Download Predictability of Recalcitrance in Neovascular Age

ORIGINAL CLINICAL STUDY
Predictability of Recalcitrance in Neovascular Age-Related
Macular Degeneration With Indocyanine
Green Angiography
Ryan B. Rush, MD*Þþ and Sloan W. Rush, MDÞþ
Purpose: This study aimed to evaluate the utility of indocyanine green
(ICG) angiography in predicting recalcitrance in neovascular age-related
macular degeneration (nAMD).
Design: A retrospective case series.
Methods: The charts of treatment-naive subjects with nAMD undergoing antiYvascular endothelial growth factor (anti-VEGF) therapy
during a 6-month period were retrospectively reviewed. The study group
consisted of subjects with persistent retinal edema on optical coherence
tomography (OCT) despite 6 consecutive monthly anti-VEGF injections. The control group was age-matched to the study group and
consisted of subjects who demonstrated complete resolution of retinal
edema on OCT after 3 or fewer monthly anti-VEGF injections.
Results: There were 42 study cases and 42 controls included in the
analysis. The baseline visual acuity, central macular thickness on OCT,
and choroidal neovascularization (CNV) surface area on ICG angiography were statistically similar between the study and control groups.
The CNV surface area on ICG angiography 2 months after starting
consecutive monthly anti-VEGF injections increased from a baseline
of 1.78 T 0.86 to 2.66 T 0.92 mm2 in the study group (P = 0.008) and
decreased from a baseline of 1.94 T 0.97 to 1.12 T 0.05 mm2 in the
control group (P = 0.04); this change in CNV size on ICG angiography
from baseline to 2-month follow-up was statistically significant between
the study and control groups (P G 0.0001).
Conclusions: Change in CNV surface area on ICG angiography can
predict which subjects with nAMD are likely to have persistent retinal
edema on OCT after 6 or more consecutive monthly anti-VEGF injections.
Key Words: indocyanine green angiography, macular degeneration,
recalcitrance
(Asia Pac J Ophthalmol 2015;4: 187Y190)
P
ersistent retinal edema on optical coherence tomography
(OCT) was found in more than 50% of patients with
neovascular age-related macular degeneration (nAMD) undergoing ranibizumab (Lucentis; Genetech, Inc, South San Francisco,
Calif ) and bevacizumab (Avastin; Genetech, Inc) therapy after
1 and 2 years in the Comparison of Age-related Macular Degeneration Treatments Trials.1,2 Patients with persistent retinal edema
despite consecutive monthly antiYvascular endothelial growth
factor (anti-VEGF) injections may be considered to have treatmentresistant choroidal neovascularization (CNV), and several strategies
have evolved to deal with the following clinical situations: injections more frequently than every 4 weeks,3 higher anti-VEGF
From the *Southwest Retina Specialists; †Panhandle Eye Group; and
‡Department of Surgery, Texas Tech University Health Science Center,
Amarillo, TX.
Received for publication November 2, 2014; accepted January 16, 2015.
The authors have no funding or conflicts of interest to declare.
Reprints: Ryan B. Rush, MD, Southwest Retina Specialists, 7411 Wallace
Blvd., Amarillo, TX, 79106. E-mail: [email protected].
Copyright * 2015 by Asia Pacific Academy of Ophthalmology
ISSN: 2162-0989
DOI: 10.1097/APO.0000000000000111
Asia-Pacific Journal of Ophthalmology
&
dosages,4 anti-VEGF combined with photodynamic therapy,5 and
changing anti-VEGF drugs. Several retrospective studies6Y10 and 1
prospective study11 have reported that switching from ranibizumab
or bevacizumab to aflibercept (VEGF-Trap Eye/Eylea; Regeneron,
Tarrytown, NY) can reduce or eliminate persistent retinal edema on
OCT in patients with recalcitrant nAMD.
