Advertisement

Effect of a Fluocinolone Acetonide Insert on Recurrence Rates in Noninfectious Intermediate, Posterior, or Panuveitis

Three-Year Results
Open AccessPublished:April 16, 2020DOI:https://doi.org/10.1016/j.ophtha.2020.04.001

      Purpose

      To examine the 36-month efficacy and safety of a 0.2 μg/day fluocinolone acetonide insert (FAi) to treat noninfectious uveitis of the posterior segment (NIU-PS).

      Design

      Phase 3, prospective, double-masked, multicenter study (clinicaltrials.gov, NCT01694186).

      Participants

      Adults (≥18 years old) with a diagnosis of NIU-PS in ≥1 eye for ≥1 year and ≥2 recurrences of uveitis requiring systemic corticosteroid, immunosuppressive treatment, or intraocular corticosteroids.

      Methods

      Participants were randomized 2:1 to FAi or sham (injection plus standard of care) treatment.

      Main Outcome Measures

      The primary outcome was the difference between the proportion of FAi-treated and sham-treated patients who had a uveitis recurrence. Secondary outcomes included time to first recurrence, number of recurrences, best-corrected visual acuity (BCVA) change from baseline, resolution of macular edema, and number of adjunctive treatments.

      Results

      One hundred twenty-nine participants (n = 87 FAi-treated; n = 42 sham-treated) were enrolled. Over 36 months of treatment, cumulative uveitis recurrences were significantly reduced with FAi compared with sham (65.5% vs. 97.6%, respectively; P < 0.001); time to first recurrence was commensurately longer (median 657.0 and 70.5 days, respectively; P < 0.001). The number of recurrences per eye was significantly lower in the FAi-treated compared with the sham-treated group (mean 1.7 vs. 5.3, respectively, P < 0.001). At 36 months, more FAi-treated eyes had a ≥15-letter increase in BCVA from baseline and fewer FAi-treated eyes had investigator-determined macular edema at month 36 compared with sham-treated eyes (33.3% vs. 14.7% and 13.0% vs. 27.3% for BCVA and macular edema, respectively). Fewer FAi compared with sham-treated participants required adjunctive treatments (57.5% vs. 97.6%, respectively). Intraocular pressure (IOP) was similar for both study groups at month 36 (mean ± standard deviation 14.5±5.1 and 14.8±5.3, respectively), and approximately half as many eyes in the FAi-treated group when compared with the sham-treated group underwent IOP-lowering surgery (5.7% vs. 11.9%). Cataract surgery was required more frequently over 36 months in the FAi-treated compared with the sham-treated group (73.8% vs. 23.8% of eyes, respectively).

      Conclusions

      Fluocinolone acetonide insert–treated eyes had significantly reduced uveitis recurrence rates throughout the study duration, significantly increased recurrence-free durations, fewer recurrence episodes among those with recurrences, less adjunctive therapy, and an acceptable side-effect profile compared with sham-treated eyes.

      Abbreviations and Acronyms:

      AC (anterior chamber), BCVA (best-corrected visual acuity), CI (confidence interval), CST (central subfield thickness), DEX (dexamethasone), DME (diabetic macular edema), FAi (fluocinolone acetonide insert), IOP (intraocular pressure), NIU-PS (noninfectious uveitis of the posterior segment), SD (standard deviation)
      Noninfectious uveitis of the posterior segment (NIU-PS), which consists of intermediate, posterior, and panuveitis, comprises a group of ocular inflammatory diseases that is responsible for up to 20% of blindness in the developed world.
      • Chen S.-C.
      • Sheu S.-J.
      Recent advances in managing and understanding uveitis.
      It frequently runs a chronic relapsing course over years, resulting in ocular tissue damage and consequent loss of visual function. Generally, NIU-PS is treated with corticosteroids: In severe cases, repeated intravitreal or systemic corticosteroid treatments may be necessary, which may cause treatment-limiting adverse events.
      • Weijtens O.
      • Schoemaker R.C.
      • Romijn F.P.
      • et al.
      Intraocular penetration and systemic absorption after topical application of dexamethasone disodium phosphate.
      ,
      • Nguyen Q.D.
      • Callanan D.
      • Dugel P.
      • et al.
      Treating chronic noninfectious posterior segment uveitis: the impact of cumulative damage.
      Topical corticosteroid eye drops are not typically absorbed into the vitreous at sufficiently high concentrations to provide effective posterior segment treatment and therefore are not an option for most patients.
      • Weijtens O.
      • Schoemaker R.C.
      • Romijn F.P.
      • et al.
      Intraocular penetration and systemic absorption after topical application of dexamethasone disodium phosphate.
      Although immunosuppressive drugs may also be used to treat uveitis, these therapies have their own potential treatment-limiting adverse events. Given these clinical constraints, intravitreal corticosteroid implants that can provide continuous release may offer a treatment option for uveitis that is effective and reduces the risk and impact of common adverse events;
      • Kim J.S.
      • Knickelbein J.E.
      • Nussenblatt R.B.
      • Sen H.N.
      Clinical trials in noninfectious uveitis.
      ,
      • You C.
      • Sahawneh H.F.
      • Ma L.
      • et al.
      A review and update on orphan drugs for the treatment of noninfectious uveitis.
      such formulations can have a positive effect on patient adherence by reducing the need for self-administered anti-inflammatory drugs.
      A 0.19 mg fluocinolone acetonide insert (FAi; Alimera Sciences Ltd, Aldershot, Hampshire, UK; EyePoint Pharmaceuticals Inc, Watertown, MA) continuously releases a sub-microgram dose (0.2 μg/day) of the corticosteroid fluocinolone acetonide to the posterior segment over a 36-month period. A previous pilot study showed that an injectable intraocular fluocinolone acetonide sustained drug-delivery system can effectively treat uveitis over 2 years.
      • Jaffe G.J.
      • Lin P.
      • Keenan R.T.
      • et al.
      Injectable fluocinolone acetonide long-acting implant for noninfectious intermediate uveitis, posterior uveitis, and panuveitis: two-year results.
      The aims of the clinical study described were to examine the long-term, 36-month efficacy, and safety of the 0.2 μg/day FAi to treat NIU-PS; the 12-month results of this study have been reported.
      • Jaffe G.J.
      • Foster C.S.
      • Pavesio C.E.
      • et al.
      Effect of an injectable fluocinolone acetonide insert on recurrence rates in chronic noninfectious uveitis affecting the posterior segment: twelve-month results.

