Two Cases of Chronic Central Serous Chorioretinopathy Successfully Treated with Systemic Interferon Alpha

ABSTRACT

Chronic central serous chorioretinopathy (CSCR) is a sight threatening disease that can lead to legal blindness. Verteporfin photodynamic therapy is the main treatment for chronic CSCR, however, there has been a critical worldwide shortage of verteporfin. Other medical treatments have been attempted with variable efficacy. Interferons have shown efficacy in treating uveitis and associated macular edema. We report 2 cases of treatment refractory chronic CSCR successfully treated with subcutaneous injection of interferon alpha with significant anatomical and functional improvement. To our knowledge, this is the first report observing the therapeutic potential of systemic interferon alpha in the treatment of chronic CSCR. A large randomized controlled clinical trial would help to better evaluate the safety and efficacy of systemic PEG-IFNα2a in treating chronic CSCR, and further define the optimal dose, treatment interval and duration.

KEYWORDS:

• Chronic central serous chorioretinopathy (CSCR)
• diagnosis and treatment
• interferon alpha treatment
• retina
• uveitis

Central serous chorioretinopathy (CSCR) is the fourth most common non-surgical retinopathy causing vision loss.¹ The majority of patients are young men with an average age between 39–51 years, especially in Western European and Asian descents.^2,3 Common symptoms are blurred vision, hypermetropia, micropsia, metamorphopsia, scotoma, dyschromatopsia and reduced contrast sensitivity, leading to a decrease in the quality of life. CSCR is characterized by a neurosensory retinal detachment with accumulation of fluid in the subretinal space between photoreceptor segments and retinal pigment epithelium (RPE). The acute form which is usually resolved within 3–4 months is characterized by the presence of subretinal fluid (SRF), clinically detectable on fundus examination and on Optical coherence tomography (OCT), with focal or multifocal RPE alterations including small pigment epithelial detachments (PEDs), and leakage through the RPE on fluorescein angiography (FA).³ Despite spontaneous resolution of SRF, many patients can develop significant clinical sequelae, including atrophy of the RPE or retina, and subretinal neovascularization.¹ Chronic CSCR with persistent SRF lasting more than 4–6 months represents a chronic chorioretinopathy typically with a widespread track of RPE atrophy. It can be a sight threatening disease leading to legal blindness in over 10% of patients.⁴ The pathophysiology of CSCR remains poorly understood, however, a variety of risk factors have been reported: exogenous or endogenous corticosteroids,^5,6 type A personality,⁷ genetic predisposition,^8,9 cardiovascular disease and hypertension,¹⁰ and gastroesophageal reflux.¹¹ Different treatments have been proposed for chronic CSCR with variable efficacy. Verteporfin photodynamic therapy (PDT) is the main treatment for chronic CSCR, which requires specialized equipment and expertise.^1,2,12,13 In July 2021, there was a worldwide shortage of verteporfin, making PDT no longer available until the shortage resolved.¹⁴ Other interventions have been studied but the efficacy largely varies, including: thermal laser,¹⁵ micropulse diode laser,¹⁶ mineralocorticosteroid receptor antagonists,^17,18 anti-vascular endothelial growth factor (VEGF) intravitreal injections,¹⁹ low-dose oral methotrexate,²⁰ rifampicin,²¹ oral mifepristone,²² oral or topical carbonic anhydrase inhibitors (CAIs).^23–25 Interferons (IFNs) have shown efficacy in the treatment of uveitis associated with Behcet’s disease,^26–28 cystoid macular edema (CME) secondary to acute retinal necrosis,²⁹ post-infectious CME associated with presumed intraocular tuberculosis,^30,31 refractory non-infectious uveitic CME.^32–35 In the present study, we report 2 cases of treatment refractory chronic CSCR successfully treated with subcutaneous injection of interferon alpha (IFN α), including IFN α2b and subsequently the pegylated interferon alpha-2a (PEG-IFN α2a) after IFN α2b was no longer being produced.

