IMPACT OF RETINAL FLUID-FREE MONTHS ON OUTCOMES IN NEOVASCULAR AGE-RELATED MACULAR DEGENERATION: A Treatment Agnostic Analysis of the HAWK and HARRIER Studies : RETINA

Secondary Logo

Journal Logo

Original Study

IMPACT OF RETINAL FLUID-FREE MONTHS ON OUTCOMES IN NEOVASCULAR AGE-RELATED MACULAR DEGENERATION

A Treatment Agnostic Analysis of the HAWK and HARRIER Studies

Eichenbaum, David MD*,†; Brown, David M. MD; Ip, Michael MD§; Khanani, Arshad M. MD¶,**; Figueroa, Marta S. MD, PhD††; McAllister, Ian L. MBBS, PhD‡‡; Laude, Augustinus MBChB§§; B, Guruprasad MBBS, MD¶¶; Tang, Shuhan PhD¶¶; Gmeiner, Benjamin PhD***; Clemens, Andreas MD, PhD†††; Souied, Eric MD, PhD‡‡‡

Author Information
Retina 43(4):p 632-640, April 2023. | DOI: 10.1097/IAE.0000000000003699
  • Open

Subretinal fluid, (SRF), intraretinal fluid (IRF), and subretinal pigment epithelium fluid, visualized using optical coherence tomography, are important biomarkers used by retina specialists to identify neovascular age-related macular degeneration (nAMD) activity and guide treatment decisions.1 Treat-and-extend and pro re nata regimens use retinal fluid criteria for this purpose.

The Comparison of AMD Treatments Trials study showed that the presence of retinal fluid is not uncommon, even in regularly treated eyes. Over half of patients who received monthly ranibizumab in the study had some retinal fluid at Month 12.2 However, numerous studies have demonstrated that less frequently treated eyes have worse visual acuity than those eyes that have more intensive treatment.3 Regarding the need to treat with the aim of completely drying the retina, there is general agreement that the presence of IRF (at baseline [BL] or recurring) is predictive of worse visual function,4–17 but the scientific literature offers greater debate relating to the impact of the presence of SRF on prognosis in nAMD.9,18,19

In the Comparison of AMD Treatments Trials study of the anti–vascular endothelial growth factor (anti-VEGF) agents ranibizumab and bevacizumab in nAMD, a prospective cohort study within the randomized controlled trial suggested that eyes with residual IRF had worse visual acuity than those without IRF, whereas eyes with SRF had better visual acuity than eyes without SRF.9 These findings were broadly supported by analyses of the HARBOR and FLUID studies.18,19 In contrast, a post hoc analysis of the Vascular Endothelial Growth Factor VEGF Trap-Eye: Investigation of Efficacy and Safety in Wet Age-Related Macular Degeneration 1 and 2 studies suggested that more intensive treatment is beneficial in patients with persistent fluid, regardless of the fluid compartment.20 In addition, a volumetric analysis of HAWK and HARRIER data has demonstrated that a lower level of any fluid (IRF or SRF) is associated with better visual outcomes.21

The Phase 3 HAWK and HARRIER trials compared the efficacy and safety of the anti-VEGF agents brolucizumab and aflibercept for the treatment of nAMD,22,23 and provide a useful dataset for further elucidation of the association of retinal fluid with visual and morphologic outcomes. In this post hoc analysis of patient-level data from HAWK and HARRIER, the degree of fluid control is quantified in terms of the total number of monthly visits at which the patient was retinal fluid-free (IRF and/or SRF). The aim of this treatment-agnostic analysis is to assess the effect of the total number of retinal fluid-free months after the loading phase on best-corrected visual acuity (BCVA) and central subfield thickness (CSFT) outcomes in the HAWK and HARRIER trials.