      Methods

      This phase 3, prospective, double-masked, multicenter study (clinicaltrials.gov: NCT01694186) randomized subjects with NIU-PS to treatment with 0.2 μg/day FAi or sham (sham injection plus standard of care) treatment. The study was conducted at 33 sites in the United States, United Kingdom, Germany, Hungary, Israel, and India. The Institutional Review Board at each center approved the study, which was performed in accordance with the tenets of the Declaration of Helsinki.
      Patients were eligible for study inclusion if they provided written informed consent and were at least 18 years of age, had a diagnosis of NIU-PS in at least 1 eye (with or without anterior uveitis) for a minimum of 1 year, and had at least 2 separate recurrences of uveitis requiring systemic corticosteroid or immunosuppressive treatment, or intraocular or periocular corticosteroids, or had received in the 12 months preceding study entry (1) systemic therapy (corticosteroid or other systemic treatment) for a minimum of 3 months or (2) at least 2 intraocular or periocular corticosteroid injections to manage uveitis.
      Details of the study design have been described
      • Jaffe G.J.
      • Foster C.S.
      • Pavesio C.E.
      • et al.
      Effect of an injectable fluocinolone acetonide insert on recurrence rates in chronic noninfectious uveitis affecting the posterior segment: twelve-month results.
      and are summarized in Figure 1. Eyes were randomized 2:1 to FAi- and sham-treated groups; the goal was to enroll 120 eyes. To minimize bias, at all sites, one investigator administered study treatments and performed study day 1 assessments, and a second investigator, masked to the assigned treatment, performed all study assessments after study day 1. Study personnel were also asked to use every reasonable effort to maintain the study masking.
      For subjects with unilateral uveitis, the study eye was the affected eye. For subjects with bilateral uveitis, the study eye was the more severely affected eye fitting the inclusion/exclusion criteria (Jaffe et al 2019, Appendix 1 [available at www.aaojournal.org] for further information on these criteria) or, if the eyes are symmetrically affected, the study eye was the right eye.
      Primary analysis was planned at 6 months, and secondary analyses were planned at 12 and 36 months. The primary outcome was the difference in proportion of FAi- and sham-treated study eyes that had a uveitis recurrence at 6 months. The study design had an 89% power (2-sided type I error rate = 0.05) to detect a difference between a sham group recurrence-free rate of 0.600 and a FAi-treated group recurrence-free rate of 0.880.
      In the study, a protocol-defined recurrence was defined by 1 of 3 possible criteria: a ≥2-step increase in the number of anterior chamber (AC) cells per high-powered field (1.6×, using a 1-mm beam), as assessed by slit-lamp examination, compared with any visit before month 6; an increase in vitreous haze of ≥2 steps compared with any visit before month 6; or a deterioration in best-corrected visual acuity (BCVA) of at least 15 letters compared with any visit before month 6.
      An analysis was performed to account for nonprotocol-defined recurrences. For this analysis, recurrence was assumed if a previously nonrecurrent study eye was treated with a prohibited local or systemic anti-inflammatory medication, defined as systemic, injectable, or topical corticosteroids, or systemic immunosuppressants, even when protocol-defined recurrence criteria defined were not met. A recurrence was also assumed if the participant missed an ophthalmic assessment at the 6-, 12-, or 36-month visit. The term “imputed” was used to describe these assumed recurrences. To prevent postprocedural inflammatory reactions from being reported as uveitis recurrences, assessments for uveitis recurrence began after the day 7 visit.
      Secondary outcomes included the proportion of study eyes that had a uveitis recurrence from baseline to 36 months, time to first recurrence, number of recurrences, BCVA change from baseline, resolution of macular edema (assessed using OCT imaging), and number of adjunctive treatments received. The differences between the FAi-treated and sham-treated groups were analyzed for statistical significance, although the study was not powered to detect differences in these secondary parameters.
      Other outcomes that were also captured comprised changes in BCVA, vitreous haze, AC cell count, intraocular pressure (IOP), IOP-related events, and cataract development at 24 and 36 months.

