Doc Ophthalmol DOI 10.1007/s10633-008-9136-7
ORIGINAL RESEARCH ARTICLE
Electrophysiological evaluation of retinal photoreceptor function after repeated bevacizumab injections Andreas Stahl Æ Nicolas Feltgen Æ Antje Fuchs Æ Michael Bach
Received: 30 January 2008 / Accepted: 3 June 2008 Springer-Verlag 2008
Abstract Bevacizumab (Avastin, Genentech) was one of the first anti-VEGF substances used to treat macular edema or choroidal neovascularization in patients with vascular ocular pathologies. However, only few studies evaluate the safety of intravitreal bevacizumab injections in regard to retinal photoreceptor function. We evaluated retinal function after repeated (2–3) monthly injections of bevacizumab in a prospective case series of 10 patients with various retinal diseases. Study endpoints were visual acuity (VA) using ETDRS charts and 3 full-field electroretinography sessions with a flash intensity range of 0.0005–2 cds/m2. V-log-I b-wave amplitudes were fitted by a Naka-Rushton model. No significant changes in scotopic or photopic ERG measures were observed between baseline ERG and last follow-up ERG. Individual patients showed transient alterations of ERG measures on the first follow-up visit. Mean visual acuity was stable over the time course of the study (logMAR = 0.42 at baseline and logMAR = 0.48 at last follow-up). In conclusion, three monthly repeated injections of bevacizumab do not affect mid-term electrophysiological retinal function. Transient alterations in ERG readings of individual patients 1 week after intravitreal bevacizumab injection may be attributed to short-term disruption of the retinal A. Stahl N. Feltgen A. Fuchs M. Bach (&) Universita¨ts-Augenklinik, Killianstr. 5, 79106 Freiburg, Germany e-mail:
[email protected]
equilibrium through the trauma of injection. Evaluation of patients receiving more than three injections of anti-VEGF substances should be the focus of future studies. Keywords Anti-VEGF Toxicity Bevacizumab Avastin Age-related macular degeneration ERG
Introduction Recent advances in the clinical application of antiangiogenic strategies have led to a breakthrough in the treatment of ocular diseases associated with either neovascularization or exsudation of retinal and choroidal blood vessels. Bevacizumab (Avastin, Genentech) was one of the first substances used for the treatment of ocular vascular pathologies [5, 7, 9, 13]. To date, numerous reports have described the beneficial effects of bevacizumab in wet age-related macular degeneration (AMD) and vascular leakage secondary to retinal vein occlusion [1, 6, 19]. Although bevacizumab treatment is widely used in clinical settings for various ocular vascular pathologies, there is only little data available on the safety of bevacizumab on CNV-related angiogenesis [4]. Bevacizumab and the closely related ranibizumab (Lucentis, Novartis) are anti-VEGF antibodies binding all isoforms of the growth factor VEGFA. As VEGF not only has effects on vascular permeability and vasoproliferation but also acts as a survival
123
Doc Ophthalmol
factor for neuronal cells [8, 20], concerns have been raised about possible side effects of intravitreal VEGFinhibition on retinal photoreceptor or glia cells [11]. Although ranibizumab, the Fab-fragment of bevacizumab, is now clinically approved for intravitreal injection in patients with wet AMD, bevacizumab is still widely used as off-label treatment for patients with other VEGF-related ocular pathologies. A number of recent reports have evaluated the safety of a single bevacizumab injection [10, 12, 14, 17, 21]. However, studies evaluating the cumulative long-term risk for retinal photoreceptor function after repeated bevacizumab injections are still missing. In the present case series we therefore evaluate photoreceptor function after repeated intravitreal injections of bevacizumab using ERG and visual acuity testing. Our aim is to assess even mild damage that might occur in a patient when receiving repeated injections of bevacizumab. Rare point-incidences would not be caught given our small sample size.