Recently, Rush et al12 described the clinical utility of
indocyanine green (ICG) angiography in subjects with nAMD
undergoing bevacizumab therapy. However, subjects switching
anti-VEGF drugs or receiving combination therapy during the
12-month study interval were purposefully excluded from that
study, which resulted in the removal of the most treatmentresistant subjects from the analysis. Therefore, it remains unknown how CNV size on ICG angiography changes in those
subjects with nAMD recalcitrant to anti-VEGF therapy. The
purpose of this study was to determine if change in CNV size
on ICG angiography early in the course of treatment can predict
the persistence of retinal edema on OCT after 6 consecutive
monthly anti-VEGF injections in subjects with nAMD.
MATERIALS AND METHODS
The Southwest Retina Specialists Institutional Review
Board (IORG0007600/IRB00009122) approved this retrospective medical record review of patients with nAMD treated from
March 2011 to March 2014 at a single private-practice institution. All research constituents observed the tenets of the Declaration of Helsinki and were directed in agreement with human
research regulations and standards.
The inclusion criteria were as follows: (1) 55 years or
older; (2) a diagnosis of treatment-naive subfoveal nAMD
confirmed by clinical examination, OCT, fluorescein angiography (FA), and ICG angiography; (3) baseline Snellen bestcorrected visual acuity (BCVA) between 20/25 and 20/200;
(4) a 6-month study period fully documented with BCVA, examination details, OCT, FA, and ICG angiography; (5) ICG
angiography that must have been performed at baseline, 2month follow-up, and 6-month follow-up; and (6) a pro re
nata regimen adhered to as described later.
The exclusion criteria were as follows: (1) macular photocoagulation, photodynamic therapy, or an intravitreal injection other than anti-VEGF therapy was given to the study eye
during the study period; (2) an ocular media opacity was affecting BCVA or obstructing imaging acquisition; (3) intraocular
surgery was performed during the study period or within 3 months
before subject enrollment; (4) retinal diseases other than nAMD
(epiretinal membrane, polypoidal choroidal vasculopathy, diabetic retinopathy, etc), uncontrolled glaucoma, or active uveitis
were present at enrollment or developed during the study period;
(5) a previous vitrectomy had been performed at any time to the
study eye; and (6) the CNV margins were unable to be measured
by ICG angiography (absence of a distinct CNV complex and/or
‘‘hot’’ spots without distinguishable margins). Polypoidal choroidal vasculopathy was defined as the presence of 1 or multiple
Volume 4, Number 4, July/August 2015
www.apjo.org
Copyright © 2015 Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
187
Rush and Rush
Asia-Pacific Journal of Ophthalmology
focal areas of hyperfluorescence from inner choroidal vessels
within the first 6 minutes after injection of ICG with or without an
associated branching vascular network and without an associated
classic or occult CNV complex.13
Baseline examinations included BCVA, slit-lamp assessment of the anterior segments and posterior segments, OCT, FA,
as well as ICG angiography. All subjects received a loading
dose of 3 consecutive monthly anti-VEGF injections, followed
by consecutive monthly injections until the macula was dry
(without intraretinal or subretinal fluid) on OCT. Once this
measure was achieved, the injection was deferred and the patient returned for follow-up 4 weeks later. If retinal edema on
OCT recurred or macular hemorrhage developed after injection
deferral, anti-VEGF treatment was resumed. The subjects failing to achieve a dry macula on OCT after 6 consecutive monthly
anti-VEGF injections were put into the study group. The subjects achieving a dry macula on OCT after the 3 consecutive
monthly anti-VEGF loading doses were put into the control
group. The subjects in the study group were permitted to change
anti-VEGF medications so long as they continued receiving
consecutive monthly injections without ever attaining a dry
macula on OCT.
The OCT, FA, and ICG angiography procedures were
performed using the Heidelberg Spectralis system (Heidelberg,
Germany). The detection of retinal edema and the central
macular thickness were determined by OCT. The surface
areaYmeasuring software of the Heidelberg Spectralis was used
to determine CNV size with ICG angiography, and this technique in the setting of nAMD has been described in a previous
study.12 Briefly, early-frame to midframe ICG angiography
images were viewed, and a test image was chosen for CNV size
analysis. The CNV margins were manually encircled using the
‘‘Inlay’’ function, and the surface area was then computed. Sideby-side OCT and FA images of the ICG angiography test image
were permitted to establish the precise location of the CNV. All
the CNV measurements on ICG angiography were independently evaluated by 2 masked observers (R.B.R. and S.W.R.). If
CNV size was not within 10% between the 2 masked observers,
the subject was excluded from the analysis.