Case description

Case 1

A 48-year-old Caucasian man with history of polyarthralgia was diagnosed with bilateral multifocal choroiditis and panuveitis by an outside provider. He was evaluated by an outside rheumatologist who did not identify any evidence of a systemic rheumatologic disease. His workup was normal for QuantiFERON Gold, syphilis, HLA-B27, Rheumatoid Factor, ANCA, anti-SSA/SSB, ESR, CRP, compliments C4/C3, complete blood count, complete metabolic panel, and computed tomography (CT) of the chest. He was started on high dose oral prednisone trial (80 mg daily) and noticed initial improvement in both vision and arthralgia. However, he subsequently developed bilateral central scotoma, worsened vision and subretinal fluid, and thus was quickly tapered off prednisone within 1 month. Mycophenolate 1500 mg twice daily was started concurrently with prednisone. He was referred to our institution for a second opinion.

Upon his first visit in our institution, his best corrected visual acuity (BCVA)was 20/30-2 OD, 20/40-2 OS. Slit lamp biomicroscopy revealed bilaterally normal anterior segment and occasional vitreous cell without haze. Fundoscopy showed multifocal retinal pigment epithelial detachments (PEDs) with subretinal fluid in macula and tracks of RPE atrophy extending from posterior pole inferiorly and nasally with shallow inferior subretinal fluid in both eyes (Figure 1A,B). Multifocal hypoautofluorescent mottled areas around optic nerve and macula with hyperautofluorescent borders, tracking inferonasally from optic nerve area, were seen in bilateral fundus autofluorescence images (Figure 1C,D). OCT demonstrated multifocal shallow subretinal fluid with multiple hyperreflective PEDs involving fovea, peripapillary region and vascular arcades in both eyes (Figure 2). Additionally, mild overlying intraretinal fluid (IRF) in macula and peripapillary region was observed in the left eye (Figure 2D–F). Fluorescein angiography showed a window defect around optic nerve and the areas of RPE atrophy, tracking inferiorly into periphery, with multifocal leakage along vascular arcades in both eyes (Figure 3A,B). Indocyanine Green Angiography (ICG) showed hypocyanescence in areas of PEDs seen on OCTs and hypercyanescence in peripapillary area in both eyes. No polyps or choroidal neovascular membranes was observed in either eye (Figure 3C–F). Overall presentation was more consistent with bilateral chronic multifocal chronic CSCR, possibly worsened under oral prednisone use, without evidence of active multifocal choroiditis. He was first trialed on acetazolamide 500 mg twice daily but could not tolerate it due to severe side effects. He was then switched to spironolactone 50 mg twice daily for 2 months without improvement. Meanwhile, he was noted to have more vitreous cell and worsened IRF in the left eye after he self-discontinued mycophenolate due to side effects (Figure 4A,B). Therefore, he was switched to weekly subcutaneous injection of 20 mg methotrexate together with 250 mg acetazolamide three times daily. Vitreous cell and IRF both improved in the left eye with trace remaining subretinal fluid in the peripapillary areas of both eyes (Figure 4C,D). The patient then self-discontinued acetazolamide because he did not perceive any vision improvement. IRF in the left eye continued to worsen despite addition of biweekly 40 mg Adalimumab to methotrexate (Figure 4E,F). He then received intravitreal anti-VEGF injections (initially bevacizumab and then monthly aflibercept) in the left eye from outside retina specialist with nearly resolution of IRF, and stable trace SRF bilaterally (Figure 4G,H). The patient was unable to get pre-authorization for monthly aflibercept injection, and it became burdensome given his young age and the frequency of injection needed. In addition, he was noted to have worsened vitreous cell and IRF when he self-discontinued mycophenolate, and vitreous cell improved significantly when he was on methotrexate monotherapy. We believed there was an inflammatory component in addition to chronic CSCR, and thus immunomodulatory therapy would be beneficial. He was then switched from methotrexate/adalimumab combined therapy to subcutaneous injection of IFN α2b (2 million units three times weekly) with resolution of bilateral SRF/IRF while off intravitreal aflibercept injections (Figure 4I,J). At this point, his vision has improved to 20/20 OD, 20/30 OS. Given side effects of flu-like symptoms, IFN α2b was gradually reduced to weekly dosing with stable images and vision. Later, weekly IFN α2b was replaced with weekly PEG-IFN α2a 180 mcg due to manufacture discontinuation of IFN α2b. Images remained stable while vision improved to 20/20 OU. Patient denied any side effects from PEG-IFN α2a, however, he subsequently self-discontinued the medication after 5 months of treatment due to excellent vision. He no longer maintains follow up with our uveitis clinic after successful treatment.