Methods

Patient Population

HAWK (NCT02307682) and HARRIER (NCT02434328) were two 2-year, randomized, double-masked, multicenter active-controlled Phase 3 trials conducted at 408 sites in North, Central, and South America; Europe; Asia; Australia; and Japan.23 These trials were conducted in accordance with the principles of the Declaration of Helsinki, International Conference on Harmonization E6 Good Clinical Practice Consolidated Guideline, and other regulations as applicable, and were compliant with the Health Insurance Portability and Accountability Act of 1996. All trial participants provided written informed consent, and independent ethics committee/institutional review board approval was obtained for these trials. Eligible patients were aged 50 years or above, were treatment naive, with choroidal neovascularization lesions secondary to nAMD with presence of subfoveal fluid, and with BCVA of 78 to 23 Early Treatment Diabetic Retinopathy Study letters.23 This post hoc analysis, which was not included in the studies' prespecified statistical analysis plans, included pooled patient-level data from the brolucizumab 6 mg (n = 718) and aflibercept 2 mg (n = 715) arms of HAWK and HARRIER.

Fluid-free Categories

Data were analyzed for the study period between the end of the loading phase and the end of the study (Weeks 12–96, a maximum of 22 visits). Patients were assigned to one of five categories based on the total number of monthly visits during this period at which they were observed to be without retinal fluid, described as fluid-free visits (FFVs): Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (“almost always dry,” 15–21 FFV), and Category 5 (“always dry,” 22 FFV) (Figure 1).

F1
Fig. 1.:
Allocation of a patient to an FFV category. In this example, the patient had a total of seven FFVs (indicated by green ticks) between Week 12 and 96, and so would have been assigned to Category 2.

Apart from Categories 1 (0 FFV) and 5 (22 FFV), categories were equal in number in terms of ranges of FFV (seven each: 1–7; 8–14, and 15–21 FFV). Fluid-free visits did not need to occur consecutively to be counted as part of the total number of FFV.

Fluid Compartment Definitions

Analyses were performed using three definitions of fluid-free status as confirmed by the reading center, based on fluid compartment: absence of SRF and IRF (IRF&SRF-neg), absence of IRF with SRF absent or present (IRF-neg) and absence of SRF with IRF absent or present (SRF-neg).

Endpoints

The endpoints of this analysis were the total number of FFV from Weeks 12 to 96, change in BCVA from BL by fluid definition and fluid-free category, and change in CSFT from BL by fluid definition and fluid-free category.

Statistical Analysis

For statistical analysis, categories of total number of FFV were summarized with the number of observations, the number of observations for each treatment group, and the corresponding frequency and percent. The endpoints of change from BL in BCVA and CSFT were analyzed by fitting analyses of covariance models with categories of total number of FFV, values at BL and age categories (<75, ≥75 years) as fixed effects. Least square means for each total number of FFV group and group differences together with corresponding two-sided 95% confidence intervals were derived from the fitted model. Last observation carried forward imputation was used for the efficacy comparison in BCVA and CSFT. Central subfield thickness variability was analyzed at 8 weeks after most recent injection, by fluid category according to fluid-free definition IRF&SRF-neg.

Results

Fluid-free Category Allocation

Patients treated with brolucizumab or aflibercept (n = 1,433) were allocated to fluid categories by fluid compartment definition and treatment arm (Table 1). Category 4 (15–21 FFV) was the most commonly allocated fluid category regardless of drug or fluid definition. The highest proportion of patients was allocated to Category 5 (always dry) when the definition of fluid-free was IRF-neg (36.7%), whereas the highest proportion of patients was allocated to Category 1 (never dry) when the definition of fluid-free was IRF&SRF-neg (5.6%). Clinical characteristics of patients within each fluid category, according to the IRF&SRF-neg definition of fluid-free status, are summarized in Table 2.

Table 1. - Fluid Category Allocation by Fluid-free Definition and Treatment Arm
Definition of Fluid-free Fluid Category, n (%)
1
Never Dry
2 3 4 5
Always Dry
IRF&SRF-neg
Brolucizumab 6 mg (n = 718) 26 (3.6) 128 (17.8) 140 (19.5) 323 (45.0) 101 (14.1)
Aflibercept 2 mg (n = 715) 54 (7.6) 155 (21.7) 151 (21.1) 274 (38.3) 81 (11.3)
 Combined (n = 1,433) 80 (5.6) 283 (19.7) 291 (20.3) 597 (41.7) 182 (12.7)
IRF-neg
Brolucizumab 6 mg (n = 718) 23 (3.2) 77 (10.7) 70 (9.7) 287 (40.0) 261 (36.4)
Aflibercept 2 mg (n = 715) 16 (2.2) 80 (11.2) 80 (11.2) 274 (38.3) 265 (37.1)
 Combined (n = 1,433) 39 (2.7) 157 (11.0) 150 (10.5) 561 (39.1) 526 (36.7)
SRF-neg
Brolucizumab 6 mg (n = 718) 5 (0.7) 82 (11.4) 110 (15.3) 326 (45.4) 195 (27.2)
Aflibercept 2 mg (n = 715) 35 (4.9) 116 (16.2) 132 (18.5) 285 (39.9) 147 (20.6)
 Combined (n = 1,433) 40 (2.8) 198 (13.8) 242 (16.9) 611 (42.6) 342 (23.9)
Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV).
IRF-neg, absence of IRF with SRF either absent or present; IRF&SRF-neg, absence of both SRF and IRF; SRF-neg, absence of SRF with IRF either absent or present; FFV, fluid-free visits; IRF, intraretinal fluid; SRF, subretinal fluid.

