Ocular histoplasmosis syndrome (OHS) is a systemic disease thought to be caused by the inhalation and systemic spread of the dimorphic fungal organism Histoplasma capsulatum. A mild systemic infection occurs after the initial infection and results in granulomatous inflammation and resultant encapsulation primarily in the lungs and the choroid. The classic ocular findings for OHS include peripheral chorioretinal scarring (“histo spots”), peripapillary atrophy and hyperplasia, and maculopathy (or choroidal neovascularization [CNV]). The development of CNV is the primary ocular complication leading to vision loss.1
OHS most commonly affects young adults between the ages of 20 years and 50 years.2 In the U.S., more than 2,000 young adults suffer significant visual impairment every year.3 If left untreated within the foveal avascular zone, as many as two-thirds may suffer vision loss below 20/200.4
The development of CNV in OHS is very similar to that of age-related macular degeneration (AMD). Treatments include off-label use of agents also administered for AMD: anti-vascular endothelial growth factor (VEGF) therapies such as ranibizumab (Lucentis; Genentech, South San Francisco, CA), bevacizumab (Avastin; Genentech, South San Francisco, CA), aflibercept (Eylea; Regeneron, Tarrytown, NY), intravitreal corticosteroids, macular surgery, and on-label photodynamic therapy (PDT) with verteporfin (Visudyne; Novartis, Basel, Switzerland).1 The purpose of this pilot study was to explore the safety and efficacy of two treatment regimens for the administration of ranibizumab for the treatment of CNV due to OHS.
Patients and Methods
This was an open-label, phase 1/2 study of intravitreally administered ranibizumab in subjects aged 18 years and older with active CNV secondary to OHS and best-corrected visual acuity (BCVA) at baseline (Visit 1) of 20/25 to 20/400 (80 to 20 ETDRS letters, with subjects 2 meters from the chart) (Figure 1). The study was approved by an institutional review board, conducted in accordance with Good Clinical Practices, and registered on ClinicalTrials.gov (NCT 00955630). All subjects provided informed consent before enrollment in the study and were examined and treated by one of four retinal specialists. The diagnosis of CNV secondary to OHS was made using patient history, ophthalmoscopic examinations, and spectral-domain optical coherence tomography (OCT) (Cirrus HD-OCT; Zeiss, Oberkochen, Germany), as well as angiographic evaluations. Subjects were excluded from the study if they met any of the following criteria: Pregnancy (positive pregnancy test) or lactation, premenopausal women not using adequate contraception, prior enrollment in any study, any other condition that the investigator believed would pose a significant hazard to the subject if the investigational therapy were initiated, participation in another simultaneous medical investigation or trial, history of allergy to sodium fluorescein, disciform scarring that would prohibit effective treatment of the lesion, or prior treatment with subfoveal thermal laser. Other prior treatments (juxtafoveal or extrafoveal thermal laser, PDT, or bevacizumab) were permitted if the lesion had a symptomatic recurrence and there was no evidence of subfoveal disciform scarring or subfoveal atrophy from thermal laser. In addition, no prior treatments were permitted within 30 days of baseline.
Study design. Subjects with active choroidal neovascularization (CNV) secondary to ocular histoplasmosis syndrome (OHS) were screened and enrolled on day 1 and received an initial treatment with intravitreal ranibizumab (0.5 mg). Group A was retreated at month 1 and month 2 and then could receive pro re nata (PRN) treatment per specific criteria. Group B received PRN treatment from month 1 onward. At monthly visits, adverse events, best-corrected visual acuity (BCVA), central subfield thickness, and ophthalmic examinations were assessed.
Subjects were randomized to one of two groups (Figure 1). Group A received three monthly injections of 0.5 mg ranibizumab through the pars plana followed by as-needed, or pro re nata (PRN), dosing based on criteria described below. Group B received an initial injection of 0.5 mg ranibizumab through the pars plana on day 1 and were treated as needed throughout the study.
Retreatment criteria were based on changes in visual acuity (VA) and/or anatomic measures including: loss of two or more lines (10 ETDRS letters) in BCVA from prior visit; any increase, persistence, or presence of subretinal fluid on OCT; persistent and/or new leakage on fluorescein angiography (FA); or new or persistent subretinal hemorrhage felt to be related to active CNV.
If a subject did not respond to ranibizumab (as evidenced by VA loss of 15 letters or more, persistent fluid on OCT exceeding the central subfield retinal thickness [CST] at baseline by 100 μm on three consecutive visits, or increasing subretinal hemorrhage), treatment with PDT, and/or intravitreal triamcinolone acetate (IVTA), either tissue plasminogen activator and intravitreal gas to displace the subretinal hemorrhage or submacular surgery was administered.
