Polypoidal Choroidal Vasculopathy Masquerading as Neovascular Age-Related Macular Degeneration Refractory to Ranibizumab




Purpose


To report a neovascular age-related macular degeneration pattern refractory to ranibizumab.


Design


Retrospective, observational case series.


Methods


Between March and May 2009, cases with neovascular age-related macular degeneration refractory to ranibizumab were investigated with indocyanine green angiography. We identified 12 eyes of 12 patients with polypoidal choroidal vasculopathy. Refractory to treatment were defined cases with persistent subretinal or intraretinal fluid, or both, after 3 or more consecutive monthly ranibizumab injections regardless of best-corrected visual acuity.


Results


All patients identified were white, of whom 6 were male. Mean age ± standard deviation at presentation was 75 ± 5.6 years (range, 64 to 81 years); diagnosis, based on fluorescein angiography, comprised occult choroidal neovascularization (CNV) in 8 eyes, and 1 case each of classic-no-occult CNV, minimally classic CNV, predominantly classic CNV, and retinal angiomatous proliferation. Eight cases had switched from courses of other therapy (5 pegaptanib, 1 photodynamic therapy, 1 photodynamic therapy then pegaptanib, 1 bevacizumab). After a mean follow-up of 10.2 ± 4.8 months (range, 3 to 18 months) and 7.6 ± 3.9 ranibizumab injections (range, 3 to 14 injections), indocyanine green angiography revealed polypoidal choroidal vasculopathy lesions in all cases.


Conclusions


Neovascular age-related macular degeneration refractory to a course of ranibizumab injections may harbor polypoidal choroidal vasculopathy. In such cases, indocyanine green angiography is a valuable tool for revealing polypoidal lesions.


Monthly administered intravitreous pharmacotherapy with ranibizumab, a vascular endothelial growth factor (VEGF) inhibitor, has been shown to improve mean visual acuity safely in eyes with neovascular age-related macular degeneration (AMD). Similar results to those reported by the randomized control trials have been achieved through an optical coherence tomography (OCT)-guided variable-dosing regimen, with fewer ranibizumab injections. In clinical practice, however, some patients fail to respond to ranibizumab, leading to repeated treatments. Currently, the cause of this poor response is not well understood. We herein describe a neovascular AMD pattern refractory to ranibizumab treatment.


Methods


This was a retrospective analysis of 12 eyes of 12 patients with neovascular AMD refractory to ranibizumab in whom polypoidal choroidal vasculopathy (PCV) was diagnosed between March and May 2009 in St. Paul’s Eye Unit. Under our care, patients with 25 letters or more of best-corrected visual acuity (BCVA) on Early Treatment Diabetic Retinopathy Study charts and subfoveal or juxtafoveal neovascular AMD of any subtype with signs of recent disease progression (ie, drop in BCVA; presence of fresh hemorrhage, exudates, or subretinal or intraretinal fluid, or both) are eligible for ranibizumab pharmacotherapy. After a loading course of 3 consecutive monthly intravitreal ranibizumab injections, patients are followed up monthly. At each visit, patients undergo BCVA assessment, dilated funduscopy, and OCT scanning. BCVA is measured by externally accredited examiners using the Early Treatment Diabetic Retinopathy Study charts at 4 and 1 m and is recorded as letters read at 1 m. Unless otherwise indicated, stereo fluorescein angiography (FA) is performed at baseline and at 3-month intervals thereafter. The decision for retreatment is based on the evidence of subretinal or intraretinal fluid on OCT, or both; BCVA loss of 5 letters or more and fewer than 20 Early Treatment Diabetic Retinopathy Study letters compared with the previous visit; and signs of leakage on FA. Regardless of BCVA change, cases are considered refractory to treatment with evidence of persistent subretinal or intraretinal fluid on OCT, or both; fresh hemorrhage; or new serosanguineous retinal pigment epithelium (RPE) detachment after 3 or more consecutive monthly ranibizumab injections. In addition to the standard management, cases refractory to ranibizumab therapy are evaluated with stereo indocyanine green angiography (ICGA).