Table 2. - Patient Demographics/BL Characteristics and Change From BL in BCVA and CSFT at Key Time Points According to the IRF&SRF-Neg Definition of Fluid-free Status
BL characteristics Fluid Category, n (%)
1
Never Dry (n = 80)
2 (n = 283) 3 (n = 291) 4 (n = 597) 5
Always Dry (n = 182)
Mean age, years (SD) 75.4 (9.17) 75.1 (9.01) 75.2 (8.62) 76.2 (8.48) 76.3 (7.77)
Gender male, n (%) 39 (48.8) 139 (49.1) 126 (43.3) 250 (41.9) 75 (41.2)
Lesion type, n (%)
 Predominantly classic 32 (40.0) 106 (37.6) 100 (34.4) 220 (37.0) 63 (34.8)
 Minimally classic 6 (7.5) 27 (9.6) 23 (7.9) 59 (9.9) 22 (12.2)
 Occult 42 (52.5) 149 (52.8) 168 (57.7) 315 (53.0) 96 (53.0)
Mean area of CNV lesion, mm2 (SD) 4.6 (4.1) 3.9 (3.8) 4.0 (4.2) 3.4 (3.6) 2.9 (3.2)
Mean BCVA, letters (SD) 56.1 (14.8) 60.0 (13.5) 61.8 (13.0) 61.1 (13.2) 61.0 (12.7)
Mean CSFT, µm (SD) 521.5 (174.2) 489.7 (167.0) 464.8 (161.8) 456.3 (154.0) 430.1 (141.2)
Mean change from BL in BCVA and CSFT
 Mean BCVA change from BL at week 48, letters (SD) 3.3 (12.3) 5.0 (14.1) 6.5 (14.0) 8.4 (12.0) 8.9 (11.0)
 Mean BCVA change from BL at week 96, letters (SD) 1.2 (15.1) 3.4 (16.0) 6.2 (14.8) 7.5 (13.9) 7.6 (13.9)
 Mean CSFT change from BL at week 48, µm (SD) −123.6 (176.3) −140.4 (165.7) −139.6 (139.7) −187.9 (140.3) −183.4 (139.0)
 Mean CSFT change from BL at week 96, µm (SD) −128.7 (183.7) −136.3 (175.0) −153.1 (150.0) −195.1 (146.6) −187.8 (139.3)
Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV).
BCVA, best-corrected visual acuity; CNV, choroidal neovascularization; CSFT, central sub-field thickness; FFV, fluid-free visits; IRF, intraretinal fluid; SD, standard deviation; SRF, subretinal fluid.

Change in Best-Corrected Visual Acuity by Fluid-free Category

Figure 2 shows change in BCVA from BL for the pooled population by fluid definition and fluid-free category. Across all three fluid definitions, patients allocated to Categories 4 and 5 consistently had the greatest gains in BCVA at Week 96, whereas patients allocated to Categories 1 and 2 consistently had the least gains in BCVA. Both SRF and IRF were deleterious to vision. BCVA outcomes at Week 96 were better in the “always dry” and “almost always dry” categories compared with the “never dry” category, regardless of whether the fluid in question was IRF, SRF or both (Figure 2B). At Week 96, the least square mean (95% confidence interval) difference in BCVA from BL between Category 5 (always dry) and Category 1 (never dry) was 8.1 (4.4, 11.9), 12.9 (8.3, 17.5), and 6.6 (1.9, 11.3) Early Treatment Diabetic Retinopathy Study letters, respectively, for fluid definitions IRF&SRF-neg, IRF-neg, and SRF-neg (Figure 3).

