Intravitreal Bevacizumab with or without Triamcinolone for Wet Age-related Macular Degeneration: Twelve-month Results of a Prospective, Randomized Investigation.

PURPOSE: The purpose of this study is to compare the long-term outcomes of intravitreal bevacizumab (IVB) with a combination therapy including IVB/intravitreal triamcinolone acetonide (IVB/IVTA) in neovascular age-related macular degeneration (AMD).
MATERIALS AND METHODS: This prospective, randomized clinical trial was conducted on 136 eyes of 136 patients with neovascular AMD. Eyes were randomly assigned to receive IVB alone (71 eyes) or in combination with IVTA (65 eyes). In the IVB group, three consecutive injections of 1.25 mg/0.05 ml of bevacizumab were administered 1 month apart, whereas in the IVB/IVTA group, 4 mg/0.05 mL of triamcinolone acetonide was added to bevacizumab in the first injection. Additional IVB injections were administered in eyes demonstrating active choroidal neovascularization. Best-corrected visual acuity (BCVA) and optical coherence tomography were performed at baseline as well as at all follow-up visits.
RESULTS: No differences were seen between the patients receiving IVB and those receiving IVB/IVTA in terms of baseline BCVA (P = 0.97) and baseline central macular thickness (CMT) (P = 0.77). BCVA improved, and CMT reduced significantly in both study arms at almost all follow-up intervals. IVB/IVTA intervention, compared with IVB, was statistically more effective in improving BCVA (P = 0.01) and in reducing CMT (P = 0.02) after 12 months. The average number of reinjections was 1.25 ± 0.92 in the IVB group and 1.06 ± 1.01 in the IVB/IVTA group (P = 0.44).
CONCLUSION: Our results suggest that the synergistic effect of intravitreal triamcinolone and IVB for treatment of neovascular AMD shows itself most apparently after 8 months of follow-up.

RCT Entities:   Population Interventions Outcomes

Introduction

Age-related macular degeneration (AMD) is the leading cause of severe visual impairment in subjects aged 50 years and above in developed countries.[12] It can be classified into neovascular (wet) and nonneovascular (dry) form. Although the neovascular form of AMD represents only 10%–20% of all cases of AMD, it accounts for 80%–90% of AMD-associated severe visual loss.[3] The neovascular form is caused by growth of abnormal blood vessels also known as choroidal neovascularization (CNV). These vessels are very fragile and tend to leak fluid and blood which leads to damage of the photoreceptors in the macula. The inflammation pathways as well as the angiogenic reactions contribute to the development of neovascularization by increasing the vascular permeability.[45] Vascular endothelial growth factor (VEGF) has been demonstrated to be the most important angiogenic factor responsible for CNV in AMD. Thus, it has been postulated that a combined treatment that suppresses both systems (inflammation and angiogenesis) would produce sustainable effects and may have advantages compared with the monotherapeutic approaches.[67] Indeed, combination therapies have showed positive effects in the therapy of wet AMD and have become increasingly common in recent years.[891011]

In the current investigation, we compared the efficacy of three consecutive intravitreal bevacizumab (IVB) injections as anti-VEGF drug with that of the same regimen combined with intravitreal triamcinolone acetonide (IVTA) as anti-inflammatory substance during the first injection in eyes afflicted with neovascular AMD.

Materials and Methods

This prospective, randomized, double-masked clinical trial adhered to the tenets of the Declaration of Helsinki and was approved by the Ethics Committees of the Urmia University of Medical Sciences. Written informed consent was obtained from all patients before enrollment. Consecutive cases aged 50 years or older with fluorescent angiography-diagnosed active subfoveal CNV due to AMD were recruited from April 2005 to March 2015 in Imam Khomeini Hospital, a referral center for eye diseases in Northwest Iran. Patients with all angiographic subtypes of AMD were included in the current investigation. Other inclusion criteria were as follows: visual acuity (VA) at baseline of 20/40–20/400 (Snellen equivalent), absence of other ocular diseases that could affect the BCVA, minimum age of 50, and minimum follow-up of 12 months. The exclusion criteria comprised the presence of retinal vascular occlusion, glaucoma, diabetic retinopathy, or any macular disease other than AMD. Participants could have received photodynamic therapy (PDT) before being recruited for the investigation; however, the period of time between the last PDT and the study recruitment must beat least 12 months. Patients were randomly assigned 1:1 to receive either IVB (IVB group) or a combination of IVB and IVTA (IVB/IVTA group). Treatment allocation was determined by picking an envelope containing a preprepared randomization label generated by a biostatistician with block size 4.

