To evaluate the 5-year efficacy of photodynamic therapy (PDT) in patients with polypoidal choroidal vasculopathy (PCV).
Forty-two eyes of 36 patients with PCV followed up for at least 60 months after PDT were reviewed. All eyes were primarily treated with PDT. Main outcome measure was best-corrected visual acuity (BCVA; logMAR [logarithm of minimal angle of resolution]) at baseline and at each follow-up visit. We also classified the eyes into 3 groups: improved (improvement ≥0.3 logMAR), decreased (deterioration ≥0.3 logMAR), and stable.
During the mean follow-up duration, 73.64 ± 13.47 months, the mean number of PDT was 2.21 ± 1.62 treatments. Recurrence was noted in 33 eyes (78.6%) during follow-up. The mean baseline BCVA was 0.78 ± 0.48 logMAR (20/120 Snellen equivalent), and the final BCVA at 60 months was 0.67 ± 0.52 logMAR (20/93 Snellen equivalent) ( P = .050, paired t test). On the final evaluation at 60 months, the mean BCVA was improved in 14 eyes (33.3%), stable in 23 eyes (54.8%), and decreased in 5 eyes (11.9%).
At 60 months after initial PDT, 88.1% of PCV patients showed stable or improved BCVA after PDT. Despite a high recurrence rate, PDT remained effective for 5 years, and represents a good therapeutic approach to PCV.
Since its introduction as a new clinical entity in 1990, polypoidal choroidal vasculopathy (PCV) has been increasingly recognized as a major cause of vision loss throughout the world. PCV is characterized by orange-red protrusion on funduscopy and polyp-like pooling of dye on indocyanine green angiography (ICGA). Abnormal branching vascular networks and pigment epithelial detachment (PED) are frequently associated with polypoidal lesions. In presumed exudative age-related macular degeneration (AMD), PCV accounts for 8%-13% of cases in white populations and 10%-54% of cases in Asian populations.
PCV is characterized by a slower progression and better visual prognosis than wet AMD, although subretinal fibrosis and retinal pigment epithelium (RPE) atrophy can lead to significant and permanent vision loss. A study regarding the natural history of PCV reported that half of eyes examined had hemorrhagic episodes, recurrent leakage, or severe RPE atrophy during and after long-term follow-up periods (24-54 months).
Among the currently available treatment modalities, photodynamic therapy (PDT) with verteporfin seems to be most effective for PCV. PCV is known to have a better therapeutic response to PDT and less responsiveness to anti–vascular endothelial growth factor (VEGF) therapy, compared with AMD. Additionally, several studies have reported complete regression of polyps in 73%-99% of patients and stabilized or improved visual acuity (81%-100%) after PDT. In a recent ICGA-guided, exploratory randomized controlled trial, the efficacies of verteporfin PDT alone or combined with ranibizumab vs ranibizumab monotherapy for PCV were evaluated. The authors reported that both verteporfin PDT alone and combined with 0.5 mg ranibizumab were superior to ranibizumab monotherapy in achieving complete regression of polyps (71.4% and 77.8%, respectively, vs 28.6%) at the 6-month follow-up.
Long-term follow-up studies over 24-36 months corroborate this superior efficacy and safety of PDT in PCV patients. However, they also reported PCV recurrence rates of up to 77% over this period, suggesting long-term follow-up may be necessary in PCV patients, because of high frequencies of recurrent polypoidal lesions, as well as enlargement and neovascular changes involving abnormal vascular networks.
In the current study, we evaluated the long-term follow-up outcomes of PDT with verteporfin in PCV patients. We evaluated the visual outcomes, recurrences, and pattern of abnormal branching vascular networks and polyps using ICGA at regular intervals over 5 years after the initial PDT.
Enrollment of Study Subjects
We retrospectively reviewed 42 eyes of 36 patients who underwent PDT for PCV at the Vitreoretinal Service clinic of Yonsei University Medical Center between January 1, 2000 and April 30, 2007. Informed consent was obtained from all participants. This retrospective study was performed with the approval of the Institutional Review Board of Yonsei University College of Medicine (Reference No.4-2012-0319) and conducted in accordance with the tenets of the Declaration of Helsinki.