• • • •
Opacities of lens or cornea that would preclude significant VA improvement Not able to give informed consent History of allergic reaction to bevacizumab Pregnancy
Study endpoints Primary study endpoints were changes in electroretinogram (ERG) recordings after bevacizumab injection compared to baseline ERG prior to treatment. Baseline ERG was obtained few hours before injection, first follow-up ERG was one week after the first injection, and second follow-up ERG was 3 weeks after the third injection. Secondary study endpoints were best-corrected visual acuity (VA) measured with ETDRS charts at same dates as ERG and complication rate. Patient examination
Patients and methods Patients and study design Ten patients receiving intravitreal bevacizumab injection for various ocular pathologies were included in this study. Our study design is that of a nonrandomized interventional case series. Inclusion criteria •
• • •
Angioproliferative disease of retinal or choroidal blood vessels—wet AMD with classic or occult choroidal neovascularization (CNV, based on fundus pictures combined with the angiographic findings following standard international nomenclature), retinal angiomatous proliferation (RAP), parafoveal telangiectasis, CNV secondary to chronic central serous chorioretinopathy (CSC) Best-corrected VA of B1.0 logMAR, corresponding to C0.1 decimal VA Age older than 18 years Patient able to give informed consent
Exclusion criteria •
Patients with retinal vein occlusions or other conditions affecting retinal function outside the macula
123
The following data were recorded at first visit: indication for bevacizumab injection as seen on fluorescein angiography (FF450, Zeiss Germany), ophthalmologic and medical history, patient age and sex, best-corrected VA (ETDRS charts), and full ocular examination. On each follow-up visit, VA and ERG were performed and possible side effects of the injection were ruled out. On the last visit, full visual examination and fluorescein angiography were performed and re-injections were scheduled depending on the results. Ethics The study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All patients gave their informed consent prior to their inclusion in the study with specific emphasis on the off-label character and possible systemic side effects as well as unknown long-term ocular complications of bevacizumab. ERG Recording The pupils were dilated and the patients adapted for 20 min in a fully darkened room. Using a Nicolet
Doc Ophthalmol
stimulator and in-house software, a flash intensity series was applied in half-log unit increments: 0.00054, 0.0018, 0.0054, 0.018, 0.054, 0.18, 0.54, 1.8 cd s/m2. Recordings contaminated by artifacts were repeated; the aim was to acquire at least two artifact-free records per intensity. After the brightest flash, the patient was light-adapted by the 30 cd/m2 background dome for 10 min. Then, using a 1.8 cd s/ m2 flash, the photopic ERGs and finally the 30.1 Hz flicker responses were recorded. Analysis The raw ERG recordings (Fig. 1, top) were visually inspected and all artifact-free peaks marked. In most cases this resulted in more than one value for the wave feature per intensity, which were averaged. The flicker ERG was quantified via a Fourier analysis of an artifact-free range of the 1-s flicker record. The scotopic a- and b-waves were fitted by Naka Rushton functions [3, 16]:
The demographic data for all patients enrolled in this study are displayed in Table 2. All patients had suffered significant decrease in VA from different ocular pathologies with VA ranging from 0.2 to 1.0 logMAR prior to injection. Statistics All statistical calculations were performed using the ‘R’ system [15]. The characteristic ERG measures were analyzed with repeated-measures ANOVAs using the factors ERG_PARAMETER and SESSION and within-error factor PATIENT. The ERG_ PARAMETERs used were the three Naka-Rushton parameters characterizing the scotopic b-wave, and the amplitudes and peak time of the photopic response (single flash and flicker). Since there were no significant findings, there was no need to adjust for multiple testing.
Results
n ERG ¼ rmax I n =ðI n þ k50 Þ
where I represents the luminous intensity, rmax represents the maximal response (not reached for the a-wave, sometimes for the b-wave), n represents the slope, and k50 the intensity of half-maximal response. The calculations were performed in the R statistical system [15]. To ensure conversion of the fit, these parameters were constrained as follows: 1:0 rmax 600;
0:01 n 10;
0:004 k50 15
The results were used for the graphical representations and for statistical analysis, since it would be unpractical and reduce statistical power (correction for multiple testing) if all possible amplitudes were statistically tested. In the case of the a-wave, the Naka-Rushton model is not appropriate and the fit parameters (rmax, n and k50) were not interpreted. Bevacizumab injection All patients underwent intraocular injection of 1.25 mg bevacizumab (Avastin) in 0.05 ml total volume over the pars plana according to the recommendations of the German Retinological Society (http://retinologie.org/). The time sequence of examinations and injections are delineated in Table 1.
