Adv Ther

https://doi.org/10.1007/s12325-020-01236-x

ORIGINAL RESEARCH

Efficacy and Safety of Intravitreal Aflibercept Treat-

and-Extend Regimens in Exudative Age-Related
Macular Degeneration: 52- and 96-Week Findings
from ALTAIR
A Randomized Controlled Trial

Masahito Ohji . Kanji Takahashi . Annabelle A. Okada . Masato Kobayashi .

Yoshimi Matsuda . Yasuhiro Terano . for the ALTAIR Investigators

Received: December 17, 2019

Ó The Author(s) 2020

ABSTRACT Methods: Adults aged at least 50 years old with

exudative AMD and best-corrected visual acuity
Purpose: To evaluate efficacy and safety of in- (BCVA) of 73–25 Early Treatment Diabetic
travitreal injections of aflibercept (IVT-AFL) Retinopathy Study (ETDRS) letters were inclu-
treat-and-extend (T&E) dosing regimens in ded. Patients received three monthly doses of
treatment-naı̈ve patients with exudative age- IVT-AFL 2 mg. At week 16, patients were ran-
related macular degeneration (AMD). domized 1:1 to IVT-AFL T&E with either 2- or
4-week adjustments. The primary endpoint was
mean change in BCVA from baseline to
week 52. Outcomes were assessed at weeks 52
and 96.
Results: Baseline characteristics were compara-
Electronic supplementary material The online
version of this article (https://doi.org/10.1007/s12325-
ble between the groups (n = 123 each). Over
020-01236-x) contains supplementary material, which is 52 weeks, mean number of injections was 7.2
available to authorized users. and 6.9 and mean last injection interval was
10.7 and 11.8 weeks, for the 2- and 4-week
M. Ohji (&) groups, respectively. From baseline, mean
Department of Ophthalmology, Shiga University of change in BCVA was ? 9.0 and ? 8.4 letters
Medical Science, Seta Tsukinowa-cho, Otsu, Shiga,
Japan (week 52) and ? 7.6 and ? 6.1 letters (week 96);
e-mail: [email protected] mean change in central retinal thickness was
- 134.4 lm and - 126.1 lm (week 52) and
K. Takahashi
- 130.5 lm and - 125.3 lm (week 96). Last
Department of Ophthalmology, Kansai Medical
University, School of Medicine, Hirakata City, injection interval before week 52 was at least
Osaka, Japan 12 weeks in 42.3% and 49.6% of patients and
56.9% and 60.2% before week 96. Over
A. A. Okada
Department of Ophthalmology, Kyorin University, 96 weeks, mean number of injections was 10.4
School of Medicine, Mitaka-shi, Tokyo, Japan (both groups). The safety profile of IVT-AFL was
consistent with previous reports.
M. Kobayashi
Bayer Yakuhin Ltd, Marunouchi 1-chome, Chiyoda-
Conclusions: IVT-AFL administered using two
ku, Tokyo, Japan different T&E regimens for treatment-naı̈ve
exudative AMD improved functional and ana-
Y. Matsuda
Y. Terano
tomic outcomes at week 52 and outcomes were
Bayer Yakuhin Ltd, Umeda 2-chome, Kita-ku, Osaka,
Japan
 Adv Ther

maintained to week 96. Outcomes were similar DIGITAL FEATURES

between the 2- and 4-week groups.
Trial Registration: ClinicalTrials.gov identifier,
This article is published with digital features,
NCT02305238.
including a summary slide and video abstract,
to facilitate understanding of the article. To
Keywords: Aflibercept; Anti-vascular endothe- view digital features for this article go to https://
lial growth factor agents; Exudative age-related doi.org/10.6084/m9.figshare.11603511.
macular degeneration; Ophthalmology; Treat-
and-extend
INTRODUCTION
Key Summary Points
Age-related macular degeneration (AMD) is a
leading cause of blindness worldwide and
Why carry out this study?
accounts for 10.9% of visual impairment in
The goal of proactive flexible treat-and- Japan [1, 2]. AMD-related vision loss is primarily
extend (T&E) regimens is to reduce the associated with the exudative form of AMD [3],
treatment burden associated with anti- which is characterized by abnormal growth of
vascular endothelial growth factor (VEGF) new blood vessels in the macula [4].
therapy, while maintaining a patient’s Overexpression of vascular endothelial
visual acuity gains. growth factor (VEGF) plays a significant role in
the development of new blood vessels [4, 5].
The aim of this study was to evaluate the
Treatment of exudative AMD involves admin-
efficacy and safety of intravitreal
istering anti-VEGF agents, such as aflibercept
aflibercept (IVT-AFL) T&E dosing
and ranibizumab, which are an effective treat-
regimens in treatment-naı̈ve patients with
ment option for patients with exudative AMD [4].
exudative age-related macular
Although rapid functional and anatomic
degeneration (AMD).
improvements are evident in the first year of
What was learned from the study? treatment, regression to baseline after initial
gains is common in a clinical setting [6–10].
IVT-AFL administered using either 2- or Improvements achieved during the first and
4-week adjustment T&E regimens in second year of protocol-mandated treatment
treatment-naı̈ve patients with exudative are difficult to maintain in clinical practice
AMD improved functional (best-corrected [6–8]. Anti-VEGF treatments are associated with
visual acuity ? 9.0 and ? 8.4 letters) and high clinic and patient burden, which may
anatomic outcomes (central retinal reduce long-term compliance and increase
thickness - 134.4 and - 126.1 lm) at healthcare costs [6, 7]. If the intervals between
52 weeks; functional and anatomic treatments are too short, patients may be over-
outcomes were maintained over 96 weeks. treated, potentially leading to unnecessary
A large proportion of patients (35.1% and treatment burden and an increased risk of
40.5%) had an intended injection interval adverse events (AEs) associated with intravitreal
of 16 weeks at week 52. injection procedure (such as conjunctival
hemorrhage and endophthalmitis). Conversely,
The incidence of treatment-emergent if the intervals between treatments are too long,
adverse events was consistent with the patients may be undertreated, potentially lead-
known safety profile of IVT-AFL. ing to inferior functional and anatomic out-
IVT-AFL T&E regimens were efficacious comes [9]. Therefore, the goal of anti-VEGF
and safe over 96 weeks of treatment using treatment for exudative AMD is to improve and
either 2- or 4-week adjustments. maintain functional and anatomic gains over
and beyond the first year of treatment, while
minimizing treatment burden on patients [7, 9].
Adv Ther