Outcome Measures and Statistical Analysis
The primary outcome of the study was a change in CNV
surface area on ICG angiography from baseline to 2-month
follow-up in the study and control groups. Statistical calculations were performed using the JMP software package from the
SAS Institute. Snellen BCVA was converted into a logarithm of
the minimal angle of resolution (logMAR) for the statistical
analysis. Numerical means were compared using 1-way analysis
of the variance whereas nominal variables were compared using
&
Volume 4, Number 4, July/August 2015
W2 and likelihood ratios where appropriate. A probability of less
than 0.05 was considered statistically significant.
RESULTS
There were 47 study cases that met the criteria for inclusion. Of these 47 cases, 5 were disqualified from analysis
because of a discordance of more than 10% for the CNV size between the masked observers, thereby allowing for an interobserver
agreement rate of 89.3% (42/47) for the study group. There were
42 control cases age-matched (T3 years) to the 42 study subjects.
The baseline characteristics of the study group were as follows: the
mean T SD age was 79.1 T 8.6 with 52% male (22/42) and 48%
female (20/42). The lens status was pseudophakic for 67% (28/42)
and phakic for 33% (14/42). The study group’s baseline BCVA was
0.55 T 0.24 logMAR (Snellen visual acuity, 20/71), central macular
thickness on OCT was 431.0 T 88.6 Km, and CNV surface area on
ICG angiography was 1.78 T 0.86 mm2. There were 30.9% (13/42)
of study subjects with classic/predominantly classic, 28.5% (12/42)
with minimally classic, and 40.4% (17/42) with occult CNV
on baseline FA. The baseline characteristics of the control group
were as follows: the mean T SD age was 78.4 T 8.1 with 57% male
(24/42) and 43% female (18/42). The lens status was pseudophakic
for 62% (26/42) and phakic for 38% (16/42). The control group’s
baseline BCVA was 0.50 T 0.22 logMAR (Snellen visual acuity,
20/63), central macular thickness on OCT was 403.4 T 98.2 Km,
and CNV surface area on ICG angiography was 1.94 T 0.97 mm2.
There were 28.5% (12/42) of control subjects with classic/
predominantly classic, 23.8% (10/42) with minimally classic, and
47.6% (20/42) with occult CNV on baseline FA. There were no
significant differences in any of the baseline characteristics between the groups.
The study group was treated with the following anti-VEGF
medications during the study period: 22 with bevacizumab
alone, 6 with aflibercept alone, 3 with ranibizumab alone, and
11 with more than 1 anti-VEGF agent. The control group was
treated with the following anti-VEGF medications during the
study period: 27 with bevacizumab alone, 9 with aflibercept
alone, and 6 with ranibizumab alone. The mean T SD number of
anti-VEGF injections delivered during the 6-month study period was 4.1 T 0.8 in the control group.
The CNV surface area on ICG angiography at 2-month
follow-up was 2.66 T 0.92 mm2 in the study group, which
was significantly enlarged from the baseline value (P = 0.008).
The CNV surface area on ICG angiography at 2-month followup was 1.12 T 1.05 mm2 in the control group, which was
significantly reduced from the baseline value (P = 0.04). The
change in CNV size on ICG angiography from baseline to 2month follow-up was significantly different between the study
and control groups (P G 0.0001). There were 76.2% (32/42) of
FIGURE 1. The treatment-naive baseline FA (A), ICG angiography (B), and OCT (C) images of a 79-year-old study patient with
baseline Snellen visual acuity of 20/60. The CNV is encircled in white with the corresponding surface area calculation on the ICG
angiography image.
