Two-year Outcomes of Intravitreal Aflibercept Injection for Neovascular Age-related Macular Degeneration with “Observe before Treat-and-Extend” Method
Article information
Abstract
Purpose
To evaluate 2-year outcomes of intravitreal aflibercept injection for neovascular age-related macular degeneration (nAMD) treated with “observe before treat-and-extend (O-TAE)” strategy in the real-world setting.
Methods
This retrospective study included treatment-naive nAMD patients treated with aflibercept using O-TAE regimen and followed up for more than 2 years. Patients were observed bimonthly to check recurrence after three monthly loading injections. In case of recurrence, treatment was resumed using the TAE regimen starting from the fourth injection. In case of nonrecurrence, observation was continued. Best-corrected visual acuity (BCVA), central macular thickness (CMT), number of injections, TAE intervals, and proportion of recurrence after dry-up following three loadings were analyzed.
Results
A total of 38 eyes of 34 patients were included. Follow-up period was 37.0 ± 11.0 months. BCVA by logarithm of minimal angle of resolution improved from 0.33 ± 0.29 at baseline to 0.24 ± 0.23 in the first year (p = 0.010) and 0.25 ± 0.22 in the second year (p = 0.054). CMT decreased significantly from 357.43 ± 74.53 μm at baseline to 269.62 ± 48.12 μm in the first year (p < 0.001) and 279.14 ± 54.64 μm in the second year (p < 0.001). Number of injections were 5.11 ± 1.69 in the first year and 3.84 ± 2.39 in the second year. The percentage of eyes with a TAE interval of ≥12 weeks was 37.0% in the first year and 34.4% in the second year. Of the 36 eyes that dried up after three loadings, 28 eyes (77.8%) recurred, and the average period of recurrence was 6.5 months. The remaining eight eyes (22.2%) had no recurrence during the mean follow-up period of 29.7 months.
Conclusions
This study showed that the newly suggested O-TAE strategy can reduce the treatment burden significantly reducing the number of injections while improving BCVA and CMT in the first and second year.
Neovascular age-related macular degeneration (nAMD) is one of the leading causes of vision loss worldwide [1]. Despite many studies have shown great treatment results for nAMD with the introduction of anti–vascular endothelial growth factor (anti-VEGF) therapy [2–4], the high cost of the drugs and frequent visits place a burden to patients [5].
Subsequently, treatment regimens such as pro re nata (PRN, as needed) and treat-and-extend (TAE) were introduced. The PRN method requires monthly follow-up visit and injections in case of recurrence. There were reports that it was difficult to maintain improvement in vision if monthly follow-up visits were not made [6–8]. As an alternative, the TAE regimen was proposed to maintain vision improvement while reducing the inconvenience of monthly visits and the number of injections [9–11]. Several studies have reported that the TAE method has a vision improvement effect similar to monthly injections and can reduce the number of visits [12–14]. As such, TAE strategy is widely used as a treatment schedule for nAMD [15].
Here, we introduce “observe before TAE (O-TAE)” regimen for treatment of nAMD. The O-TAE regimen is a modification of TAE that has an observation period to detect recurrence before starting TAE. Although there are several studies reported long-term clinical real-world results for aflibercept using conventional TAE therapy in nAMD [16–20], there has been few reports on clinical outcomes of O-TAE strategy in nAMD. Jeong et al. [21] suggested a similar strategy called “modified observe-and-plan” regimen according to the individual recurrence interval in a 1-year follow-up retrospective study. The purpose of the present study was to evaluate 2-year clinical outcomes of intravitreal aflibercept injection for nAMD treated with O-TAE strategy.
Materials and Methods
Ethics statement
This study was approved by the Institutional Review Board of Ewha Womans University Medical Center (No. 2024-05-048). The requirement of informed consent was waived due to the retrospective nature of the study. The authors adhered to the tenets of the Declaration of Helsinki throughout the study.
Study design
This was a single centered, retrospective cohort study.
Study population
Inclusion criteria was as follows: (1) patients who were diagnosed with nAMD from March 1, 2019, to May 31, 2022, and treated with intravitreal aflibercept injections at our medical center (Ewha Womans University Mokdong Hospital, Seoul, Korea; and Ewha Womans University Seoul Hospital, Seoul, Korea); (2) treatment-naive patients; (3) treated with O-TAE strategy; and (4) followed up more than 2 years (i.e. the possible latest follow-up was May 31, 2024).