Diagnosis of PCV was established by the presence of branched vascular network(s) terminating in polypoidal-like lesions located under the RPE on ICGA. The angiograms of all cases were evaluated independently by 4 clinicians (A.N.S., C.J.P., H.H., S.P.H.). Final diagnosis was based on the concurrent opinion of 3 of 4 clinicians. PCV was classified angiographically, after Cackett and associates, according to polyp size, location, formation, and number of discrete polyp areas (foci). Size of polyp was based on the largest diameter of polyp measured in micrometers. Location of polyp was divided into 4 categories: peripapillary (within 1 disc diameter from the optic nerve head), subfoveal, juxtafoveal (< 200 μm from the fovea), and extrafoveal. Lesion formation was classified into 3 groups: single, cluster (≥ 2 polyps in a group configuration), string (≥ 3 polyps in a line), or a combination thereof. Number of discrete areas (foci) was classified as single or multiple. Patient data were retrieved, including gender, age at presentation, race, eye laterality, initial study eye diagnosis, fellow eye diagnosis, past medical and ocular history, BCVA (at presentation, at ranibizumab initiation, and at PCV diagnosis), OCT findings, number of ranibizumab treatments, and follow-up duration.




Results


The characteristics and clinical data of the 12 PCV cases are listed in Table 1 . All patients were white, of whom half were male. At presentation, mean age ± standard deviation was 75 ± 5.5 years (range, 64 to 81 years). Diagnosis at baseline, based on FA results, comprised occult choroidal neovascularization (CNV) in 8 eyes, and 1 case each of classic-no-occult CNV, minimally classic CNV, predominantly classic CNV, and retinal angiomatous proliferation. Four cases were treatment-naïve lesions, whereas 8 cases had switched from courses of other therapy with a mean follow-up before ranibizumab initiation of 15 ± 12 months (range, 3 to 40 months). In detail, 5 patients had a course of pegaptanib intravitreal pharmacotherapy (mean, 6.4 ± 2; range, 3 to 8), 1 patient had 3 photodynamic therapies, 1 patient had 1 photodynamic therapy (PDT) followed by 3 pegaptanib intravitreal injections, and 1 patient had 1 bevacizumab intravitreal injection. Fellow eye diagnosis at presentation comprised 4 cases with dry AMD, 4 cases with end-stage neovascular AMD, 2 cases with isolated serous RPE detachment, 1 with epiretinal membrane, and 1 without any macular disease.



TABLE 1

Twelve Cases with Ranibizumab-Refractory Neovascular Age-Related Macular Degeneration and Polypoidal Choroidal Vasculopathy Lesions
































































































































































Case No. Gender Age (yrs) Eye Study Eye Diagnosis Fellow Eye Diagnosis BCVA at Presentation Initial Treatment BCVA at Ranibizumab Initiation No. of Ranibizumab Injections BCVA at PCV Diagnosis
1 M 64 Left MC CNV End-stage nAMD 49 Pegaptanib 34 14 59
2 F 75 Left Occult CNV ERM 64 Bevacizumab 64 14 61
3 F 76 Right RAP Serous RPE detachment 58 Ranibizumab 58 3 65
4 F 65 Left Occult CNV End-stage nAMD 33 Pegaptanib 28 8 43
5 M 79 Left Occult CNV End-stage nAMD 61 Ranibizumab 61 3 64
6 M 79 Left Occult CNV End-stage nAMD 68 Ranibizumab 68 8 66
7 F 79 Right Occult CNV Serous RPE detachment 54 Pegaptanib 55 4 69
8 M 77 Left Classic CNV Dry AMD 40 PDT, Pegaptanib 39 4 27
9 M 81 Left Occult CNV Dry AMD 25 Pegaptanib 31 11 29
10 F 75 Right PC CNV Physiologic 53 PDT 45 6 59
11 M 75 Right Occult CNV Dry AMD 48 Ranibizumab 48 7 61
12 F 69 Left Occult CNV Dry AMD 74 Pegaptanib 64 9 63

AMD = age-related macular degeneration; BCVA = best-corrected visual acuity (assessed with Early Treatment Diabetic Retinopathy Study charts and expressed in number of letters seen); CNV = choroidal neovascularization; ERM = epiretinal membrane; F = female; M = male; MC = minimally classic; nAMD = neovascular AMD; PC = predominately classic; PCV = polypoidal choroidal vasculopathy; PDT = photodynamic therapy; RAP = retinal angiomatous proliferation; RPE = retinal pigment epithelium; yrs = years.


