Fluocinolone Acetonide Intravitreal Implant 190 μg (ILUVIEN®) in Vitrectomized versus Nonvitrectomized Eyes for the Treatment of Chronic Diabetic Macular Edema

Pars plana vitrectomy (PPV) has been suggested as a potential treatment option for diabetic macular edema (DME) due to the characteristics of the vitreous in diabetic patients [1]. When the vitreous gel is removed and replaced by balanced saline solution, oxygen transport to ischemic retinal areas is improved, as is clearance of vascular endothelial growth factor (VEGF) and other cytokines, thus reducing the edema and neovascularization [2]. Many studies have also addressed the advantages of PPV in DME patients with associated vitreoretinal traction, since it can also play an important role in the generation and/or maintenance of DME [3,4,5]. However, the chronic and multifactorial pathogenesis of DME usually requires continuous pharmacological treatments.

In vitrectomized eyes, drugs are believed to be more rapidly distributed throughout the eye than in an eye with an intact vitreous gel. There is evidence that the clearance of anti-VEGF agents and corticosteroids (i.e., triamcinolone) increases after vitrectomy, reduces the drug exposure, and impacts the treatment's success [3,4,5,6,7,8,9].

The continuous release of fluocinolone acetonide (FAc; 0.2 µg/day) from a sustained drug delivery system could potentially enhance the effect of vitrectomy in patients with DME. The 0.2 µg/day FAc intravitreal implant (ILUVIEN®, Alimera Sciences Inc., Alpharetta, GA, USA) is a nonbioerodible microimplantable cylindrical tube (3.5 × 0.37 mm) made from polyimide and loaded with 190 μg of FAc. It is approved for the treatment of vision impairment associated with chronic DME, considered insufficiently responsive to available therapies [10]. The implant releases 0.2 μg/day FAc for up to 36 months and offers an alternative therapeutic strategy, providing sustained delivery of the corticosteroid to maximize its anti-inflammatory, angiostatic, and antipermeability effects [11,12,13].

The purpose of this study is to compare the outcomes (functional, anatomical, and safety) in nonvitrectomized and vitrectomized eyes with chronic DME following the administration of the 0.2 µg/day FAc implant.

This was a noninterventional, retrospective, comparative analysis of 43 eyes with chronic DME, defined as having at least 1 year of documented DME and having received at least 1 prior DME therapy. All subjects gave their informed consent, and the study protocol complies with the institute's committee on human research. All eyes were treated with a single FAc implant and treated per standard practices, and informed consent was obtained from all subjects prior to the injection of the 0.2 μg/day FAc implant. The treated patients were divided into 2 groups: 24 eyes which had undergone PPV prior to 0.2 µg/day FAc (group 1) and 19 eyes which had not been vitrectomized (group 2). Demographic data are summarized in Table 1.

The main outcome measure was the mean change in best corrected visual acuity (BCVA) reported as an Early Treatment Diabetic Retinopathy Study (ETDRS) letter score. Secondary outcome measures included: (a) the change in central subfield foveal thickness (CSFT) using spectral-domain optical coherence tomography (Spectralis® Tracking Laser Tomography, Heidelberg, Germany); (b) the change in BCVA between phakic and pseudophakic eyes; (c) the percentage of patients achieving ≥20/40 vision; (d) the percentage of patients with an improvement in BCVA ≥15 ETDRS letters; and (e) the changes in intraocular pressure (IOP) using the Goldmann applanation tonometry method.

Patients were examined prior to administration of the 0.2 µg FAc implant (baseline) and then during the first week and month after injection and quarterly thereafter. All changes were calculated by subtracting the baseline values from the last observed value. For all parameters, within-group comparisons were conducted using a paired Student t test, and an unpaired Student t test was used to compare between groups. For IOP, a correlation between IOP at last observation and vitreous status was performed using the Fisher Z transformation. Statistical differences were defined as a p value ≤0.05. All data are reported as means ± standard error unless stated otherwise.

Following intravitreal injection of the 0.2 μg/day FAc implant there was a mean follow-up period of 8.5 ± 1.6 months (median, 6.0; range, 1-21). Twenty-four eyes were enrolled in group 1, and 19 eyes were enrolled in group 2.

Demographic data and baseline values for groups 1 and 2 are reported in Table 1, and prior DME therapies are reported in Table 2.

There was no difference in baseline BCVA values between groups 1 and 2 (Table 1). Following administration of the 0.2 μg/day FAc implant there was a mean increase in BCVA of +16.9 ± 3.39 letters (p < 0.001) in group 1 and +8.2 ± 4.62 letters (p = 0.092) in group 2, but the comparison between the groups revealed no statistical difference (p = 0.130) (Fig. 1).