      Results

      In total, 129 participants were enrolled in the study; 87 eyes of these subjects were randomized to the FAi-treated group and 42 to the sham-treated group. Baseline demographics and participant characteristics are shown in Table 1. In the FAi-treated group there was a longer uveitis duration, slightly lower percentage of female patients, a greater percentage of eyes without cells in the AC, and more eyes with central subfield thickness (CST) less than 300 μm at baseline compared with the sham-treated group.
      Table 1Baseline Demographics and Patient Characteristics
      FAi (n = 87)Sham Control (n = 42)
      Age, mean ± SD (yrs)48.3±13.948.3±13.7
      Sex (%)
       Female57.569.0
       Male42.531.0
      Systemic uveitis treatment (% eyes)50.650.0
      Duration of uveitis, mean ± SD (yrs)7.8±6.75.6±6.8
      Vitreous haze absent (% eyes)25.319.0
      Vitreous haze grade (% eyes)
       0/0.5+55.250.0
       1/2+44.850.0
       3/4+0.00.0
      AC cells absent (% eyes)62.147.6
      AC cells grade (% eyes)
       0/0.5+88.578.6
       1/2+11.521.4
       3/4+0.00.0
      BCVA, mean ± SD (ETDRS letters)66.9±15.564.9±15.5
      CST (% eyes)
       <300 μm42.533.3
       ≥300 μm55.264.3
      Macular edema present (% eyes)56.565.9
      Lens status (% eyes)
       Pseudophakic51.750.0
       Phakic48.350.0
      IOP, mean ± SD (mmHg)13.9±3.113.6±3.2
      AC = anterior chamber; BCVA = best-corrected visual acuity; CST = central subfield thickness; ETDRS = Early Treatment Diabetic Retinopathy Study; FAi = fluocinolone acetonide insert; IOP = intraocular pressure; SD = standard deviation.

       Uveitis Recurrence

      Over 36 months of treatment, the FAi-treated group had a significantly reduced recurrence rate when compared with those in the sham-treated group. The protocol-defined recurrence rate in the FAi group compared with the sham-treated group was 5.7% compared with 28.6% (a 5-fold difference; P < 0.001), and the cumulative uveitis recurrence rate (that included imputed recurrences) was 65.5% in the FAi-treated group compared with 97.6% in the sham-treated group (P < 0.001).
      At month 36, the time to first recurrence in FAi-treated eyes was significantly longer compared with sham-treated eyes (median 657.0; 95% confidence interval [CI], 395–1051] days and 70.5; 95% CI, 57–91 days, respectively; P < 0.001). Probability of recurrence was greater in the sham group compared with the FAi group throughout the study (Fig 2); the difference was significant at the 6-, 12-, and 36-month time points (Table 2).
      Figure thumbnail gr2
      Figure 2Probability of recurrence by days from treatment (intention-to-treat population).
      Table 2Proportion of Subjects with Uveitis Recurrence in the Study Eye over 36 Months
      Time Since TreatmentFAi (n = 87) n (%)Sham Control (n = 42) n (%)Odds Ratio (95% CI)P Value
      6 mos24 (27.6)38 (90.5)24.94 (8.04–77.39)<0.001
      12 mos33 (37.9)41 (97.6)67.09 (8.81–511.06)<0.001
      36 mos57 (65.5)41 (97.6)21.58 (2.83–164.70)<0.001
      CI = confidence interval; FAi = fluocinolone acetonide insert.
      The number of recurrences per eye occurring over 36 months was also significantly lower in the FAi-treated group compared with the sham-treated group (mean ± standard deviation [SD], 1.7±2.4 vs. 5.3±3.8, respectively, P < 0.001), and a higher proportion of eyes in the FAi-treated group had no uveitis recurrence compared with the sham-treated group (34.5% vs. 2.4%). In addition, a greater proportion of eyes in the FAi-treated group compared with the sham-treated group had uveitis recur only once in 3 years (33.3% vs. 11.9%, respectively). Unsurprisingly, frequent uveitis recurrence during the 3 years (defined as >5 recurrences) occurred in a lower proportion of eyes in the FAi-treated group compared with the sham-treated group (8.0% vs. 40.5%, respectively).
      Observed protocol-defined uveitis recurrences occurred in a greater percentage of the sham-treated eyes, whereas the percentage of eyes with an imputed recurrence was more similar in the 2 groups (59.8% and 69.0%, respectively) (Table 3). A greater proportion of imputed recurrences were treated with systemic medications in the FAi-treated group compared with the sham-treated group (28.7% vs. 11.9%); however, a lesser proportion of local injections were administered (8.0% vs. 38.1%).
      Table 3Recurrences and Reasons for Imputing an Event
      FAi (n = 87)Sham control (n = 42)
      Recurrence rate within 36 mos (%)65.597.6
       Observed5.728.6
       Imputed59.869.0
      Reason for imputed event (%)
       Missing data4.60.0
       Systemic corticosteroid or immunosuppressant28.711.9
       Intraocular/periocular corticosteroid8.038.1
       Topical corticosteroid18.419.0
      FAi = fluocinolone acetonide insert.

