Dealing with the Verteporfin Shortage: Treatment Options and Outcomes in Patients with Chronic and Non-Resolving Central Serous Chorioretinopathy

Central serous chorioretinopathy (CSC) is a pachychoroid disease that is characterized by a neuroretinal detachment due to accumulation of serous subretinal fluid (SRF). If SRF is located in the macula, permanent loss of visual acuity (VA), color vision and contrast sensitivity can occur. Structural abnormalities in eyes with CSC are initially mainly found in the choroid and include the following pachychoroid features: choroidal hyperpermeability, thickening of the choroid and dilated large choroidal veins in Haller’s layer (pachyvessels) with concomitant attenuation of the choriocapillaris. These choroidal anomalies may lead to retinal pigment epithelium (RPE) dysfunction and focal RPE defects, causing the subsequent accumulation of SRF [1, 2]. CSC can be categorized in acute CSC (aCSC), non-resolving CSC, and chronic CSC (cCSC). In aCSC, SRF resolves spontaneously within 4 months, and patients have a good visual prognosis. Non-resolving CSC is defined as aCSC with SRF persisting for longer than 4 months. In cCSC, irreversible changes in the RPE or photoreceptor layers are seen, due to persistent SRF [3, 4]. Resolution of SRF is warranted, to prevent irreversible vision loss.

Half-dose photodynamic therapy (HD-PDT) is proven to be effective in cCSC, and it is the treatment of first choice when SRF persists for 2–6 months [3, 5]. In HD-PDT, a photosensitizing benzoporphyrin derivate known as verteporfin (Visudyne^®; Alcami Carolinas Corporation, Charleston, SC, USA) is intravenously administered, followed by a laser treatment. Other treatments for CSC are laser photocoagulation, subthreshold micropulse laser (SML), and oral eplerenone. The last two mentioned treatments have been shown to be inferior to HD-PDT in large randomized controlled trials [6, 7]. Additionally, it was shown that treatment with steroid eye drops might also have a potentially beneficial effect in CSC and in peripapillary pachychoroid syndrome (PPS), another pachychoroid-related disease [8‒10].

As of July 2021, there is a worldwide shortage of verteporfin [11, 12]. In the Netherlands, only 156 vials of verteporfin were distributed to the Dutch hospitals in the first year of the shortage, whereas 766 vials were distributed in the year before. Due to the nationwide shortage of verteporfin, a national committee was established that consisted of retinal specialists from clinics that perform PDT. This committee allocated verteporfin according to a prioritization schedule that was adjusted depending on the stock of verteporfin [11, 12]. Verteporfin was assigned only to CSC patients with persistent subfoveal fluid and a leakage point that was not accessible with laser photocoagulation (focal leakage point within 375 μm from the fovea on fluorescein angiography) [13]. Verteporfin was allocated if SRF was present for 3 months in case the contralateral eye had a VA <20/40 Snellen equivalent. In CSC eyes with a VA of >20/40 Snellen in the fellow eye, verteporfin was accorded when SRF was present ranging from 4 to 8 months (depending on inventory).

Whereas before verteporfin shortage most CSC cases with persistent SRF would have been treated with PDT immediately, this was not feasible as a consequence during the shortage. Based on findings on multimodal imaging a more patient-tailored approach was adopted such as a wait-and-see policy or other treatment modalities including SML, laser photocoagulation or treatment with prednisolone eye drops.

In this study, we evaluate the management of CSC during shortage of verteporfin. We report and compare the outcomes of patients with non-resolving CSC and cCSC who were referred to a tertiary referral center in a 1-year period prior to and a 1-year period during verteporfin shortage.

In this retrospective study, all medical records of patients referred for CSC between June 1, 2020, and May 31, 2022, to the Radboud University Medical Center (Nijmegen, The Netherlands) were screened. Patients were divided into two groups. The first group contained patients who had the baseline visit during times of full availability of verteporfin (group 1; June 1, 2020–May 31, 2021). The second group consisted of patients referred to the outpatient clinic in times of verteporfin shortage (group 2; June 1, 2021–May 31, 2022).

This study was conducted in accordance with the Tenets of the Declaration of Helsinki. Due to the retrospective nature of this study, the need to obtain written informed consent was waived by the Institutional Review Board (MREC Oost-Nederland, Application No. 2022-15947).