At baseline, all eligible participants underwent an ophthalmologic examination including best-corrected VA (BCVA) measurement with early treatment diabetic retinopathy study chart, fundus examination with a noncontact 78-diopter lens, slit-lamp examination, intraocular pressure (IOP) measurement (Goldmann applanation tonometry), fluorescein angiography (FA) (Imagenet; Topcon Corporation, Tokyo, Japan) and optical coherence tomography (OCT) (Stratus OCT 3; Zeiss Jena GmbH, Jena, Germany).

Both groups received mandated therapy with three intravitreal injections on day 0, month 1 and month 2. Patients in the IVB group were treated with three consecutive bevacizumab (1.25 mg/0.05) (Avastin®, Genentech/Roche, San Francisco, USA) injections. In the IVB/IVTA group, 4 mg/0.05 mL of triamcinolone acetonide (Triamhexal®; Hexal AG, Holzkirchen, Germany) was added to bevacizumab in the first injection. By the way, the second and third injections in the IVB/IVTA group consisted of bevacizumab only. IVB injections were performed with a 30-G needle through the superior temporal pars plana 3.75 mm and 3.25 mm from the limbus in phakic and pseudophakic eyes, respectively. Triamcinolone acetonide was injected in the same manner using another syringe inferotemporally. The eyes were patched for 6 h. Patients were instructed to take ciprofloxacin pills (500 mg) two times a day for 5 days thereafter. All participants also received an eye drop of betamethasone (0.5%) four times a day for 5 days. All subjects were examined 24 h after injection particularly for signs of intraocular inflammation. Follow-ups were scheduled at 1, 2, 4, 6, 8, 10, and 12 months after injections. BCVA and OCT were recorded in every visit. Need for retreatment with IVB injection was scheduled if any of the following criteria was present: (i) VA loss ≥5 letters from the previous visit with OCT or fluorescein angiographic evidence of fluid in the macula; (ii) new or persistent subretinal or intraretinal macular hemorrhage; (iii) an increase in OCT central retinal thickness ≥100 μm; (iv) evidence of persistent fluid on OCT; or (v) new area of classic CNV or increased CNV area or persistence of leakage on angiography since the previous visit. IOP was evaluated at baseline, 1 week after injection and at months 6 and 12. FA was repeated at the 3rd month and when considered necessary thereafter. The following clinical signs were used to make the diagnosis of glaucoma: elevated IOP (>21 mmHg), asymmetric or progressive disc cupping, optic nerve cupping, visual field abnormalities, or anterior segment changes (such as corneal enlargement or Haab's striae, alone or in combination). The injecting ophthalmologist (N. S. A) was not involved in the collection of data, whereas the evaluating ophthalmologists (Q. M and M. A) were masked to treatment assignment. All other personnel and the patients were not aware of treatment assignment as well.

All analyses were performed using SPSS software (version 17.0, SPSS Inc., Chicago, USA). Mann–Whitney U-test and Chi-square test were utilized for comparing quantitative and qualitative variables, respectively. Changes relative to baseline in VA and central macular thickness (CMT) were evaluated by paired t-test within the investigated groups. Analysis of covariance was employed to compare CMT and BCVA between the two groups. The survival time before the first retreatment was compared by log-rank test between the two groups. P < 0.05 was considered as statistically significant. The statistician who performed the analysis was masked to the details of the investigation.

Results

Between April 2005, and March 2015, one hundred forty-two eyes of 142 patients were initially enrolled in the current investigation. As shown in Figure 1 and 71, cases were assigned to each of the groups. Of these, six patients-all of whom in the IVB/IVTA arm of the study-refused to refer again for at least five follow-up visits and hence were excluded from the analysis. Therefore, a total of 136 patients including 74 (54.4%) male and 62 (45.6%) female participants with the mean age of 71.1 ± 8.4 years were investigated [Figure 1]. No major differences were seen between the patients receiving IVB and those receiving IVB/IVTA in terms of baseline characteristics [Table 1].

