Graefes Arch Clin Exp Ophthalmol DOI 10.1007/s00417-013-2553-7
RETINAL DISORDERS
Clinical outcomes after switching treatment from intravitreal ranibizumab to aflibercept in neovascular age-related macular degeneration Florian M. Heussen & Qing Shao & Yanling Ouyang & Antonia M. Joussen & Bert Müller
Received: 26 August 2013 / Revised: 3 December 2013 / Accepted: 9 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To describe the treatment response to aflibercept in patients with exudative age-related macular degeneration that showed insufficient or diminishing treatment effects under ranibizumab. Methods From December 2012 till June 2013 all patients receiving intravitreal injections of aflibercept after previous treatment with ranibizumab were collected in a database and retrospectively reviewed. Clinical data such as visual acuity or central subfield retinal thickness on optical coherence tomography (OCT) scans were analyzed for the time frame before, during, and shortly after the aflibercept injections. Of particular interest was the comparison of clinical features under ongoing ranibizumab treatment to the time during aflibercept treatment. Results Seventy-one eyes of 65 patients were included in the study. All eyes had previous ranibizumab injections in their medical history, the average number of which was nine (range 3-43). For the total group the mean visual acuity (VA) before the first ranibizumab injection was 0.54 logMAR, and after the last ranibizumab injection was 0.57 logMAR. Mean VA changed from 0.47 logMAR before the first aflibercept injection to 0.25 logMAR after the last aflibercept injection. Central subfield retinal thickness (CSRT) on OCT changed from a mean of 417.28 μm to 349.52 μm under ranibizumab treatment and from 338.76 μm to 272.00 μm under aflibercept treatment. Interestingly, 33 % of cases that did not show a F. M. Heussen (*) : Q. Shao : Y. Ouyang : A. M. Joussen : B. Müller Department of Ophthalmology, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany e-mail: [email protected] Q. Shao Department of Ophthalmology, Shanghai First People’s Hospital, Shanghai, China
functional improvement under ranibizumab therapy gained visual acuity after aflibercept treatment. Conclusion Aflibercept appears to be an effective choice for patients with neovascular age-related macular degeneration who were resistant to previous therapy of ranibizumab. The longevity of this effect still remains questionable. Keywords Age-related macular degeneration . Anti-VEGF treatment . Ranibizumab . Aflibercept . Tachyphylaxis Abbreviations OCT Optical coherence tomography SD-OCT Spectral-domain OCT FA Fluorescein angiography VA Visual acuity AMD Age-related macular degeneration CSRT Central subfield retinal thickness
Introduction The advent of ranibizumab (Lucentis®) in the treatment of exudative age-related macular degeneration is widely regarded as a major breakthrough in our field. For the first time, patients could realistically expect to regain vision under treatment where previous approaches would cause a slowdown in visual deterioration at best [1–6]. However, even with this seemingly fail-safe treatment, there are cases that simply do not respond well functionally or morphologically. Taking it one step further, in a few cases there seems to be no effect at all [7]. The overall number of treatment-resistant cases may be small in view of all treated eyes; however, it is this particular group of patients which causes the most concern amongst clinicians. While some of these treatment failures may be due to misdiagnosis and therefore mistreatment, there have
Graefes Arch Clin Exp Ophthalmol
been reports about a lacking response to ranibizumab in exudative AMD. Some authors suspected a tachyphylaxislike effect of the intravitreal anti-VEGF injections — a claim that was supported by an increase in effectiveness when the drug in use was switched from ranibizumab to bevacizumab and vice versa [8]. Recently, aflibercept (Eylea®), a newly approved drug for intravitreal delivery in exudative AMD, has entered the market [9]. While details of the pharmacokinetics and drug mechanism may very well be substantially different from ranibizumab, from a clinician’s point of view, both drugs fall into the same category of VEGF-binding substances [10–15]. Since Eylea® may molecularly offer a slightly different approach to the same target, the question arises whether it may be beneficial in cases where Lucentis® has failed to achieve satisfying results. To elucidate this problem, we performed a review of cases treated at our institution.