Patients were included in this study if they presented with subfoveal SRF on spectral domain optical coherence tomography (SD-OCT), due to cCSC or non-resolving CSC. Diagnosis was based on previous described definitions [4, 7]. cCSC was defined as hyperfluorescent leakage and the presence of RPE window defects on fluorescence angiography (FA) with corresponding hyperfluorescent areas on indocyanine green angiography (ICGA), if ICGA was available. For cCSC, subjective visual symptoms or SRF on SD-OCT had to be present for at least 6 weeks [6, 7]. Diagnosis of non-resolving CSC was made if subjective visual symptoms or SRF on SD-OCT were present for at least 4 months in combination with hyperfluorescent areas on ICGA and presence of RPE window defects on FA [4]. Exclusion criteria were evidence of other retinal diseases that could cause SRF; presence of a choroidal neovascularization secondary to CSC; treatment within 6 months before the baseline visit with one of the following: photodynamic therapy, SML, focal laser photocoagulation, mineralocorticoid receptor antagonists; and a follow-up period of less than 4 months after the baseline visit (Fig. 1).

Data were collected from the baseline visit to the first visit after 4 months of follow-up. The following data from the medical records were collected: age; sex; steroid use; previous performed treatments for cCSC (HD-PDT, laser photocoagulation, SML or mineralocorticoid receptor antagonists); corrected VA (either best corrected VA or patients own refraction, measured in Snellen and converted to ETDRS letter scores according to a previously published method [14]); findings on slit-lamp biomicroscopy and on fundoscopy; and treatment strategy. Imaging characteristics were analyzed on images obtained during regular clinical visits, containing SD-OCT, OCT angiography (OCT-A), FA and ICGA. All imaging was performed on the Spectralis HRA + OCT device (Heidelberg Engineering, Heidelberg, Germany). SRF was recorded as subfoveal SRF in case it was present inside 750 µm from the foveal center and this was graded by two independent graders (S.Y. and F.M.v.d.T.).

Treatment strategies that were performed in this cohort were as follows: HD-PDT, SML, laser photocoagulation, prednisolone eye drops and a wait-and-see policy. HD-PDT was performed according to a previously described protocol [15]: 3 mg/m² verteporfin was administered intravenously in 10 min. At 5 min after the infusion, PDT laser treatment was performed with a fluency of 50 J/cm², a wavelength of 689 nm and a duration of 83 s. The treatment was guided by the location of hypercyanescent areas on ICGA. SML was performed with an 810-nm diode laser (Iridex, Mountain View, USA) and hypercyanescent areas on ICGA were treated. Most SML treatments were performed according to a previously described protocol with a power of 1,800 mW, a duty cycle of 5%, a frequency of 500 Hz, a treatment spot of 150 µm and an exposure time of 0.2 s per spot [15]. Two other protocols for SML treatment were also performed, one with a power of 400 mW, a treatment spot of 500 µm and an exposure time of 1 s, the other with a power of 1,200 mW, a treatment spot of 200 µm and an exposure time of 0.3 s. Laser photocoagulation was based on the focal leakage point on FA and was performed using a Pascal^® laser (Topcon Medical Laser Systems, Santa Clara, CA, USA) with a spot size of 100–200, a power of 80–200 mW and an application time of 10–20 ms. Prednisolone acetate eye drops 10 mg/mL (Pred Forte^®) were prescribed for three times a day (every 8 h). Prior to the off-label treatment with prednisolone eye drops, benefits, risks and alternative treatments were discussed with the patient.

In the statistical analysis, quantitative variables are presented as the mean and standard deviation (SD) and categorical variables as counts and percentages. VA measurements were converted to approximate ETDRS letter scores using the formula: ETDRS = 85 + 50 × log10(snellen_fraction) [14]. Normal distribution of data was assessed for continuous variables by the Kolmogorov-Smirnov test. If groups were normally distributed, mean values were compared by the Student’s t test. Otherwise, median values were tested with the Mann-Whitney U test. The chi-square test was used to test for differences in categorical variables. A p value <0.05 was considered significant.

A total of 211 patients were referred to the Radboud University Medical Center for non-resolving CSC or cCSC between June 1, 2020, and May 31, 2022, and 79 patients finally met the inclusion criteria. Most patients were excluded because subfoveal SRF had resolved by the time they were seen at our clinic. The selection of the included cases is presented in a flowchart (Fig. 1). Among the cases that met inclusion, 32 patients (36 eyes) were in group 1 and 47 patients (49 eyes) were in group 2.