F2
Fig. 2.:
Change in BCVA from BL for pooled population by fluid definition and category. Graphs show least square means with covariate adjustment for BL BCVA and age, and bars show SE, for three definitions of fluid-free status: IRF&SRF-neg, IRF-neg, and SRF-neg. Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV). BCVA, best-corrected visual acuity; BL, baseline; FFV, fluid-free visits; IRF, intraretinal fluid; IRF-neg, absence of IRF with SRF either absent or present; IRF&SRF-neg, absence of both SRF and IRF; SE, standard error of the mean; SRF, subretinal fluid; SRF-neg, absence of SRF with IRF either absent or present.
F3
Fig. 3.:
Best-corrected visual acuity change from BL at Weeks 48, 72, and 96 compared between Categories 2 to 5 and Category 1 (reference). Forest plots show least square means and 95% confidence intervals of difference in BCVA change for pooled population by fluid definition and category, for three definitions of fluid-free status: IRF&SRF-neg, IRF-neg, and SRF-neg. Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV). BCVA, best-corrected visual acuity; FFV, fluid-free visits; IRF, intraretinal fluid; IRF-neg, absence of IRF with SRF either absent or present; IRF&SRF-neg, absence of both SRF and IRF; SRF, subretinal fluid; SRF-neg, absence of SRF with IRF either absent or present.

Change in Central Subfield Thickness by Fluid-free Category

The least squares mean change in CSFT from BL for the pooled population by fluid definition and fluid-free category is shown in Figure 4. Across all three fluid definitions, patients allocated to “always dry” and “almost always dry” (Categories 4 and 5) consistently had the greatest reductions in CSFT at Week 96, whereas patients allocated to Categories 1, 2, and 3 consistently had the least reductions in CSFT. The greatest differences in change in CSFT at Week 96 between the always dry and never dry categories were observed for fluid definition SRF-neg (Figure 4C). A “sawtooth pattern” of month-to-month fluctuations in CSFT was visible for each fluid definition and fluid-free category, but this was much more exaggerated for the least dry categories (especially IRF&SRF-neg Category 1 and SRF-neg Category 1) and was minimal for IRF&SRF-neg and SRF-neg Category 4 and 5 patients (Figure 4). Central subfield thickness variability at 8 weeks after most recent injection was analyzed by fluid category according to fluid-free definition IRF&SRF-neg (Figure 5). The results showed less CSFT variability over time in “always dry” (Category 5) and “almost always dry” (Category 4). Mean SD in these categories was 8.9 and 16.9, respectively, compared with 33.1 in the “never dry” category.

F4
Fig. 4.:
Change in CSFT from BL for pooled population by fluid definition and category. Graphs show least square means with covariate adjustment for BL CSFT and age, and bars show SE, for three definitions of fluid-free status: IRF&SRF-neg, IRF-neg, and SRF-neg. Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV). CSFT, central subfield thickness; FFV, fluid-free visits; IRF, intraretinal fluid; IRF-neg, absence of IRF with SRF either absent or present; IRF&SRF-neg, absence of both SRF and IRF; SE, standard error of the mean; SRF, subretinal fluid; SRF-neg, absence of SRF with IRF either absent or present.
F5
Fig. 5.:
Central subfield thickness variability at 8 weeks after most recent injection for the pooled population, by fluid category according to fluid-free definition IRF&SRF-neg. Fluid categories are based on the total number of monthly FFV during Weeks 12 to 96: Category 1 (“never dry,” 0 FFV), Category 2 (1–7 FFV), Category 3 (8–14 FFV), Category 4 (15–21 FFV), and Category 5 (“always dry,” 22 FFV). CSFT, central subfield thickness; FFV, fluid-free visits; IRF, intraretinal fluid; IRF&SRF-neg, absence of both IRF and SRF; SRF, subretinal fluid.