Methods This study was a retrospective, consecutive case series. Patients with neovascular AMD previously treated with ranibizumab and transitioned to aflibercept were collected at our institution between December 21, 2012 and June 5, 2013. Inclusion criteria were a diagnosis of exudative AMD confirmed by fluorescein angiography and spectral-domain optical coherence tomography (SD-OCT), previous injections with ranibizumab and subsequent injections with aflibercept in the same eye. Patients with a diagnosis of polypoidal choroidal vasculopathy (PCV) and retinal angiomatous proliferations (RAP) were not included for the purpose of this study. Injections at our institute are routinely performed as three monthly injections followed by a checkup 1 month later. In case of persistent retinal fluid or visual deterioration, another set of three monthly injections is performed. As all injections have to be approved by the patient’s health insurance provider, sometimes only single injections might have been approved. Approval for data collection and analysis was obtained from the institutional review board of the CharitéUniversitätsmedizin Berlin. The research adhered to the tenets set forth in the Declaration of Helsinki. The medical charts of all patients in question were reviewed, and clinical data such as basic demographics and visual acuity (VA) were recorded for the relevant time points. SD-OCT scans from the same time points were reviewed in the Heidelberg Eye Explorer software (Heyex Software Version 1.7.10). Correct positioning of the scan on the fovea was ensured. The automated segmentation of scans covering the central subfield was manually corrected to ensure proper segmentation of the inner limiting membrane and the RPE band. The retinal thickness of the central subfield in the ETDRS grid was then read from the screen (Fig. 1). Time
points that were elected for the study included: 1) the visit before the first ranibizumab injection, when the decision for intravitreal injections was made, 2) the visit before the last three ranibizumab injections, 3) the follow-up visit after the last three injections of ranibizumab, 4) the visit before the first aflibercept injection, when the decision for intravitreal aflibercept injections was made (may coincide with No 3), 5) the follow-up visit after the first aflibercept injection, 6) the visit after the second aflibercept injection, 7) the follow-up visit after the third aflibercept injection, and 8) the follow-up visit after the fourth aflibercept injection. For subgroup analyses, we defined functional response as any gain in visual acuity after intravitreal treatment, and morphological response as any decrease in central retinal thickness. Treatment failure therefore was functionally defined as lack of gain of at least 1 letter in VA, and morphologically defined as lack of decrease of CRST by at least 1 micron. Statistical analyses to compare the parameters were performed using a 2-tailed paired t-test, a Wilcoxon–Mann–Whitney-test, or a chi-square-test where appropriate.
Results Ninety-three eyes of 86 patients received aflibercept injections. Seventy-one eyes of 65 patients were transitioned from ranibizumab to aflibercept between December 21, 2012 and June 5, 2013. Four eyes had been previously treated by bevacizumab before the ranibizumab therapy. Twenty-two eyes were excluded for lack of previous treatment with ranibizumab. The mean patient age was 77 years (range 43–95). Twentyfour males and 41 females were included. On average, all eyes received nine ranibizumab injections (range 3–43) or 3.25 injections per year before switching to aflibercept therapy. The mean time between last ranibizumab injection and first aflibercept injection was 77 days (range 20–422 days, median 37 days). After excluding four patients whose interval was over 1 year, the mean time was 58 days (range 20–259 days, median 35 days). All 71 eyes received at least one aflibercept injection. Sixty-six eyes received at least two aflibercept injections, 45 eyes had three aflibercept injections, and 12 eyes had four aflibercept injections. The average number of aflibercept injections was 2.73 (range 1–4). A summary of the basic demographics is shown in Table 1. Functional results The mean logMAR visual acuity of the total group was 0.54 before the first ranibizumab injection, 0.55 before the last three ranibizumab injections (p=0.629), and 0.57 after last ranibizumab injection (p=0.442). Before the first aflibercept
Graefes Arch Clin Exp Ophthalmol Fig. 1 The standard ETDRS macula thickness plot by optical coherence tomography (OCT). a The infrared images from the Spectralis scan with the thickness maps in false colors with proper centration of the ETDRS grid on the fovea. b The mean retinal thickness in the central subfield shown in ETDRS grid is 318 μm. c OCT B scan showing the segmentation of internal limiting membrane (ILM) and retinal pigment epithelium band (RPE) which had to be adjusted manually. The area between ILM and RPE was used to calculate retinal thickness