Patients demographics and visual acuity Ten patients from our outpatient department were included in the study according to the inclusion criteria outlined above. Mean age of enrolled patients was 72 years; mean visual acuity (VA) at presentation was logMAR = 0.42. Eight patients received three monthly injections. Patients ID 1 and 9 received only two injections due to disease characteristics (ID 1) or due to failure to complete follow-up examinations (ID 9). Patient ID 6 received photodynamic therapy (PDT) one day prior to the first bevacizumab injection. No complications like increase in intraocular pressure, retinal detachment, or endophthalmitis were observed on follow-up examinations. Baseline visual acuity (VA) was recorded on the day of the first injection. The first follow-up was scheduled one week after the first bevacizumab injection, the second follow-up three weeks after the final injection. VA values for both baseline as well as first and second follow-up are shown in Table 2. Electroretinographic recordings ERG data for one exemplary patient (ID 10) is presented in Fig. 1. The resulting scotopic a- and b-
123
Doc Ophthalmol
123
Doc Ophthalmol b Fig. 1 ERGs from a typical subject (ID 10) before the first
injection (left pair of columns, OD/OS) and on the second follow-up visit three weeks after the third injection (right columns). The right eye was injected. The top 8 traces are the scotopic recordings covering an intensity range from 0.00054 to 1.8 cd s/m2. Time traces run from -50 to +200 ms, and from 0 to 1,000 ms for flicker stimuli. Amplitudes are -400 to +400 lV for scotopic, and -100 to +100 lV for photopic. Further down are the photopic recordings to the 1.8 cd s/m2 flash and flicker recordings with the corresponding Fourier spectra. At the bottom the V-log-I functions for the scotopic aand b-waves with their Naka-Rushton fit are depicted for the injected OD
wave amplitudes are plotted vs. intensity at the bottom (V-log-I functions). The data points represent the mean when repeat data was available for that intensity. The continuous lines represent the corresponding Naka-Rushton fit result (see Methods). Figure 2 displays V-log-I functions for all ten patients. Between different patients strong interindividual variability can be observed both in the baseline and in the follow-up ERG recordings. The intra-individual variability between baseline and follow-up ERGs, however, is less pronounced. For
some patients (ID 1, 2, and 10) the b-wave recordings of the first follow-up recording differ slightly from the baseline recording. On the second follow-up recording three weeks after the third injection, however, the b-waves again closely resemble their respective baseline values. The data presented in Fig. 2 suggest stable retinal function after repeated bevacizumab injections. In order to test the hypothesis that for individual patients repeated bevacizumab injections do not lead to significant changes in the electrophysiologic function of the retina, the scotopic ERG parameters (Fig. 3) and the photopic amplitudes and peak times of the single flash and the flicker response (Fig. 4) were tested using repeated-measures ANOVA with patient as the within-factor. The statistical analysis showed no significant changes in ERG recordings after repeated intravitreal injections of bevacizumab. There was no significant effect of session for the scotopic measures (ANOVA results; a-wave, amplitude: P = .36, peak time: P = .15; b-wave, rmax: P = .70, slope: P = .14, k50: P = .19) nor for the photopic amplitudes and peak times of the single flash or flicker response (all P-values above .6).
Table 1 Time sequence of examinations and injections Time (weeks)
0
1
VA
X
X
ERG
X
X
Injection
X
4
8
11 X X
X
X
Discussion In our study population, repeated intravitreal injections of 1.25 mg bevacizumab did not induce
Table 2 Demographic data and visual acuity (VA) measurements for all ten patients Patient ID
Vascular pathology
Sex
Patient age (years)
Session 1 (baseline) VA (logMAR)
Session 2 (1st follow up) VA (logMAR)
Session 3 (2nd follow up) VA (logMAR)
1
CNV from CSC
F
47
0.5
0.3
0.2
2
Parafov. teleang.
M
57
0.3
0.2
0.3
3
Occult CNV
F
66
0.3
0.3
0.2
4
Occult CNV
F
77
0.5
0.5
0.6
5
Occult CNV
M
87
1.0
1.5
1.6
6
Classic CNV
M
86
0.4
0.6
0.7
7
Occult CNV
F
70
0.2
0.1
0.2
8
Occult CNV
M
71
0.4
Missing
0.3
9
Occult CNV
F
81
0.3
0.3
Missing
10
RAP
M
77
0.3
0.5
0.2
Baseline VA was measured prior to the first bevacizumab injection, first VA follow up was 1 week after the first injection, and second VA follow up was 3 weeks after the last injection. CNV = Choroidal neovascularization, CSC = Central serous chorioretinopathy, RAP = Retinal angiomatous proliferation
123
Doc Ophthalmol
Fig. 2 V-log-I functions from all patients (b-wave is always the upper one) for all three recordings. The ordinate always runs from 0 to 460 lV, the abscissa from 0.00045 to 2.0 cd s/m2. Top: before the first injection; middle: 1 week after the first injection; bottom: 3 weeks after the last injection. Interindividual variability between patients is higher (horizontal)
Fig. 3 Boxplots of the scotopic a-wave parameters (left) and b-wave NakaRushton parameters (right) rmax (maximal response), n (slope), and k50 (half-max intensity) for the 3 sessions (before the first injection, after the first injection, and after the last injection). Corresponding to the ANOVA findings, there is no trend across sessions. The k50 parameter shows the widest variability between subjects. [Plot details: the median is indicated by the horizontal thick lines, the notches give a 95% confidence interval for the difference in two medians, the box covers the 25–75% percentile range, the ‘‘antennas’’ indicate the range, outliers are indicated by circles.]