Mean BCVA was 52 ± 14 letters (range, 25 to 74 letters) and 50 ± 13 letters (range, 28 to 68 letters) at presentation and at initiation of ranibizumab treatment, respectively, representing a nonsignificant drop of 2 ± 5 letters ( P = .14, Student t test). After a mean follow-up of 10.2 ± 4.8 months (range, 3 to 18 months) and 7.6 ± 3.9 ranibizumab injections (range, 3 to 14 injections), mean BCVA was 55 ± 14 letters (range, 27 to 69 letters). This represents an improvement by 5 ± 10 letters since initiation of ranibizumab ( P = .07). Best-corrected visual acuity improved by 15 letters in 2 patients (17%), whereas 10 patients (83%) lost or gained fewer than 15 letters. No patient lost more than 15 letters with ranibizumab treatment.


Mean central foveal thickness (CFT) on OCT was 349 ± 141 μm and 316 ± 136 μm at presentation and at initiation of ranibizumab treatment, respectively. At diagnosis of PCV, mean CFT was 370 ± 140 μm, indicative of resistance to ranibizumab treatment. Because CFT was measured with different OCT machines, we cannot quantify CFT changes precisely. However, qualitative analysis of OCT scanning revealed persistent retinal exudation despite continuous ranibizumab treatments.


The average size of the largest polyp was 340 ± 160 μm (range, 120 to 770 μm). Polypoidal choroidal vasculopathy lesions were found in multiple discrete areas (foci) in 9 eyes (75%). The breakdown of formation of lesion was: single polyp in 9 eyes (75%), cluster of polyps in 6 eyes (50%), and a string of polyps in 1 eye (8%). In this cohort, we did not detect any peripapillary PCV lesions. In 6 eyes (50%), we detected foveal PCV lesions, whereas in 5 eyes (42%), we detected extrafoveal PCV lesions. One eye had both foveal and extrafoveal polyps. Details on PCV classification for each patient are presented in Table 2 . PCV lesions on the fellow eye were found in 5 cases (42%). Four of those eyes were diagnosed clinically with end-stage neovascular AMD and 1 eye was diagnosed with serous RPE detachment.



TABLE 2

Polypoidal Choroidal Vasculopathy Classification a in 12 Cases with Ranibizumab-Refractory Neovascular Age-Related Macular Degeneration


















































































Case No. Size (μm) Location Formation Discrete Areas
1 400 Extrafoveal Single/cluster 2
2 210 Foveal Single 1
3 210 Foveal Single 4
4 380 Extrafoveal Cluster 2
5 770 Extrafoveal Cluster 2
6 420 Foveal/extrafoveal Cluster/string 2
7 340 Foveal Single/cluster 4
8 430 Foveal Single 1
9 230 Extrafoveal Single 3
10 250 Foveal Single 1
11 120 Foveal Single 5
12 330 Extrafoveal Single/cluster 2

Adapted from Cackett P, Wong D, Yeo I. A classification system for polypoidal choroidal vasculopathy. Retina 2009;29:187–191.

a This classification system for polypoidal choroidal vasculopathy comprises polyp size, location, formation, and number of discrete polyp areas.



Case 1


A 64-year-old white man with hypertension was referred by his optician for left eye suspected neovascular AMD. The patient reported metamorphopsia with concomitant vision drop of few weeks’ duration. On presentation, BCVA was 49 letters (20/100 Snellen equivalent). CFT was 379 μm. He was diagnosed with minimally classic CNV lesion ( Figure 1 ) and started with pegaptanib intravitreal pharmacotherapy. The fellow eye showed signs of end-stage neovascular AMD. After 6 consecutive pegaptanib injections, BCVA dropped to 34 letters (20/200) with evidence of leakage on FA and persistent subretinal and intraretinal fluid on OCT (CFT, 348 μm). He then was switched to ranibizumab intravitreal pharmacotherapy. After 14 consecutive monthly ranibizumab injections, BCVA improved to 59 letters (20/63 Snellen); FA showed leakage with concomitant persistent exudation on OCT (CFT, 288 μm). He subsequently underwent ICGA, which revealed PCV lesions ( Figure 2 ). The size of the largest polyp was 400 μm. Two discrete areas of PCV lesions (foci) were identified (a single PCV lesion and a cluster of 4 polyps), both being extrafoveal in location. Spectral-domain OCT showed a relative hypereflectivity under the RPE detachment, suggestive of the PCV lesions, with concomitant intraretinal and subretinal fluid ( Figure 2 ).


Jan 17, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Polypoidal Choroidal Vasculopathy Masquerading as Neovascular Age-Related Macular Degeneration Refractory to Ranibizumab

Full access? Get Clinical Tree

Get Clinical Tree app for offline access