Fluid-free Category Allocation by Anti–Vascular Endothelial Growth Factor Therapy

In addition to the treatment agnostic analysis, the fluid-free category allocation analysis was also performed for the two subgroups of patients that received brolucizumab 6 mg (n = 718) and aflibercept 2 mg (n = 715) (Table 1). Numerically more patients in the aflibercept arm than the brolucizumab arm were allocated to “never dry” (Category 1) for the fluid definitions IRF&SRF-neg (7.6% vs. 3.6%) and SRF-neg (4.9% vs. 0.7%). Meanwhile, numerically, more patients in the brolucizumab arm than the aflibercept arm were allocated to “always dry” and “almost always dry” (Categories 4 and 5) for the fluid definitions IRF&SRF-neg (45.0% and 14.1% vs. 38.3% and 11.3%, respectively) and SRF-neg (45.4% and 27.2% vs. 39.9% and 20.6%, respectively).

Discussion

The presence or absence of retinal fluid on optical coherence tomography images is a key morphologic feature used to guide treatment decisions in nAMD. In this treatment-agnostic analysis of data from the HAWK and HARRIER studies, we examined the association between total retinal fluid-free months and visual and morphologic outcomes.

The fluid category allocation analysis showed that most patients in the HAWK and HARRIER studies achieved good anatomical outcomes. In the pooled population, most patients were “always dry” or “almost always dry” (Categories 5 and 4, respectively, with at least 15 of 22 FFV), ranging from 54.4% to 75.8%, depending on the fluid definition.

We found that patients with nAMD in Categories 4 and 5 had better BCVA outcomes after 2 years of anti-VEGF treatment, irrespective of fluid categorization, compared with patients with fewer FFV. As may be predicted, these patients also had greater and more stable reductions in CSFT. The least squares mean difference in BCVA gain at Week 96 between the driest and least dry categories was over a line of vision (6.6–12.9 letters depending on fluid definition), indicating a clinically meaningful difference.

Our findings are not in complete alignment with some previously published analyses. For example, a post hoc analysis of the HARBOR study reported that visual outcomes at 2 years were better in eyes with residual SRF versus those with resolved SRF, and worse in eyes with residual IRF versus resolved IRF.18 In our analysis, patients with any residual fluid (indicated by FFV of 21 or less) had consistently worse BCVA outcomes than those with no fluid (FFV of 22), regardless of the fluid definition. However, it is interesting to note that in our study, both the benefit of having no fluid and the negative impact of having persistent fluid appear more marked in the IRF-neg analysis compared with the SRF-neg analysis (Figure 2, B and C). This observation concurs with the prevailing opinion that IRF is more detrimental to vision than SRF.

Similarly, the results of the FLUID study suggest that some SRF can be tolerated without negatively impacting visual outcomes.19 However, in FLUID, both the “intensive” arm (in which patients were treated if they had any IRF, SRF or both) and “relaxed” arm (in which patients were treated if they had SRF >200 µm at the foveal center, or any IRF) received treatment with anti-VEGF therapy and even the “relaxed” arm was relatively intolerant of fluid. As a result, patients in both arms received more frequent injections than many receive in clinical practice (15.8 and 17 injections over 24 months for the relaxed and intensive arms, respectively). In spite of this, the visual gains in the FLUID study were unusually low, possibly because of a relatively high BL BCVA leading to a ceiling effect in potential scope for improvement in vision. Both arms achieved an approximate 3-letter gain at 24 months compared with 6 to 9 letter average gains with other treat-and-extend studies.19

A recurring feature of those analyses performed to date that have shown that eyes with SRF achieve greater gains in BCVA compared with eyes who are SRF-free is the inclusion in the study design of a pro re nata arm. In a pro re nata regimen, the decision to retreat is postponed until the appearance of fluid, which can have a confounding effect on the outcomes of these analyses. It means that even a patient with good visual acuity who is responding well to treatment is likely to have multiple recurrences of fluid over the course of a study, and the association of visual gains with SRF may be correlative more than causative.

Unlike previous analyses that analyzed fluid in cross-sectional, associative methodology, the current analysis analyzes the total cumulative impact of absence of retinal fluid overall (IRF and SRF) and by IRF and SRF on an individual patient after the loading phase, providing an alternative way of looking at the cumulative impact of fluid on vision over the course of the entire study.