123
than the intra-individual variability between sessions (vertical). Possible adverse effects of the injections would be apparent when comparing the sessions from top to bottom; none are obvious. Patients are arranged from left to right by ID of Table 2
Doc Ophthalmol Fig. 4 Boxplots of the photopic P1 (left) and flicker response (right). Top: amplitudes, Bottom: peak times. The flicker amplitudes were derived via Fourier analysis and represent the magnitude of the first harmonic; thus the values are less than half compared to peak-trough measures. Corresponding to the ANOVA findings, there is no trend across sessions
alterations of electrophysiological retina function, either for rod-related activity (Fig. 3) or for conerelated activity (Fig. 4). This finding is in line with earlier observations for single bevacizumab injections [10, 12, 14, 17, 21]. In some patients, we observed temporary alteration of ERG readings between baseline and first follow-up visit. On the second follow-up visit, however, ERG readings had returned to the individual patient’s baseline values. The temporary ERG alterations noted on the first follow-up visit may be due to the fact that the first follow-up ERG was recorded one week after injection, whereas the second follow-up ERG was recorded 3 weeks after the last injection. The variations seen in the first follow-up ERG might therefore reflect short-term modifications of the retinal equilibrium due to the trauma of intravitreal injection. The reversibility of the observed ERG alterations in individual patients suggests that these short-term modifications are not affecting long-term retinal function. The stability of
the Naka-Rushton parameters over the observation period emphasizes the fact that three monthly injections of bevacizumab do not affect overall retinal function. Changes in visual acuity (VA) were not the focus of this study and have been more thoroughly evaluated in other studies [2, 18]. VA values in our study population remained stable over the time of the study with a mean VA of logMAR = 0.42 at baseline and a final VA of logMAR = 0.48 after the last injection. In conclusion, this study does not reveal any photoreceptor damage after up to three repeated bevacizumab injections. Evaluation of patients receiving more than three injections should be the focus of future studies. Such studies should also compare bevacizumab treatment to other intravitreally injected VEGF-inhibitors concerning long-term retinal toxicity. Acknowledgments We thank Prof. Lutz L. Hansen and PD Dr. Hansjuergen Agostini for inspiration and advice.
123
Doc Ophthalmol
References 1. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA, Nasir MA, Giust MJ (2006) Intravitreal bevacizumab (Avastin) for neovascular age-related macular degeneration. Ophthalmology 113:363–372, e365 2. Emerson MV, Lauer AK, Flaxel CJ, Wilson DJ, Francis PJ, Stout JT et al (2007) Intravitreal bevacizumab (Avastin) treatment of neovascular age-related macular degeneration. Retina 27:439–444. doi:10.1097/IAE.0b013e31804b3e15 3. Fulton AB, Rushton WA (1978) The human rod ERG: correlation with psychophysical responses in light and dark adaptation. Vision Res 18:793–800. doi:10.1016/00426989(78)90119-0 4. Inan UU, Avci B, Kusbeci T, Kaderli B, Avci R, Temel SG (2007) Preclinical safety evaluation of intravitreal injection of full-length humanized vascular endothelial growth factor antibody in rabbit eyes. Invest Ophthalmol Vis Sci 48:1773–1781. doi:10.1167/iovs.06-0828 5. Iturralde D, Spaide RF, Meyerle CB, Klancnik JM, Yannuzzi LA, Fisher YL et al (2006) Intravitreal bevacizumab (Avastin) treatment of macular edema in central retinal vein occlusion: a short-term study. Retina 26:279–284. doi: 10.1097/00006982-200603000-00005 6. Jaissle GB, Ziemssen F, Petermeier