The goal of flexible treatment strategies, subject, a clinical study was required to explore
including pro re nata (PRN) and proactive treat- the optimal T&E regimen(s) that could be applied
and-extend (T&E) regimens, is to reduce the in real-world clinical practice.
treatment burden associated with anti-VEGF The aim of this randomized study was to
therapy, while maintaining visual acuity (VA) examine the efficacy and safety of IVT-AFL
gains [7, 9, 11]. T&E is a proactive, individualized administered in two different T&E regimens
dosing strategy whereby the injection interval (2- and 4-week adjustments) in patients with
can be gradually extended if functional and exudative AMD, while allowing a minimum
anatomic stability is maintained, and the inter- interval of 8 weeks and a maximum interval of
val shortened if deterioration is observed, in 16 weeks.
order to minimize the risk of disease recurrence
rather than in response to it [6]. For intravitreal
aflibercept (IVT-AFL) injection, proactive T&E METHODS
approaches are effective for maintaining func-
tional outcomes in patients with exudative Study Design
AMD, while reducing the burden of clinic visits
[6]. IVT-AFL T&E regimens offer patients an ALTAIR was a 96-week, randomized, open-label,
individualized treatment approach whereby the phase 4 study (ClinicalTrials.gov identifier,
patient’s injection interval is decided on the basis NCT02305238) that investigated the efficacy
of functional and anatomic evaluations. This and safety of repeated doses of IVT-AFL with two
flexible approach avoids both overtreatment and different T&E approaches in patients with
undertreatment, while optimizing functional exudative AMD. The study was conducted at 41
and anatomic outcomes. Additionally, patients centers across Japan between December 2014
receive an injection at every visit and therefore and November 2017, in accordance with the
physicians can adapt the treatment plan Declaration of Helsinki and the International
accordingly [7]. The functional and anatomic Conference on Harmonization guidelines E6:
improvements achieved during the initial dosing Good Clinical Practice. The protocol and any
phase of PRN regimens are often not sustained amendments were approved by the independent
unless frequent monitoring and strict retreat- ethics committee or institutional review board
ment criteria are applied. Therefore, individual- (IRB) at each study site (see supplementary
ized proactive T&E regimens are generally material). There was no central IRB involved in
preferred over PRN regimens [7]. the study and the protocol was reviewed and
The optimal T&E regimen for the treatment of approved by the IRB at each participating center.
exudative AMD is yet to be determined, particu- All patients provided written informed consent.
larly the ideal amount of time that the interval
can be adjusted by at any given time and the Participants
maximum injection interval offered [7, 11].
Recommended guidelines for patients in the Adults aged at least 50 years old with exudative
Asia–Pacific region suggest that the IVT-AFL changes due to active subfoveal choroidal neo-
injection interval can be extended by 4-week vascularization (CNV) lesions secondary to
increments following the initial three doses, to a AMD, including juxtafoveal lesions that affec-
maximum interval of 12 weeks for patients with ted the fovea, as evidenced by fluorescein
inactive disease [11]. To date, no randomized angiography (FA) in the study eye, were
controlled studies have examined in detail the included. Patients were required to have
outcomes associated with 4-week adjustments best-corrected visual acuity (BCVA) of 73–25
and injection intervals beyond 12 weeks. Further Early Treatment Diabetic Retinopathy Study
evaluation could be beneficial for identifying (ETDRS) letters (approximately 20/40–20/320
patients that are suitable for 2- or 4-week adjust- Snellen equivalent) in the study eye. If both
ments and injection intervals of up to 16 weeks eyes met the inclusion criteria, the eye with the
[11]. Given the lack of published evidence on this worst BCVA was selected as the study eye. Both
 Adv Ther

eyes could be treated, but only one eye per extension or shortening thereafter was set as
patient was analyzed. Exclusion criteria are lis- 2 weeks as a conservative measure to ensure the
ted in Table S1 in the supplementary material. best possible VA outcomes for these patients.
The injection interval was maintained in both
Randomization and Interventions groups without change if the criteria for treat-
ment adjustment were not met and residual
The study design is summarized in Fig. S1 in the fluid was decreased from the previous visit. The
supplementary material. This study was con- criteria for shortening, maintaining, or extend-
ducted as a randomized trial to enable a ing the injection interval are shown in Table 1.
descriptive comparison of the two treatment The minimum and maximum injection inter-
arms without introducing selection bias for one vals were 8 and 16 weeks, respectively.
T&E strategy over the other. Block randomiza-
tion was stratified by baseline BCVA (\ 55 and Study Endpoints
C 55 ETDRS letters) and exudative AMD subtype
(presence or absence of polypoidal choroidal The primary endpoint was mean change in
vasculopathy [PCV], as decided by the investi- BCVA (ETDRS letters) from baseline to week 52.
gator). Patients received three initial monthly Secondary endpoints included, but were not
doses of IVT-AFL 2 mg in the study eye. Patients restricted to, the proportion of patients who lost
received IVT-AFL at week 16 and were random- fewer than 15 letters (vision maintenance) or
ized 1:1 to receive IVT-AFL in a T&E regimen gained at least 15 letters, the proportion of
with either a 2-week (IVT-AFL-2W) or 4-week patients without fluid on OCT, and the mean
(IVT-AFL-4W) adjustment, to minimize poten- change in central retinal thickness (CRT) from
tial imbalances during treatment initiation. baseline to week 52. The mean number of
Patients received an IVT-AFL injection at IVT-AFL injections, the mean injection interval,
every pre-scheduled treatment visit. The timing the last injection interval (the interval between
of treatment visits following IVT-AFL injection the last two injections), and the intended injec-
at week 16 was determined at the previous visit tion interval (determined by the investigator as
by the treating physician on the basis of evalu- the next planned interval and based on pre-
ation and pre-defined treatment criteria. defined treatment criteria evaluating the efficacy
Patients were evaluated at weeks 52 and 96, of the previous injection) were recorded. End-
regardless of treatment schedule. Ophthalmic points were assessed at weeks 52 and 96.
evaluations were conducted at every treatment AEs were treatment-emergent events if they
and evaluation visit, except for safety follow-up, occurred or worsened after the first IVT-AFL
using VA testing with ETDRS letter score, slit dose and at most 30 days after the last dose. All
lamp and indirect ophthalmoscopy, and optical AEs were reported in case report forms and
coherence tomography (OCT). Fundus photog- coded using Medical Dictionary for Regulatory
raphy, FA, and indocyanine green angiography Activities version 19.1. An adjudication of AEs
were conducted at the screening visit and at according to the Antiplatelet Trialists’ Collabo-
weeks 52 and 96. Safety evaluations were con- ration (APTC) criteria was performed.
ducted at every treatment and evaluation visit,
including the follow-up.
Patients who met the criteria for treatment Statistical Analysis
adjustment in the IVT-AFL-2W group had their
injection interval extended or shortened by The sample size of 240 randomized patients was
2 weeks, while patients who met the criteria for calculated on the basis of an assumed standard
treatment adjustment in the IVT-AFL-4W group deviation (SD) of 12.5 ETDRS letters in BCVA
had their injection interval extended or short- change from baseline to week 52, non-inferior-
ened by 4 weeks. If patients in the IVT-AFL-4W ity delta of 5 letters, and a two-sided alpha of
group underwent an interval shortening of 0.05 between the two treatment arms. While
4 weeks (due to disease activity), any possible the objective of this study was to describe the
Adv Ther

