To evaluate the safety and tolerability and treatment efficacy of high-dose (2.0 mg) intravitreal ranibizumab for recalcitrant radiation retinopathy.
A phase I to II open-label, nonrandomized prospective clinical trial was performed on 10 eyes of 10 patients with recalcitrant radiation retinopathy who were failing standard dose anti–vascular endothelial growth factor (VEGF) therapy. External beam or plaque brachytherapy–associated retinopathy was characterized by persistent macular edema or leakage on optical coherence tomography or fluorescein angiography. Intravitreal 2.0 mg ranibizumab was given monthly up to 12 months and monitored for tolerability and change in best-corrected visual acuity (BCVA), central foveal thickness, and clinical signs of radiation retinopathy.
Seven patients completed the 1-year study and received all 12 injections; 3 withdrew from the study due to worsening retinopathy (1 after external beam, 2 following plaque). Treatment was well-tolerated with no severe adverse reactions. A total of 70% had stable (n = 3) or improved (n = 4) BCVA. Mean change in BCVA was +3.3 letters at 6 months and +0.7 letters at 1 year. Mean improvement in central foveal thickness (CFT) was −19.3% (range −57 to +15%) at 1 year. Initial mean CFT was 428 µm (range 192-776); final mean CFT was 333 µm (range 190-532). A total of 80% demonstrated a statistically significant (p<0.05) reduction in CFT.
Regardless of radiation source, intravitreal injections of 2.0 ranibizumab induced significant reductions in macular edema and maintained or improved BCVA in most patients who were failing standard dose anti-VEGF therapy.
Eur J Ophthalmol 2013; 23(6): 850 - 856
Article Type: ORIGINAL RESEARCH ARTICLE
AuthorsPaul T. Finger, Kimberly J. Chin
- • Accepted on 01/06/2013
- • Available online on 28/06/2013
- • Published in print on 15/10/2013
This article is available as full text PDF.
Radiation therapy is an effective tool in the fight against eye cancer. However, successful radiotherapy for uveal melanoma or orbital tumors can be marred by radiation-related vision loss. Radiation retinopathy is a progressive disease, first characterized by exudation of vascular components, followed by vascular occlusions, hypoxia, neovascularization, and finally scarification. This natural course results in vision loss.
Prior to the advent of anti–vascular endothelial growth factor (VEGF) therapy, all patients with radiation maculopathy were expected to lose vision. In 2007, we reported the first case series in which intravitreal anti-VEGF injections were found to successfully decrease vascular transudation associated with radiation maculopathy (RM) (1). Now there are many worldwide reports that radiation-induced macular retinal vessel transudation can be controlled with periodic intravitreal anti-VEGF medication (2-3-4-5-6-7-8-9-10).
However, over the last 6 years our clinical experience has uncovered a subset of patients who do not respond to standard anti-VEGF therapy. In these cases of recalcitrant disease, anti-VEGF therapy induces an initial improvement in retinal edema but the macula does not regain a normal contour, and/or the signs of RM (hemorrhages, cotton-wool spots, microangiopathy) worsen despite monthly dosing.
This investigator-sponsored trial sought to determine whether an increased dose (2.0 mg) of ranibizumab (Lucentis, Genentech, Inc., South San Francisco, California, USA) could benefit patients with recalcitrant retinopathy. In this study, we examined the safety and tolerability of high-dose ranibizumab as well as treatment effectivity based on outcomes of visual acuity and central foveal thickness (CFT) on optical coherence tomography (OCT).
MATERIALS AND METHODS
Federal Drug Administration (FDA), Investigational New Drug, and New York Eye Cancer Center Institutional Review Board approvals were prospectively obtained. This clinical trial was Health Insurance Portability and Accountability Act–compliant, and was registered with www.clinicaltrials.gov. After informed consent was obtained, 10 patients from a single clinical practice were enrolled in this phase I/II, open-label, prospective, clinical trial. As this drug is not commercially available, the drug was supplied at no charge to the patients by Genentech, Inc.
Study entry criteria
Patients eligible for this study were required to have 1) a clinical diagnosis of RM; 2) prior treatment with intravitreal anti-VEGF therapy with incomplete response; 3) been irradiated more than 6 months and no more than 10 years prior to enrollment; 4) an initial visual acuity of 20/400 or better in the study eye; and 5) age >21 years. Exclusion criteria included pregnancy, aphakia, uncontrolled glaucoma in the study eye, or history of glaucoma filtering surgery.