188
www.apjo.org
* 2015 Asia Pacific Academy of Ophthalmology
Copyright © 2015 Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
Asia-Pacific Journal of Ophthalmology
&
Volume 4, Number 4, July/August 2015
Recalcitrant Macular Degeneration
FIGURE 2. The FA (A), ICG angiography (B), and OCT (C) images of the same patient presented in Figure 1 at the 2-month follow-up
evaluation after receiving 2 bevacizumab injections. Notice that the CNV size on ICG angiography has enlarged by more than 33%
from baseline. The Snellen visual acuity remains 20/60.
subjects with enlarged CNV size and 64.2% (27/42) of subjects
with more than 33% enlargement in CNV size on ICG
angiography in the study group at 2-month follow-up; in
comparison, there were 19.0% (8/42) of subjects with enlarged
CNV size, and 7.1% (3/42) of subjects with more than 33%
enlargement in CNV size on ICG angiography in the control
group at 2-month follow-up. These differences between the
groups were significant (P = 0.022 and P = 0.005, respectively).
The study group’s BCVA was 0.48 T 0.19 logMAR
(Snellen visual acuity, 20/60), central macular thickness on
OCT was 363.6 T 75.6 Km, and CNV surface area on ICG angiography was 2.82 T 1.16 mm2 at the end of the 6-month study
period. The final BCVA was not significantly different from
the baseline value (P = 0.39), but the final central macular
thickness on OCT was significantly reduced from the baseline
value (P = 0.035), and the final CNV size on ICG angiography
was significantly larger than the baseline value (P = 0.002)
in the study group. The control group’s BCVA was 0.30 T 0.16
logMAR (Snellen visual acuity, 20/39), central macular thickness on OCT was 278.5 T 68.9 Km, and CNV surface area
on ICG angiography was 0.88 T 0.84 mm 2 at the end of the
6-month study period. The control group’s final BCVA, central
macular thickness on OCT, and CNV size on ICG angiography
were significantly improved from baseline (P G 0.001 for all). See
Figures 1 to 3 for a case example from the study group.
DISCUSSION
The structure of CNV on ICG angiography may vary substantially according to the stage of development or regression.14,15 The Heidelberg Spectralis confocal scanning laser
ophthalmoscope technology provides high-quality images that
allow clinicians to detect and precisely measure CNV surface area
on ICG angiography regardless of the CNV stage; similar to a
previous report12 using this method of CNV size determination,
masked observers in this study were also able to agree on more
than 80% of test images within 10% of the lesion size. This high
interobserver concordance rate supports the notion that the CNV
measurement technique used by this study provides a less subjective and more reproducible method compared with CNV classification systems of the past that graded hyperfluorescence and dye
leakage with digital videoangiography.16Y23 This is the first study to
evaluate the utility of ICG angiography as a clinical tool to predict
the likelihood of future recalcitrance in subjects with nAMD early
in the course of treatment.
The most significant finding of this study is the enlargement in CNV size on ICG angiography observed from baseline
to 2 months in the study group compared with the reduction in
CNV size on ICG angiography from baseline to 2 months in the
control group. A previous study12 demonstrated an increase of
33% or more in CNV size on ICG angiography from baseline
to 2-month follow-up in just 5.7% of their overall cohort
of subjects with nAMD; the overall study population in that
study had a decrease in CNV size on ICG angiography from 1.9
T 2.5 mm2 at baseline to 1.66 T 2.11 mm2 at 2-month follow-up.
Similar to this previous study, the control group of our current
study had an increase of 33% or more in CNV size on ICG
angiography from baseline to 2-month follow-up in a very
small number of cases (7.1%). However, most subjects (64.2%)
in the study group had an increase of 33% or more in CNV
size on ICG angiography from baseline to 2-month follow-up,
and this was statistically significant compared with the control group. Therefore, the typical subject with nonrecalcitrant
nAMD undergoing anti-VEGF therapy has a decrease in CNV
size from baseline to 2-month follow-up and rarely has an increase of 33% or more in CNV size on ICG angiography from
baseline to 2-month follow-up. The clinical application of this
finding is as follows: subjects demonstrating an enlargement
in CNV size on ICG angiography from baseline to 2-month
FIGURE 3. The FA (A), ICG angiography (B), and OCT (C) images of the same patient presented in Figures 1 and 2 after receiving 6
consecutive monthly bevacizumab injections. Notice that the CNV size on ICG angiography has further increased and subretinal fluid on
OCT persists. The Snellen visual acuity has decreased to 20/70. The pronounced enlargement in CNV surface area on ICG angiography
from baseline to 2-month follow-up predicted this subject’s recalcitrance to bevacizumab therapy.