Figure 1

Flow diagram of patient participation

Table 1

Baseline and demographic characteristics of the study groups

Figures 2 and 3 display the time course of BCVA and CMT, respectively, over the12-month study period. Improvement in BCVA was already observed at the first follow-up visit (month 1) and reached its maximum 6 months after primary injection in both groups. Compared with baseline, mean changes in BCVA were statistically significant in both groups at almost all follow-up points [Table 2]. The only exception in this trend was observed at month 1 for IVB arm. Indeed, average BCVA improved by 8.7 letters in the IVB group and 14.6 letters in the IVB/IVTA group after 12 months. BCVA findings also revealed that compared with the IVB, IVB/IVTA intervention was statistically more effective from the 6th month to the end of the study [Table 2]. This difference remained significant even after adjustments for age and baseline BCVA. At the end of study, no cases of moderate vision loss (15 or more letters) or severe vision loss (30 or more letters) were observed in either group of patients. On the other hand, 24 (36.9%) patients in the IVB/IVTA group and 14 (19.7%) cases in the IVB group gained 15 or more letters (P = 0.041). At the 12-month follow-up, there were 10 patients in the IVB/IVTA group and4 cases in the IVB group with 30 or more letters improvement in their BCVA values (P = 0.112).

Figure 2

Best-corrected visual acuity letter scores in each study group at baseline and at 1, 2, 4, 6, 8, 10, and 12 months after intervention

Figure 3

Central macular thickness in each study group at baseline and at 1, 2, 4, 6, 8, 10, and 12 months after intervention

Table 2

Within and between-group comparison of best-corrected visual acuity and central macular thickness values at different time points

There was a statistically significant reduction in CMT at all follow-up sessions compared to baseline values in both groups of patients [Table 2]. In fact, mean CMT decreased by 90.3 μm in the IVB group and 138 μm in the IVB/IVTA group after 12 months. CMT reduction was statistically more pronounced in IVB/IVTA group than the IVB group at months 8, 10, and 12 [Table 2]. However, no significant difference was found in the earlier follow-ups.

When narrowing the analysis to pseudophakic eyes, like the whole population, a trend toward greater treatment benefit was observed in the IVB/IVTA arm at later follow-up points. In this subgroup of patients, mean BCVA levels were 4.0, 4.1, and 4.3 letters higher in the IVB/IVTA arm compared with the IVB arm at the 8th, 10th and 12th month of the study, respectively (P < 0.05 for all). The better anatomical response of pseudophakic eyes to the IVB/IVTA management was also evident from 8 months onward. Compared to the IVB/IVTA group, the average CMT levels of pseudophakic eyes in the IVB group were 34.2, 32.3, and 34.5 μm thicker at months 8, 10, and 12, respectively (P < 0.05 for all). It should be noted that the overall changes in mean BCVA and CMT at the study visits were not statistically significant between the phakic and pseudophakic eyes, neither in the subgroups nor in the total population (P > 0.05 for all).

On the other hand, confining the analysis to patients who had previously been treated by PDT revealed a statistically borderline difference in terms of mean CMT levels between the two arms of the study in favor of the IVB/IVTA-treated group after 12 months (P = 0.059). At 1 year, the average CMT levels decreased by 93.3 μm in the IVB arm and 115.2 μm in the IVB/IVTA arm. Although the VA of eyes with PDT history increased in both study groups after 12 months, no significant difference was observed between the treatment arms (P = 0.117) (data not shown).

A total of 153 reinjections were performed, 84 in the IVB arm, and 69 in the IVB/IVTA arm. Figure 4 displays the number of reinjections in each group during study. The average (standard deviation [SD]) number of retreatments was 1.25 (0.92) in the IVB group and 1.06 (1.01) in the IVB/IVTA group (P = 0.438, Mann–Whitney test). In the IVB/IVTA group, 24 (36.9%) eyes were needless of retreatment, whereas in the IVB group 18 (25.4%) eyes did not require to be retreated within 12 months. Meanwhile, 21 (32.3%), 14 (21.5%), 4 (6.2%), and 2 (3.1%) eyes in the IVB/IVTA group needed one, two, three, and four extra treatments, respectively. Moreover, one, two, three, and four extra injections were required in 31 (43.7%), 15 (21.1%), 5 (7.0%), and 2 (2.8%) eyes in the IVB group, respectively. In addition, the average (SD) time to the first reinjection was 6.60 (2.02) months in the IVB arm and 6.78 (1.89) months in the IVB/IVTA arm. Survival time (Kaplan–Meier analysis) to the first required retreatment failed to establish a statistical difference between the investigated groups (P = 0.886, Mantel-Cox test).