Because there was a statistical difference in the pseudophakic:phakic ratio between groups, and the mean change in BCVA in group 2 was found not to be statistically significant, a subanalysis carried out on group 2 (nonvitrectomized) showed that lens status had no effect on the mean change in BCVA (p = 0.404) (Fig. 2).

A gain of ≥15 letters, from baseline to the last observation, was achieved in 37.5 and 36.8% in group 1 and in group 2, respectively (Table 3).

BCVA at baseline and at the last observation point showed the majority of patients having a BCVA between 34 and 68 letters in both groups. After the 0.2 μg/day FAc implant there was a distribution change, with more patients achieving gains in vision that placed them in the good vision group (i.e., a BCVA between 69 and 85 letters or achieving ≥20/40 vision). Indeed, in vitrectomized eyes there was an increase from 8.3% at baseline to 29.2% at the last observation, and in the nonvitrectomized eyes there was an increase from 0 to 15.8% (Table 3; Fig. 3a, b).

After administering the 0.2 μg/day FAc implant, the mean change in CSFT, from baseline, was: group 1, -217.7 ± 40.8 µm (p < 0.001); group 2, -155.6 ± 43.4 µm (p = 0.002) (Fig. 4). No statistical difference was found between the groups (p = 0.306).

IOP distribution and IOP-lowering medication at baseline and last observation are represented in Table 4. The mean change, at the last observation, in IOP was +1.6 ± 0.7 mm Hg (p = 0.023) and +0.8 ± 1.3 mm Hg (p = 0.547) from baseline values of 14.9 ± 0.52 and 15.3 ± 0.67 mm Hg (Table 5) in group 1 and group 2, respectively. There were no statistically significant differences in the mean changes of IOP between the groups at the last observation point (p = 0.544) or any time point up to 21 months (Table 5). No significant correlation between vitrectomy status and IOP was found (p = 0.431).

Prior to ILUVIEN, there were 6 (31.6%) nonvitrectomized and 9 (37.5%) vitrectomized eyes being treated with IOP medication. Over the subsequent 24 months, there were 7 (36.8%) nonvitrectomized and 4 (16.75%) vitrectomized eyes that required medication to manage new cases of raised IOP. Moreover, only 1 patient in the nonvitrectomized group presented an IOP >30 mm Hg at month 3. This patient had been treated with 4 different IOP-lowering eyedrops previously to FAc implantation. The IOP was successfully managed by cyclophotocoagulation guided with transillumination.

In total there were 8 patients with a phakic lens - 1 in the vitrectomized group and 7 in the nonvitrectomized group. Cataract extractions were conducted in 4 eyes - 1 phakic, vitrectomized eye (phacoemulsification performed at month 6) and 3 phakic, nonvitrectomized eyes (phacoemulsification performed at month 6 in 2 eyes and at month 3 in 1 eye) after administration without worsening or recurrence of DME. No other treatment-related adverse events were reported.

Additional therapies were carried out after the FAc implant had been administered - in group 1, 2 eyes (8.3%) were administered ranibizumab or laser, and in group 2, 5 eyes (26.3%) received ranibizumab or aflibercept.

This real-life study compared the efficacy and safety outcomes of a 0.2 µg/day FAc implant in vitrectomized (group 1) and nonvitrectomized (group 2) eyes with chronic DME. Our findings show that despite the vitreous status, patients receiving a single 0.2 μg/day FAc implant had continuous and sustained exposure to treatment and had both functional and anatomical improvements over the study period. BCVA improved by +16.9 and +8.2 letters, whereas CSFT was reduced by -217.7 and -155.6 μm in group 1 and group 2, respectively. Although improvements in BCVA and CSFT seem to have been greater in group 1 there were no statistically significant differences between groups during the follow-up period. Additionally, an improvement in vision ≥20/40 of 29.2 and 15.8% was obtained in the vitrectomized group and nonvitrectomized group, respectively. To our knowledge, this is the first real-life study that shows that the 0.2 μg/day FAc implant (a) stabilizes/improves visual acuity in over 95% of vitrectomized eyes studied, (b) provided functional gains in visual acuity (i.e., ≥20/40) in previously vitrectomized eyes, and (c) reveals outcomes in vitrectomized eyes similar to those reported in the FAME studies where previously vitrectomized eyes were excluded [11,12,13].

The current results confirm the current knowledge concerning the use of a single 0.2 μg/day FAc implant in vitrectomized eyes. Meireles et al. [14] reported the results from 26 eyes that had undergone prior vitrectomy and had a mean follow-up period of 8.5 months (median, 6.0 months). Results showed improvements in visual acuity (+11.7 letters from baseline; range, -19 to +40; p < 0.0004) and CSFT (-233.5 µm from baseline; range, -678 to 274 µm), but, unlike the current study, there was no comparison versus nonvitrectomized eyes [10]. This was, however, performed by El-Ghrably et al. [15]. Subgroup analysis of 12 vitrectomized and 10 nonvitrectomized eyes achieving a 12-month follow-up showed the effectiveness of a single 0.2 μg/day FAc implant in both groups [15]. The current study provides new insights into vitrectomized eyes by assessing for 2 reasons - the first is by assessing the treatment history prior to the 0.2 μg/day FAc implant in vitrectomized and nonvitrectomized eyes (Tables 1, 2) as well as determining functional improvements in visual acuity (i.e., the stabilization/improvement in vision as well as the achievement of driving vision).

