PCV has been described as an occult CNV since the 1990s, and it was considered to be related to AMD [12]. In 2003 Yuzawa et al. identified typical choroidal features in patients with PCV, detected by ICGA. Those choroidal vessel abnormalities were proposed to be pathogenetic of PCV [13].
In 2013 Kawamura et al. described two different patterns of PCV [9]. The first pattern was named polypoidal CNV: it showed up in FA and ICGA as a vascular network of feeder and draining vessels, new vessels growing through a break in the RPE and regular inner choroid. They also described a second pattern of PCV named typical PCV, where feeder and draining vessels were not detectable and this form of neovascularization would originate from choroidal vascularization abnormalities on atherosclerotic hyalinization.
Pang and Freund in 2015 were the first to indicate type 1 PCV neovascularization as a possible neovascular feature of pachychoroid condition, naming it pachychoroid neovasculopathy. According to these authors, pachychoroid drives a spectrum of different clinical presentations from PPE and CSC, to CNV and PCV along a continuous pathogenic evolution [1].
A thick choroid could therefore be the primum movens of a vascular condition which leads to a neoangiogenic process [5]. The dilation of vessels in choroidal Haller’s layer determines choroidal compression in Sattler’s layer and the thinning of choriocapillaris. Secondary hypoxic changes in RPE are responsible for epitheliopathy, serous detachment or atrophy. Manayath et co-workers showed that the coexistence of PCV and RPE changes, due to pre-existing associated CSC, may support the association between these two conditions in a pathogenetic relationship with a thick choroid [5].
PCV associated with pachychoroid should therefore be differentiated from similar neovascularization associated to AMD [10, 11, 14] and it should be indicated as PPCV. Based on Gass classification, PCV is considered a type 1 subretinal neovascularization which grows underneath the RPE [15]. This anatomic location may explain the inferior response to treatment with the anti-VEGF drugs in comparison to other forms of NV, due to an inadequate delivery of drugs to the correct site [16-17].
PCV was first treated with vPDT, which has good efficacy in closing polyps and reducing exudation, but recurrences of disease activity, i.e. exudation or hemorrhages, account for even more than 50% of cases during the first year of observation [18-19].
After the introduction of anti-VEGF drugs in clinical practice, the first study comparing vPDT and anti-VEGF was the LAPTOP trial, showing a superiority of 0.5mg Ranibizumab compared to vPDT on naïve PCV patients, in terms of visual acuity (VA) improvement [20]. The PEARL study, which focused on anti-VEGF treatment in eyes with hemorrhagic or exudative PCV, showed a low efficacy after monthly treatment with 0.5 mg Ranibizumab, with only 31% of polypoidal complexes decreased in size after 1 year [21].
The EPIC study showed a good response to IVA labelled protocol, with a polyp closure rate of 67% in treated eyes at the 6-month endpoint. Patients included in this study had high VA at baseline and there was no statistically significant improvement in VA at the endpoint, while 90% of this group showed VA stability [22].
Although previously evaluated [23], the effectiveness of combining vPDT with intravitreal therapy (IVT) with anti-VEGF drugs in PCV treatment was stated by Everest II and Planet studies with evidence level I. Everest II demonstrated that Ranibizumab plus standard vPDT combined therapy was statistically superior in terms of VA recovery and free-from-disease interval when compared to ranibizumab monotherapy, after a 12-month follow-up. Complete polypoidal regression was also higher in the combined therapy group (69.3% vs. 34.7%) [24]. Choroidal response to treatment was later reported as a higher reduction in mean choroidal thickness in the combined group (55.2 µm) compared to the ranibizumab monotherapy cohort (30 µm) [25].
The PLANET study examined the effect of a combined therapy with Aflibercept plus vPDT compared to Aflibercept monotherapy within a 12-month follow-up. It demonstrated the non-inferiority of monotherapy treatment in comparison to the combined treatment. VA improvement after 52 weeks of follow-up was similar in both groups (10.9 vs 10.7 ETDRS letters). The closure rate of polypoidal lesions was similar in both groups (44.8% vs 38.9%). Choroidal thickness was not analyzed. No clear information about the vPDT setting parameters, such as the dose of irradiation, fluence and spot dimension, were provided. Moreover, the combined treatment was assigned and performed after a loading dose of 3 monthly IVA, not as per protocol but only if some strict criteria were fulfilled [26].
Generally, neither the first EVEREST II report nor the PLANET study reported a basal choroidal condition or differentiated the PCV form.