Figure 4

Number of injections in each group

The mean IOP levels did not significantly change in the two study groups after 6 and 12 months. Moreover, the within group comparisons of IOP were not statistically significant at any time point (1st week, 6th month, and 12th month) in either group. However, IOP was increased in two participants (both from the IVB/IVTA group) to about 30 mmHg 1 week after the first injection. IOP was controlled in these cases by topical eye drops. In addition, three patients (all from the VB/IVTA group) developed glaucoma during the 1st week after the initial injection which resolved with medication. There were no cases of traumatic cataract, endophthalmitis, retinal detachment, severe ocular inflammation, or retinal tear. In addition, no cases of systemic adverse event such as myocardial infarction or stroke were recorded.

Discussion

The long-term outcomes of this clinical study revealed that the combination therapy with IVB/IVTA is effective for treating AMD in terms of improving BCVA and decreasing CMT. Furthermore, the percentage of eyes gaining 15 or more letters in VA was significantly higher in the IVB/IVTA group than in the IVB group. However, compared to IVB treatment, adding triamcinolone to bevacizumab did not obviate the need for retreatment significantly. The combined therapy also failed to significantly lengthen the injection-free survival time.

There are several studies in the literature evaluating the impacts of combined intravitreal injection of bevacizumab and triamcinolone on AMD-affected eyes overtime. In a retrospective case series, Colucciello [12] treated 30 eyes of 27 patients afflicted with exudative AMD with a single combined intravitreal injection of bevacizumab (1.25 mg/0.05 mL) and triamcinolone (2 mg/0.05 mL). Their results revealed a statistically significant reduction in foveal thickness as well as in subfoveal fluid volume (P < 0.01) at 8 weeks after the injection. The same group conducted another interventional case-series study on 16 eyes of 16 patients with exudative AMD who had been previously treated with intravitreal mono-injections of bevacizumab (1.5 mg) without significant reduction in macular exudation or improvement in VA.[13] Participants of this investigation underwent a combined intravitreal injection of bevacizumab (1.5 mg) and triamcinolone acetonide (about 20 mg). A significant improvement (P = 0.03) was observed in VA at 3 months after treatment. A similar prospective clinical trial by Tao and Jonas.[11] revealed that IVB (1.5 mg) combined with intravitreal high-dose triamcinolone (20–25 mg) more effectively improves vision and reduce macular thickness in eyes with exudative AMD after unsuccessful IVB monotherapy. It has also been revealed that combined therapy (IVTA: 4 mg/0.1 ml, followed by IVB: 1.25 mg/0.05 ml 1 week later) is an effective and safe therapeutic procedure to treat CNV associated with large pigment epithelial detachment in AMD.[9] The findings of these studies demonstrate that triamcinolone acetonide, either at low (2 mg) or high (up to 25 mg) concentration can yield encouraging results in treating neovascular AMD when it is administered intravitreally in combination with a low dose of bevacizumab. Similar results were obtained with a moderate concentration (4 mg) of triamcinolone in the current study.

There are also investigations in the literature which have compared the therapeutic effects of intravitreal injections of bevacizumab/triamcinolone with that of the intravitreal injections of bevacizumab.[810] In a prospective, randomized clinical trial performed in Iran, 55 eyes receiving three consecutive intravitreal injections of bevacizumab (1.25 mg/0.05 mL) were compared with 60 eyes receiving the same regimen of combined therapy with a single intravitreal injection of triamcinolone (2 mg/0.05 mL) during the first IVB injection.[8] Due to clinical findings, additional IVB injections were administered in eyes with active CNV. Significant visual improvement and CMT reduction were observed in both arms of the study compared with the baseline values. Although the addition of IVTA seemed to be effective in reducing CMT in all follow-up sessions, no significant difference was found between two groups in this regard. On contrary, visual improvement seemed to be accelerated by the addition of IVTA at weeks 6 and 12, but this difference disappeared at the last follow-up visit (week 18). The same group had performed a similar randomized clinical trial three years earlier, in which 92 eyes of 90 patients with subfoveal and juxtafoveal CNV secondary to AMD had been randomly assigned to receive IVB (1.25 mg) alone or in combination with IVTA (2 mg).[10] Results demonstrated that both treatments (IVB vs. IVB/IVTA) were well tolerated for the management of neovascular AMD, but no significant difference was observed between the two modalities 6 weeks after treatment in terms of reduction of macular thickness, visual improvement, and improved leakage on FA. Lack of additive effect of triamcinolone acetonide in these two studies may have originated from inadequate dosage, low sample size, or short follow-up.[5] Particularly, a trend toward greater visual improvement had been seen with combined therapy in both studies.