In order to extract cases that met the above inclusion criteria, we proceeded in the order of the flowchart in Fig. 1. First, using clinical data warehouse search, we retrieved the cases of intravitreal aflibercept injections performed from March 1, 2019, to July 31, 2022. A total of 1,750 cases were extracted. Among these, duplicate patients were removed based on registration number, leaving only one for each duplicate case; 388 patients remained. By searching their electronic medical records, 189 eyes of 178 patients with confirmed diagnosis of nAMD were extracted. Among the 189 eyes, 151 eyes were excluded. Exclusion criteria were the following: (1) previous history of nAMD treatment (anti-VEGF, laser, or photodynamic therapy; n = 72); (2) follow-up period of less than 2 years (n = 26); (3) regimen other than O-TAE (n = 15); (4) started treatment with bevacizumab or ranibizumab (n = 14); (5) switched to other anti-VEGF (n = 12); (6) cataract operation during follow- up periods (n = 6); (7) vitrectomy history or vitrectomy during follow-up periods (n = 5); and (8) other ocular disease of affecting visual acuity (n = 1). The remaining 38 eyes were finally enrolled and analyzed in the study.
Treatment strategy and subgroups
Here, we suggest a so-called O-TAE regimen. Patients included in the study received three monthly loading intra-vitreal injections of aflibercept 2 mg/0.05 mL (Eylea, Bayer Pharma) for nAMD. Patients were observed bimonthly to check recurrence. The recurrence was defined as new retinal hemorrhage or presence of subretinal and/or intraretinal fluid on spectral-domain optical coherence tomography (OCT; Spectralis, Heidelberg Engineering). In case of recurrence or if fluid remains after three loading injections, treatment was resumed using the TAE regimen starting from the fourth injection. In cases where there was no recurrence, follow-up observation was continued. When starting the TAE regimen, the injection interval started from 8 weeks and was adjusted by 2 weeks depending on whether there was recurrence. The injection interval was extended by 2 weeks without recurrence and shortened by 2 weeks with recurrence. In the present study, there were some cases without recurrence after three loading injections. Cases that started the TAE regimen were divided into two subgroups according to the minimum and maximum TAE intervals chosen by two physicians (SCC and HJK): group A with TAE interval from a minimum of 8 weeks to a maximum of 12 weeks and group B with TAE interval from a minimum of 4 weeks to a maximum of 16 weeks.
Clinical outcome measurements
At each visit, all patients underwent complete eye examinations, including measurement of the best-corrected visual acuity (BCVA; decimal values) and intraocular pressure, slit-lamp biomicroscopy, dilated fundus examinations, wide fundus photography (Nikon, Optos California), and OCT. Central macular thickness (CMT) was measured at each visit by OCT. The diagnosis of nAMD was confirmed with dilated fundus examination or wide fundus photography, OCT, fluorescein angiography, and indocyanine green angiography. BCVA, CMT, number of injections, TAE intervals, proportion of recurrence after drying up following three loading injections, average time to recurrence, and average follow-up period for nonrecurrence were analyzed.
Statistical analysis
Statistical analyses were performed using IBM SPSS ver. 27.0 (IBM Corp). The data are presented as the mean ± standard deviation. The BCVAs were converted to logarithm of the minimal angle of resolution (logMAR) for analysis. The BCVA and CMT were compared between baseline and the first or the second year using the Wilcoxon signed rank test. The Mann-Whitney U-test was used to analyze the difference between group A and group B. A p-value of 0.05 denoted statistically significance.
Results
A total of 38 eyes of 34 patients (21 men, 13 women) with treatment-naive nAMD were included. The mean age of the included patients was 72.0 ± 9.8 years. Baseline mean BCVA by logMAR was 0.33 ± 0.29 and baseline mean CMT was 357.43 ± 74.53 μm. Of the total 38 eyes, 28 eyes were typical nAMD and the remaining 10 eyes were polypoidal choroidal vasculopathy (PCV). All of the nine eyes of retinal angiomatous proliferation met the exclusion criteria and were not included in the study. The remaining demographic data are presented in Table 1. For a total of 30 eyes, the TAE regimen was started. Among these, one case which recurred 34.3 months after 3 loading injections was excluded from the subgroup analysis. Finally, 18 eyes were classified as group A and 11 eyes were classified as group B. Except for lens status, there were no significant differences in baseline factors between group A and group B (Supplementary Table 1). The proportion of phakic lenses was relatively higher in group A than in group B (p = 0.048).
