Study finds no VA benefit by adding corticosteroid implant to anti-VEGF injections

By Cheryl Guttman Krader; Reviewed by Raj K. Maturi, MD

Results from the Diabetic Retinopathy Clinical Research Network (DRCR.net) Protocol U are providing some insights on the management of diabetic macular edema (DME) that does not respond optimally to regular anti-VEGF injections.

The study found that many patients will benefit from continuing with anti-VEGF treatment using ranibizumab 0.3 mg (Lucentis, Genentech). While the combination therapy with the addition of the sustained, dexamethasone drug-delivery system 0.7 mg (Ozurdex, Allergan) resulted in greater reduction in retinal thickness compared with anti-VEGF monotherapy, it was not associated with any functional benefit and exposed patients to the known corticosteroid-related risks.

Raj K. Maturi, MD, presented the results at Retina Subspecialty Day 2017 and they were published concurrently in JAMA Ophthalmology.

“About 40% of patients with DME continue to have retinal thickening and reduced vision despite regular anti-VEGF injections,” said Dr. Maturi, clinical associate professor of ophthalmology, Indiana University School of Medicine; private practice, Midwest Eye Institute, both in Indianapolis. “The results of Protocol U suggest that many of these individuals will improve by continuing the same treatment for another 3 to 6 months.

“Previous studies have consistently shown that corticosteroids reduce retinal thickening in eyes with DME, and in Protocol U, there was greater reduction in macular thickness when adding the corticosteroid implant to anti-VEGF therapy,” Dr. Maturi added. “That benefit, however, must be balanced by the lack of functional benefit and the high incidence of intraocular pressure (IOP) rise that also occurs with this treatment.”

Results of a preplanned subgroup analysis of best-corrected visual acuity (BCVA) outcomes based on phakic status suggested that corticosteroid-induced cataractous changes may have limited visual acuity improvement in the combination therapy group. However, Protocol U was not sufficiently sized to determine if pseudophakic patients would have greater functional benefit with combination treatment, Dr. Maturi said.

“The study was originally designed to enroll only pseudophakic patients, but it was expanded to include phakic eyes because enrollment was very slow,” Dr. Maturi pointed out. “Therefore, we believe it would be difficult to conduct another larger study to investigate any benefit of combination therapy in pseudophakic patients.”

Study design

Protocol U was a phase II study conducted at 40 sites across the United States. Patients were eligible for enrollment if they had electronic Early Treatment Diabetic Retinopathy Study (e-ETDRS) BCVA between 78 letters and 24 letters (approximate Snellen equivalent, 20/32 to 20/320) and central-involved DME on clinical exam and optical coherence tomography (OCT), based on sex-specific criteria, despite at least 3 anti-VEGF injections in the prior 20 weeks. 

Patients entered a run-in phase during which they received 3 monthly injections of ranibizumab. Eyes with persistent edema and vision loss were stratified according to changes in BCVA and retinal thickness during the run-in period and randomized 1:1 to receive ranibizumab with sham or dexamethasone implant injection.

“Of 236 patients enrolled in to the run-in phase, 129 were randomized to treatment,” Dr. Maturi reported.

Patients returned for monthly follow-up visits. Ranibizumab was injected if the BCVA was <84 letters (approximate Snellen equivalent, ≤20/25) or the OCT central subfield thickness (CST) was at or above the protocol-defined, sex-specific cutoff. A second dexamethasone implant or sham injection could be given beginning at week 12, and patients were followed to week 24.

Outcomes

The combination and monotherapy groups were well-balanced in most of their baseline characteristics. During the run-in phase, BCVA improved by a mean of 3 letters and CST was reduced by about 50 µm in both groups.

Mean CST at randomization was 375 µm in eyes randomized to combination treatment and 396 µm for the ranibizumab group. CST was ≥350 µm in about 50% of eyes in both groups. The percentage of phakic eyes was greater in the combination versus monotherapy group (60% versus 50%).

Study retention was excellent. All patients in the monotherapy group completed the 24-week visit as did 97% of patients in the combination group. Mean number of ranibizumab injections was 5.7 in the ranibizumab group and 5.6 in the combination group, and all but 2 patients in both groups received a second sham or dexamethasone injection.

Change in BCVA from the randomization visit to 24 weeks was analyzed as the primary outcome measure. It averaged 3.0 letters in the ranibizumab group and 2.7 letters in the combination group. Th adjusted treatment group difference was not statistically significant.

Results of preplanned subgroup analyses showed that patients receiving the dexamethasone implant gained 5.1 BCVA letters if they were pseudophakic but only 1.1 letters if they were phakic. Mean BCVA gain in eyes treated with ranibizumab only was greater in phakic versus pseudophakic eyes (4.1 letters versus 2.0 letters), Dr. Maturi reported.

The proportion of eyes achieving a BCVA gain of ≥15 letters was significantly greater in the combination group than in eyes receiving monotherapy with ranibizumab (11% versus 2%, P = 0.03). However, combination treatment was also associated with a higher rate of BCVA loss ≥10 letters, and the difference showed a trend to being not statistically significant (13% versus 6%, P = 0.09).

Analysis of OCT data showed significantly greater reduction in CST in eyes receiving combination therapy compared with those that continued on anti-VEGF treatment alone (110 µm versus 62 µm, P < 0.001).  The percentage of eyes achieving a “flat retina” (normal CST values based on the sex-specific OCT criteria) was also significantly greater in the combination treatment group (52% versus 31%, P = 0.02).

The safety review showed that the incidence of ocular adverse events was twofold higher with combination therapy compared with ranibizumab alone (63% versus 31%; P < 0.001). The difference was due to a high rate of increased IOP in the combination group (29%). There were no reports of increased IOP in eyes treated with anti-VEGF monotherapy.

“Twenty percent of eyes with the dexamethasone implant needed topical treatment for elevated IOP,” Dr. Maturi reported.
During the 24-week randomization period, 3 (5%) eyes in the combination group, but no eyes receiving ranibizumab alone, underwent cataract surgery.

Raj K. Maturi, MD

e. rmaturi@gmail.com

This article was adapted from a presentation that Dr. Maturi delivered at the Retina Subspecialty Day prior to the 2017 American Academy of Ophthalmology meeting. Dr. Maturi receives fees from Allergan and Genentech to conduct studies.