Radiation maculopathy was defined by new-onset intraretinal hemorrhage, intraretinal microangiopathy, neovascularization, cotton-wool spots, vascular sheathing, or macular edema. Incomplete response was defined as persistent or worsening leakage on OCT or fluorescein angiography and/or persistent clinical signs of retinopathy (exudate, hemorrhage, cotton-wool spots, microangiopathy) despite monthly anti-VEGF therapy.
Intravitreal injections of 2.0 mg ranibizumab (recombinant humanized anti-VEGF monoclonal antibody fragment, rhuFab V2 [Genentech, Inc., South San Francisco, California, USA]) were given. This drug provides 4 times the strength of standard ranibizumab (0.5 mg).
In this series, eyes were prepared with topical proparacaine, then betadine. Then a subconjunctival injection of xylocaine for focal anesthesia was performed. The injecting surgeon wore a surgical mask over his nose and mouth. While the anesthetic took effect, ranibizumab was drawn from the bottle into a 1-cc syringe; then, a 30-gauge needle was placed for injection. After a second application of antibiotic drops, an eyelid speculum was placed, followed by transscleral injection through the pars plana. Optic nerve perfusion was checked by indirect ophthalmoscopy. After 30 minutes (±10 minutes), intraocular pressures were checked by Goldmann tonometry. Patients were prescribed antibiotic steroid drops to be taken 4 times a day for 7 days. Patients returned for a safety visit to check for any adverse side effects 7 days (±3 days) after the first injection.
Treatments were continued if clinical examination revealed improved or persistent RM and OCT showed improved or stable macular and retinal edema. Treatments were discontinued and patients withdrawn from the study per the principal investigator’s discretion (P.T.F.) if it was determined that the retinopathy was worsening despite monthly dosing and the patient could seek alternative treatments.
Main outcome measures
Outcome measures recorded at baseline and then monthly were best-corrected visual acuity (BCVA) using Early Treatment Diabetic Retinopathy Study charts in Collaborative Ocular Melanoma Study–certified examination rooms. Fundus photography and CFT on spectral-domain OCT were performed at each visit. Fluorescein angiography and ultrasound imaging (for uveal melanoma patients) were examined at baseline and at months 3, 6, 9, and 12. Blood pressure and any adverse events were recorded at each visit.
As per FDA protocol, subjects were assigned to one of two groups in an alternating manner. Group 1 (n = 5) subjects received injections every 30 days (±7 days) through 12 months (maximum 12) and group 2 (n = 5) subjects received injections every 30 days (±7 days) for the first 4 months and every month thereafter until month 12 (maximum of 12 injections) if the subjects had signs of intraretinal hemorrhage or macular edema as determined by ophthalmoscopic examination, fluorescein angiography, color fundus photography, and/or OCT when compared to their last visit.
We calculated the rates of change of CFT between initial treatment month and follow-up treatment months for every patient. The nonparametric Wilcoxon signed rank test was then used to test if the patient’s CFT significantly changed with time. In the test, the median change of foveal thickness was assumed to be zero (no change). Actual median change of foveal thickness was considered to be significantly different from hypothetical zero median changes if p value was smaller than 0.05. We also used the Kaplan-Meier method to estimate the cumulative change rates of foveal thickness over time. To better present the status of change, we defined 3 outcomes of foveal thickness decreasing at least 10%, 15%, and 20% during the follow-up time, and then drew the corresponding curves.
There were 7 women and 3 men, with a mean age of 55 years (median 54 years, range 31-87). Three patients had hypertension, and 1 had non-insulin-dependent diabetes. Eight patients had received radiation therapy for posterior uveal melanoma, 1 for lacrimal gland adenoid cystic carcinoma (ACC), and 1 for adenocarcinoma of the skull base. For the patients with melanoma, the mean apical tumor height and mean largest tumor basal diameter was 5.2 mm (range 2-11.3) and 11.6 mm (range 7-17), respectively. All patients had a history of anti-VEGF therapy (bevacizumab 1.25 mg and/or ranibizumab 0.5 mg), and were receiving active therapy at the time of study enrollment.