For the total 38 eyes, the BCVA improved significantly in the first year and improved by marginal significance in the second year, compared with the baseline (Table 2 and Fig. 2). The BCVA by logMAR improved from 0.33 ± 0.29 at baseline to 0.24 ± 0.23 in the first year (p = 0.010), and 0.25 ± 0.22 in the second year (p = 0.054). There was no significant difference in BCVA between group A and group B at baseline, first year, and second years. Among the 10 eyes of PCV patients, none showed severe deterioration signs such as submacular hemorrhage or breakthrough hemorrhage.
The CMT decreased significantly in the first year, and second year compared with the baseline (Table 3 and Fig. 3). The CMT decreased significantly from 357.43 ± 74.53 μm at baseline to 269.62 ± 48.12 μm in the first year ( p < 0.001), and 279.14 ± 54.64 μm in the second year (p < 0.001). There was no significant difference in CMT between group A and B at baseline, 1 year, and 2 years.
The number of injections until the first year of follow-up was 5.11 ± 1.69 and the number of injections from the first to the second year of follow-up was 3.84 ± 2.39 (Table 4 and Fig. 4). The number of injections in the second year was significantly lower than in the first year (p < 0.001). There was no significant difference in the number of injections between group A and B in the first and second year. There was no significant difference in the number of injections between the eyes with typical nAMD and PCV in the first and second year (Supplementary Table 2).
When groups A and B were combined, the proportions of TAE interval were as shown in Fig. 5. In the second year, the percentage of eyes with a TAE interval of 8, 9–11, 12–15, and 16 weeks was 34.5%, 31.0%, 31.0%, and 3.4%, respectively. The percentage of eyes with a TAE interval of 12 weeks or more was 37.0% in the first year and 34.4% in the second year.
Among the total of 38 eyes, 36 eyes (94.7%) dried up (i.e., had no intraretinal or subretinal fluid) after 3 loading injections, and the remaining two eyes (5.3%) had some fluid (Table 5). Of the 36 eyes that dried up, 28 eyes (77.8%) recurred and the average period from the third loading injection to the recurrence was 6.5 ± 7.1 months. Of the 36 eyes that dried up, eight eyes (22.2%) had no recurrence during a mean follow-up period of 29.7 ± 8.1 months covering the third loading injection to the last follow-up. There was no significant difference in the proportion of the recurrence after dry-up nor in mean recurrence time between typical nAMD and PCV (Supplementary Table 3). None of the baseline factors was correlated with the recurrence after dry-up following three loading injections (Supplementary Table 4).
Discussion
In this study, O-TAE strategy showed improvement in BCVA and CMT in the first and second year compared to the baseline. The average number of injections was 5.1 in the first year and 3.8 in the second year. The percentage of eyes with a TAE interval of 12 weeks or more was 37.0% in the first year and 34.4% in the second year. Of the 36 eyes that dried up after three loading injections, 28 eyes (77.8%) recurred and the average period to recurrence was 6.5 months. The remaining eight eyes (22.2%) had no recurrence during the mean follow-up period of 29.7 months. None of the baseline factors was correlated with the recurrence.
