Predictors of AMD Treatment Response

Abstract

It is important that we understand which factors dictate the visual acuity outcome in people receiving anti-vascular endothelial growth factor (anti-VEGF) for choroidal neovascularization (CNV) secondary to age-related macular degeneration (AMD).1Lim J. Wickremasinghe S.S. Xie J. et al.Delay to treatment and visual outcomes in patients treated with anti-vascular endothelial growth factor for age-related macular degeneration.Am J Ophthalmol. 2012; 153: 678-686Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar, 2Lux A. Llacer H. Heussen F.M. Joussen A.M. Non-responders to bevacizumab (Avastin) therapy of choroidal neovascular lesions.Br J Ophthalmol. 2007; 91: 1318-1322Crossref PubMed Scopus (137) Google Scholar, 3Teper S.J. Nowinska A. Pilat J. et al.Involvement of genetic factors in the response to a variable-dosing ranibizumab treatment regimen for age-related macular degeneration.Mol Vis. 2010; 16: 2598-2604PubMed Google Scholar In this study, we sought to assess whether optical coherence tomography (OCT) features at baseline and after initial treatment correlated with visual outcome at 12 months. This study was approved by the Human Research and Ethics Committee of the Royal Victorian Eye and Ear Hospital.Treatment-naïve patients diagnosed with subfoveal CNV secondary to AMD were prospectively consecutively recruited from the Medical Retina Clinic at the Royal Victorian Eye and Ear Hospital and the private rooms of the Eastern Retinal Service in Box Hill, Melbourne, Australia. All patients received 3, monthly injections of ranibizumab and/or bevacizumab, with subsequent retreatment decisions at monthly visits based on a modification of the SUSTAIN study.4Holz F.G. Amoaku W. Donate J. et al.Safety and efficacy of a flexible dosing regimen of ranibizumab in neovascular age-related macular degeneration: the SUSTAIN study.Ophthalmology. 2011; 118: 663-671Abstract Full Text Full Text PDF PubMed Scopus (332) Google Scholar In brief, retreatment was performed if there was either persistence or recurrence of intraretinal fluid or subretinal fluid on OCT, loss of ≥1 lines of vision, or the presence of new macular hemorrhage on clinical examination.At baseline, OCT was assessed for the presence of an intraretinal fluid (IRF) component (cysts or edema), subretinal fluid (SRF), pigment epithelial detachment (PED), and a significant band of reflectance at the retinal pigment epithelial (RPE) level, because it has been suggested that this is a sign of irreversible damage to the outer retina/RPE.5Fleckenstein M. Charbel Issa P. Helb H.M. et al.High-resolution spectral domain-OCT imaging in geographic atrophy associated with age-related macular degeneration.Invest Ophthalmol Vis Sci. 2008; 49: 4137-4144Crossref PubMed Scopus (233) Google Scholar After 3 injections, eyes without IRF or SRF were considered “dry,” as were eyes with PED without IRF or SRF. Regardless of the presence of PED, if either IRF or SRF were present, eyes were considered as having “persistent fluid” (Fig 1A, B, available at http://aaojournal.org). The presence or absence of a hyperreflective RPE band was considered separately as it could be present irrespective of the nature of the fluid status (Fig 1C, D).At baseline, 144 of 214 eyes (67.2%) had IRF with or without SRF and 70 (32.8%) eyes had SRF alone. Of 58 eyes (27.1%) that had PED, 38 had associated IRF, and 20 had additional SRF alone (Table 1, available at http://aaojournal.org). Best-corrected logMAR visual acuity (BCVA) at baseline was significantly worse in eyes with IRF (0.79 [Snellen equivalent, 20/122]; 95% confidence interval [CI], 0.68–0.90) compared with eyes with SRF alone (0.55 [20/72]; 95% CI, 0.34–0.75; P = 0.006). The presence of PED did not significantly impact on BCVA. Data at the 12-month visit were available for 187 eyes (87.4% of recruited eyes). No differences were seen with respect to the baseline population and the 187 eyes with complete data available at 12 months (P>0.10). Over 12 months, 26.2% of eyes gained ≥2 lines of BCVA compared with 19.3% ≥3 line gain in the SUSTAIN study.4 51.3% remained stable (±2 lines) and 22.5% lost ≥2 lines. The location of fluid at baseline did not significantly influence the likelihood of BCVA improvement at 12 months, although eyes with SRF at baseline had better BCVA (0.54 [20/69]; 95% CI, 0.34–0.75) compared with those with an IRF component (0.78 [20/120]; 95% CI, 0.63–0.93; P = 0.07). After the 3 initial injections, 63.1% of eyes were classified as being