Figure 1. Multimodal imaging on initial presentation of Patient 1. A-B. Ultra wide-field fundus images showed multifocal retinal pigment epithelial detachments (PEDs) with subretinal fluid in macula and tracks of RPE atrophy extending from posterior pole inferiorly and nasally with shallow inferior subretinal fluid in both eyes. C-D. Fundus autofluorescence images revealed multifocal hypoautofluorescent mottled areas around optic nerve and macula with hyperautofluorescent borders, tracking inferonasally from optic nerve area in both eyes.

Figure 2. Macular OCT on initial presentation of Patient 1. A-C. OCT showed multifocal shallow subretinal fluid with multiple hyperreflective PEDs involving fovea and peripapillary region in the right eye. D-F. OCT showed multifocal shallow subretinal fluid with multiple hyperreflective PEDs involving fovea and peripapillary region with overlying intraretinal fluid in the left eye.

Figure 3. Fluorescein angiography and Indocyanine Green Angiography (ICG) on initial presentation of Patient 1. Fluorescein angiography revealed window defect around optic nerve and the areas of RPE atrophy, tracking inferiorly into periphery, with multifocal leakage along vascular arcades in the right eye (A) and the left eye (B). ICG showed hypocyanescence in area of PEDs seen on OCTs and hypercyanescence in peripapillary area. No polyps or choroidal neovascular membranes was observed in either eye: Early phase right eye (C) and left eye (D). Late phase right eye (E) and left eye (F).

Figure 4. Serial macular OCT obtained during the treatment course of Patient 1. The same cuts of representative OCT images in each eye from various time points were used for comparison. A-B. stable SRF in the right eye, worsened IRF OS after the patient self-discontinued mycophenolate. C-D. Combined therapy of methotrexate and acetazolamide led to significantly improved IRF OS and moderately improved SRF OU with residual trace SRF at peripapillary areas OU. E-F. Worsened IRF OS on combined therapy of methotrexate and adalimumab when the patient was off acetazolamide. G-H. Patient self-discontinued methotrexate but remained on adalimumab. Trace SRF OU remained stable, IRF OS mostly resolved with monthly intravitreal aflibercept injection. I-J. After switching to subcutaneous injection of IFN α2b, bilateral SRF/IRF fully resolved while remaining off intravitreal aflibercept injections.