       Adjunctive Treatments

      Fewer participants in the FAi-treated group compared with the sham-treated group required adjunctive treatments over 36 months of follow-up (Fig 3); the percentage of eyes receiving any adjunctive medication in the FAi-treated group was 57.5% compared with 97.6% in the sham-treated group. The mean number of adjunctive treatments per eye in the FAi-treated group was 0.48 compared with 1.52 in the sham-treated group.
      Figure thumbnail gr3
      Figure 3Study participants receiving at least 1 adjunctive treatment (systemic treatment, local injections, or topical ophthalmic treatment) over a 36-month period. FAi = fluocinolone acetonide insert.

       Visual Acuity Outcomes

      At 36 months, 33.3% of eyes in the FAi-treated group had a ≥15-letter increase in BCVA from baseline compared with 14.7% of eyes in the sham-treated group. Furthermore, there was a significantly greater mean change in BCVA over 36 months: +9.1 Early Treatment Diabetic Retinopathy Study letters (SD = 13.0) for the FAi-treated group compared with +2.5 letters (SD = 14.2) for the sham-treated group (P = 0.020; Fig 4). Conversely, fewer patients in the FAi-treated group had a ≥15 letter decrease in BCVA compared with the sham group (1.4% vs. 8.8%, respectively).
      Figure thumbnail gr4
      Figure 4Mean best-corrected visual acuity (BCVA) change in treated eyes over a 36-month period. ETDRS = Early Treatment Diabetic Retinopathy Study; FAi = fluocinolone acetonide insert.

       Control of Edema

      There was a tendency for fewer eyes in the FAi-treated group to have investigator-determined macular edema at month 36 compared with sham-treated eyes (13.0% vs. 27.3%, respectively); however, this difference did not reach statistical significance (P = 0.079). The reduction in CST and its subsequent maintenance occurred more quickly in the FAi group than in the sham group; this difference in CST in favor of the FAi-treated group was observed from day 28 onward (Fig 5).
      Figure thumbnail gr5
      Figure 5Mean central subfield thickness (CST) in treated eyes over a 36-month period. FAi = fluocinolone acetonide insert.

       Vitreous Haze and Anterior Chamber Cells

      A similar percentage of eyes in the FAi- and sham-treated groups had no vitreous haze or AC cells at month 36 (vitreous haze absence: 88.9% vs. 91.2%, respectively; AC cell absence: 84.7% vs. 85.3%, respectively); however, FAi-treated eyes achieved an absence of vitreous haze and AC cells sooner than sham-treated ones (vitreous haze absent at 6 months: 80.5% vs. 59.5%; AC cells absent at 6 months: 89.7% vs. 64.3%, respectively; Fig 6).
      Figure thumbnail gr6
      Figure 6Percentage of eyes without (A) vitreous haze or (B) anterior chamber (AC) cells over 36 months. FAi = fluocinolone acetonide insert.

       Safety: Intraocular Pressure and Cataract

      Intraocular pressure was well controlled in both study groups (Fig 7); IOP was similar for both study groups at month 36, and comparable proportions of eyes in the 2 groups had IOP-related events over 36 months (Table 4). Most IOP elevations were treated with IOP-lowering medication; 42.5% of eyes in the FAi-treated group received IOP-lowering medication compared with 33.3% of eyes in the sham-treated group. The proportion of eyes in the FAi-treated group that underwent IOP-lowering surgery was approximately half that in the sham-treated group (5.7% vs. 11.9%). As expected, cataract surgery was required more frequently over 36 months in the FAi-treated group compared with the sham-treated group (73.8% of eyes vs. 23.8% of eyes, respectively). Cataract removal had a similar effect on visual acuity outcomes for both treatment groups when measured at the first assessment after removal (+20.3 letters for the FAi-treated group vs. +23.4 letters for the sham-treated group).
      Figure thumbnail gr7
      Figure 7Intraocular pressure (IOP) over 36 months’ follow-up in the fluocinolone acetonide insert (FAi)- and sham-treated groups.
      Table 4Intraocular Pressure–Related Events over 36-Month Follow-up
      FAi (n = 87)Sham control (n = 42)
      IOP at month 36 (mmHg)
       Mean ± SD14.5±5.114.8±5.3
       Change from baseline ± SD0.8±5.01.4±5.7
      IOP events up to month 36 (% of eyes)
       >25 mmHg24.123.8
       >30 mmHg16.111.9
       Change from baseline ≥12 mmHg23.026.2
       IOP-lowering medication42.533.3
       IOP-lowering surgery5.711.9
       Hypotony10.311.9
      Cataract
       Phakic at baseline (n)4221
       Cataract surgeries (n, %)31 (73.8)5 (23.8)
      FAi = fluocinolone acetonide insert; IOP = intraocular pressure; SD = standard deviation.