dry, whereas 26.2% had IRF and 10.7% had residual SRF only. The factors influencing BCVA change at 12 months are shown in Table 2 (available at http://aaojournal.org). After 3 injections, eyes that were dry had better BCVA at 12 months (0.61 [20/80]; 95% CI, 0.52–0.70) compared with those with residual IRF (0.95 [20/180]; 95% CI, 0.78–1.11; P = 0.05). Eyes with SRF alone (0.65 [20/90]; 95% CI, 0.39–0.91) were similar to those that were dry. The presence of persistent fluid was associated with a reduced chance of visual improvement at 12 months (18.6% vs 31.4%; P = 0.04).At baseline, 6 eyes demonstrated RPE hyperreflectance. This increased to 29 (15.5%) after 3 injections. Regardless of the presence or absence of fluid, hyperreflectivity conferred a greater likelihood of BCVA loss (48.3% vs 19.7%; P = .006) and poorer final BCVA (0.89 [20/155]; 95% CI, 0.69–1.10) compared with eyes without hyperreflectance (0.65 [20/90]; 95% CI, 0.56–0.73; P = 0.02). Two thirds (66%) of eyes with PED at baseline were noted to have residual PED after initial injections with no difference in final BCVA dependent on the PED.After 3 treatments, eyes with a residual IRF component were associated with a greater chance of BCVA loss at 12 months compared with those who were dry (odds ratio [OR], 6.08; 95% CI, 1.53–32.56; P = 0.01), whereas those with RPE hyperreflectivity were also more likely to lose BCVA (OR, 4.48; 95% CI, 1.11–16.41; P = 0.03; Table 3, Table 4, available at http://aaojournal.org). Our findings suggest that the presence of IRF fluid at baseline leads to worse presenting acuity and confers a significantly worse prognosis for visual outcome.Failure to clear fluid from the retina after the loading dose injections also correlates with poorer BCVA outcome and with a greater chance of losing vision at month 12 (Fig 2A, B, available at http://aaojournal.org). This may relate to disruption of retinal architecture with IRF as opposed to SRF, where the retina may be relatively intact.Failure to resolve fluid from the retina after initial treatment may be a marker of irreversible structural damage to the photoreceptor/RPE complex. Most eyes with persistent fluid after 3 injections also had persistent fluid at 12 months, suggesting that the ability to dry the retina is determined early in the course of treatment. Early in the course of treatment, RPE hyperreflectance also correlates with BCVA loss at 12 months. Most eyes with this feature had significant subretinal hemorrhage at baseline. Of the initial 214 eyes, 29 had significant hyperreflectance at the RPE level. Of these, some degree of atrophy or subretinal fibrosis was identified in 21 eyes, and of the remaining 8 eyes, 7 had residual subretinal hemorrhage, which by 12 months revealed atrophy (n = 2) or subretinal fibrosis (n = 5).Table 5 (available at http://aaojournal.org) details findings of eyes with RPE hyperreflectance. No association was seen with presence of PED at baseline or after 3 treatments and BCVA change at 12 months. Anti-VEGF treatment however, did reduce the frequency of PED compared with baseline.In conclusion, eyes with IRF after initial treatment may require more aggressive treatment strategies to achieve best possible outcome, whereas a significant RPE hyperreflective band limits the potential for visual improvement. It is important that we understand which factors dictate the visual acuity outcome in people receiving anti-vascular endothelial growth factor (anti-VEGF) for choroidal neovascularization (CNV) secondary to age-related macular degeneration (AMD).1Lim J. Wickremasinghe S.S. Xie J. et al.Delay to treatment and visual outcomes in patients treated with anti-vascular endothelial growth factor for age-related macular degeneration.Am J Ophthalmol. 2012; 153: 678-686Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar, 2Lux A. Llacer H. Heussen F.M. Joussen A.M. Non-responders to bevacizumab (Avastin) therapy of choroidal neovascular lesions.Br J Ophthalmol. 