K, Szurman P, Ladewig M, Gelisken F et al (2006) Bevacizumab for treatment of macular edema secondary to retinal vein occlusion. Ophthalmologe 103:471–475. doi:10.1007/s00347-006-1355-2 7. Jorge R, Costa RA, Calucci D, Scott IU (2007) Intravitreal bevacizumab (Avastin) associated with the regression of subretinal neovascularization in idiopathic juxtafoveolar retinal telangiectasis. Graefes Arch Clin Exp Ophthalmol 245:1045–1048. doi:10.1007/s00417-006-0468-2 8. Lazarovici P, Marcinkiewicz C, Lelkes PI (2006) Cross talk between the cardiovascular and nervous systems: neurotrophic effects of vascular endothelial growth factor (VEGF) and angiogenic effects of nerve growth factor (NGF)-implications in drug development. Curr Pharm Des 12:2609–2622. doi:10.2174/138161206777698738 9. Lazic R, Gabric N (2007) Intravitreally administered bevacizumab (Avastin) in minimally classic and occult choroidal neovascularization secondary to age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol 245:68–73. doi:10.1007/s00417-006-0466-4 10. Maturi RK, Bleau LA, Wilson DL (2006) Electrophysiologic findings after intravitreal bevacizumab (Avastin) treatment. Retina 26:270–274. doi:10.1097/00006982200603000-00003 11. Michels S (2006) Is intravitreal bevacizumab (Avastin) safe? Br J Ophthalmol 90:1333–1334. doi:10.1136/bjo. 2006.102293
123
12. Moschos MM, Brouzas D, Apostolopoulos M, Koutsandrea C, Loukianou E, Moschos M (2007) Intravitreal use of bevacizumab (Avastin) for choroidal neovascularization due to ARMD: a preliminary multifocal-ERG and OCT study. Multifocal-ERG after use of bevacizumab in ARMD. Doc Ophthalmol 114:37–44. doi:10.1007/s10633-006-9036-7 13. Ng EW, Adamis AP (2005) Targeting angiogenesis, the underlying disorder in neovascular age-related macular degeneration. Can J Ophthalmol 40:352–368 14. Pai SA, Shetty R, Vijayan PB, Venkatasubramaniam G, Yadav NK, Shetty BK et al (2007) Clinical, anatomic, and electrophysiologic evaluation following intravitreal bevacizumab for macular edema in retinal vein occlusion. Am J Ophthalmol 143:601–606. doi:10.1016/j.ajo.2006.12.037 15. R Development Core Team (2007) R: A language and environment for statistical computing. Accessed 31 July 2006, from (http://www.R-project.org) 16. Severns ML, Johnson MA (1993) The care and fitting of Naka-Rushton functions to electroretinographic intensityresponse data. Doc Ophthalmol 85:135–150. doi:10.1007/ BF01371129 17. Shetty R, Pai SA, Vincent A, Shetty N, Narayana KM, Sinha B, Shetty BK (2007) Electrophysiological and structural assessment of the central retina following intravitreal injection of bevacizumab for treatment of macular edema. Doc Ophthalmol 116(2):129–135. doi: 10.1007/ s10633-007-9090-9) 18. Spaide RF, Laud K, Fine HF, Klancnik JM Jr, Meyerle CB, Yannuzzi LA et al (2006) Intravitreal bevacizumab treatment of choroidal neovascularization secondary to agerelated macular degeneration. Retina 26:383–390. doi: 10.1097/00006982-200604000-00001 19. Stahl A, Agostini H, Hansen LL, Feltgen N (2007) Bevacizumab in retinal vein occlusion-results of a prospective case series. Graefes Arch Clin Exp Ophthalmol 245:1429– 1436. doi:10.1007/s00417-007-0569-6 20. Wang Y, Galvan V, Gorostiza O, Ataie M, Jin K, Greenberg DA (2006) Vascular endothelial growth factor improves recovery of sensorimotor and cognitive deficits after focal cerebral ischemia in the rat. Brain Res 1115:186–193. doi:10.1016/j.brainres.2006.07.060 21. Ziemssen F, Luke M, Messias A, Beutel J, Tatar O, Zrenner E, Bartz-Schmidt KU (2007) Safety monitoring in bevacizumab (Avastin) treatment: retinal function assessed by psychophysical (visual fields, colour vision) and electrophysiological (ERG/EOG) tests in two subgroups of patients. Int Ophthalmol 28(2):101–109. doi: 10.1007/ s10792-007-9122-1