In a recent study by Sakurada et coll. regarding PCV treatment, choroidal thickness was a relevant parameter to be considered together with VA, in order to evaluate the effects of vPDT+Ranibizumab and vPDT+Aflibercept combined therapies. Better VA at the endpoint was statistically associated with higher subfoveal choroidal thickness at baseline and with greater CCT reduction irrespective of treatment modalities [27]. Other studies would suggest that eyes with hyperpermeable choroid would respond better to vPDT when compared with AMD related PCV [28] but poorly to mono-therapy with anti-VEGF [11]; Yanagi et coll. addressed PCV related to pachychoroid as a subgroup of PCV, associating different clinical responses to treatment with peculiar pathogenesis and underlining the need for a lower number of IVTs when this treatment is combined with vPDT [10].
In our study we showed the clinical results of combined vPDT+IVA therapy in patients affected by PPCV (figure 1 and 2).
The combined treatment rationale is based on the different targets of these therapies. The vPDT was applied for its photochemical thrombosis and remodeling of the choroidal network, acting on aneurysmal vascular origin and aneurysmal vessel closure, thus reducing choroidal congestion and hyperpermeability [18]; however, it is expected to induce overexpression of VEGF [23]. Laser spot dimension was set in order to evaluate the effect on neovascular lesions and on correspondent choroidal areas. Its effect on choroidal permeability and congestion was evaluated as a reduction in choroidal thickness and choroidal hyperpermeability [5,10]. IVA works by blocking VEGF, neoangiogenesis and subsequently retinal exudation; its effect was evaluated as changes in retinal fluids and thicknesses.
The response to treatment was demonstrated by an improvement in all the parameters taken into account.
CDVA progressively improved during the time of follow-up, despite relapses and IVA retreatment, as a result of recovery from chorioretinal function and resolution of retinal fluids, reaching statistical significance after 6 and 12 months. Almost all the eyes improved in CDVA after 12 months; one eye showed a relapse of activity at the 12-month follow-up visit.
IRF and SRF presence reduced as a response to anti-VEGF drug’s anti-edema activity and by polypoidal closure secondary to the direct effect of PDT. Moreover, a reduction in Haller’s choroidal vascular thickness and in choroidal hyperpermeability could determine a decrease of SRF and, therefore, lower stress on RPE. It is noteworthy that SRF was present in 5.3% of patients after 12 months of follow-up, and IRF in 10.5%; therefore, our data showed a significant reduction in SRF in the majority of patients and a reduction in IRF as well, even though the latter was not proven to be statistically significant. Since SRF and IRF were evaluated as qualitative parameters, their clinical interpretation should be related to quantitative parameters such as CRT and MMT, which were all significantly reduced after 12 months.
CRT followed the CDVA trend, and its highest reduction was after 4.5 and 6 months. MMT reduction was statistically significant after the first month of follow up. Its value accounts for a larger area consisting of fluids, than that included in CRT, where polypoidal lesions were mostly located: in agreement with the literature, we attributed macular thickness reduction to the synergistic action of vascular obliteration vPDT-induced on polypoidal lesions and anti-VEGF-mediated fluid reabsorption.
Choroidal thicknesses (CCT and MCT) displayed a statistically significant reduction after the first month of follow-up, which stabilized during the following months. These data are directly associated with vascular structural modifications vPDT-induced, with a reduction of RPE and choriocapillaris damage caused by minor vascular congestion and vessel caliper [29]. The effect on chorioretinal structures would cause a reduction of SRF, which was related more to the stress on RPE rather than to the polypoidal activity. IVA would maintain the initial effect of vPDT, keeping the retina dry.
The number of IVA (mean 4.6 IVA/year) administered is consistent with the EVEREST II study (Ranibizumab+vPDT+PRN protocol: 5.2 IVT/year, VS monotherapy group: 7.3 IVT/year), but not comparable to the PLANET study due to the strict year-long treatment regimen adopted (8 IVT/year + rescue vPDT) [24-26]. Comparisons with other studies is not possible due to differences in protocols, selection criteria and follow-up. In our opinion, PPCV is better controlled by combined therapy which allows for a fewer number of IVA.
A limitation of the study was the low number of patients recruited: the cohort was however highly selected among patients fulfilling all the pachychoroid criteria. Furthermore, no control group with monotherapy treatment was present. CDVA showed high standard deviation, due to retinal haemorrhages at the baseline but also to high VA recovery after disease activity resolution. Our hypothesis would be strongly reinforced by morphological analysis of choroidal changes, which are only indirectly proved by choroidal thicknesses. However, while ICGA only shows hyperpermeability, attempts at a quantitative analysis of choroidal vessels, as the pachychoroid index, are still not validated [30].