Case 2

A 64-year-old Caucasian man with history of bilateral chronic CSCR diagnosed 15 years ago was referred to our institution. Upon his first visit with us, BCVA was 20/25-2 OD and 20/100 OS. Slit lamp biomicroscopy revealed bilateral intraocular lens replacement with normal anterior segment and vitreous without evidence of inflammation. Fundoscopy showed bilateral multifocal extrafoveal RPE changes, exudates, and nasal scarring (Figure 5A,B). Fundus autofluorescence images revealed stippled hypoautofluorescent areas with edges of hyperautofluorescence in macula, nasal to optic disc, and scattered similar lesions along arcades in both eyes (Figure 5C,D). In the right eye, OCT showed IRF with hard exudates in nasal macula and peripapillary region, multifocal disruption of ellipsoid zones and RPE irregularity (Figure 6A–C). In the left eye, IRF with hard exudates was mostly seen in nasal macula and peripapillary region. Subfoveal SRF with multifocal disruption of ellipsoid zones and RPE irregularity were also observed (Figure 6D–F). FA showed window defects in areas with RPE atrophy without vascular leakage, but ICG revealed multifocal hypercyanescent areas with a few hypocyanescent spots in both eyes (Figure 7). Overall, this was consistent with bilateral multifocal chronic CSCR without evidence of active intraocular inflammation. Prior to referral, he had failed multiple treatments, including thermal laser, series of intravitreal anti-VEGF injections, and oral acetazolamide. We first trialed him on 1 month of spironolactone 25 mg twice daily, but it failed to provide any improvement. He then underwent multiple sections of PDT in both eyes in the following year without substantial improvement. Later, choroidal neovascular membrane with polypoidal appearance was found in the left eye by OCT angiography (Figure 8). Patient started monthly intravitreal aflibercept injections in the left eye with a local retina provider, however, no benefit was found. Although we did not observe any inflammatory activity on ocular examination and multimodal imaging, failure to respond to multiple treatments of chronic CSCR and the development of IRF raised the concern of possible underlying multifocal choroiditis. A prior small study has used low dose oral methotrexate in patients without evidence of intraocular inflammation to successfully treat chronic CSCR.²⁰ Considering the anecdotal success of treating chronic CSCR with IFNα/PEG-IFN α2a in Case 1, we discussed a trial of immunomodulatory therapies with this patient, and he wished to proceed. After the patient tested negative for QuantiFERON Gold and FTA/RPR, various immunomodulatory therapies were trialed together with more sessions of PDT and intravitreal aflibercept injection, including 3 month of oral methotrexate 20 mg weekly, 3 months of mycophenolate 1000 mg twice daily, and 3 months of combined therapy of mycophenolate and adalimumab 40 mg biweekly. However, no consistent improvement of SRF/IRF was observed (Figures 9 and 10). The patient was instructed to stop all treatments including intravitreal injections, and then started PEG-IFN α2a 180 mcg weekly subcutaneous injection after 7 weeks of washout period for adalimumab. Significant improvements on SRF and IRF were observed in both eyes after 4 doses of weekly PEG-IFN α2a injection, along with vision improvement to 20/20 OD, 20/70 OS (Figures 9 and 10). Due to insurance issues, the patient had to stop PEG-IFN α2a for 2 months, which led to rapidly worsened SRF and IRF in both eyes. Eventually, he was able to restart weekly PEG-IFN α2a. After 1 month of PEG-IFN α2a monotherapy, bilateral multifocal SRF/IRF nearly resolved (Figures 9 and 10) and vision was stable at 20/20 OD, 20/70 OS. Due to moderate neutropenia and thrombocytopenia, the dose of PEG-IFN α2a was reduced to 90 mcg weekly and then 90 mcg biweekly. Patient was able to maintain stable vision and images while continuing reduced dose of PEG-IFN α2a. Thrombocytopenia resolved, and neutropenia continued to improve with dose reduction.

Figure 5. Multimodal imaging on initial presentation of Patient 2. A-B. Ultra wide-field fundus images showed multifocal extrafoveal RPE changes, exudates and nasal scarring in both eyes. C-D. Fundus autofluorescence images revealed stippled hypoautofluorescent areas with edges of hyperautofluorescence in macula and nasal to optic disc, scattered similar lesions along arcades in both eyes.

Figure 6. Macular OCT on initial presentation of Patient 2. A-C. in the right eye, OCT showed IRF with hard exudates in nasal macula and peripapillary region, multifocal disruption of ellipsoid zones and RPE irregularity. D-F. in the left eye, OCT showed IRF with hard exudates mostly in nasal macula and peripapillary region, SRF under fovea, multifocal disruption of ellipsoid zones and RPE irregularity.

Figure 7. Fluorescein angiography and Indocyanine Green Angiography (ICG) on initial presentation of Patient 2. FA showed window defects in areas with RPE atrophy without vascular leakage in the right eye (A, B) and the left eye (D, E). ICG revealed multifocal hypercyanescent areas with a few hypocyanescent spots in the right eye (C) and the left eye (F).

Figure 8. OCT angiography revealed the development of choroidal neovascular membrane (CNVM) in the left eye. A: Right eye: mild non-perfusion in the superficial vascular complex and deep capillary complex, no CNVM. B. Left eye: moderate non-perfusion in the deep capillary complex with CNVM.