Table 1 Criteria for shortening, maintaining, or extending the IVT-AFL injection interval

Criteria for shortening the treatment interval

When any of the following criteria are met for the study eye, the subsequent treatment interval was shortened:
New or persistent fluid with unchanged or increased fluid volume from measurement at the previous treatment visit as
indicated by OCT
Loss of C 5 ETDRS letters from the previous visit in conjunction with recurrent fluid on OCT
An increase in CRT of C 100 lm at the central 1 mm compared with the lowest previous value measured by OCT
New-onset neovascularization as determined at the investigator’s discretion based on review of fundus examination
and multi-imaging assessment if deemed necessary
New macular hemorrhage
New fluid or persistent intra- or subretinal fluid with unchanged or increased fluid volume from the previous visit as
indicated by total OCT scan area (all volumetric fluid assessments were derived from multiple cross-sectional images
and extracted from the OCT report)
Criteria for maintaining the treatment interval
If none of the criteria for shortening were met and residual fluid had decreased from the previous visit, then the
treatment interval was maintained without change, even with persistent fluid
Criteria for extending the treatment interval
If none of the criteria for shortening were met and there was no fluid on OCT, then the interval was extended
CRT central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, IVT-AFL intravitreal aflibercept,
OCT optical coherence tomography

outcomes achieved with both treatment regi- descriptively, as no confirmatory statistical

mens, the sample size was calculated with a analysis was intended.
power of 86%, which allowed for descriptive Statistical evaluation was performed using
statistical comparison. The two-sided 95% con- Statistical Analysis Software v9.2 or higher (SAS
fidence intervals (CI) of mean change in BCVA Institute Inc., Cary, NC, USA). Detailed statisti-
were estimated per treatment arm using one- cal methodology is available in the supple-
sample t statistics. Any treatment differences for mental methods.
the two treatment arms were estimated using an
analysis of covariance model with treatment
arm and exudative AMD subtype as fixed effect RESULTS
and baseline BCVA as a covariate. Secondary
visual outcomes were summarized descriptively, Patients
and their two-sided 95% CIs were estimated
using normal approximation; any point esti- In total, 288 patients were enrolled and 255
mate treatment differences were estimated patients completed screening. Of these, 254
using the Mantel–Haenszel (MH) method strat- patients received at least one of the three initial
ified by baseline BCVA (\ 55 or C 55 ETDRS monthly doses of IVT-AFL and comprised the
letters) and exudative AMD subtype. Patients in safety analysis set. In total, of 247 patients
the IVT-AFL-4W group who had their interval treated with IVT-AFL, 124 patients were ran-
shortened remained in the 4-week group for domized to the 2-week group and 123 patients
data analysis. All statistical analyses were to the 4-week group. Eight patients were not
exploratory and outcomes were summarized randomized. One patient in the IVT-AFL-2W
 Adv Ther

Fig. 1 Patient disposition. *A total of eight patients were logistic difficulties (n = 1), or no IVT-AFL injection
not randomized because of withdrawal by patient (n = 3), (n = 1). 2W/4W 2-/4-week adjustment, AE adverse event,
protocol violation (n = 2), physician decision (n = 1), IVT-AFL intravitreal aflibercept

group was not included in the full analysis set Treatment Exposure
(FAS) as the post-randomization assessment was
not available because of premature discontinu- The mean (SD) number of IVT-AFL injections
ation (death) immediately after randomization was 7.2 (0.9) in the IVT-AFL-2W group and
at week 16 (Fig. 1). Overall, 85.8% (n = 212) of 6.9 (1.0) in the IVT-AFL-4W group (baseline to
patients completed the study. The main reasons week 52) and 3.6 (1.6) in the IVT-AFL-2W group
for discontinuation were progressive disease and 3.7 (1.4) in the IVT-AFL-4W group (weeks
(n = 8), withdrawal by patient (n = 7), and AEs 54–96). Overall, from baseline to week 96, the
(n = 6) (Fig. 1). mean (SD) number of IVT-AFL injections was
Patient baseline demographics were similar almost identical between the IVT-AFL-2W and
between the two groups (Table 2). The overall IVT-AFL-4W groups (10.4 [2.6] and 10.4 [2.4],
mean (SD) age was 74.0 (8.0) years, baseline respectively).
BCVA was 55.0 (12.5) ETDRS letters, and most The mean (SD) last interval at week 52 was
patients were male (72.4%). 10.7 (2.7) weeks and 11.8 (3.7) weeks, and
at week 96 was 12.2 (3.6) weeks and
Adv Ther