In addition to the conventional standard TAE [22–24], various modified treatment regimens were introduced in the previous studies [21,25–30] (Table 6 and Fig. 6A–6E). “Modified TAE” regimens for nAMD were described in previous studies [25–27]. Ohnaka et al. [25] introduced modified TAE regimen, which consists of an induction phase, an observation phase, and a TAE phase (Fig. 6B). During the induction phase, patients received ≥3 monthly intravitreal aflibercept loading injections until no signs of disease activity. Patients were monitored monthly until disease activity appeared to determine the “recurrence interval” during the observation phase. The recurrence interval was defined as the time from when no signs of disease activity was achieved during the induction phase to when the first signs of recurrence appeared. During TAE phase, patients received monthly injections until no signs of disease activity. Then, the next injection was administered after initial “treatment interval” which was reduced by 2 weeks from the recurrence interval. Subsequent injections were administered with an interval reduced by 2 weeks for occasions with disease activity, and with an interval increased by 2 weeks for cases without disease activity [25]. “Observe-and-plan” regimen for treatment of nAMD was described first in the study on ranibizumab by Mantel et al. [28] and analyzed later in the study on aflibercept by Parvin et al. [29] and Subhi et al. [30]. After three loading injections, a monthly observation visit was conducted to determine the recurrence interval. Then, the recurrence interval was reduced by 2 weeks to determine the retreatment interval for the next two or three injections without intermittent ophthalmic examination during “treatment plan” phase (Fig. 6C) [28,29]. The available choice of retreatment intervals was limited to 1 to 3 months. Recurrences that were observed at ≥4 months since last injection was an indication to treat every 3 months. Follow-up assessment visits were performed with the same interval from the last injection of the “treatment plan.” The adjustment of the intervals in the subsequent treatment plans was performed according to the disease activity (Fig. 6C). Compared to the “modified TAE” or “observe-and-plan,” our O-TAE regimen is different in that observation was performed every 2 months after the third injection, and if recurrence occurred, TAE was started from the fourth injection with the treatment interval of 8 weeks. Jeong et al. [21] suggested a “modified observe-and-plan” strategy according to the interval of individual recurrence in a 1-year follow-up retrospective study (Fig. 6D). After three monthly loading injections of aflibercept, all patients were monitored monthly during the observation phase to determine whether a dry macula was attained in their study. Patients who showed exudative activity within 3 months were categorized to early-recurrence group. The other patients were allocated to the late-recurrence group. The early-recurrence group was treated with TAE regimen with adjusted interval of 2 weeks within 8 to 12 weeks. The late-recurrence group was treated as needed. TAE was applied only to the early-recurrence group showing disease activity within 3 months after 3 monthly loading injections in their study. Our O-TAE regimen is different from the “modified observe-and-plan” in that if recurrence occurred, TAE was started from the fourth injection regardless of the time of recurrence (Fig. 6E). Our study is also different in that patients were observed after three monthly loading injections with an interval of 2 months until recurrence (Fig. 6E). Comparison of the conventional TAE and other modified variations of the treatment regimen is summarized in Table 6 [21–30] and Fig. 6.
In the current study, BCVA and CMT improved in the first and second year compared to the baseline. There have been few previous studies on O-TAE analyzed in our study. It can be said that the method applied to the early-recurrence group of the “modified observe-and-plan” by Jeong et al. [21] is similar to the O-TAE regimen of our study. As in our study, BCVA and CMT were significantly improved in the first year, compared to the baseline in their study.
The mean number of injections was 5.1 ± 1.7 in the first year and 3.8 ± 2.4 in the second year in our study. This is lower than that of conventional TAE. Ishibashi et al. [19] reported retrospective study result of 4-year outcomes of intravitreal aflibercept injection for nAMD using TAE regimen in Japan in 39 eyes of 39 patients. TAE interval ranged from a minimum of 4 weeks to a maximum of 20 weeks in their study. BCVA and CMT improved significantly compared to baseline and was maintained until the fourth year in the study. The average number of injections was 7.9, 6.0, 5.5, and 5.4 in the first, second, third, and fourth year, respectively. The mean number of injections in the first and second year was lower in our study than in their study. Choi et al. [31] reported 1-year outcomes of the intravitreal aflibercept using TAE method in 36 treatment- naive nAMD in Korean population. The treatment interval was shortened or extended to a minimum of 8 weeks and a maximum of 16 weeks depending on disease activity. The average number of injections per year was 6.8 ± 0.5 in their study. The mean number of injections in the first year was lower in our study than in their study. In our O-TAE regimen, instead of immediately starting the fourth injection with TAE after 2 months following three loading injections, there was an observation phase to check recurrence after three monthly loadings. Therefore, the mean number of injections in the first and second year is reduced compared to the conventional TAE. As shown in Table 5, after drying up with three loading injections, recurrence did not occur in 22.2%, which presumably lowered the overall average number of injections. The significance of O-TAE is that it can reduce the treatment burden by significantly reducing the number of injections while improving BCVA and CMT.