The mean time from radiation to development of radiation retinopathy and initiation of anti-VEGF therapy was 34.2 months (range 3-69) and the mean time from radiation retinopathy to study enrollment was 29.4 months (11-50) (
TUMOR CHARACTERISTICS AND RADIATION DATA FOR STUDY PATIENTS
|Patient number||Radiation type||Tumor location||Initial tumor thickness, mm||Initial tumor LBD, mm||Dose to fovea, Gy||Dose to apex, Gy||Time from radiation to RR, mo||Time from RR to study enrollment, mo|
|EA = equator anterior; IMRT = intensity modulated radiation therapy; NA = not applicable; P = tumor centered posterior; Pd-103 = palladium-103; PE = tumor centered at equator anterior location; RR = radiation retinopathy; LBD = largest basal dimension.|
|1||Proton beam||Lacrimal gland||NA||NA||NA||NA||22||43|
All uveal melanoma patients underwent palladium-103 brachytherapy over 5-7 days. Mean dose to tumor apex was 80 Gy (range 58.8-89.4) and mean dose to fovea was 81 Gy (30.6-137.6) (
The lacrimal gland ACC patient (patient 1) had undergone proton beam radiation therapy (72 CGE/36 fractions), and the skull base adenocarcinoma patient (patient 3) had undergone intensity modulated radiation therapy (IMRT) 6 MV photons at 200 cGy for 25 fractions (total dose 50 Gy).
Seven patients completed the 1-year study and received all 12 injections (
Patient 2 with recalcitrant radiation retinopathy following plaque brachytherapy for uveal melanoma. (Top left) Despite 12 months of anti–vascular endothelial growth factor therapy, macular edema persists with a central foveal thickness (CFT) = 436 µm on optical coherence tomography (OCT). (Top right) Pre-high-dose fluorescein angiogram in arteriovenous phase demonstrates persistent macular and retinal edema. (Bottom left) After 12 monthly injections of 2.0 mg ranibizumab, the macula resumes a normal contour on OCT, with a CFT = 228 µm. (Bottom right) After 12 monthly injections of 2.0 mg ranibizumab, fluorescein angiogram in arteriovenous phase demonstrates a marked reduction in macular and retinal edema.
Three patients were discontinued from the study per principal investigator (P.T.F.) discretion due to worsening retinopathy despite monthly therapy. They elected to receive alternate treatment (one each at months 7, 8, and 9). Of the 3 patients who discontinued the study, 2 were from group 1 and 1 was from group 2. This was not found to be significant in that the groups were treated with the same number of injections (monthly) despite the creation of 2 study groups (arms) for the FDA protocol. One patient had undergone IMRT (skull base adenocarcinoma) and 2 had undergone plaque brachytherapy. Including those withdrawn from the study, the mean follow-up period was 9 months.
Primary outcome: safety and tolerability
Ranibizumab 2.0 mg was well-tolerated for up to 1 year of monthly treatment. There were no cases of ocular or systemic allergy, endophthalmitis, elevated blood pressure, vascular ischemic events, or other severe adverse event as noted at each visit. There were no cases of death or metastasis during the study.
Secondary outcome: treatment effect at 1 year
Initial median BCVA was 20/45 (range 20/20-20/200) and final median BCVA was 20/60 (range 20/20-20/125). At the conclusion of the study, 70% were found to have stable (n = 3) or improved (n = 4) BCVA. The mean change in BCVA as compared to baseline was +3.3 letters at 6 months and +0.7 letters (range −9 to +8) at 1 year. This represents essentially equivalent BCVA compared to baseline. Three patients lost letters of acuity at 1 year.
When we examined the 12-month OCT findings for each patient prior to high-dose study intervention, 50% of patients had worsening edema and the other 50% had stable but persistent edema (
Change in central foveal thickness (CFT) pre- and post-high-dose 2.0-mg ranibizumab. Time “0 months” represents the baseline CFT prior to study intervention.
Overall, 8 of 10 patients showed a decreased change of CFT at every time point during the study period. Compared with baseline CFT, the median decreased CFT in each of these 8 patients was statistically significant (p<0.05). However, the remaining 2 (patients 6 and 8) showed a slight increased trend in CFT, and the median increased change in one patient was statistically significant (p<0.05).
Cumulative Kaplan-Meier change rates of central foveal thickness (CFT). This graph demonstrates the cumulative change rate if the change is defined as a decrease in CFT of at least 10%, 15%, or 20%.