The inclusion criteria were: (1) symptomatic macular PCV with subfoveal leakage on fluorescein angiography (FA); (2) baseline best-corrected visual acuity (BCVA) of 0.2 logMAR (logarithm of the minimal angle of resolution) or worse; and (3) presence of branching vascular networks and polypoidal lesions in ICGA. Exclusion criteria were: (1) any treatment for PCV prior to PDT, including laser photocoagulation, transpupillary thermotherapy, or radiotherapy; (2) other concomitant ocular diseases, such as diabetic retinopathy, high myopia, vein or artery occlusion, or epiretinal membrane; (3) any systemic contraindications to verteporfin or angiographic dyes; (4) idiopathic or secondary choroidal neovascularization; (5) absence of signs of AMD in the study and fellow eyes; or (6) RPE tears or ripping.
PCV diagnosis was based on ICGA findings on the choroidal vasculature, branching vascular networks, and polypoidal lesions using the Heidelberg Retina Angiograph system (Heidelberg Engineering, Heidelberg, Germany) with confocal scanning laser ophthalmoscope. Greatest linear dimension (GLD) was measured at baseline and at 60-month follow-up.
All patients received a complete ocular examination, including BCVA testing using a Snellen visual acuity chart, slit-lamp biomicroscopy, dilated fundus examination with indirect ophthalmoscope, color fundus photography, digital FA, ICGA, and optical coherence tomography (OCT) at baseline. BCVA was measured with a standard Snellen visual acuity chart at 6 meters and then converted to the logMAR visual acuity for statistical analysis.
The patients were observed 1 month after PDT, and then at 3-month intervals. At every visit, the BCVA testing, dilated fundus examination, and OCT were performed. In addition to these routine examinations, if patients complained of worsening vision or metamorphopsia, additional FA and ICGA were performed. We collected the examination data from the baseline visit and the 1-, 3-, 6-, 12-, 24-, 36-, 48-, and 60-month follow-ups and interpreted them retrospectively.
When BCVA was converted to logMAR vision for statistical analysis, we classified the patients into 3 groups: improved, stable, and decreased BCVA. An improvement of 0.3 or more in logMAR visual acuity was defined as improved VA, and a reduction of at least 0.3 logMAR visual acuity was defined as decreased BCVA.
PDT was performed according to the protocols of the Treatment of Age-Related Macular Degeneration with Photodynamic Therapy (TAP) and Verteporfin in Photodynamic Therapy (VIP) study groups. All patients received a 6 mg/m 2 infusion of verteporfin (Visudyne; Novartis AG, Bulach, Switzerland) over 10 minutes, followed by diode laser delivery to the choroidal neovascularization at 689 nm, 15 minutes after infusion start. A total light intensity of 50 J/cm 2 and light dose rate of 600 mW/cm 2 for 83 seconds was used, with the aim of covering the entire PCV lesion. An additional 500 μm surrounding the PCV lesion, covering the borders on each side, was included in the treatment area.
Re-treatment was considered based on FA, ICGA, and OCT findings. We defined the recurrence of PCV as the reappearance of active PCV lesions on ICGA with subfoveal leakage on FA after at least 6 months without treatment. Earlier detection of active PCV with leakage before 6 months was considered as persistence of the original lesion, rather than recurrence.
PDT combined with anti-VEGF agents was administered when polypoidal lesions with exudative changes were observed by ICGA, and an intravitreal anti-VEGF monotherapy was given when only exudative changes were observed with no definite polypoidal lesions. The combination treatment regimen consisted of a session of PDT guided by ICGA and an intravitreal injection of anti-VEGF agent within 7 days after PDT or on the same day. Bevacizumab (Avastin; Genentech, San Francisco, California, USA) or ranibizumab (Lucentis; Genentech) was injected as 3 consecutive intravitreal injections, and additional injections were administered depending on the presence of an active lesion from FA, ICGA, and OCT findings. All intravitreal injections were performed in the operating room under sterile conditions. Prior to the procedure, informed consent regarding the off-label use of bevacizumab and its possible complications was obtained from the patients.
SPSS software was used for statistical analysis (version 18.0; SPSS Inc, Chicago, Illinois, USA). The mean BCVA at baseline was compared with the mean BCVA at each follow-up by a paired t test. For nonparametric analysis, the Mann-Whitney U test was used. Results with P < .05 were considered statistically significant.