Table 2 Patient baseline characteristics and demographics (FAS)

Characteristic IVT-AFL-2W adjustment IVT-AFL-4W adjustment Total
n 5 123 n 5 123 N 5 246
Age, mean (SD), years 73.0 (7.9) 75.0 (8.1) 74.0 (8.0)
Sex, n (%)
Male 87 (70.7) 91 (74.0) 178 (72.4)
Baseline BCVA score, mean (SD) 54.8 (13.1) 55.3 (12.0) 55.0 (12.5)
Baseline CRT, mean (SD), lm 386.2 (159.2) 370.3 (120.0) 378.3 (141.0)
Type of exudative AMD, n (%)a
Typical AMD 75 (61.0) 75 (61.0) 150 (61.0)
Polypoidal choroidal vasculopathy 46 (37.4) 44 (35.8) 90 (36.6)
Retinal angiomatous proliferation 4 (3.3) 9 (7.3) 13 (5.3)
Type of CNV lesions on FA, n (%)b
Classic CNV 35 (28.5) 42 (34.1) 77 (31.3)
Classic CNV and occult CNV 14 (11.4) 17 (13.8) 31 (12.6)
Occult CNV 72 (58.5) 62 (50.4) 134 (54.5)
Pigment epithelial detachment, n (%) 83 (67.5) 78 (63.4) 161 (65.4)
Subretinal fluid, n (%) 104 (84.6) 104 (84.6) 208 (84.6)
Intraretinal fluid, n (%) 43 (35.0) 44 (35.8) 87 (35.4)
Hemorrhage, n (%) 67 (54.5) 61 (49.6) 128 (52.0)
Subretinal hemorrhage 56 (45.5) 50 (40.7) 106 (43.1)
Intraretinal hemorrhage 28 (22.8) 30 (24.4) 58 (23.6)
2W/4W 2-/4-week adjustment, AMD age-related macular degeneration, BCVA best-corrected visual acuity, CNV choroidal
neovascularization, CRT central retinal thickness, FA fluorescein angiography, FAS full analysis set, IVT-AFL intravitreal
aflibercept, SD standard deviation
a
Patients could be classified into more than one group
b
Unknown in the IVT-AFL-2W (n = 2) and IVT-AFL-4W (n = 1) adjustment groups and one patient had no CNV in
the IVT-AFL-4W group

12.5 (3.6) weeks in the IVT-AFL-2W and 41.5% (IVT-AFL-2W) and 46.3% (IVT-AFL-4W)
IVT-AFL-4W groups, respectively. The last of patients (Fig. 2).
injection interval before week 52 was at least The intended injection interval at the last
12 weeks in 42.3% (IVT-AFL-2W) and 49.6% visit up to week 52 was at least 12 weeks in
(IVT-AFL-4W) of patients and before week 96 56.8% (n/N = 63/111; IVT-AFL-2W) and 57.8%
was at least 12 weeks in 56.9% (IVT-AFL-2W) (n/N = 67/116; IVT-AFL-4W) of patients and was
and 60.2% (IVT-AFL-4W) of patients. The last extended to 16 weeks for 35.1% (n/N = 39/111;
injection interval before week 52 was 16 weeks IVT-AFL-2W) and 40.5% (n/N = 47/116;
in 0.0% (IVT-AFL-2W) and 40.7% (IVT-AFL-4W) IVT-AFL-4W) of patients.
of patients and before week 96 was 16 weeks in Up to week 96, the injection interval in the
IVT-AFL-2W and IVT-AFL-4W groups was
 Adv Ther

Fig. 2 The last injection interval up to a Week 52 and b Week 96 (FAS)

extended to 16 weeks for 43.1% (n = 53) and Efficacy

54.5% (n = 67) of patients, respectively, and
maintained at 16 weeks until study completion Functional Outcomes
for 41.5% (n = 51) and 42.3% (n = 52) of The mean change in BCVA (ETDRS letters) from
patients, respectively. The proportion of baseline to week 52 was 9.0 letters (95% CI
patients who stayed at an 8-week injection 6.4–11.5) in the IVT-AFL-2W group and 8.4
interval, who never had an extension, was letters (95% CI 6.0–10.8) in the IVT-AFL-4W
27.6% (n = 34) in the IVT-AFL-2W group and group (least squares [LS] mean difference of
22.0% (n = 27) in the IVT-AFL-4W group. Up to - 0.4 [95% CI - 3.8 to 3.0]). From baseline to
week 96, the proportion of patients with inter- week 96, the mean change in BCVA was 7.6
vals of 12 weeks and beyond who maintained at letters (95% CI 5.0–10.3) and 6.1 letters (95% CI
least a 12-week interval was 48.0% (n = 59) for 3.1–9.0) in the IVT-AFL-2W and IVT-AFL-4W
both treatment arms. groups, respectively (LS mean difference of
- 1.4 [95% CI - 5.3 to 2.4]) (Fig. 3).
Adv Ther

Fig. 3 Mean change in BCVA (ETDRS letter score) in visual acuity, CI confidence interval, ETDRS Early
IVT-AFL-2W and IVT-AFL-4W groups from baseline to Treatment Diabetic Retinopathy Study, FAS full analysis
week 96 (FAS). Last observation carried forward analysis. set, IVT-AFL intravitreal aflibercept, LSM least squares
2W/4W 2-/4-week adjustment, BCVA best-corrected mean

Fig. 4 Mean change in central retinal thickness (lm) from CI confidence interval, CRT central retinal thickness,
baseline to week 96 (FAS). Last observation carried IVT-AFL intravitreal aflibercept, LSM least squares mean
forward analysis. 2W/4W 2-/4-week adjustment,