      Discussion

      In this study, the FAi-treated group had significantly fewer uveitis recurrences over 36 months and fewer recurrences per study eye compared with the sham-treated group; the FAi group required fewer adjunctive treatments and had a longer median time to onset of recurrence. More than one-third (34.5%) of FAi-treated eyes remained free of recurrence or did not need any adjunctive medication for 36 months compared with 2.4% of eyes in the sham group. In addition, more FAi- than sham-treated eyes had uveitis recur only once in 3 years. Over this period, the FAi-treated group had a significantly greater improvement in BCVA, less visual acuity loss, and a reduction in CST that was maintained from day 28.
      These long-term results confirm and extend the month 6, 12, and 24 results that were reported previously
      • Jaffe G.J.
      • Lin P.
      • Keenan R.T.
      • et al.
      Injectable fluocinolone acetonide long-acting implant for noninfectious intermediate uveitis, posterior uveitis, and panuveitis: two-year results.
      • Jaffe G.J.
      • Foster C.S.
      • Pavesio C.E.
      • et al.
      Effect of an injectable fluocinolone acetonide insert on recurrence rates in chronic noninfectious uveitis affecting the posterior segment: twelve-month results.
      • Cai C.X.
      • Skalak C.
      • Keenan R.T.
      • et al.
      Time to disease recurrence in noninfectious uveitis following long-acting injectable fluocinolone acetonide implant.
      and demonstrate that uveitis recurrence rates were significantly reduced compared with sham-treated eyes over the entire 36 months of this study (P < 0.001 at 6, 12, and 36 months). Most notably, there were far fewer observed recurrences in the FAi-treated group, supporting the beneficial effect of treatment on the observable clinical end points of cellular infiltrate in the AC, vitreous haze, and BCVA. Indeed, where reanalyses using alternative definitions of recurrence have been conducted that exclude topical corticosteroid use as a proxy for recurrence (per US Food and Drug Administration guidance), the difference in recurrences between the FAi- and sham-treated groups was even greater (56.3% vs. 92.9%, respectively). On the basis of these favorable study results, the Food and Drug Administration approved FAi for the treatment of NIU-PS.
      EyePoint Pharmaceuticals
      YUTIQTM (fluocinolone acetonide intravitreal implant) 0.18 mg, for intravitreal injection: highlights of prescribing information.
      A significantly lower recurrence rate was observed in FAi-treated eyes compared with sham-treated eyes, whether recurrence rate was calculated strictly according to the protocol definition or whether imputed recurrences (as defined previously) were included. The approximately 1.5-fold magnitude of the recurrence rate difference between FAi-treated and sham-treated eyes was less than the 5-fold magnitude of the recurrence rate difference, when protocol-defined recurrence rates were determined. In addition, missed visits, which occurred in 4 instances in the FAi-treated group, but in none of the sham-treated group, were imputed as recurrences. Because it is unknown whether these subjects truly had a uveitic recurrence, it is possible that the imputed recurrences based on missed visits could have overestimated the cumulative recurrence rate in FAi-treated eyes. Likewise, treatment of the fellow eye was considered an imputed recurrence, as was systemic anti-inflammatory agents for systemic anti-inflammatory disease. Inclusion of these events as a study eye recurrence could have overestimated the recurrence rate in the FAi-treated eyes. Accordingly, for these reasons, which resulted in a smaller magnitude of the cumulative recurrence rate difference between FAi- and sham-treated eyes, when compared with protocol-defined recurrence rates, respectively, we would consider the inclusion of imputed recurrences to be a more conservative estimate of uveitis recurrence rate.
      Treatment with local or systemic anti-inflammatory agents was one of the reasons for an imputed uveitis recurrence. There are several possible reasons why the subject might have been treated with one of these agents: inflammation that did not meet the minimum threshold for a recurrence (e.g., a 1-step increase in AC cells or vitreous haze, or <15-letter decrease in Early Treatment Diabetic Retinopathy Study visual acuity letters compared with any month before month 6). Presumably, the investigator considered this mild inflammation to be a recurrence, even though it did not meet the strict recurrence threshold. It is also likely that the investigator treated eyes with anti-inflammatory therapy for macular edema attributed to inflammation for which he believed that treatment was necessary, even though recurrent macular edema was not a protocol-defined recurrence criterion.
      Although the proportion of imputed (nonprotocol-defined) recurrences was similar in the FAi- and sham-treated groups, the treatment reasons for these imputed recurrences were different. A higher proportion of FAi-treated eyes had recurrence imputed due to treatment with systemic therapy, whereas many more eyes in the sham-treated group had recurrence imputed due to treatment with local corticosteroid injections. The reasons for these differences are not entirely clear, and further investigations are warranted to understand these differences.