Clinical assessments of retinopathy (exudates, retinal hemorrhage, microaneurysm) using comparative color photography and fluorescein angiography were made compared to baseline. As a general finding, this subset of patients with previously treated retinopathy were primarily noted to have retinal edema and microangiopathy rather than hemorrhage. Specific comparative results revealed that compared to baseline, of the 6 patients with baseline exudate, 1/6 improved, 3/6 were stable, and 2/6 worsened. For retinal hemorrhage (baseline n = 4), 3/4 improved and 1/4 worsened. For microangiopathy (baseline n = 10), 4/10 improved, 4/10 were stable, and 2/10 worsened. Clinical assessments of macular edema using color photography and fluorescein angiography were also made, and 7/10 improved, 1 was stable, and 2 worsened.
The results of the present study suggest that a higher dose (2.0 mg) of ranibizumab can benefit patients with recalcitrant radiation retinopathy who are failing standard anti-VEGF therapy. Regardless of the radiation source (proton beam, IMRT, or plaque brachytherapy), most patients experienced improvements in macular edema as best illustrated on comparative OCT. Even though these patients were failing standard dosing, the general trend was an initial improvement in CFT after the first injection (90%), followed by less dramatic improvement, and then a plateau in CFT. At the conclusion of the study, most (80%) were found to have a statistically significant improvement in CFT as compared to baseline. However, the CFT improvement on OCT did not correlate to significant (less than one line) improvements in visual acuity.
Recent studies have also demonstrated that high-dose ranibizumab has beneficial potential (11, 12). Fung et al (11) (pilot study) studied 9 eyes with recalcitrant age-related macular degeneration (AMD) over 6 months and noted an improvement in BCVA and CFT using 2.0 mg ranibizumab as compared to 0.5 mg. The multicenter Super-dose Anti-VEGF (SAVE) trial studied 2.0 mg ranibizumab for 87 patients with recalcitrant AMD (those failing standard 1.25 mg bevacizumab or 0.5 mg ranibizumab monthly dosing for at least 9 months) (12). They found that the high dose led to statistically significant improvements in visual acuity and macular anatomy within the first 3 months of treatment. However, similar to the present study, their longer-term results failed to demonstrate significant improvement in visual acuity (in press 2013). Therefore, the higher dose of 2.0 mg ranibizumab is not expected to become commercially available.
It is important to note that the patients enrolled in our study were selected with advanced, recalcitrant disease and thus poor prognosis for vision. These cases included those with large tumors, a relatively high macular radiation dose, and/or a subfoveal or juxtapapillary tumor location. Of these 3 factors, the only controllable factor is radiation dose. At The New York Eye Cancer Center, we work closely with our radiation oncologists to employ the lowest possible curative dose and dose rates. We also counsel patients that the role of anti-VEGF therapy is not to reverse the damage created by radiation; rather its role is to suppress a progressive disease and thus preserve vision for as long as possible. From this perspective, this study showed that high-dose ranibizumab was effective in being able to maintain or improve visual acuity in 70% of patients for a mean of 9 months.
As high-dose (2.0 mg) ranibizumab was not available at the conclusion of the study, these patients had elected to continue treatment with high-dose bevacizumab. More recently, off-label high-dose bevacizumab, including 2.0 mg and 2.5 mg bevacizumab, has become available through the use of larger dose volumes (0.08 mL and 0.1 mL, respectively). However, the increased drug volume can cause transient increased intraocular pressures with resultant temporary decreased vision.
Any study of intravitreal anti-VEGF therapy must consider the well-documented intraocular risks, including but not limited to endophthalmitis, uveitis, and vitreous hemorrhage. In addition, there has been concern in AMD studies over the broad inhibition of VEGF-A in which frequent dosing may lead to geographic atrophy and poor visual outcome (13); however, this must be weighed against the natural course of radiation retinopathy.
Weaknesses of this study include that it is a small case series with no control group and thus no statistically significant comparison. Though a nonhomogenous group including both melanoma and nonmelanoma etiologies, all study patients had secondary unilateral radiation retinopathy.
Reviewers have pointed out that without therapeutic intervention, radiation retinopathy can run a course that leads to reabsorption of retinal exudates and hemorrhages and decreased macular thickness over time. However, untreated long-term radiation retinopathy almost invariably leads to permanent visual field defects and loss of useful vision.
Herein, we have shown that 2.0 mg monthly ranibizumab was tolerated for 1 year. Decreased CFT and stable vision was noted in most cases. These agents may offer select patients with recalcitrant retinopathy the potential to prolong useful vision when standard therapy fails. However, the safety, longer-term tolerability, and effectivity of high doses of anti-VEGF agents requires further investigation.
Study drug and funding for statistical analysis was supplied by Genentech, Inc., South San Francisco, California, USA.