There is no consensus between the physicians on the optimal dose of IVTA; however, it has been reported that 4 mg of this corticosteroid will generate a vitreous concentration approximately 20,000 times the one required to entirely saturate all receptors of triamcinolone in the cell cytoplasm.[14] Moreover, it has been claimed that the drug persistence in the vitreous will probably extend by increasing the levels of triamcinolone from 4 to 20 mg.[1415] However, higher concentrations of triamcinolone seems to be associated with multiple ocular complications including increased IOP, endophthalmitis, pseudoendophthalmitis, cataract formation, and increased risk of retinal detachment.[14] Besides, more severe IOP elevation has been reported in those eyes receiving repeat injections of IVTA.[16] Hence, a moderate dose (4 mg) of triamcinolone was administered as adjunctive treatment to bevacizumab only at the time of the initial injection in the present analysis. This concentration is two folds higher than that which had been previously used by other researchers,[810] and thus, is expected to achieve a longer duration of action. In fact, our results verified that a single injection of 4 mg IVTA could establish a significant benefit at least up to 12 months. However, it had no effect on the need for repeat injections of bevacizumab compared to IVB group.

It has been suggested that cataract surgery may increase the risk of development and progression of advanced AMD.[17] This phenomenon was attributed to increased inflammation, increased light toxicity, and postoperative cystoid macular edema.[18] On the other hand, multiple investigations have demonstrated that intravitreally injected drugs such as bevacizumab can be detected in the anterior chamber shortly after the injection.[1920] Thus, it is possible that the absence of the barrier function of the natural crystalline lens would result in more drug diffusion into the anterior chamber.[21] Our 12-month results, however, showed that, there is no significant difference between phakic and pseudophakic eyes in terms of VA and central foveal thickness. This finding is compatible with some other studies.[2223] Nevertheless, a larger number of cases and a longer follow-up are required to confirm these findings.

PDT selectively occludes vessels within CNV by non-thermal photothrombosis.[24] It has also been revealed that PDT leads to choriocapillaris hypoperfusion in the irradiated region which recovers within 3 months.[2526] Such effects may contribute to atrophy of the neurosensory retina and retinal pigment epithelial and therefore cause subsequent central scotomata.[27] Hence, combination therapy with an anti-VEGF agent plus PDT to manage CNV secondary to AMD has been growing in popularity in recent years. In the current investigation, history of PDT did not improve the visual outcomes of treatment and its effect on the reduction of CMT was marginal. Similar results were obtained in some previous studies.[82829] However, the current study was not designed to address this question. Besides, low number of patients with PDT history mandates caution in drawing broad conclusions.

The present analysis, not only investigated more patients but also had longer follow-up time than previous ones. Nevertheless, it suffers from a number of limitations. Most notably, CMT was measured by a time-domain OCT apparatus which has been reported to have lower reproducibility compared to spectral domain OCT in patients with wet AMD.[30] In addition, further studies with higher volume samples and multiple doses of bevacizumab/triamcinolone are needed to evaluate more accurately the effect of IVTA on neovascular AMD. A much longer follow-up period is also advised because it is rational to imagine that some treatment-related adverse events could arise much later. Moreover, restricting the study to only treatment-naive patients would produce more unbiased results.

Conclusion

According to the visual-gain achievement, low retreatment rates and sustainable CNV eradication result, it seems that the administered dosage of bevacizumab-triamcinolone combined therapy could be a useful treatment option for wet AMD. However, complications arising from this study warrant further evaluation.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.