In total, 42 eyes of 36 patients who completed 5-year follow-up visits after the initial PDT were analyzed. The patients had a mean age of 66.76 ± 7.81 years, and the group consisted of 29 male and 7 female subjects. The mean follow-up duration was 73.64 ± 13.47 months. Baseline FA showed occult choroidal neovascularization in all 42 eyes. Baseline ICGA showed a dilated network of inner choroidal vessels with terminal hyperfluorescent polypoidal lesions in all cases. The patients’ clinical details are listed in Table 1 .
|Age (y; mean ± SD)||66.76 ± 7.81|
|Mean follow-up duration (mo)||73.64 ± 13.47|
|Mean number of PDT treatments||2.21 ± 1.62|
|Mean number of anti-VEGF treatments||5.05 ± 5.55|
|Baseline visual acuity, logMAR (Snellen equivalent)||0.78 ± 0.48 (20/120)|
|Visual acuity at 60 months, logMAR (Snellen equivalent)||0.67 ± 0.52 (20/93)|
|Number (%) of eyes with PCV recurrence after first PDT ± anti-VEGF treatment||33 (78.6%)|
|Mean recurrence interval after first remission (mo)||25.15 ± 21.48|
The mean BCVA was 0.78 ± 0.48 logMAR (20/120 Snellen equivalent) at baseline and 0.67 ± 0.52 logMAR (20/93 Snellen equivalent) at the 60-month follow-up. The changes of mean BCVA at each follow-up visit are depicted in Figure 1 . The mean BCVA showed a tendency of improvement for 6 months after initial PDT and then stabilized through the 60-month follow-up visit. The mean BCVA showed a significant improvement between 1 month and 3 months after the initial PDT, from 0.73 ± 0.43 logMAR (20/107 Snellen equivalent) to 0.68 ± 0.47 logMAR (20/95 Snellen equivalent) ( P = .043, paired t test).
Compared with the baseline BCVA, the mean BCVA at each follow-up significantly improved (3-month, P = .015; 6-month, P = .005; 12-month, P = .004; 24-month, P = .007; 36-month, P = .007; 48-month, P = .034; paired t test). However, whereas the mean BCVAs at the 1- and 60-month follow-ups were improved with respect to the baseline BCVA, the results were not statistically significant ( P = .222 and P = .050, respectively; paired t test).
After assessment of BCVA at each follow-up visit, we categorized the patients into 3 groups, based on their BCVA change from baseline: improved, stable, and decreased visual acuity. The mean BCVA had improved in 12 (28.6%), 13 (31.0%), 14 (33.3%), and 14 (33.3%) eyes at 12, 24, 36, and 48 months, respectively; it remained stable in 28 (66.6%), 26 (61.9%), 24 (57.1%), and 24 (57.1%) eyes at 12, 24, 26, and 48 months, respectively; and it deteriorated in 2 (4.8%), 3 (7.1%), 4 (9.6%), and 4 (9.6%) eyes at 12, 24, 36, and 48 months, respectively ( Figure 2 ). On the final evaluation at 60 months, the mean BCVA was improved in 14 eyes (33.3%), stable in 23 eyes (54.8%), and decreased in 5 eyes (11.9%). The mean BCVA changes observed over follow-up visits are depicted in Figure 3 .
Recurrence of Polypoidal Choroidal Vasculopathy
Recurrences of PCV were noted in 33 eyes (78.6%) during follow-up. The mean number of PDT treatments given to eyes with recurrent PCV was 2.45 ± 1.72, and that of anti-VEGF injections was 6.42 ± 5.51. The mean number of PDT treatments in eyes without recurrence was 1.33 ± 0.71; none of the eyes without recurrence received anti-VEGF injections. The eyes with recurrent PCV received significantly more PDT treatments and anti-VEGF injections than those without recurrence ( P < 0.001, P = .025, respectively; Mann-Whitney U test).
Visual Outcome in Recurred Polypoidal Choroidal Vasculopathy
The mean BCVA at baseline was 0.81 ± 0.44 logMAR (20/129 Snellen equivalent) for the eyes with recurrence and 0.67 ± 0.60 logMAR (20/93 Snellen equivalent) for the eyes without recurrence. At the 60-month follow-up, the mean BCVA had significantly improved to 0.71 ± 0.47 logMAR (20/102 Snellen equivalent) for the eyes with PCV recurrence ( P = .048, Wilcoxon signed rank test), and to 0.52 ± 0.63 logMAR (20/66 Snellen equivalent) for the eyes without ( P = .012, Wilcoxon signed rank test). At each follow-up, the mean BCVA was better in the eyes without recurrence ( Figure 4 ); however, this difference was not statistically significant for any of the follow-up visits ( Table 2 ).