* 2015 Asia Pacific Academy of Ophthalmology
www.apjo.org
Copyright © 2015 Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
189
Asia-Pacific Journal of Ophthalmology
Rush and Rush
follow-up, especially when an increase of 33% or more occurs, are statistically likely to be recalcitrant to anti-VEGF
therapy with persistent retinal edema on OCT despite 6 or
more consecutive monthly injections. Having now identified
these treatment-resistant subjects at a very early stage in the
course of treatment, the clinician may then consider alternative
treatment options such as combination therapy, a switch in antiVEGF medication, higher-dosage ranibizumab, or anti-VEGF
therapy more frequently than once per month at a stage sooner than would otherwise have been considered to improve longterm outcomes for this challenging subgroup.
Weaknesses of this study include the retrospective collection
of data, the use of logMAR visual acuity, the small number of study
subjects, limited follow-up, and the observer-dependent determination of CNV size on ICG angiography. The information in our
study subjects’ medical records was not sufficient to match the
2 groups for ‘‘maturity’’ of CNV by symptom duration before
treatment initiation. It should be cautioned that the results of this
study may not apply to patients with CNV presenting as ‘‘hot spots’’
without distinguishable margins on ICG angiography; the subjects
with hot spots in this study often had polypoidal choroidal
vasculopathy and were excluded from analysis per the exclusion
criteria. In summary, our study demonstrated that a change in CNV
surface area on ICG angiography from baseline to 2-month followup can predict which subjects with nAMD are likely to have persistent retinal edema on OCT at 6-month follow-up despite consecutive monthly anti-VEGF injections during that period. The
clinician should be alerted to the likelihood of recalcitrance at an
early stage in the treatment course of nAMD to consider alternative
treatment options sooner when CNV size enlarges from baseline to
2-month follow-up, particularly if 33% or more. Future research is
needed to further establish the predictive value and utility of ICG
angiography in subjects with nAMD undergoing anti-VEGF therapy before the authors can recommend widespread routine implementation in clinical practice.
REFERENCES
1. CATT Research Group. Ranibizumab and bevacizumab for neovascular
age-related macular degeneration. N Engl J Med. 2011;364:1897Y1908.
2. Comparison of Age-related Macular Degeneration Treatments Trials
(CATT) Research Group, Martin DF, Maguire MG, et al. Ranibizumab
and bevacizumab for treatment of neovascular age-related macular
degeneration: two-year results. Ophthalmology. 2012;119:1388Y1398.
3. Stewart MW, Rosenfeld PJ, Penha FM, et al. Pharmacokinetic rationale
for dosing every 2 weeks versus 4 weeks with intravitreal ranibizumab,
bevacizumab, and aflibercept (vascular endothelial growth factor
Trap-eye). Retina. 2012;32:434Y457.
4. Brown DM, Chen E, Mariani A, et al. Super-dose anti-VEGF (SAVE)
trial: 2.0 mg intravitreal ranibizumab for recalcitrant neovascular
macular degeneration-primary end point. Ophthalmology.
2013;120:349Y354.
5. Tozer K, Roller AB, Chong LP, et al. Combination therapy for
neovascular age-related macular degeneration refractory to
antiYvascular endothelial growth factor agents. Ophthalmology.
2013;120:2029Y2034.
&
Volume 4, Number 4, July/August 2015
8. Yonekawa Y, Andreoli C, Miller JB, et al. Conversion to aflibercept for
chronic refractory or recurrent neovascular age-related macular
degeneration. Am J Ophthalmol. 2013;156:29Y35.