In this study all patients receiving the FAc implant had had an insufficient response to prior DME therapies, and the analysis of groups 1 and 2 revealed some subtle differences indicative of disease state and its progression. Indeed, statistical comparisons showed more eyes in the vitrectomized group had received prior panretinal photocoagulation (100% [vitrectomized] vs. 78.9% [nonvitrectomized]; p = 0.031) which may suggest a worsening in the overall diabetic retinopathy status.

Moreover, the increased percentage of eyes receiving intravitreal injections of triamcinolone acetonide, the most used steroid (p = 0.033; 95.8 vs. 68.4%, respectively) in the vitrectomized group may also be indicative of disease progression to a more proinflammatory state. From these findings we hypothesize that patients with previous vitrectomy to treat DME may already be at a chronic stage of the disease [16] and, therefore, should be treated earlier with an FAc implant complying with the approved indication or once an insufficient response to triamcinolone or dexamethasone has been established.

The advantages of PPV seem clear in DME patients with associated vitreoretinal traction, but the procedure remains controversial in nontractional cases, although several theories in favor of vitrectomy have been proposed: the removal of pathological vitreous and subclinical traction at the macula; the elimination of inflammatory mediators that could increase vessel permeability; the improvement in oxygen concentration in the vitreous cavity; and retinal vessel changes with normalization of the macular blood flow and decreased leakage [7].

Nevertheless, it is discussed that PPV could lead to more rapid drug diffusion and clearance from the vitreous cavity in vitrectomized eyes, therefore limiting the exposure of intravitreal therapies and reducing the treatment success [7] of therapies such as anti-VEGFs [8,17,18,19] and maybe even dexamethasone implants [20,21,22,23,24,25,26], although the authors acknowledge that the treatment regimen and duration for a dexamethasone implant in vitrectomized eyes still remain controversial [27].

Additionally, 8.3% of the eyes in group 1 and 26.3% in group 2 needed additional therapies for DME. Presently there is little or no guidance from the published literature on the use of additional therapies or when the decision to add them should be made [28]; a combination therapy approach in DME is being increasingly discussed, since there are multiple mechanisms which become more important in the chronic stages of DME. Some authors have suggested that the combination of anti-VEGF therapies with steroids would provide a more optimal pharmacotherapy in DME [29].

As expected and described in other studies [14,15,30,31,32], there was an increase in IOP during the mean follow-up period. The mean change in IOP was +1.6 mm Hg (p = 0.023) and +0.8 mm Hg (p = 0.547) from baseline in group 1 and group 2, respectively. Though group 1 presented a statistically significant mean change in IOP, no correlation was found between IOP and the vitreous status (p = 0.430). The increase in IOP was easily manageable with IOP-lowering medication in all eyes, except 1 eye from the nonvitrectomized group, which had to undergo cyclophotocoagulation. Nevertheless, IOP increases did not affect long-term efficacy outcomes, and IOP remained below 21 mm Hg in the vast majority of patients (95.8% in group 1 and 89.5% in group 2).

Cataract formation was developed in 1 phakic eye in group 1 and 3 phakic eyes in group 2 with subsequent cataract extraction during the follow-up period and without apparent impairment of functional outcomes, similarly to what was observed in the FAME studies [11].

Our analysis has some limitations that are related in part to the retrospective nature of the data and the small cohort. Another limitation is the different follow-up period of analysis between patients and the lack of general information concerning the underlying metabolic status.

The present study is the first real-life clinical comparison of the effectiveness and safety of the 0.2 μg/day FAc implant in vitrectomized and nonvitrectomized eyes with insufficient response to available therapies in Portugal, and, irrespective of vitrectomy, similar outcomes were achieved in chronic DME patients. Our results support that a 0.2 μg/day FAc implant is being used earlier in vitrectomized eyes as shown indirectly through the increased injection of steroids relative to anti-VEGFs (Table 1) and a possible indicator of an increased proinflammatory state as vitrectomy will have been conducted in eyes with long-standing macular edema. Patients will continue to be monitored to assess longer-term real-world benefits as the 0.2 μg/day FAc implant provided therapy that lasted for up to 3 years in its pivotal randomized controlled trials.

Medical writing assistance was supported by Alimera Sciences.

The authors have no conflict of interests to declare.