      The decreased number of uveitis recurrences and the delayed median time to first recurrence in the FAi-treated group compared with the sham-treated group are evidence that inflammation was better controlled in the FAi- than the sham-treated group. There are no direct comparative data for other injectable delivery systems. The dexamethasone (DEX) insert is another example of an injectable intravitreal corticosteroid implant to treat NIU-PS, but differences in study designs (duration and end points) and enrolled patient populations preclude quantitative comparisons between the DEX insert and the FAi, for example, uveitis recurrence was not recorded in a previous study of a DEX insert,
      • Lowder C.
      • Belfort R.
      • Lightman S.
      • et al.
      Dexamethasone intravitreal implant for noninfectious intermediate or posterior uveitis.
      nor were between recurrence rates in implant-treated eyes relative to controls.
      The FAi differs from the DEX insert in its long-term (36 months vs. 3–6 months, respectively), microdose, sustained release, from which patients with recurrent uveitis may achieve additional benefit from a sustained-release single injection technology.
      • Jaffe G.J.
      • Lin P.
      • Keenan R.T.
      • et al.
      Injectable fluocinolone acetonide long-acting implant for noninfectious intermediate uveitis, posterior uveitis, and panuveitis: two-year results.
      In clinical practice, the features of these corticosteroid inserts are likely to substantially affect treatment decisions and long-term outcomes. In a study of uveitis treatment with the DEX insert,
      • Lowder C.
      • Belfort R.
      • Lightman S.
      • et al.
      Dexamethasone intravitreal implant for noninfectious intermediate or posterior uveitis.
      there was a reduction in BCVA after week 8, which may indicate a waning effect of DEX on BCVA at this time point. The difference in duration of action between the DEX insert and FAi may be reasonably expected to result in similar temporal treatment effects in uveitis, and the shorter duration of drug release afforded by the DEX insert would imply a greater recurrence rate over an extended follow-up time with this treatment method.
      Fewer participants in the FAi-treated group compared with the sham-treated group required adjunctive systemic corticosteroid or immunosuppressive therapy, intraocular corticosteroids, or topical corticosteroids. These clinically important data are further evidence that inflammation was better controlled in the FAi-treated group when compared with the sham-treated group. Uveitis often runs a chronic course that requires excellent patient compliance to self-administer systemic or topical medications, often over many years, which many patients find difficult. Suboptimal compliance could lead to suboptimal uveitis control. Furthermore, the increased number of local injections needed in the sham-treated group might suggest that these patients would need to return more frequently for clinic visits, which would increase the burden on both patient and provider. Finally, by decreasing or eliminating adjunctive therapy, side effects associated with these treatments can be minimized or eliminated. For example, by decreasing systemic therapy, potentially toxic systemic side effects could be minimized or eliminated, and by reducing the frequency of periocular or intraocular injections, complications associated with these procedures could be minimized.
      Retinal structural and functional benefits were observed for patients with the FAi. The FAi-treated group experienced a rapid and sustained reduction of macular edema (CST) over the 36 months of the study, compared with the sham-treated group, and a smaller proportion of FAi-treated eyes had edema at 36 months compared with sham-treated eyes. Likewise, patients who received the FAi had a rapid and sustained BCVA improvement over the 36 months of study and substantially more FAi-treated eyes had a ≥15-letter BCVA gain compared with the sham group. Conversely, fewer eyes had a ≥15-letter BCVA loss. Collectively, these efficacy outcomes suggest that it is more beneficial to maintain steady, long-term uveitis control, as produced by the FA, than can be achieved by providing more frequent, episodic treatment for each recurrence, as was needed in the sham group.
      The adverse event (AE) rates were higher with the FAi than in the treated-sham group; however, the AE rates with the FAi, with the exception of cataract, did not appear to be substantially higher than in the sham-treated group (Appendix 1 and Table S1, available at www.aaojournal.org). Hypotony was a transient phenomenon associated with the injection procedure in eyes receiving the FAi implant, and there was no meaningful difference in the incidence of hypotony between FAi- and sham-treated eyes. On the basis of these data, the risk of hypotony is not considered significant enough to override the benefit of the FAi to treat NIU-PS.
      Intraocular pressure was generally well controlled in both study groups; most frequently, IOP-lowering medications were used to control IOP elevations. The lower rate of IOP-lowering surgery in the FAi-treated group compared with the sham-treated group is notable and contrasts with previous studies of the surgically implanted Retisert (Bausch + Lomb, Bridgewater, NJ) implant, which resulted in a higher rate of IOP surgery compared with the DEX implant.
      • Jaffe G.J.
      • Martin D.
      • Callanan D.
      • et al.
      Fluocinolone acetonide implant (Retisert) for noninfectious posterior uveitis: thirty-four-week results of a multicenter randomized clinical study.
      ,
      • Arcinue C.A.
      • Cerón O.M.
      • Foster C.S.
      A comparison between the fluocinolone acetonide (Retisert) and dexamethasone (Ozurdex) intravitreal implants in uveitis.
      Results of 3 phase 2b/3 trials evaluating the safety and efficacy of the Retisert implant (0.59 mg fluocinolone acetonide) to treat NIU-PS revealed that 10.9% of Retisert-treated eyes required IOP-lowering surgery within 1 year of treatment and 32.0% of eyes within 3 years of treatment.