|PCV Recurrence||No Recurrence||P Value a|
|Patient age at diagnosis (years)||66.88 ± 7.17||66.33 ± 10.34||.763|
|GLD at baseline (μm)||2811.57 ± 1008.52||2843.00 ± 1064.60||.944|
|GLD at 60 months (μm)||3741.92 ± 1459.61||2550.00 ± 1020.02||.016|
|Number of PDT treatments||2.45 ± 1.72||1.33 ± 0.71||.025|
|Number of anti-VEGF injections||6.42 ± 5.51||0.00||<0.001|
|BCVA at baseline, logMAR (Snellen equivalent)||0.81 ± 0.44 (20/129)||0.67 ± 0.64 (20/93)||.224|
|BCVA at 1 month (logMAR) (Snellen equivalent)||0.73 ± 0.37 (20/107)||0.72 ± 0.65 (20/104)||.526|
|BCVA at 3 months (logMAR) (Snellen equivalent)||0.70 ± 0.43 (20/100)||0.61 ± 0.64 (20/81)||.348|
|BCVA at 6 months (logMAR) (Snellen equivalent)||0.68 ± 0.42 (20/95)||0.56 ± 0.54 (20/72)||.181|
|BCVA at 12 months (logMAR) (Snellen equivalent)||0.68 ± 0.46 (20/95)||0.56 ± 0.65 (20/72)||.262|
|BCVA at 24 months (logMAR) (Snellen equivalent)||0.69 ± 0.46 (20/97)||0.53 ± 0.66 (20/67)||.152|
|BCVA at 36 months (logMAR) (Snellen equivalent)||0.67 ± 0.47 (20/93)||0.51 ± 0.67 (20/64)||.152|
|BCVA at 48 months (logMAR) (Snellen equivalent)||0.71 ± 0.47 (20/102)||0.51 ± 0.68 (20/64)||.127|
|BCVA at 60 months (logMAR) (Snellen equivalent)||0.71 ± 0.48 (20/102)||0.52 ± 0.67 (20/66)||.181|
Recurrence Interval of Polypoidal Choroidal Vasculopathy
We also analyzed the PCV recurrence intervals. There was 1 recurrence in 15 eyes, 2 recurrences in 12 eyes, and 3 recurrences in 6 eyes. Initial recurrences were noted in 3 eyes within 12 months of baseline PDT treatment; in 9 eyes between 13 and 24 months; in 12 eyes between 25 and 36 months; in 4 eyes between 37 and 48 months; and in 3 eyes between 49 and 60 months. Furthermore, 2 eyes showed an initial recurrence at 70 months and 84 months after the initial PDT session. In eyes with 2 or 3 instances of PCV recurrence, the mean interval between first and second recurrences was 21.64 ± 14.76 months (maximum 52 months), and 21.25 ± 11.30 months (maximum 36 months) between second and third recurrences.
Greatest Linear Dimension of Recurred Polypoidal Choroidal Vasculopathy
The GLD observed in PCV eyes with at least 1 recurrence after PDT was 2811.57 ± 1008.52 μm at baseline and 3741.92 ± 1459.61 μm at the 60-month follow-up. In the 9 eyes without any recurrence during the follow-up period, the extent of branching vascular network had decreased at the 60-month follow-up (2550.00 ± 1020.02 μm), with respect to baseline (2843.00 ± 1064.60 μm). The GLD between the eyes with and without recurrence was not significant at baseline ( P = .944, Mann-Whitney U test), but was statistically significant at 60-month follow-up ( P = .016, Mann-Whitney U test).
Abnormal Branching Vascular Network of Recurred Polypoidal Choroidal Vasculopathy
The extent of abnormal branching vascular network was stable in 5 eyes and enlarged in 28 eyes at the 60-month follow-up with respect to baseline observations. Among the 28 eyes showing an expanded abnormal branching vascular network, 25 of them had new polypoidal lesions in the periphery of their abnormal vascular networks, in regions the network had expanded into after PDT. In the remaining 3 eyes, there were no new polyps visible with ICGA, but leakage persisted from the branching vascular network. Among the 5 eyes with stable or similar abnormal branching vascular networks, new polyps adjacent to a previous polyp were observed in 1 eye, new polyps at the periphery of the branching vascular network were observed in 2 eyes, and persistent leakage from the branching vascular network itself was seen for 2 eyes.
None of the patients developed systemic or ocular complications associated with PDT or the supplemental verteporfin, nor did any of the patients develop systemic complications, such as cerebrovascular accidents, attributable to the intravitreal anti-VEGF injections. No ocular complications, including cataracts, increased intraocular pressure, endophthalmitis, or retinal detachment, were detected, nor were any RPE tears noted.