections of ranibizumab (Lucentis, Genentech) appear to be safe and well tolerated over long-term use in patients with neovascular age-related macular degeneration (AMD). In a study following patients for 4 years, no systemic safety signals emerged. Although problems linked to vascular endothelial growth factor inhibition were suggested, the lack of a control group makes it difficult to prove a correlation, said David S. Boyer, MD, who presented 3-year results from Extension Study 2508 of ranibizumab.

Describing the background of the multicenter, open-label extension trial, Dr. Boyer, a retina-vitreous specialist in private practice in the Los Angeles area, said that the drug's manufacturer first sponsored three open-label phase I/II safety studies of ranibizumab (studies 1770, 2128, and 2425) using a lyophilized formula of the drug rather than its current formulation.

Patients participating in the initial studies received a single dose, multiple doses, or an escalating dose. If patients lost fewer than 15 letters during these studies, then they could continue receiving treatment on a compassionate use basis in the extension study.

The primary outcome was the incidence and severity of ocular and nonocular adverse events, and the secondary outcome was best-corrected visual acuity measured by number of letters on the ETDRS chart. VA was evaluated as a safety measure rather than a gauge of effectiveness because there were periods of time when patients were not receiving medication, Dr. Boyer said.

Of the 120 patients in the early safety trials, 70 were eligible to continue in the compassionate use study; 67 received treatment, and the remaining three were enrolled after participation became optional and never required further injections.

The mean duration of follow-up in the extension study was 2.7 years, and the median was 3.1 years. Adding the time spent in the initial studies, the mean and median follow-up periods were 3.8 and 4.3 years, respectively.

Injection variability

The frequency of injections varied depending on whether patients were treated before or after the amendment to the protocol that shifted injections from a mandatory monthly schedule to discretionary. When patients were treated only if their vision became unstable, injections were given every 3 to 4 months. About 28% of patients required no treatment during the extension study.

Variability also occurred in the total number of injections. In the initial study, patients received an average of seven injections, Dr. Boyer said, whereas in the extension study, the range was from one to 44.

In all, some patients received as many as 52 injections.

Reviewing the adverse events, Dr. Boyer told the audience that the most prominent problem was vitritis, which occurred in approximately 12% of the treated eyes but in none of the fellow eyes.

"This was, I felt, very high and may indicate that prolonged use may contribute to immunogenicity and there may be a problem," Dr. Boyer said. "But you have to take this with a grain of salt because the lyophilized form [of ranibizumab] was used in the beginning, and that may be a lot more sensitizing."

Retinal hemorrhages also were common, occurring in 27% of the study eyes versus 9% of the fellow eyes. Adverse events classified as serious included VA reduction, severe iridocyclitis, and, in one case, retinal hemorrhage.

The systemic effects included arterial thrombotic events (ATEs), which occurred in seven patients. Individuals in this subgroup experienced multiple events, including hypertension in 36% of these patients as well as noncerebral, nonocular hemorrhages in four patients. The hypertension rate may be lower than the average in the general population for this age group, Dr. Boyer reported.

The rate of ATEs was 4.1% in the first year of the extension study. No pattern or association could be found between these events and the number or frequency of injections, however, Dr. Boyer said.

He also mentioned that gains in VA experienced since the start of the initial ranibizumab studies were maintained in the extension trial.