injection, the mean VA was 0.67 (p=0.067, compared to after the last ranibizumab injection). After the first aflibercept injection, the VA changed into 0.65 (p=0.426). This initial gain of VA was maintained over the various follow-up times. The duration of time without treatment between the last ranibizumab injection and the first aflibercept injection did not significantly affect the visual outcome in our patient cohort when excluding the four outliers mentioned above (time interval >1 year). More detailed subgroup analyses based on the number of aflibercept injections received are summarized in Table 2. We performed further analyses based on functional success as defined in the methods section. Of the 71 eyes, 32 showed functional success under the prior ranibizumab therapy, but were then later switched to aflibercept due to a weakening effect or persistent intraretinal and/or subretinal fluid. The remaining 39 eyes did not show any functional success under ranibizumab therapy. Ten out of 32 eyes in the group with Table 1 Demographic characteristics of patients No. eyes
71
No. patients (men/women) Age, mean (range), years No. prior ranibizumab injections, mean (range) No. injections, mean (range) Interval of switching, mean (range), days No. eyes with one aflibercept injection No. eyes with two aflibercept injections No. eyes with three aflibercept injections No. eyes with four aflibercept injections
65 (24/41) 77 (43–95) 9 (3–43) 2.7 (1–4) 58 (20–259) 5 21 33 12
No. number
initial functional success improved in visual acuity after initiating aflibercept therapy. Similarly, 13 out of 39 eyes gained in visual acuity with aflibercept, although ranibizumab had not given functional improvement beforehand. Summaries of the number of patients and absolute VA values are given in Tables 3 and 4. Morphological results The mean central subfield retinal thickness of all eyes measured by OCT before the first ranibizumab injection was 417.28 μm (range 229–850), before the last three ranibizumab injections was 389.32 μm (range 208–643, p=0.054), and after the last ranibizumab injection was 349.52 μm (range 206–579, p
RETINAL DISORDERS
Clinical outcomes after switching treatment from intravitreal ranibizumab to aflibercept in neovascular age-related macular degeneration Florian M. Heussen & Qing Shao & Yanling Ouyang & Antonia M. Joussen & Bert Müller
Received: 26 August 2013 / Revised: 3 December 2013 / Accepted: 9 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To describe the treatment response to aflibercept in patients with exudative age-related macular degeneration that showed insufficient or diminishing treatment effects under ranibizumab. Methods From December 2012 till June 2013 all patients receiving intravitreal injections of aflibercept after previous treatment with ranibizumab were collected in a database and retrospectively reviewed. Clinical data such as visual acuity or central subfield retinal thickness on optical coherence tomography (OCT) scans were analyzed for the time frame before, during, and shortly after the aflibercept injections. Of particular interest was the comparison of clinical features under ongoing ranibizumab treatment to the time during aflibercept treatment. Results Seventy-one eyes of 65 patients were included in the study. All eyes had previous ranibizumab injections in their medical history, the average number of which was nine (range 3-43). For the total group the mean visual acuity (VA) before the first ranibizumab injection was 0.54 logMAR, and after the last ranibizumab injection was 0.57 logMAR. Mean VA changed from 0.47 logMAR before the first aflibercept injection to 0.25 logMAR after the last aflibercept injection. Central subfield retinal thickness (CSRT) on OCT changed from a mean of 417.28 μm to 349.52 μm under ranibizumab treatment and from 338.76 μm to 272.00 μm under aflibercept treatment. Interestingly, 33 % of cases that did not show a F. M. Heussen (*) : Q. Shao : Y. Ouyang : A. M. Joussen : B. Müller Department of Ophthalmology, Charité-Universitätsmedizin Berlin, Augustenburger Platz 1, 13353 Berlin, Germany e-mail: [email protected] Q. Shao Department of Ophthalmology, Shanghai First People’s Hospital, Shanghai, China
functional improvement under ranibizumab therapy gained visual acuity after aflibercept treatment. Conclusion Aflibercept appears to be an effective choice for patients with neovascular age-related macular degeneration who were resistant to previous therapy of ranibizumab. The longevity of this effect still remains questionable. Keywords Age-related macular degeneration . Anti-VEGF treatment . Ranibizumab . Aflibercept . Tachyphylaxis Abbreviations OCT Optical coherence tomography SD-OCT Spectral-domain OCT FA Fluorescein angiography VA Visual acuity AMD Age-related macular degeneration CSRT Central subfield retinal thickness