BCVA was assessed with ETDRS refraction at monthly intervals along with dilated funduscopic examination and OCT testing. FA and fundus photographs (FF450 fundus camera; Carl Zeiss Meditec, Jena, Germany) were obtained at baseline and months 3, 6, 9, and 12. The images (OCT and FA) were reviewed by the authors who were masked to the treatment regimens. As previously treated patients were included in the protocol, an attempt to classify angiography patterns as to classic versus occult was not made.
The primary outcome measures were the incidence of adverse events (AEs) and serious adverse events (SAEs) through month 12. Key secondary outcome measures included mean changes in BCVA and CST and the total number of injections required from day 1 through month 12. No formal statistical testing was planned for this phase 1/2 study.
A total of 21 eyes from 21 subjects qualified for the study. All subjects were included for the safety analyses and two were excluded from the efficacy analyses due to discontinuation from the study (Subject 21, Group B was lost to follow-up after the first injection and Subject 8, Group A, withdrew after three injections at month 7). Of the 19 eyes included in the efficacy analyses, nine had treatment prior to study enrollment, including bevacizumab (n = 7), PDT with or without IVTA (n = 3), and thermal laser (n = 2). Three of the nine patients received more than one type of prior therapy.
The safety population comprised white subjects with a mean age of 50.0 years (range: 19 years to 79 years) and was 62% female, 38% male. Nine subjects had received treatment prior to baseline (intravitreal bevacizumab, seven subjects; PDT, three subjects; IVTA, two subjects; laser, one subject). The right eye was the study eye for 60% of subjects. The location of the CNV was subfoveal in 57%, juxtafoveal in 29%, and extrafoveal in 5% of cases. Two subjects had mixed lesion locations, one case each of juxtafoveal/subfoveal and juxtafoveal/peripapillary. Demographic and baseline characteristics were reasonably well-balanced between the groups.
At the baseline visit, the mean BCVA for the two treatment groups were similar: Group A, 71 ETDRS letters ± 8.2 ETDRS letters, and Group B, 66.3 ETDRS letters ± 8.4 ETDRS letters. The mean CSTs for Group A (358.6 μm ± 65.4 μm) and Group B (349.1 μm ± 111.4 μm) were equivalent, although the range for Group A was greater that of Group B (242 μm to 557 μm versus 247 μm to 442 μm).
Visual Acuity Results
The mean BCVA improved by approximately two ETDRS lines in both groups beginning with the second visit and was maintained through month 12. (Figure 2A). The mean change from baseline BCVA was similar for subjects treated with three initial doses or one dose followed by PRN treatment. (Figure 2A). Group A appeared to recover vision slightly more rapidly than Group B at month 1; however, there were no statistically significant differences between the groups based on an unpaired t-test.
Visual acuity and retinal thickness by visit. Best-corrected visual acuity (BCVA) at 2 meters was evaluated, central subfield thickness (CST) was determined using optical coherence tomography, and dilated ophthalmoscopic exams were performed at each visit. Mean BCVA (Panel A) and mean change in BCVA improved to a similar extent in Groups A and B. Mean CST was reduced slightly more rapidly in Group A than in Group B at month 1, but the difference was not statistically significant (t-test). ETDRS = Early Treatment of Diabetic Retinopathy Study; PRN = pro re nata; SEM = standard error of the mean.
At month 12, the vision of each subject had improved by one or more lines and 63% of the subjects in the study had VA of 20/20 or better in the study eye. Prior treatment did not limit VA improvement as four of five subjects in Group A and three of four subjects in Group B who had received therapy for CNV prior to the study were able to read 80 letters or more at the last visit.
The improvement in VA was paralleled by anatomical improvements (Figure 2B). The mean CST was also reduced for both treatment regimens. The mean CST for Group A was slightly lower at months 1 through 3; however, these differences missed statistical significance (P = .1580, P = .0797, and P = .1449 for months 1, 2, and 3, respectively.) For the remainder of the study, the mean CSTs were essentially the same.
Improvements in retinal structure on OCT and reduced fluorescein leakage were observed in nearly all subjects in Group A and Group B. Figure 3 provides OCT images (Figure 3A) and fluorescein angiograms (Figure 3B) for a typical patient in Group A (three injections followed by PRN treatment). The subfoveal cyst observed in this patient's OCT at baseline resolved by month 3, and mean CST was normalized throughout the remainder of the study. Late fluorescein leakage was essentially eliminated by month 3 (the first post-baseline angiogram obtained).