The proportion of patients who lost fewer [95% CI - 9.5 to 3.2]) (Fig. S2a in the supple-
than 15 ETDRS letters was 96.7% (95% CI mentary material). The proportion of patients
93.6–99.9) in the IVT-AFL-2W group and 95.9% who gained at least 15 ETDRS letters was 32.5%
(95% CI 92.4–99.4) in the IVT-AFL-4W group at (95% CI 24.2–40.8) and 30.9% (95% CI
week 52 (MH-adjusted difference of - 0.9% 22.7–39.1) in the IVT-AFL-2W and IVT-AFL-4W
[95% CI - 5.7 to 4.0]). At week 96, 95.1% groups, respectively, at week 52 (MH-adjusted
(95% CI 91.3–98.9) and 91.9% (95% CI 87.0–96.7) difference of - 1.4% [95% CI - 12.7 to 9.8]).
of patients in the IVT-AFL-2W and IVT-AFL-4W At week 96, 28.5% (95% CI 20.5–36.4) and
groups lost fewer than 15 ETDRS letters, 31.7% (95% CI 23.5–39.9) of patients in the
respectively (MH-adjusted difference of - 3.1% IVT-AFL-2W and IVT-AFL-4W group, respectively,
 Adv Ther

Table 3 Safety overview at week 96 (SAS)

Number of patients (%) IVT-AFL-2W IVT-AFL-4W Randomization
adjustment adjustment failurea
n = 124 n = 123 n=7
Any TEAE 85 (68.5) 86 (69.9) 5 (71.4)
Mild 62 (50.0) 55 (44.7) 2 (28.6)
Moderate 15 (12.1) 22 (17.9) 1 (14.3)
Severe 8 (6.5) 9 (7.3) 2 (28.6)
Ocular TEAE (study eye)
Any ocular TEAE (study eye) C 2%b 26 (21.0) 38 (30.9) 0
Cataract 7 (5.6) 10 (8.1) 0
Conjunctival hemorrhage 4 (3.2) 8 (6.5) 0
Dry eye 3 (2.4) 6 (4.9) 0
Retinal pigment epithelium tear 3 (2.4) 0 0
Non-ocular TEAE
Any non-ocular TEAE C 3%c 65 (52.4) 69 (56.1) 5 (71.4)
Constipation 4 (3.2) 7 (5.7) 0
Large intestine polyp 0 4 (3.3) 0
Nasopharyngitis 26 (21.0) 20 (16.3) 0
Influenza 2 (1.6) 4 (3.3) 0
Contusion 1 (0.8) 4 (3.3) 0
Hypertension 1 (0.8) 4 (3.3) 1 (14.3)
Any serious TEAEs 19 (15.3) 20 (16.3) 3 (42.9)
Ocular SAE in study eye 3 (2.4) 2 (1.6) 0
Non-ocular SAE 16 (12.9) 16 (13.0) 3 (42.9)
Any TEAE leading to discontinuation of study drug 1 (0.8) 2 (1.6) 0
d
APTC arterial thromboembolic events 1 (0.8) 2 (1.6) 0
Non-fatal myocardial infarction
Acute myocardial infarction 0 1 (0.8) 0
Myocardial infarction 0 1 (0.8) 0
Non-fatal stroke 0 1 (0.8) 0
Vascular death 1 (0.8) 0 0
Any death 2 (1.6) 1 (0.8) 0
2W/4W 2-/4-week adjustment, APTC Antiplatelet Trialists’ Collaboration, IVT-AFL intravitreal aflibercept, SAE serious
adverse event, SAS safety analysis set, TEAE treatment-emergent adverse event
a
Randomization failure was due to physician decision (n = 1), logistical difficulties (n = 1), protocol violation (n = 2),
and withdrawal by patient (n = 3)
b
Ocular TEAEs C 2% in either IVT-AFL treatment arm
c
Non-ocular TEAEs C 3% in either IVT-AFL treatment arm
d
Three events were reported in two patients
Adv Ther

gained at least 15 ETDRS letters (MH-adjusted were nasopharyngitis (21.0% and 16.3%) and
difference of 3.4% [95% CI - 7.9 to 14.7]) constipation (3.2% and 5.7%). No cases of
(Fig. S2b in the supplementary material). endophthalmitis were reported.
The only serious ocular TEAE in the study
Anatomic Outcomes eye was cataract, which was reported in 2.4%
The mean change in CRT from baseline to and 1.6% of patients in the IVT-AFL-2W and
week 52 was - 134.4 lm (95% CI - 162.2 to IVT-AFL-4W groups, respectively. Three
- 106.6) in the IVT-AFL-2W group and patients (one patient in the IVT-AFL-2W group
- 126.1 lm (95% CI - 147.1 to - 105.1) in the and two patients in the IVT-AFL-4W group)
IVT-AFL-4W group (LS mean difference of discontinued IVT-AFL because of a TEAE up to
- 5.8 lm [95% CI - 24.3 to 12.7]). Mean week 96. Two deaths (intracranial hemorrhage
change in CRT from baseline to week 96 was and completed suicide) were reported in the
- 130.5 lm (95% CI - 158.7 to - 102.4) in the IVT-AFL-2W group, both of which were assessed
IVT-AFL-2W group and - 125.3 lm (95% CI as not related to IVT-AFL. One death (pneu-
- 146.6 to - 104.1) in the IVT-AFL-4W group (LS monia) was reported in the IVT-AFL-4W group
mean difference of - 9.0 [95% CI - 27.7 to 9.7]) (not related to IVT-AFL). The proportion of
(Fig. 4). APTC events was similar between the treatment
The proportion of patients without fluid on arms (0.8% and 1.6%) and consistent with pre-
OCT was 53.7% (IVT-AFL-2W) and 62.6% vious studies [12, 13].
(IVT-AFL-4W) at week 16, and 68.3% (95% CI
60.1–76.5) (IVT-AFL-2W) and 69.1% (95% CI
60.9–77.3) (IVT-AFL-4W) at week 52 (MH-
DISCUSSION
adjusted difference of - 1.0% [95% CI
The ALTAIR study showed that IVT-AFL in a
- 10.6 to 12.7]). At week 96, 67.5% (95% CI
T&E regimen with 2-week or 4-week adjust-
59.2–75.8) of patients in both treatment arms
ments improved functional and anatomic out-
had no fluid on OCT (MH-adjusted difference of
comes in treatment-naı̈ve patients with
0.4% [95% CI - 11.4 to 12.2]).
exudative AMD over 52 weeks, while reducing
the treatment burden. The mean change in
Safety BCVA from baseline to week 52 was 9.0
(IVT-AFL-2W) and 8.4 (IVT-AFL-4W) letters and
An overview of the main safety findings at mean change in CRT from baseline to week 52
week 96 is provided in Table 3. The proportion was - 134.4 lm (IVT-AFL-2W) and - 126.1 lm
of treatment-emergent AEs (TEAEs) was (IVT-AFL-4W). Functional and anatomic out-
comparable between the IVT-AFL-2W and comes were maintained to week 96; the mean
IVT-AFL-4W groups (68.5% and 69.9%, respec- change in BCVA from baseline to week 96 was
tively), and these were predominantly mild and 7.6 (IVT-AFL-2W) and 6.1 (IVT-AFL-4W) letters
moderate in severity. The incidence of ocular and mean change in CRT from baseline to
TEAEs was higher in the IVT-AFL-4W group week 96 was - 130.5 lm (IVT-AFL-2W) and
(30.9%) compared with the IVT-AFL-2W group - 125.3 lm (IVT-AFL-4W). The mean number
(21.0%) (Table 3). However, the incidence of of IVT-AFL injections was the same in the two
non-ocular TEAEs was similar between the groups (10.4). More than half of patients (57%
IVT-AFL-2W and IVT-AFL-4W groups (52.4% in the 2-week and 58% in the 4-week group) had
and 56.1%, respectively). The most common an intended injection interval of at least
ocular TEAEs in the IVT-AFL-2W and 12 weeks up to week 52 and the majority of
IVT-AFL-4W groups, respectively, were cataract patients (57% [IVT-AFL-2W] and 60%
(5.6% and 8.1%), conjunctival hemorrhage [IVT-AFL-4W]) had a last injection interval of at
(3.2% and 6.5%), dry eye (2.4% and 4.9%), and least 12 weeks at week 96. More than 40% of
retinal pigment epithelium tear (2.4% and 0%), patients (42% [IVT-AFL-2W] and 46%
whereas the most common non-ocular TEAEs [IVT-AFL-4W] had a last injection interval of
 Adv Ther