In an earlier volumetric analysis of the HAWK and HARRIER studies, it was shown that patients with lower fluid volumes are associated with better visual outcomes, irrespective of the fluid status (IRF or SRF).21 The current study also demonstrated that eyes that were “always dry” or “almost always dry” were associated with reduced CSFT fluctuations. Although some of the increased levels of month-to-month fluctuations in CSFT visible for the least dry categories in Figure 4 may be explained by the fact that brolucizumab-treated patients in these categories were likely treated q8w, whereas those in the mostly or always dry categories will have likely been treated q12w, the subsequent analysis looking at variability 8 weeks after the most recent injection addresses this shortcoming. Previous analyses of data from Comparison of AMD Treatments Trials/Inhibition of VEGF in Age-related choroidal Neovascularisation and HAWK/HARRIER have demonstrated that greater variability in retinal thickness is associated with worse BCVA outcomes24,25 and development of fibrosis and macular atrophy.25

Given the above, it seems that lower CSFT variability is reflective of the total cumulative retinal fluid absence after the loading phase and its association with vision. Further investigation of this finding in different nAMD studies and in different regimens (pro re nata, treat-and-extend, fixed) would be of merit to confirm the importance of minimizing long-term CSFT variability and how best this may be achieved.

Although the main focus of this analysis was treatment-agnostic, a treatment-specific fluid-free category allocation analysis was additionally performed to determine whether the assigned anti-VEGF therapy has any impact on the number of FFV. This analysis showed that a numerically greater proportion of patients receiving brolucizumab were “always dry” or “almost always dry” compared with those receiving aflibercept, reflecting the superior anatomical outcomes reported in the original analysis of these studies.22,23 However, no statistical comparison was made between therapies and therefore no conclusions should be drawn from this analysis.

The limitations of this analysis include the post hoc nature of the work. Also, treatment-specific analysis of BCVA or CSFT outcomes was not possible because patient numbers in the categories were not matched and/or were very low, and hence such an analysis would be statistically inappropriate. Finally, a limitation of most nAMD data sets is the relatively short-term nature of the data. Further analyses of ongoing longer-term prospective data sets could validate whether tolerance or intolerance of fluid over time has a long-term effect on visual acuity.

In conclusion, our analysis shows that absence of retinal fluid at more visits after loading phase has a positive association with visual and anatomic outcomes in nAMD patients, regardless of fluid type. With additional supporting evidence, “always dry” or “almost always dry” IRF and SRF status may be considered as a key treatment goal in order to reduce CSFT variability and improve visual outcomes in patients with nAMD undergoing anti-VEGF therapy.

Acknowledgments

The authors thank Jennifer Green (Green Ink Communications Ltd, United Kingdom) for medical writing and editorial assistance toward the development of this article.