Introduction The advent of ranibizumab (Lucentis®) in the treatment of exudative age-related macular degeneration is widely regarded as a major breakthrough in our field. For the first time, patients could realistically expect to regain vision under treatment where previous approaches would cause a slowdown in visual deterioration at best [1–6]. However, even with this seemingly fail-safe treatment, there are cases that simply do not respond well functionally or morphologically. Taking it one step further, in a few cases there seems to be no effect at all [7]. The overall number of treatment-resistant cases may be small in view of all treated eyes; however, it is this particular group of patients which causes the most concern amongst clinicians. While some of these treatment failures may be due to misdiagnosis and therefore mistreatment, there have
Graefes Arch Clin Exp Ophthalmol
been reports about a lacking response to ranibizumab in exudative AMD. Some authors suspected a tachyphylaxislike effect of the intravitreal anti-VEGF injections — a claim that was supported by an increase in effectiveness when the drug in use was switched from ranibizumab to bevacizumab and vice versa [8]. Recently, aflibercept (Eylea®), a newly approved drug for intravitreal delivery in exudative AMD, has entered the market [9]. While details of the pharmacokinetics and drug mechanism may very well be substantially different from ranibizumab, from a clinician’s point of view, both drugs fall into the same category of VEGF-binding substances [10–15]. Since Eylea® may molecularly offer a slightly different approach to the same target, the question arises whether it may be beneficial in cases where Lucentis® has failed to achieve satisfying results. To elucidate this problem, we performed a review of cases treated at our institution.
Methods This study was a retrospective, consecutive case series. Patients with neovascular AMD previously treated with ranibizumab and transitioned to aflibercept were collected at our institution between December 21, 2012 and June 5, 2013. Inclusion criteria were a diagnosis of exudative AMD confirmed by fluorescein angiography and spectral-domain optical coherence tomography (SD-OCT), previous injections with ranibizumab and subsequent injections with aflibercept in the same eye. Patients with a diagnosis of polypoidal choroidal vasculopathy (PCV) and retinal angiomatous proliferations (RAP) were not included for the purpose of this study. Injections at our institute are routinely performed as three monthly injections followed by a checkup 1 month later. In case of persistent retinal fluid or visual deterioration, another set of three monthly injections is performed. As all injections have to be approved by the patient’s health insurance provider, sometimes only single injections might have been approved. Approval for data collection and analysis was obtained from the institutional review board of the CharitéUniversitätsmedizin Berlin. The research adhered to the tenets set forth in the Declaration of Helsinki. The medical charts of all patients in question were reviewed, and clinical data such as basic demographics and visual acuity (VA) were recorded for the relevant time points. SD-OCT scans from the same time points were reviewed in the Heidelberg Eye Explorer software (Heyex Software Version 1.7.10). Correct positioning of the scan on the fovea was ensured. The automated segmentation of scans covering the central subfield was manually corrected to ensure proper segmentation of the inner limiting membrane and the RPE band. The retinal thickness of the central subfield in the ETDRS grid was then read from the screen (Fig. 1). Time
points that were elected for the study included: 1) the visit before the first ranibizumab injection, when the decision for intravitreal injections was made, 2) the visit before the last three ranibizumab injections, 3) the follow-up visit after the last three injections of ranibizumab, 4) the visit before the first aflibercept injection, when the decision for intravitreal aflibercept injections was made (may coincide with No 3), 5) the follow-up visit after the first aflibercept injection, 6) the visit after the second aflibercept injection, 7) the follow-up visit after the third aflibercept injection, and 8) the follow-up visit after the fourth aflibercept injection. For subgroup analyses, we defined functional response as any gain in visual acuity after intravitreal treatment, and morphological response as any decrease in central retinal thickness. Treatment failure therefore was functionally defined as lack of gain of at least 1 letter in VA, and morphologically defined as lack of decrease of CRST by at least 1 micron. Statistical analyses to compare the parameters were performed using a 2-tailed paired t-test, a Wilcoxon–Mann–Whitney-test, or a chi-square-test where appropriate.