16 weeks up to week 96. There were no new effective in the first year of treatment and con-
safety events with the IVT-AFL T&E regimens tinuously efficacious in the second year using
compared with previous studies [12, 13]. either a 2- or 4-week adjustment based on
Although good outcomes have been defined criteria for interval extension, mainte-
achieved in clinical trials, numerous studies nance, or shortening. In other studies, injection
have shown the challenges of bringing this intervals were adjusted by 2-week increments
efficacy into the real world [14–17]. In clinical with a set minimum injection interval of
practice, fixed dosing is associated with burdens 4 weeks and a maximum of 12 weeks [21–23]. In
for both the patient and the clinic; therefore, in ALTAIR, 57–60% of patients achieved a last
real-world practice, regimens such as T&E injection interval of at least 12 weeks up to
(proactive) and PRN (reactive) are often adopted week 96, compared with 17–37% of patients in
to reduce treatment burden while maintaining other prospective studies of ranibizumab and
functional outcomes [7]. Utilization of proac- IVT-AFL T&E regimens [21, 24–26]. Up to
tive IVT-AFL T&E regimens, further substanti- week 96, 41–46% of patients in ALTAIR reached
ated by the results of ALTAIR, allows for a the maximum last injection interval of 16 weeks
pragmatic approach to treatment of exudative either with 2- or 4-week adjustments. Moreover,
AMD and offers benefits to both physicians and the results of the ALTAIR study suggest that
patients [7]. IVT-AFL T&E regimens with a minimum injec-
The duration of VEGF-A suppression differs tion interval of 8 weeks may provide patients
between patients [18, 19]; therefore, by titrating with good functional outcomes. The good
the injection interval on the basis of the indi- functional outcomes observed in the VIEW
vidual patient’s functional and anatomic out- studies, in which IVT-AFL was given every
comes, and adjusting the treatment if necessary, 8 weeks, further supports this [12, 13]. The
physicians can achieve optimal functional out- functional and anatomic outcomes in ALTAIR,
comes for each patient while reducing the fre- using a T&E regimen, are comparable with
quency of clinic visits. With proactive, those observed for the IVT-AFL every 4 or
individualized T&E dosing regimens, the need 8 weeks (q4 and q8; fixed dosing) arm in the
for interim monitoring is minimized. Reducing VIEW study [12, 13]. The mean change in BCVA
the number of appointments per patient and from baseline to week 52 was 8.7, 9.3, and 8.4
minimizing the need for monitoring visits letters and from baseline to week 96 was 6.9,
could help to ease clinic flow and patient bur- 7.6, and 7.6 letters in ALTAIR and VIEW q4 and
den while maintaining functional outcomes. q8 groups, respectively. The mean change in
Furthermore, planning the next injection helps CRT from baseline to week 52 was - 130 lm,
to minimize the possibility of treatment delays - 138 lm, and - 139 lm and from baseline to
and facilitates clinic management [7]. The week 96 was - 128 lm, - 128 lm, and
molecular attributes of aflibercept allow for - 133 lm in ALTAIR and VIEW q4 and q8
extended injection intervals. Previous studies of groups, respectively. In ALTAIR, patients
IVT-AFL in patients with exudative AMD have received fewer injections than in VIEW (10.4,
reported a median aqueous half-life of at least 16.0 [q4], and 11.2 [q8], respectively). Extend-
9 days [20] and a range of intraocular VEGF ing the interval by 4 weeks, to beyond 12 weeks
suppression times of up to 16 weeks [18, 19]. and up to a 16-week interval, offers potential
The results from ALTAIR indicate that, with advantages for both patients (reduced treatment
IVT-AFL T&E, the injection interval can be burden) and healthcare providers (scheduling
extended to 12 weeks and beyond in approxi- visits) compared with the alternative 2-week
mately 57–60% of patients. increments [7, 11].
ALTAIR explored interval adjustments and While there was no reading center involved
outcomes using an IVT-AFL T&E regimen in in the ALTAIR study, the investigators con-
patients with exudative AMD. Our study ducted ophthalmic examinations at the
demonstrated that an IVT-AFL T&E regimen, screening visit and at every treatment and
following initial monthly dosing, can be evaluation, and fundus photography, FA, and
Adv Ther