References

1. Schmidt-Erfurth U, Chong V, Loewenstein A, et al. Guidelines for the management of neovascular age-related macular degeneration by the European Society of Retina Specialists (EURETINA). Br J Ophthalmol 2014;98:1144–1167.
2. CATT Research Group, Martin DF, Maguire MG, Ying G-s, et al. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med 2011;364:1897–1908.
3. Mones J, Singh RP, Bandello F, et al. Undertreatment of neovascular age-related macular degeneration after 10 years of anti-vascular endothelial growth factor therapy in the real world: the need for a change of mindset. Ophthalmologica 2020;243:1–8.
4. Dervenis N, Younis S. Macular morphology and response to ranibizumab treatment in patients with wet age-related macular degeneration. Clin Ophthalmol 2016;10:1117–1122.
5. Lee H, Jo A, Kim HC. Three-dimensional analysis of morphologic changes and visual outcomes in neovascular age-related macular degeneration. Invest Ophthalmol Vis Sci 2017;58:1337–1345.
6. Ying GS, Kim BJ, Maguire MG, et al. Sustained visual acuity loss in the comparison of age-related macular degeneration treatments trials. JAMA Ophthalmol 2014;132:915–921.
7. Sagiv O, Zloto O, Moroz I, Moisseiev J. Different clinical courses on long-term follow-up of age-related macular degeneration patients treated with intravitreal anti-vascular endothelial growth factor injections. Ophthalmologica 2017;238:217–225.
8. Chatziralli I, Nicholson L, Vrizidou E, et al. Predictors of outcome in patients with neovascular age-related macular degeneration switched from ranibizumab to 8-weekly aflibercept. Ophthalmology 2016;123:1762–1770.
9. Jaffe GJ, Martin DF, Toth CA, et al. Macular morphology and visual acuity in the comparison of age-related macular degeneration treatments trials. Ophthalmology 2013;120:1860–1870.
10. Jaffe GJ, Ying GS, Toth CA, et al. Macular morphology and visual acuity in year five of the comparison of age-related macular degeneration treatments trials. Ophthalmology 2019;126:252–260.
11. Sharma S, Toth CA, Daniel E, et al. Macular morphology and visual acuity in the second year of the comparison of age-related macular degeneration treatments trials. Ophthalmology 2016;123:865–875.
12. Wickremasinghe SS, Sandhu SS, Busija L, et al. Predictors of AMD treatment response. Ophthalmology 2012;119:2413–2414.e5.
13. Kodjikian L, Decullier E, Souied EH, et al. Predictors of one-year visual outcomes after anti-vascular endothelial growth factor treatment for neovascular age-related macular degeneration. Retina 2018;38:1492–1499.
14. Segal O, Barayev E, Nemet AY, et al. Prognostic value of hyperreflective foci in neovascular age-related macular degeneration treated with bevacizumab. Retina 2016;36:2175–2182.
15. Wickremasinghe SS, Janakan V, Sandhu SS, et al. Implication of recurrent or retained fluid on optical coherence tomography for visual acuity during active treatment of neovascular age-related macular degeneration with a treat and extend protocol. Retina 2016;36:1331–1339.
16. Schmidt-Erfurth U, Waldstein SM. A paradigm shift in imaging biomarkers in neovascular age-related macular degeneration. Prog Retin Eye Res 2016;50:1–24.
17. Waldstein SM, Simader C, Staurenghi G, et al. Morphology and visual acuity in aflibercept and ranibizumab therapy for neovascular age-related macular degeneration in the VIEW trials. Ophthalmology 2016;123:1521–1529.
18. Holekamp NM, Sadda S, Sarraf D, et al. Effect of residual retinal fluid on visual function in ranibizumab-treated neovascular age-related macular degeneration: effect of retinal fluid on vision outcomes in HARBOR. Am J Ophthalmol 2022;233:8–17.
19. Guymer RH, Markey CM, McAllister IL, et al. Tolerating subretinal fluid in neovascular age-related macular degeneration treated with ranibizumab using a treat-and-extend regimen: FLUID study 24-month results. Ophthalmology 2019;126:723–734.
20. Jaffe GJ, Kaiser PK, Thompson D, et al. Differential response to anti-VEGF regimens in age-related macular degeneration patients with early persistent retinal fluid. Ophthalmology 2016;123:1856–1864.
21. Schmidt-Erfurth U, Mulyukov Z, Gerendas BS, et al. Therapeutic response in the HAWK and HARRIER trials using deep learning in retinal fluid volume and compartment analysis. Eye (Lond) 2022; doi: 10.1038/s41433-022-02077-4. Online ahead of print.
22. Dugel PU, Singh RP, Koh A, et al. HAWK and HARRIER: ninety-six-week outcomes from the phase 3 trials of brolucizumab for neovascular age-related macular degeneration. Ophthalmology 2021;128:89–99.
23. Dugel PU, Koh A, Ogura Y, et al. HAWK and HARRIER: phase 3, multicenter, randomized, double-masked trials of brolucizumab for neovascular age-related macular degeneration. Ophthalmology 2020;127:72–84.
24. Jhaveri C, Dugel P, Wykoff CC, et al. Visual and anatomical outcomes for brolucizumab and aflibercept in patients with nAMD: 96-week data from HAWK and HARRIER. Data Presented at the ASRS Annual Meeting. Chicago, USA. July 26‒30, 2019.
25. Evans RN, Reeves BC, Maguire MG, et al. Associations of variation in retinal thickness with visual acuity and anatomic outcomes in eyes with neovascular age-related macular degeneration lesions treated with anti-vascular endothelial growth factor agents. JAMA Ophthalmol 2020;138:1043–1051.
Keywords:

aflibercept; anti–vascular endothelial growth factor; brolucizumab; neovascular age-related macular degeneration; retinal fluid

Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the Opthalmic Communications Society, Inc.