Example case images for Group A. Reduction in central subfield thickness and subfoveal cyst was observed by month 1, and elimination of late-phase fluorescein leakage was observed at month 3 in this typical patient from Group A (three injections followed by pro re nata dosing). The retinal structure improved as the subfoveal cyst observed on optical coherence tomography resolved by month 1 (A). By month 3 (first post-baseline fluorescein angiogram), the late subfoveal leakage has resolved (B).
Figure 4 provides OCT images (Figure 4A) and fluorescein angiograms (Figure 4B) for a typical patient in Group B (one injection followed by PRN treatment). As in Group A, resolution of the subfoveal cyst was apparent at month 1; however, subfoveal architecture continued to gradually improve through month 6 as evidenced by the image and the mean CST. Late subfoveal fluorescein leakage was absent at month 3 and was reduced for the remainder of the study.
Example case images for Group B. Improvement from baseline in central subfield thickness (CST) and fluorescein leakage was also observed in nearly all subjects in Group B at the 1-month visit. The reduction in CST (A) and leakage (B) was generally maintained through the visits at months 3, 6, and 12. (A) Optical coherence tomography images for Patient 6 (three injections followed by pro re nata dosing). FA = fluorescein angiography.
The mean total number of injections from day 1 through month 12 was the same for the two treatment regimens (Group A: 5.7 ± 3.1, Group B: 5.8 ± 3.8). At month 1, eight of 10 subjects in Group B required retreatment, indicating that a single injection was not sufficient treatment in this population. During months 7 through 12, fewer PRN injections were required in patients treated initially with three monthly injections followed by PRN therapy (Group A: 13 injections for nine subjects) than for patients treated with one injection followed by PRN therapy (Group B: 21 injections total for 10 subjects).
Ranibizumab was safe and well-tolerated in these subjects with CNV due to OHS. There were no AEs related to drug or injection procedure. The only AEs reported during the study were nausea and body rash thought to be due to shampoo in one subject each. No systemic or ocular SAEs occurred during the study.
This study demonstrates that ranibizumab is effective and safe in the treatment of CNV due to OHS. No patients lost VA during the course of the 12-month study, and the majority returned to VA of 20/20. In addition, there was a robust anatomical improvement as measured by CST measured by OCT testing.
There was no significant difference observed between the treatment regimens with PRN treatment after the first dose or after the third dose. This differs from other similar etiologies of CNV (eg, AMD) in that these subjects can be managed with a PRN approach and monthly monitoring. The findings of this study are in agreement with the study by Heier et al.,5 in which they found that ranibizumab was effective for OHS when administered PRN following either a single injection or three injections.5
Anti-VEGF therapy has become standard therapy for CNV secondary to OHS. The results of this study further support this conclusion and provide additional experience in this patient population.
The limitations of this study are the small size of the study and the use of previously treated subjects. Nonetheless, the efficacy of ranibizumab was consistent in all subjects, and no significant safety concerns were identified.
In conclusion, intravitreal anti-VEGF therapy with ranibizumab is safe and achieves outstanding VA and anatomical results in patients with CNV secondary to OHS with fewer treatments than other conditions (eg, exudative AMD) and can be used effectively on a PRN basis without any treatment-related adverse events.
- Diaz RI, Sigler EJ, Raifieetary MR, Calzada JI. Ocular histoplasmosis syndrome. Surv Ophthalmol. 2015;60(4):279–295. doi:10.1016/j.survophthal.2015.02.005 [CrossRef]
- Presumed ocular histoplasmosis syndrome. In: Agarwal A, ed. Gass' Atlas of Macular Diseases, Vol. 1. 5th ed. Edinburgh, Scottland: Elsevier Saunders; 2012:158–176.
- Olk RJ, Burgess DB, McCormick PA. Subfoveal and juxtafoveal subretinal neovascularization in the presumed ocular histplasmosis syndrome. Visual prognosis. Ophthalmology. 1984;91(12):1592–1602. doi:10.1016/S0161-6420(84)34113-6 [CrossRef]
- Gutman FA. The natural course of active choroidal lesions in the presumed ocular histoplasmosis syndrome. Trans Am Ophthalmol Soc. 1979;77:515–541.
- Heier JS, Brown D, Ciulla T, et al. Ranibizumab for choroidal neovascularization secondary to causes other than age-related macular degeneration: A phase I clinical trial. Ophthalmology. 2011;118(1):111–118. doi:10.1016/j.ophtha.2010.04.016 [CrossRef]