Figure 9. Serial macular OCT of the right eye obtained during the treatment course of Patient 2. The same cuts of representative OCT images in the right eye from various time points were used for comparison. A-B. Persistent multifocal SRF and IRF despite multiple sessions of PDTs and various empiric treatments of CSCR, and immunomodulatory therapies, before starting PEG-IFN α2a monotherapy. C-D. Significant rapid improvement on SRF/IRF after 4 doses of weekly PEG-IFN α2a monotherapy. E-F. Rapid recurrence of SRF/IRF after the patient was off PEG-IFN α2a monotherapy. G-H. Near resolution of SRF/IRF after restarting PEG-IFN α2a monotherapy.

Figure 10. Serial macular OCT of the left eye obtained during the treatment course of Patient 2. The same cuts of representative OCT images in the left eye from various time points were used for comparison. A-B. Persistent multifocal SRF and IRF despite multiple sessions of PDTs and various empiric treatments of CSCR, and immunomodulatory therapies, before starting PEG-IFN α2a monotherapy. C-D. Significant rapid improvement on SRF/IRF after 4 doses of weekly PEG-IFN α2a monotherapy. E-F. Rapid recurrence of SRF/IRF after the patient was off PEG-IFN α2a monotherapy. G-H. Mild residual subfoveal SRF and near resolution of IRF/SRF after restarting PEG-IFN α2a monotherapy.

Discussion

Chronic CSCR with persistent SRF and significant RPE atrophy is a vision threatening disease which can lead to legal blindness in over 10% of patients.⁴ The pathophysiology of CSCR is uncertain and likely multifactorial, therefore, no effective targeted therapy has been developed. Verteporfin PDT is the mainstream treatment for chronic CSCR, however, the worldwide shortage of verteporfin since July 2021 has created an enormous impact on ophthalmic patient care because there is no equally effective treatment available. Different treatments have been attempted for chronic CSCR with variable efficacy. We have a total of 7 patients who have ever received IFNα therapy for ocular disease at our institution. Only 2 out of these 7 patients have components of chronic CSCR. To our knowledge, this is the first study reporting the successful treatment of refractory chronic CSCR with systemic IFNα therapy. IFN α2a or − 2b, a cytokine with antiviral, anti-angiogenic, and immunomodulatory effects, has shown efficacy in the treatment of several types of CME in both non-infectious and post-infectious uveitis. IFN α-2a and −2b are no longer commercially available, and now replaced by pegylated IFNα-2a. The term “pegylation” refers to the addition of nontoxic, nonimmunogenic and highly water-soluble polyethylene glycol molecules to interferon. Pegylation decreases the clearance and prolongs the half-life of interferon, which allows a longer duration of action and less frequent administration. A recent study has demonstrated the effectiveness of PEG-IFN α2a in uveitic macular edema previously responding to non-pegylated IFN α-2a and -2b.³⁶ Our first case of chronic CSCR had an inflammatory component evident by vitreous cell, which worsened significantly together with CME when the patient self-discontinued immunomodulatory therapies. The treatment course was complicated by medication non-adherence and inconsistent follow up, but IFNα was the only medication that appeared to demonstrate consistent benefit in controlling both vitreous cell and retinal edema. The second case did not appear to have any evidence of intraocular inflammation and had failed multiple empiric treatment of chronic CSCR, including multiple sessions of PDT. The visual potential was mostly limited by chronic central RPE atrophy and outer retinal damage from longstanding SRF. The significant improvement of SRF/IRF on systemic PEG-IFN α2a monotherapy therapy and subsequent rapid recurrence when off therapy strongly supported the efficacy of PEG-IFN α2a. It is unclear how systemic IFNα impacts the pathogenesis of chronic CSCR. Previously, oral methotrexate at a non-immunosuppressive low dose was proposed as an effective treatment for CSCR in a small retrospective case series of 9 patients,²⁰ although the mechanism was not understood since immunosuppression was not the goal and no patients exhibited signs of intraocular inflammation. A recent study has shown a significantly upregulated pro-inflammatory cytokine level and downregulated pro-angiogenic VEGF level in the plasma of patients with chronic CSCR compared to controls.³⁷ This provided new evidence for the possible role of plasma cytokines in the pathogenesis of CSCR and suggested the possible efficacy of systemic IFNα in CSCR treatment through modulating the cytokine pathway. In addition, an early in vitro study in 1995 revealed that IFN α2b enhanced the barrier function of bovine retinal microvascular endothelium and suggested the therapeutic potential by modifying microvascular leakage.³⁸ Some authors favor this hypothesis of action on the blood-retina barrier rather than its immunomodulatory effects given the rapid response to IFNα observed in the treatment of chronic CME in non-infectious uveitis^34,39 and refractory pseudophakic CME.⁴⁰ It is important to note that the tolerance to IFNα is variable and side effects could be severe. The most common side effect of treatment with PEG-IFNα2a included a mild flu-like illness following injections, which typically resolved within 3 days. Other minor side effects include malaise, fatigue, body aches, headaches, mild gastrointestinal disturbance, and mild injection site reaction.³² Severe side effects such as neutropenia, elevated liver function tests, severe depression, and epilepsy have been reported though the rate of treatment discontinuation for major side effects was highly variable.^35,41 We routinely obtain baseline complete blood count (CBC) and comprehensive metabolic panel (CMP) prior to initiation of IFNα therapy. We repeat these tests 1 month later and then every 2 months if they remain normal. If any abnormality or side effect is identified, we will modify the treatment by reducing the dose and treatment frequency. More frequent laboratory tests (every 2 weeks to every 1 month) are then required until the abnormality resolves. Notably, prior studies of IFNα therapy in chronic viral hepatitis have shown that 6.6% to 15% of patients develop clinical thyroid disease as a complication.^42,43 Hepatitis C virus and genetic factors are implicated in the pathophysiology, but recent data also suggest that IFNα may precipitate thyroiditis by both direct thyroid toxic effects and immune modulatory mechanisms.⁴² We do not routinely obtain thyroid function test in asymptomatic patients on IFN treatment. However, if patients develop any signs or symptoms of thyroid dysfunction, an endocrinological examination including thyroid stimulating hormones (TSH), free T4, and thyroid antibodies should be performed promptly. Another important side effect of IFNα therapy is the neuropsychiatric complications that have been well studied in patients with chronic viral hepatitis, including irritability, extreme emotional lability, severe depression, delirium, paranoia, and suicidal potential.^44,45 Failure to recognize these side effects and initiate interventions quickly could result in serious harm to patients. Because Type A personality is a risk factor for chronic CSCR, we provide counseling to our patients prior to initiation of IFNα therapy, and patients with baseline psychiatric disorders may not be good candidates for IFNα. Neither of our patients in this study experienced thyroid dysfunction or psychiatric side effects.