The mean age of the total group was 51.9 ± 10.7 years, and 20% of patients were women. The mean follow-up time was 6.2 ± 2.1 months. Baseline variables per group are shown in Table 1. There was no significant difference between the groups in time from the start of subjective symptoms to the first visit (p = 0.536), in time from the first OCT until the first visit (p = 0.825) and in the follow-up time (p = 0.231). The mean number of visits during follow-up was 3.3 ± 0.7, and this was not statistically different between the two groups (p = 0.777).

Due to the lack of verteporfin, treatment strategy was altered. The flowchart of the decision-making at the baseline visit during times of verteporfin availability and shortage is presented in Figure 2. During the shortage, there was a reduction in performed HD-PDT treatments at the baseline visit (Fig. 3).

After wait-and-see policy or a treatment at baseline visit, additional treatment was performed in a subset of patients during follow-up. Additional treatment was performed in 4 eyes (11%) in group 1, and all of these eyes received HD-PDT. In group 2, 12 (25%) eyes received additional treatment during follow-up: 9 eyes were treated with HD-PDT, 1 eye was treated twice with SML, 1 eye was treated with prednisolone eye drops, and 1 eye was treated with laser photocoagulation.

There was no statistically significant difference regarding subfoveal SRF resolution and complete SRF resolution between the two study groups. Subfoveal fluid was resorbed in 67% and 69% and complete SRF resolution occurred in 61% and 55% of the eyes in group 1 and group 2, respectively (Table 2). VA increased significantly in both groups, with 7.6 ± 10.6 (p < 0.001) ETDRS letters in group 1 and with 7.8 ± 11.6 (p < 0.001) ETDRS letters in group 2. There was no significant difference in the final VA between both groups (p = 0.637) (Table 2).

At 4 months after the baseline visit, there was a significant difference in number of performed HD-PDT, as it was performed in 89% and in 45% of the eyes in group 1 and group 2, respectively (p < 0.001). Patients in group 2 had to wait significantly longer to receive HD-PDT after the baseline visit (34.3 ± 16.4 days vs. 61.0 ± 37.0 days, p = 0.008) and the time from HD-PDT until the last follow-up was shorter (165.5 ± 79.8 days vs. 120.0 ± 90.0 days, p = 0.019). No significant difference was observed in resolution of (subfoveal) SRF and in VA after HD-PDT between group 1 and group 2 (Table 2).

In the total cohort, 54 eyes were treated with HD-PDT and the mean follow-up after HD-PDT was 4.4 ± 2.9 months. Subfoveal SRF resolution occurred in 70% of the eyes, complete SRF resolution occurred in 65% of the eyes and an improvement of 9.0 ± 12.1 ETDRS letters (p < 0.001) was observed after HD-PDT.

A wait-and-see policy was implemented in 24 eyes at the baseline visit in the total cohort. However, during follow-up, 8 of these eyes were treated with HD-PDT (5 eyes), SML (1 eye), laser photocoagulation (1 eye), or prednisolone eye drops (1 eye). Of the 16 eyes that received no other treatment during follow-up (meaning, no treatment at all), subfoveal SRF resolved in 12 eyes and complete SRF resolution was observed in 9 eyes at the last follow-up visit (Fig. 4). Spontaneous resolution of SRF after a wait-and-see policy at the baseline visit occurred in 50% of the eyes subfoveally and in 37% of the eyes, complete resolution was observed. An improvement of 7.9 ± 7.7 ETDRS letters was observed in the eyes that received no treatment during follow-up (p = 0.004).

During availability of verteporfin, 11 eyes were initially treated with prednisolone eye drops. During follow-up, 4 eyes received additional treatment with HD-PDT. Subfoveal SRF resolution occurred in 4 of the 7 eyes that were only treated with prednisolone eye drops and complete SRF resolution occurred in 2 of these 7 eyes (Fig. 5). There was no difference in VA between baseline and follow-up in the patients that were only treated with prednisolone eye drops (−0.8 ± 9.9 ETDRS letters at follow-up, p = 0.893).