Results Ninety-three eyes of 86 patients received aflibercept injections. Seventy-one eyes of 65 patients were transitioned from ranibizumab to aflibercept between December 21, 2012 and June 5, 2013. Four eyes had been previously treated by bevacizumab before the ranibizumab therapy. Twenty-two eyes were excluded for lack of previous treatment with ranibizumab. The mean patient age was 77 years (range 43–95). Twentyfour males and 41 females were included. On average, all eyes received nine ranibizumab injections (range 3–43) or 3.25 injections per year before switching to aflibercept therapy. The mean time between last ranibizumab injection and first aflibercept injection was 77 days (range 20–422 days, median 37 days). After excluding four patients whose interval was over 1 year, the mean time was 58 days (range 20–259 days, median 35 days). All 71 eyes received at least one aflibercept injection. Sixty-six eyes received at least two aflibercept injections, 45 eyes had three aflibercept injections, and 12 eyes had four aflibercept injections. The average number of aflibercept injections was 2.73 (range 1–4). A summary of the basic demographics is shown in Table 1. Functional results The mean logMAR visual acuity of the total group was 0.54 before the first ranibizumab injection, 0.55 before the last three ranibizumab injections (p=0.629), and 0.57 after last ranibizumab injection (p=0.442). Before the first aflibercept
Graefes Arch Clin Exp Ophthalmol Fig. 1 The standard ETDRS macula thickness plot by optical coherence tomography (OCT). a The infrared images from the Spectralis scan with the thickness maps in false colors with proper centration of the ETDRS grid on the fovea. b The mean retinal thickness in the central subfield shown in ETDRS grid is 318 μm. c OCT B scan showing the segmentation of internal limiting membrane (ILM) and retinal pigment epithelium band (RPE) which had to be adjusted manually. The area between ILM and RPE was used to calculate retinal thickness
injection, the mean VA was 0.67 (p=0.067, compared to after the last ranibizumab injection). After the first aflibercept injection, the VA changed into 0.65 (p=0.426). This initial gain of VA was maintained over the various follow-up times. The duration of time without treatment between the last ranibizumab injection and the first aflibercept injection did not significantly affect the visual outcome in our patient cohort when excluding the four outliers mentioned above (time interval >1 year). More detailed subgroup analyses based on the number of aflibercept injections received are summarized in Table 2. We performed further analyses based on functional success as defined in the methods section. Of the 71 eyes, 32 showed functional success under the prior ranibizumab therapy, but were then later switched to aflibercept due to a weakening effect or persistent intraretinal and/or subretinal fluid. The remaining 39 eyes did not show any functional success under ranibizumab therapy. Ten out of 32 eyes in the group with Table 1 Demographic characteristics of patients No. eyes
71
No. patients (men/women) Age, mean (range), years No. prior ranibizumab injections, mean (range) No. injections, mean (range) Interval of switching, mean (range), days No. eyes with one aflibercept injection No. eyes with two aflibercept injections No. eyes with three aflibercept injections No. eyes with four aflibercept injections
65 (24/41) 77 (43–95) 9 (3–43) 2.7 (1–4) 58 (20–259) 5 21 33 12
No. number
initial functional success improved in visual acuity after initiating aflibercept therapy. Similarly, 13 out of 39 eyes gained in visual acuity with aflibercept, although ranibizumab had not given functional improvement beforehand. Summaries of the number of patients and absolute VA values are given in Tables 3 and 4. Morphological results The mean central subfield retinal thickness of all eyes measured by OCT before the first ranibizumab injection was 417.28 μm (range 229–850), before the last three ranibizumab injections was 389.32 μm (range 208–643, p=0.054), and after the last ranibizumab injection was 349.52 μm (range 206–579, p