indocyanine green angiography were con- conjunction with the ALTAIR steering com-
ducted at the screening visit and at weeks 52 mittee, Bayer participated in the design of the
and 96. Patients in the 4-week group who had study; analysis and interpretation of the data;
their injection interval shortened by 4 weeks preparation, review, and approval of the
and were then extended by 2-week increments manuscript; and decision to submit the manu-
remained in the 4-week group for analysis. All script for publication. Additionally, Bayer was
statistical analyses were exploratory, and no responsible for the conduct of the study and
confirmatory statistical analysis was intended. oversight of the collection and management of
Although the objective of this study was to data.
describe the outcomes achieved with both
treatment regimens, the sample size allowed for Medical Writing and Editorial Sup-
descriptive statistical comparison. There was a port. Medical writing and editorial support for
greater proportion of male patients than female the preparation of this manuscript (under the
patients in the ALTAIR study, which is consis- guidance of the authors) was provided by Mia
tent with the Japanese population of the Cahill (ApotheCom, UK) and was funded by
VIEW 2 study [27]. Further analyses will evalu- Bayer Consumer Care AG, Switzerland. Mia
ate the efficacy of the two different T&E regi- Cahill has no conflicts of interest to declare. The
mens in subgroups of interest, such as those authors thank Daniel Janer, MD who provided
with PCV. input and expert medical guidance.

Authorship. All named authors meet the

International Committee of Medical Journal
CONCLUSION Editors (ICMJE) criteria for authorship for this
article, take responsibility for the integrity of
IVT-AFL administered to treatment-naı̈ve
the work as a whole, and have given their
patients with exudative AMD using two differ-
approval for this version to be published.
ent T&E regimens, with a minimum injection
interval of 8 weeks and a maximum interval of
Disclosures. Masahito Ohji has received
16 weeks, improved and maintained functional
grants, consultancy fees, honorarium, travel
and anatomic outcomes over 96 weeks, while
support and speaker fees from AbbVie Japan,
minimizing the treatment burden on patients.
Inc., Alcon Pharma K.K. (Japan), Allergan, B.L.J
To provide further practical and personalized
Ltd., Bayer Yakuhin, Ltd. (Japan), Chugai,
IVT-AFL T&E treatment regimens for patients
HOYA, Kowa, Novartis Pharma K.K., Otsuka
with exudative AMD, future investigations should
Pharmaceuticals, Pfizer Pharmaceuticals K.K,
address predictive factors (such as subtypes of
Santen Pharmaceuticals Co., Ltd., Senju Phar-
AMD lesions and fluid status) of functional/ana-
maceutical Co., Ltd., and Topcon.
tomic outcomes and injection intervals.
Kanji Takahashi has received consultancy fees,
honoraria, travel support and speaker fees from
Alcon Pharma K.K. (Japan), Bayer Yakuhin, Ltd.
ACKNOWLEDGEMENTS (Japan), B.L.J Ltd., Carl Zeiss Co., Ltd., Kowa,
Kyowa Kirin Co., Ltd., Novartis Pharma K.K.,
The authors thank all the patients and investi- Otsuka Pharmaceuticals, Pfizer Pharmaceuticals
gators who contributed to this study. K.K., Santen Pharmaceuticals Co., Ltd., and
Senju Pharmaceutical Co., Ltd.
Annabelle A. Okada has received personal fees
Funding. Funding for the study, medical
from AbbVie Japan, Inc., Astellas Japan, Bayer
writing and editorial assistance for this manu-
Healthcare AG, Daiichi-Sankyo, and Senju Phar-
script, and funding for the Rapid Service Fee and
maceutical Co., Ltd.; and grants and personal
Open Access Fee was provided by Bayer. In
fees from Alcon Pharma K.K. (Japan), Bayer
 Adv Ther

Yakuhin Ltd. (Japan), Mitsubishi Tanabe Phar- REFERENCES

ma Corporation, and Santen Pharmaceuticals
Co., Ltd. (Japan). 1. World Health Organization. Prevention of blind-
Masato Kobayashi, Yoshimi Matsuda and Ya- ness and visual impairment. WHO. 2019. http://
suhiro Terano are employees of Bayer Yakuhin www.who.int/blindness/causes/priority/en/. Acces-
Ltd. (Japan). sed Dec 2019.

2. Yamada M, Hiratsuka Y, Roberts CB, et al. Preva-

Compliance with Ethics Guidelines. The lence of visual impairment in the adult Japanese
study was conducted at 41 centers across Japan population by cause and severity and future pro-
between December 2014 and November 2017, jections. Ophthalmic Epidemiol. 2010;17:50–7.
in accordance with the Declaration of Helsinki
3. Ferris FL 3rd, Fine SL, Hyman L. Age-related macu-
and the International Conference on Harmo- lar degeneration and blindness due to neovascular
nization guidelines E6: Good Clinical Practice. maculopathy. Arch Ophthalmol. 1984;102:1640–2.
The protocol and any amendments were
approved by the independent ethics committee 4. Ambati J, Fowler BJ. Mechanisms of age-related
macular degeneration. Neuron. 2012;75:26–39.
or institutional review board (IRB) at each study
site (see supplementary material). There was no 5. Aiello LP, Avery RL, Arrigg PG, et al. Vascular
central IRB involved in the study and the pro- endothelial growth factor in ocular fluid of patients
tocol was reviewed and approved by the IRB at with diabetic retinopathy and other retinal disor-
ders. N Engl J Med. 1994;331:1480–7.
each participating center. All patients provided
written informed consent. A list of investigators 6. Freund KB, Korobelnik JF, Devenyi R, et al. Treat-
who participated in the study is provided in the and-extend regimens with anti-VEGF agents in
supplementary material. retinal diseases: a literature review and consensus
recommendations. Retina. 2015;35:1489–506.