In 5 of the 12 eyes (42%) treated with prednisolone eye drops during follow-up, the intraocular pressure (IOP) increased with 5 mm Hg or more and in 2 patients (17%), the IOP raised above 25 mm Hg. These 2 patients needed additional IOP lowering treatment, and eventually prednisolone eye drops were tapered off. Although these patients showed a reduction of SRF, they finally underwent HD-PDT.

In patients with non-resolving CSC or cCSC, HD-PDT is the therapy of first choice [3, 5]. Due to a lack of verteporfin, which is required for PDT, the management of cCSC was challenged. In this study, we provided a comprehensive overview of how the shortage was managed. HD-PDT was performed less frequently during the first year of verteporfin shortage and patients’ treatment at first visit was adapted according to the presented flowchart (Fig. 2). At a mean follow-up of 6.2 months, no changes were observed in the VA and in (subfoveal) SRF resolution between patients referred before the shortage and during the shortage of verteporfin.

During the verteporfin shortage, all cCSC patients with persistent subfoveal SRF had to be discussed in a national indication committee, in order to be treated with HD-PDT [11, 12]. The period that subfoveal SRF had to be present to receive PDT, ranged from 4 to 8 months depending on the stock of verteporfin. As a result of the shortage, a shift in the distribution of treatments occurred. At the baseline visit, a wait-and-see policy was followed in most patients, other patients were being treated with focal laser photocoagulation, SML or prednisolone eye drops. This treatment regimen is, apart from the treatment with prednisolone eye drops, in line with the recently published evidence-based treatment guideline for CSC [5]. After a follow-up period of at least 4 months, HD-PDT was still performed significantly less, as only 45% of the eyes received HD-PDT during the verteporfin shortage compared with 89% of the eyes in the year before the verteporfin shortage. In addition, the time from the baseline visit until HD-PDT was longer during the shortage of verteporfin. The treatment strategy in cCSC and non-resolving CSC was thus significantly influenced by the shortage of verteporfin and less patients underwent HD-PDT.

The complete resolution rate at a mean follow-up of 4.4 months after HD-PDT in this study was 65%. In previous studies, a complete resolution after half-dose or half-fluence PDT occurred in 51–100% of cCSC patients [6, 7, 16‒18]. HD-PDT was shown to be superior to other treatments, one randomized controlled trial reported higher SRF resolution rates after HD-PDT treatment (67.2%) when compared to high-density SML (HSML) (28.8%) at 7–8 months after treatment [7]. In another trial, spontaneous complete resolution in cCSC patients occurred in 11% of the patients after 3 months and in 20% at 6 months [19]. HD-PDT is thus superior to a wait-and-see policy and HSML. However, during the shortage of verteporfin a reduction of number of HD-PDT treatments was enforced. Although the number of eyes treated with HD-PDT was significantly lower during the verteporfin shortage (89% vs. 45%, p < 0.001), we did not observe a significant difference in subfoveal and complete SRF resolution between the two groups. Complete SRF resolution occurred in 61% and in 55% of the patients referred before the verteporfin shortage and during the verteporfin shortage, respectively. Different explanations could be feasible for this nonsignificant difference in resolution. It could possibly be due to the higher number of spontaneous resolution in this study, which was 37% in the patient group with a wait-and-see policy at the baseline visit. This number is higher than the resolution rate of 20% after 6 months in the VICI trial [19]. This could be explained by differences in inclusion criteria and a selection of treatment in this study, as only patients without leakage on fluorescein angiography (Fig. 4), or an objectified decrease in SRF between referral and first visit at our institution, were allocated to a wait-and-see policy (Fig. 2). Moreover, the resolution rate of 65% after HD-PDT in this study is slightly lower than in prospective trials [6, 7], which can partly be explained by a difference between patients included in the prospective trials and real-world patients. For example, in prospective studies, only treatment-naïve patients were included. In this study, 85% of the eyes were treatment-naive and HD-PDT could possibly be more effective in treatment-naive eyes. Due to the lower success rate in the regular clinical practice, the difference in resolution between HD-PDT and non-PDT treated patients is reduced. Finally, it is also possible that other treatment strategies are more successful than previously described. Further research is needed to study why a reduction in performed HD-PDT does not directly lead to an alteration in SRF resolution. In future studies it can be investigated whether the SRF resolution rate after HD-PDT is different for a subset of patients with specific characteristics (e.g., patients with diffuse leakage, nontreatment naïve patients).