      • Goldstein D.A.
      • Godfrey D.G.
      • Hall A.
      • et al.
      Intraocular pressure in patients with uveitis treated with fluocinolone acetonide implants.
      This reduction in IOP-lowering surgery required with the injectable FAi compared with the surgically implanted Retisert implant is likely related to the lower daily corticosteroid dose.
      There is some controversy as to the interpretation of the long-term outcome data of Retisert in comparison with the systemic approach for the management of NIU-PS.
      • Albini T.A.
      • Callaway N.F.
      • Jaffe G.J.
      • et al.
      Beg to differ.
      The MUST Trial 7-year follow-up data suggested that sustained control of inflammation results in better long-term visual outcomes than does a relapsing and retreating course, with the implication that scheduled replacement of regional therapies might be considered as a strategy.
      • Kempen J.H.
      • Altaweel M.M.
      • Holbrook J.T.
      • et al.
      With respect to regional fluocinolone acetonide therapies, the design of the FAi intravitreal insert (injectable) facilitates easier re-treatment than does the surgically implanted Retisert. However, the lower dose and release rate of the 0.19-mg insert seem to result in a higher relapse rate than the 0.59-mg surgically placed implant. Comparative studies of these 2 regional modalities might provide data on their relative effectiveness.
      An FAi similar to the one used in the current study has been studied and approved to treat diabetic macular edema (DME).
      • Campochiaro P.A.
      • Brown D.M.
      • Pearson A.
      • et al.
      Sustained delivery fluocinolone acetonide vitreous inserts provide benefit for at least 3 years in patients with diabetic macular edema.
      The incidence of IOP-lowering surgery was 4.8% in patients receiving the FAi to treat DME, compared with 5.7% of patients receiving the FAi to treat NIU-PS; a smaller proportion of sham-treated eyes in the DME study received IOP-lowering surgery compared with sham-treated eyes in the present NIU-PS study (0.5% vs. 11.9%). This difference may be due to the higher proportion of eyes treated with corticosteroid adjunctive treatment in the sham-treated group of the NIU-PS study; alternative treatments to corticosteroids were more readily available to treat DME in that sham-treated group.
      In the current study, the cataract surgery rate was higher in the FAi-treated group compared with the sham-treated group (73.8% and 23.8%, respectively); these rates are comparable to those in a previous study of FAi for the treatment of DME (80.0% and 27.3%, respectively). Studies of Retisert to treat NIU-PS have described similarly high proportions of phakic eyes that require cataract extraction.
      • Callanan D.G.
      • Jaffe G.J.
      • Martin D.F.
      • et al.
      Treatment of posterior uveitis with a fluocinolone acetonide implant: three-year clinical trial results.
      During the 6-month HURON trial, cataracts were reported in 15% of phakic eyes treated with DEX compared with 7% of control-treated eyes,
      • Lowder C.
      • Belfort R.
      • Lightman S.
      • et al.
      Dexamethasone intravitreal implant for noninfectious intermediate or posterior uveitis.
      whereas in the current study, 14.9% of FAi-treated eyes and 4.8% of sham-treated eyes had cataract at 6 months. Collectively, these data indicate that the risk of cataract surgery and cataract formation resulting from the FAi’s continuous, long-term, daily sub-microgram corticosteroid exposure is similar to the risk from other corticosteroids.
      This study has limitations. First and foremost, data on specific types of intermediate, posterior, and panuveitis were not collected, and so the relative efficacy for specific uveitis diagnoses is unclear. Furthermore, the study was likely underpowered to detect significant efficacy differences based on underlying uveitis etiology, even had those data been available. It is reassuring that the overall study population appeared to respond well to the FAi. An important limitation of the long-term results reported is that the study was originally designed and powered to investigate differences in uveitis recurrence at 6 months; consequently, the current statistical comparisons are all powered from secondary objectives. It is also important to recognize the high number of imputed recurrences, which could confound the results reported here; however, the clear separation in observed recurrences supports a benefit for FAi treatment. The treatment effect on BCVA may have been underestimated because proportionally more study eyes had a baseline visual acuity of more than 70 letters, which would have limited mean BCVA gains and the proportion of subjects whose visual acuity could have improved by 15 letters from baseline. A further limitation was that resolution of macular edema was not defined quantitatively in the protocol. Also, the use of nonstandardized measurements of macula edema was allowed according to the study protocol so that macular edema resolution could be accurately assessed for a given subject but not between subjects.
      In conclusion, notwithstanding these limitations, the collective data show that FAi-treated eyes had significantly reduced rates of uveitis recurrence over the entire study duration, significantly increased duration of recurrence-free time, fewer recurrence episodes among those with recurrences, less adjunctive therapy, and an acceptable side-effect profile when compared with sham-treated eyes. These data support the FAi as an office-based rather than a surgical procedure that provides a useful long-term treatment option for NIU-PS.