9. Ho VY, Yeh S, Olsen TW, et al. Short-term outcomes of aflibercept for
neovascular age-related macular degeneration in eyes previously
treated with other vascular endothelial growth factor inhibitors.
Am J Ophthalmol. 2013;156:23Y28.
10. Kumar N, Marsiglia M, Mrejen S, et al. Visual and anatomical outcomes
of intravitreal aflibercept in eyes with persistent subfoveal fluid
despite previous treatments with ranibizumab in patients with
neovascular age-related macular degeneration. Retina.
2013;33:1605Y1612.
11. Chang AA, Li H, Broadhead GK, et al. Intravitreal aflibercept for
treatment-resistant neovascular age-related macular degeneration.
Ophthalmology. 2014;121:188Y192.
12. Rush RB, Rush SW, Aragon AV 2nd, et al. Evaluation of choroidal
neovascularization with indocyanine green angiography in neovascular
age-related macular degeneration subjects undergoing intravitreal
bevacizumab therapy. Am J Ophthalmol. 2014;158:337Y344.
13. Koh AC, Chen LJ, Chen SJ, et al. Polypoidal choroidal vasculopathy:
evidence-based guidelines for clinical diagnosis and treatment. Retina.
2013;33:686Y716.
14. Wolf S, Knabben H, Krombach G, et al. Indocyanine-green angiography
in patients with occult choroidal neovascularization. Ger J Ophthalmol.
1996;5:251Y256.
15. Guyer DR, Yannuzzi LA, Slakter JS, et al. Classification of choroidal
neovascularization by digital indocyanine green videoangiography.
Ophthalmology. 1996;103:2054Y2060.
16. Semoun O, Guigui B, Tick S, et al. Infrared features of classic choroidal
neovascularisation in exudative age-related macular degeneration.
Br J Ophthalmol. 2009;93:182Y185.
17. Trabucchi G, Brancato R, De Molfetta V, et al. Indocyanine green and
fluorescein angiography of surgically excised macular choroidal
neovascularizations: correlations with histopathologic and
ultrastructural findings. Graefes Arch Clin Exp Ophthalmol.
1996;234:294Y299.
18. Nakajima M, Shimada H, Sato M, et al. Comparison between
indocyanine green angiography and histopathological observations of
choroidal neovascular membrane in age-related macular degeneration
[in Japanese]. Nihon Ganka Gakkai Zasshi. 1997;101:584Y592.
19. Lee BL, Lim JI, Grossniklaus HE. Clinicopathologic features of
indocyanine green angiography-imaged, surgically excised choroidal
neovascular membranes. Retina. 1996;16:64Y69.
20. Obana A, Gohto Y, Nishiguchi K, et al. Classification and visual
outcome of choroidal neovascularization in age-related macular
degeneration by indocyanine green angiography. Jpn J Ophthalmol.
2000;44:668Y676.
21. Obana A, Gohto Y, Matsumoto M, et al. Indocyanine green
angiographic features prognostic of visual outcome in the natural course
of patients with age related macular degeneration. Br J Ophthalmol.
1999;83:429Y437.
6. Bakall B, Folk JC, Boldt HC, et al. Aflibercept therapy for exudative
age-related macular degeneration resistant to bevacizumab and
ranibizumab. Am J Ophthalmol. 2013;156:15Y22.
22. Chang TS, Freund KB, de la Cruz Z, et al. Clinicopathologic correlation
of choroidal neovascularization demonstrated by indocyanine green
angiography in a patient with retention of good vision for almost four
years. Retina. 1994;14:114Y124.
7. Cho H, Shah CP, Weber M, et al. Aflibercept for exudative AMD with
persistent fluid on ranibizumab and/or bevacizumab. Br J Ophthalmol.
2013;97:1032Y1035.
23. Avvad FK, Duker JS, Reichel E, et al. The digital indocyanine green
videoangiography characteristics of well-defined choroidal
neovascularization. Ophthalmology. 1995;102:401Y405.
190
www.apjo.org
* 2015 Asia Pacific Academy of Ophthalmology
Copyright © 2015 Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.