There is currently no consensus on the optimal dose or duration of IFNα therapy in ocular diseases, and the dosing should be titrated to clinical benefits, patient tolerance, and side effect profiles. Given the small number of patients in our study, we believe a large randomized controlled clinical trial would help to better evaluate the safety and efficacy of systemic PEG-IFNα2a in treating chronic CSCR, and further define the optimal dose, treatment interval and duration.

Authorship

All authors attest that they meet the current ICMJE criteria for authorship.

Patient consent

Consent to publish the case report was not obtained. This report does not contain any personal information that could result in patient identification.

Disclosure statement

PL consults for Bausch & Lomb and has consulted for Roche. EBS serves as a consultant to Roche/Genentech, EyePoint, Acelyrin, Gilead, Alumis and Kriya, and receives research funding from Roche/Genentech, EyePoint, Acelyrin, and Gilead. The remaining authors have no financial disclosures. PL has been supported by National Eye Institute Grant K08 EY022948, a Collins Medical Trust Grant, a Research to Prevent Blindness Career Development Award, and currently by the National Eye Institute grant R01 EY032882. EBS is supported by the Department of Veterans Affairs of the United States government.

Additional information

Funding

This work was supported by the National Institutes of Health (Bethesda, MD) P30 EY010572 core grant, the Malcolm M. Marquis, MD Endowed Fund for Innovation, and an unrestricted grant from Research to Prevent Blindness (New York, NY) to Casey Eye Institute, Oregon Health & Science University.