Data Availability. The datasets generated 7. Lanzetta P, Loewenstein A, Vision Academy Steer-
during and/or analyzed during the current ing Committee. Fundamental principles of an anti-
study are available from the corresponding VEGF treatment regimen: optimal application of
intravitreal anti-vascular endothelial growth factor
author on reasonable request. therapy of macular diseases. Graefes Arch Clin Exp
Ophthalmol. 2017;255:1259–73.
Open Access. This article is licensed under a
Creative Commons Attribution-NonCommer- 8. Maguire MG, Martin DF, Ying GS, et al. Five-year
cial 4.0 International License, which permits outcomes with anti-vascular endothelial growth
factor treatment of neovascular age-related macular
any non-commercial use, sharing, adaptation, degeneration: the comparison of age-related mac-
distribution and reproduction in any medium ular degeneration treatments trials. Ophthalmol-
or format, as long as you give appropriate credit ogy. 2016;123:1751–61.
to the original author(s) and the source, provide
9. Patel PJ, Devonport H, Sivaprasad S, et al. Afliber-
a link to the Creative Commons licence, and cept treatment for neovascular AMD beyond the
indicate if changes were made. The images or first year: consensus recommendations by a UK
other third-party material in this article are expert roundtable panel, 2017 update. Clin Oph-
included in the article’s Creative Commons thalmol. 2017;11:1957–66.
licence, unless indicated otherwise in a credit 10. Rezaei KA. Global trends in retina. Chicago: Amer-
line to the material. If material is not included ican Society of Retina Specialists; 2016.
in the article’s Creative Commons licence and
your intended use is not permitted by statutory 11. Koh A, Lanzetta P, Lee WK, et al. Recommended
guidelines for use of intravitreal aflibercept with a
regulation or exceeds the permitted use, you treat-and-extend regimen for the management of
will need to obtain permission directly from the neovascular age-related macular degeneration in
copyright holder. To view a copy of this licence, the Asia–Pacific region: report from a consensus
visit http://creativecommons.org/licenses/by- panel. Asia–Pacific J Ophthalmol. 2017;6:296–302.
nc/4.0/. 12. Heier JS, Brown DM, Chong V, et al. Intravitreal
aflibercept (VEGF trap-eye) in wet age-related
Adv Ther

macular degeneration. Ophthalmology. 2012;119: macular degeneration. Invest Ophthalmol Vis Sci.
2537–48. 2017;58:406.

13. Schmidt-Erfurth U, Kaiser PK, Korobelnik JF, et al. 21. Berg K, Hadzalic E, Gjertsen I, et al. Ranibizumab or
Intravitreal aflibercept injection for neovascular bevacizumab for neovascular age-related macular
age-related macular degeneration: ninety-six-week degeneration according to the Lucentis compared
results of the VIEW studies. Ophthalmology. to Avastin study treat-and-extend protocol: two-
2014;121:193–201. year results. Ophthalmology. 2016;123:51–9.

14. Cohen SY, Mimoun G, Oubraham H, et al. Changes 22. Silva R, Berta A, Larsen M, et al. Treat-and-extend
in visual acuity in patients with wet age-related versus monthly regimen in neovascular age-related
macular degeneration treated with intravitreal macular degeneration: results with ranibizumab
ranibizumab in daily clinical practice: the LUMIERE from the TREND study. Ophthalmology. 2018;125:
study. Retina. 2013;33:474–81. 57–65.

15. Holz FG, Tadayoni R, Beatty S, et al. Multi-country 23. Yamamoto A, Okada AA, Nakayama M, Yoshida Y,
real-life experience of anti-vascular endothelial Kobayashi H. One-year outcomes of a treat-and-
growth factor therapy for wet age-related macular extend regimen of aflibercept for exudative age-re-
degeneration. Br J Ophthalmol. 2015;99:220–6. lated macular degeneration. Ophthalmologica.
2017;237:139–44.
16. Writing Committee for the UK Age-Related Macular
Degeneration EMR Users Group. The neovascular 24. Gillies MC, Hunyor AP, Arnold JJ, et al. Effect of
age-related macular degeneration database: multi- ranibizumab and aflibercept on best-corrected
center study of 92,976 ranibizumab injections: visual acuity in treat-and-extend for neovascular
report 1: visual acuity. Ophthalmology. 2014;121: age-related macular degeneration: a randomized
1092–101. clinical trial. JAMA Ophthalmol. 2019;137:372–9.

17. Zhu M, Chew JK, Broadhead GK, et al. Intravitreal 25. Guymer RH, Markey CM, McAllister IL, Gillies MC,
ranibizumab for neovascular age-related macular Hunyor AP, Arnold JJ. Tolerating subretinal fluid in
degeneration in clinical practice: five-year treat- neovascular age-related macular degeneration trea-
ment outcomes. Graefes Arch Clin Exp Ophthal- ted with ranibizumab using a treat-and-extend
mol. 2015;253:1217–25. regimen: FLUID study 24-month results. Ophthal-
mology. 2019;126:723–34.
18. Fauser S, Muether PS. Clinical correlation to differ-
ences in ranibizumab and aflibercept vascular 26. Wykoff CC, Ou WC, Brown DM, et al. Randomized
endothelial growth factor suppression times. Br J trial of treat-and-extend versus monthly dosing for
Ophthalmol. 2016;100:1494–8. neovascular age-related macular degeneration:
2-year results of the TREX-AMD study. Ophthalmol
19. Fauser S, Schwabecker V, Muether PS. Suppression Retina. 2017;1:314–21.
of intraocular vascular endothelial growth factor
during aflibercept treatment of age-related macular 27. Ogura Y, Terasaki H, Gomi F, et al. Efficacy and
degeneration. Am J Ophthalmol. 2014;158:532–6. safety of intravitreal aflibercept injection in wet
age-related macular degeneration: outcomes in the
20. Do DV, Nguyen QD. Pharmacokinetics of free Japanese subgroup of the VIEW 2 study. Br J Oph-
aflibercept in patients with neovascular age related thalmol. 2015;99:92–7.