In the current study, the percentage of eyes with a TAE interval of 12 weeks or more was 37.0% in the first year and 34.4% in the second year. In the study of Jeong et al. [21] which applied a strategy similar to our O-TAE regimen in their early-recurrence group, the percentages of patients with a TAE interval of 12 weeks were 25.0% in the early-recurrence group. In the study of Ishibashi et al. [19], the percentage of patients with a TAE interval of 12 weeks or more was 46.2%, 46.2%, 43.6%, and 46.2% in the first, second, third, and fourth year, respectively. In our study, TAE interval ranged from 8 to 12 weeks in group A and from 4 to 16 weeks in group B. In the study of Jeong et al. [21], TAE interval ranged from 8 to 12 weeks. In the study of Ishibashi et al. [19], TAE interval ranged from 4 to 20 weeks. These differences could be attributed to the difference in the above proportions among our study and theirs.
Out of the 36 eyes which dried up after three loading injections, eight eyes (22.2%) did not recur during an average follow-up period of 29.7 ± 8.1 months. This suggests that there is a certain proportion of patients who are stabilized without recurrence for a considerable period of time after three loading injections. This can be a clinical basis for having an observation period to check for recurrence after three monthly loading injections. If TAE is applied to these stable patients, there is a possibility of overtreatment. Among the 36 eyes which dried up after three loadings, the remaining 28 eyes (77.8%) recurred, taking an average of 6.5 ± 7.1 months until recurrence. In the study of Inoue et al. [32], the median period from the third loading injection to retreatment was 5 months. The range of period until recurrence after third loading injection was 1.9 to 34.3 months in our study, suggesting that long-term follow-up is important for nAMD. In the present study, among eyes which dried up after three loadings, there were no significant baseline factors between the recurrence group and the nonrecurrence group. In the study of Jeong et al. [21], no significant differences in the baseline characteristics were observed between the early- and late-recurrence groups. In the study of Kikushima et al. [33], older age and T-allele of ARMS2 A69S (rs10490924) were significant factors associated with retreatment after three loading injections. In the current study, age was not a significant factor in the recurrence. Genetic analysis was not performed in our study.
This study has several limitations. First, this study had a retrospective design and had an inherent risk of selection bias. The adjustment of TAE interval was not consistently unified and was performed according to the preferences of each physician. Nevertheless, the BCVA, CMT, number of injections were not significantly different between group A and group B. Second, the size of the study samples is small. However, by extracting data by combining clinical data warehouse and electronic medical records using the method described in the method section, the study populations in the current study were selected, without data loss from the entire neovascular AMD cohort diagnosed and treated at our medical center. In addition, exclusion criteria including treatment naïve cases were strictly applied through a thorough review of electronic medical records. Lastly, the entire population of this study was Korean, indicating that the results of the current study may not be applicable to the population of other ethnicities. The strength of this study is the investigation of the real-world outcomes of the O-TAE strategy for nAMD, which has not been studied well previously. Longer term studies with a larger number of patients on the clinical outcomes of O-TAE regimen are needed.
In conclusion, the 2-year long-term results of intravitreal aflibercept injection with O-TAE strategy showed effectiveness with functional and anatomical improvement for nAMD. In addition, O-TAE regimen was able to reduce the number of injections compared to conventional TAE. O-TAE strategy could be considered as an alternative in the treatment of nAMD.
Acknowledgements
None.
Notes
Conflicts of Interest
None.
Funding
None.
Supplementary Materials
Supplementary materials are available from https://doi.org/10.3341/kjo.2024.0083.
Supplementary Table 1. Baseline characteristics and demographics between the two subgroups
kjo-2024-0083-Supplementary-Table-1.pdfSupplementary Table 2. Number of injections in the first and second year according to the subtypes of nAMD
kjo-2024-0083-Supplementary-Table-2.pdfSupplementary Table 3. Dry-up after three loading injections and recurrence after dry-up, according to the subtypes of nAMD
kjo-2024-0083-Supplementary-Table-3.pdfSupplementary Table 4. Univariate and multivariate logistic regression analysis of variables for recurrence after three monthly aflibercept loading injections
kjo-2024-0083-Supplementary-Table-4.pdf