As it was not possible to treat patients with HD-PDT when criteria established by the national committee were not met, an alternative treatment with prednisolone eye drops was being explored. This was implemented with the goal of preventing more structural damage to the retina that can result from prolonged presence of SRF in patients unable to receive treatment [3, 4]. Previously, it was reported that a reduction of intraretinal fluid in patients with PPS occurred in all 17 eyes that were treated with topical prednisolone eye drops [9]. PPS and CSC are both diseases within the pachychoroid disease spectrum [2]. We subsequently studied treatment with prednisolone eye drops in 44 eyes with cCSC, and observed a significant reduction in central subfield thickness and foveal resolution in 22% of the eyes at 6 weeks after treatment initiation [10]. In the current study, 7 eyes were treated with monotherapy topical prednisolone and another 4 eyes were initially treated with topical prednisolone but were additionally treated with HD-PDT at a later point in time due to an incomplete response to prednisolone eye drops or an unacceptable increase in IOP. Subfoveal SRF resolution was objectified in 4 of the 7 eyes that were treated with topical prednisolone only (Fig. 5). Treatment with prednisolone eye drops seems to work well in a subset of cCSC cases, as shown in this study and in previously reported studies. Therefore, prednisolone eye drops may be useful in a subset of cCSC patients who do not respond well to HD-PDT; however, more studies are needed to further evaluate the efficacy and safety of treatment with prednisolone eye drops.

There are several limitations in this study such as the retrospective study design, the small number of eyes with monotherapy of prednisolone eye drops, micropulse laser or laser photocoagulation and the treatment allocation, in which treatments were performed according to expected efficacy, and not at random (Fig. 2). In addition, some patients were not included in the study because of a follow-up time that was shorter than 4 months; however, this number was similar between the two groups. The follow-up time was relatively short, and it remains to be seen if the non-resolving CSC and cCSC patients that were referred during verteporfin shortage will ultimately show a similar disease course as patients that were referred during full availability of verteporfin. More specifically, whether VA and number of recurrences will be comparable.

To our knowledge, this is the first study that presented a possible treatment strategy during verteporfin shortage and compared outcomes during the shortage and during availability of verteporfin. In conclusion, due to the shortage of verteporfin, HD-PDT was performed significantly less and at a later point of time for non-resolving CSC and cCSC. In this study, this did not lead to differences in both structural and functional outcomes as in terms of (subfoveal) SRF resolution and VA. Medicine shortages should be avoided, especially in cases where no alternative exists. A chosen treatment should be at a treating physicians’ discretion and should not be let by the unavailability of drugs. However, this study shows that although HD-PDT remains the mainstay treatment in cCSC, other treatment strategies, including a wait-and-see policy or prednisolone eye drops in case certain multimodal imaging characteristics are present, may also be successful.

This study was conducted in accordance with the Tenets of the Declaration of Helsinki. This study protocol was reviewed and approved by the Ethics Committee of the Radboud University Medical Center (MREC Oost-Nederland, 2022-15947). A written informed consent was not obtained for participation in this study. Due to the retrospective study design, the need to obtain written informed consent was waived by the Ethics Committee of the Radboud University Medical Center.

No competing interests exist for any author.

Femke M. van den Tillaart was supported by Stichting Ooglijders (Rotterdam, the Netherlands); Stichting A.F. Deutman Oogheelkunde research fonds (Nijmegen, the Netherlands); and UitZicht project (2021-15) (Delft, the Netherlands) that was financed by Algemene Nederlandse Vereniging ter Voorkoming van Blindheid; Stichting Blinden-Penning; Landelijke Stichting voor Blinden en Slechtzienden (LSBS); Stichting Oogfonds Nederland; Rotterdamse Stichting Blindenbelangen; Stichting Oogfonds Nederland; Stichting Retina Nederland Fonds. The funding organizations had no role in the design or conduct of this research.

Femke M. van den Tillaart, Franca Hartgers, Suzanne Yzer, and Carel B. Hoyng were involved in the design of the study. Femke M. van den Tillaart and Suzanne Yzer collected the data. Femke M. van den Tillaart analyzed the data and wrote the manuscript. Franca Hartgers, Suzanne Yzer, and Carel B. Hoyng reviewed the manuscript and approved the final version of the manuscript.

All data generated or analyzed in this study are included in this article. Further inquiries can be directed to the corresponding author.