       Acknowledgments

      The authors acknowledge the vital contribution of the study investigators and sites (Appendix 2, available at www.aaojournal.org).

      Supplementary Data

      References

        • Chen S.-C.
        • Sheu S.-J.
        Recent advances in managing and understanding uveitis.
        F1000Res. 2017; 6: 280
        • Weijtens O.
        • Schoemaker R.C.
        • Romijn F.P.
        • et al.
        Intraocular penetration and systemic absorption after topical application of dexamethasone disodium phosphate.
        Ophthalmology. 2002; 109: 1887-1891
        • Nguyen Q.D.
        • Callanan D.
        • Dugel P.
        • et al.
        Treating chronic noninfectious posterior segment uveitis: the impact of cumulative damage.
        Proceedings of an expert panel roundtable discussion. Retina. 2006; Suppl: 1-16
        • Kim J.S.
        • Knickelbein J.E.
        • Nussenblatt R.B.
        • Sen H.N.
        Clinical trials in noninfectious uveitis.
        Int Ophthalmol Clin. 2015; 55: 79-110
        • You C.
        • Sahawneh H.F.
        • Ma L.
        • et al.
        A review and update on orphan drugs for the treatment of noninfectious uveitis.
        Clin Ophthalmol. 2017; 11: 257-265
        • Jaffe G.J.
        • Lin P.
        • Keenan R.T.
        • et al.
        Injectable fluocinolone acetonide long-acting implant for noninfectious intermediate uveitis, posterior uveitis, and panuveitis: two-year results.
        Ophthalmology. 2016; 123: 1940-1948
        • Jaffe G.J.
        • Foster C.S.
        • Pavesio C.E.
        • et al.
        Effect of an injectable fluocinolone acetonide insert on recurrence rates in chronic noninfectious uveitis affecting the posterior segment: twelve-month results.
        Ophthalmology. 2019; 126: 601-610
        • Cai C.X.
        • Skalak C.
        • Keenan R.T.
        • et al.
        Time to disease recurrence in noninfectious uveitis following long-acting injectable fluocinolone acetonide implant.
        Graefes Arch Clin Exp Ophthalmol. 2020; 258: 1023-1030
        • EyePoint Pharmaceuticals
        YUTIQTM (fluocinolone acetonide intravitreal implant) 0.18 mg, for intravitreal injection: highlights of prescribing information.
        • Lowder C.
        • Belfort R.
        • Lightman S.
        • et al.
        Dexamethasone intravitreal implant for noninfectious intermediate or posterior uveitis.
        Arch Ophthalmol. 2011; 129: 545-553
        • Jaffe G.J.
        • Martin D.
        • Callanan D.
        • et al.
        Fluocinolone acetonide implant (Retisert) for noninfectious posterior uveitis: thirty-four-week results of a multicenter randomized clinical study.
        Ophthalmology. 2006; 113: 1020-1027
        • Arcinue C.A.
        • Cerón O.M.
        • Foster C.S.
        A comparison between the fluocinolone acetonide (Retisert) and dexamethasone (Ozurdex) intravitreal implants in uveitis.
        J Ocul Pharmacol Ther. 2013; 29: 501-507
        • Goldstein D.A.
        • Godfrey D.G.
        • Hall A.
        • et al.
        Intraocular pressure in patients with uveitis treated with fluocinolone acetonide implants.
        Arch Ophthalmol. 2007; 125: 1478-1485
        • Albini T.A.
        • Callaway N.F.
        • Jaffe G.J.
        • et al.
        Beg to differ.
        Ophthalmic Surg Lasers Imaging Retina. 2019; 50: 266-268
        • Kempen J.H.
        • Altaweel M.M.
        • Holbrook J.T.
        • et al.
        JAMA. 2017; 317 (Association between long-lasting intravitreous fluocinolone acetonide implant vs systemic anti-inflammatory therapy and visual acuity at 7 years among patients with intermediate, posterior, or panuveitis): 1993-2005
        • Campochiaro P.A.
        • Brown D.M.
        • Pearson A.
        • et al.
        Sustained delivery fluocinolone acetonide vitreous inserts provide benefit for at least 3 years in patients with diabetic macular edema.
        Ophthalmology. 2012; 119: 2125-2132
        • Callanan D.G.
        • Jaffe G.J.
        • Martin D.F.
        • et al.
        Treatment of posterior uveitis with a fluocinolone acetonide implant: three-year clinical trial results.
        Arch Ophthalmol. 2008; 126: 1191-1201