2007; 91: 1318-1322Crossref PubMed Scopus (137) Google Scholar, 3Teper S.J. Nowinska A. Pilat J. et al.Involvement of genetic factors in the response to a variable-dosing ranibizumab treatment regimen for age-related macular degeneration.Mol Vis. 2010; 16: 2598-2604PubMed Google Scholar In this study, we sought to assess whether optical coherence tomography (OCT) features at baseline and after initial treatment correlated with visual outcome at 12 months. This study was approved by the Human Research and Ethics Committee of the Royal Victorian Eye and Ear Hospital. Treatment-naïve patients diagnosed with subfoveal CNV secondary to AMD were prospectively consecutively recruited from the Medical Retina Clinic at the Royal Victorian Eye and Ear Hospital and the private rooms of the Eastern Retinal Service in Box Hill, Melbourne, Australia. All patients received 3, monthly injections of ranibizumab and/or bevacizumab, with subsequent retreatment decisions at monthly visits based on a modification of the SUSTAIN study.4Holz F.G. Amoaku W. Donate J. et al.Safety and efficacy of a flexible dosing regimen of ranibizumab in neovascular age-related macular degeneration: the SUSTAIN study.Ophthalmology. 2011; 118: 663-671Abstract Full Text Full Text PDF PubMed Scopus (332) Google Scholar In brief, retreatment was performed if there was either persistence or recurrence of intraretinal fluid or subretinal fluid on OCT, loss of ≥1 lines of vision, or the presence of new macular hemorrhage on clinical examination. At baseline, OCT was assessed for the presence of an intraretinal fluid (IRF) component (cysts or edema), subretinal fluid (SRF), pigment epithelial detachment (PED), and a significant band of reflectance at the retinal pigment epithelial (RPE) level, because it has been suggested that this is a sign of irreversible damage to the outer retina/RPE.5Fleckenstein M. Charbel Issa P. Helb H.M. et al.High-resolution spectral domain-OCT imaging in geographic atrophy associated with age-related macular degeneration.Invest Ophthalmol Vis Sci. 2008; 49: 4137-4144Crossref PubMed Scopus (233) Google Scholar After 3 injections, eyes without IRF or SRF were considered “dry,” as were eyes with PED without IRF or SRF. Regardless of the presence of PED, if either IRF or SRF were present, eyes were considered as having “persistent fluid” (Fig 1A, B, available at http://aaojournal.org). The presence or absence of a hyperreflective RPE band was considered separately as it could be present irrespective of the nature of the fluid status (Fig 1C, D). At baseline, 144 of 214 eyes (67.2%) had IRF with or without SRF and 70 (32.8%) eyes had SRF alone. Of 58 eyes (27.1%) that had PED, 38 had associated IRF, and 20 had additional SRF alone (Table 1, available at http://aaojournal.org). Best-corrected logMAR visual acuity (BCVA) at baseline was significantly worse in eyes with IRF (0.79 [Snellen equivalent, 20/122]; 95% confidence interval [CI], 0.68–0.90) compared with eyes with SRF alone (0.55 [20/72]; 95% CI, 0.34–0.75; P = 0.006). The presence of PED did not significantly impact on BCVA. Data at the 12-month visit were available for 187 eyes (87.4% of recruited eyes). No differences were seen with respect to the baseline population and the 187 eyes with complete data available at 12 months (P>0.10). Over 12 months, 26.2% of eyes gained ≥2 lines of BCVA compared with 19.3% ≥3 line gain in the SUSTAIN study.4 51.3% remained stable (±2 lines) and 22.5% lost ≥2 lines. The location of fluid at baseline did not significantly influence the likelihood of BCVA improvement at 12 months, although eyes with SRF at baseline had better BCVA (0.54 [20/69]; 95% CI, 0.34–0.75) compared with those with an IRF component (0.78 [20/120]; 95% CI, 0.63–0.93; P = 0.07). After the 3 initial injections, 63.1% of eyes were classified as being dry, whereas 26.2% had IRF and 10.7% had residual SRF only. The factors influencing BCVA change at 12 months are shown in Table 2 (available at http://aaojournal.org). After 3 injections, eyes that were dry had better BCVA at 12 months (0.61 [20/80]; 95% CI, 0.52–0.70) compared with those with residual IRF (0.95 [20/180]; 95% CI, 0.78–1.11; P = 0.05). Eyes with SRF alone (0.65 [20/90]; 95% CI, 0.39–0.91) were similar to those that were dry. The presence of persistent fluid was associated with a reduced chance of visual improvement at 12 months (18.6% vs 31.4%; P = 0.04). At baseline, 6 eyes demonstrated RPE hyperreflectance. This increased to 29 (15.5%) after 3 injections. Regardless of the presence or absence of fluid, hyperreflectivity conferred a greater likelihood of BCVA loss (48.3% vs 19.7%; P = .006) and poorer final BCVA (0.89 [20/155]; 95% CI, 0.69–1.10) compared with eyes without hyperreflectance (0.65 [20/90]; 95% CI, 0.56–0.73; P = 0.02). Two thirds (66%) of eyes with PED at baseline were noted to have residual PED after initial injections with no difference in final BCVA dependent on the PED. After 3 treatments, eyes with a residual IRF component were associated with a greater chance of BCVA loss at 12 months compared with those who were dry (odds ratio [OR], 6.08; 95% CI, 1.53–32.56; P = 0.01), whereas those with RPE hyperreflectivity were also more likely to lose BCVA (OR, 4.48; 95% CI, 1.11–16.41; P = 0.03; Table 3, Table 4, available at http://aaojournal.org). Our findings suggest that the presence of IRF fluid at baseline leads to worse presenting acuity and confers a significantly worse prognosis for visual outcome. Failure to clear fluid from the retina after the loading dose injections also correlates with poorer BCVA outcome and with a greater chance of losing vision at month 12 (Fig 2A, B, available at http://aaojournal.org). This may relate to disruption of retinal architecture with IRF as opposed to SRF, where the retina may be relatively intact. Failure to resolve fluid from the retina after initial treatment may be a marker of irreversible structural damage to the photoreceptor/RPE complex. Most eyes with persistent fluid after 3 injections also had persistent fluid at 12 months, suggesting that the ability to dry the retina is determined early in the course of treatment. Early in the course of treatment, RPE hyperreflectance also correlates with BCVA loss at 12 months. Most eyes with this feature had significant subretinal hemorrhage at baseline. Of the initial 214 eyes, 29 had significant hyperreflectance at the RPE level. Of these, some degree of atrophy or subretinal fibrosis was identified in 21 eyes, and of the remaining 8 eyes, 7 had residual subretinal hemorrhage, which by 12 months revealed atrophy (n = 2) or subretinal fibrosis (n = 5). Table 5 (available at http://aaojournal.org) details findings of eyes with RPE hyperreflectance. No association was seen with presence of PED at baseline or after 3 treatments and BCVA change at 12 months. Anti-VEGF treatment however, did reduce the frequency of PED compared with baseline. In conclusion, eyes with IRF after initial treatment may require more aggressive treatment strategies to achieve best possible outcome, whereas a significant RPE hyperreflective band limits the potential for visual improvement. Supplementary dataTable 1Background and Treatment Characteristics of 214 Eyes Included in the StudyCharacteristicn%Age group (yrs) <754320.1 75–8412457.9 ≥854722.0Gender Male8338.8 Female13161.2Smoking status, if known Never smoker8142.2 Smoker (past or present)11157.8Type of CNV lesion Predominantly classic5425.2 Nonpredominantly classic⁎Includes minimally classic, occult, retinal angiomatous proliferation, fibrovascular RPE detachment.16074.8Size of CNV (DA) ≥213663.5 >27836.5Baseline OCT appearance IRF with SRF14467.2 SRF7032.8 Hyperreflectivity of RPE62.8Treatment type Ranibizumab and bevacizumab combination7233.6 Ranibizumab14266.4Delay in seeking treatment, median weeks (IQR)12.0 (5.0–26.0)Baseline BCVA, median (IQR)0.60 (0.42–0.86)No. of injections at 12 months, median (IQR)5 (3–7)BCVA = best-corrected logMAR visual acuity; CNV = choroidal neovascularization; DA = disc areas; IQR = interquartile range; IRF = intraretinal fluid; OCT = optical coherence tomography; RPE = retinal pigment epithelium; SRF = subretinal fluid. Includes minimally classic, occult, retinal angiomatous proliferation, fibrovascular RPE detachment. Open table in a new tab Table 2Visual Outcome at 12 Months Based on Baseline and Treatment Characteristics of Study EyesCharacteristicChange in BCVA at 12 MonthsP ValueImproved BCVA (>2 Line Gain; n = 49; 26.2%)Stable BCVA (≤2 Line Gain or Loss; n = 96; 51.3%)Decreased BCVA (>2 Line Loss; n = 42; 22.5%)Age group, n (%), yrs0.37†Chi square test. <7514 (35.9)18 (46.2)7 (17.9) 75–8426 (24.5)58 (54.7)22 (20.8) ≥859 (21.4)20 (47.6)13 (31.0)Gender, n (%)0.01†Chi square test. Male23 (31.9)27 (37.5)22 (30.6) Female26 (22.6)69 (60.0)20 (17.4)Smoking status if known, n (%)0.38†Chi square test. Never smoker22 (31.934 (49.3)13 (18.8) Smoker (past or present)23 (23.0)52 (52.0)25 (25.0)Type of CNV lesion, n (%)0.13†Chi square test. Predominantly classic18 (36.7)20 (51.3)11 (22.4) Nonpredominantly classic⁎Includes minimally classic, occult, retinal angiomatous proliferation, fibrovascular RPE detachment.31 (22.5)76 (55.1)31 (22.5)Size of CNV, n (%), DA0.44†Chi square test. ≤234 (29.0)59 (50.4)24 (20.5) >216 (22.8)35 (50.0)19 (27.2)Delay in seeking treatment (weeks), median (IQR)11.5 (4.0–31.0)12.0 (4.5–26.0)12.0 (6.5–21.5)0.06‡Analysis of variance test.Baseline BCVA, median (IQR)0.69 (0.48–1.00)0.60 (0.33–0.78)0.60 (0.48–0.78)0.22‡Analysis of variance test.OCT appearance at baseline, n (%) IRF with or without SRF42 (29.2)66 (45.8)36 (25.0)0.49†Chi square test. SRF22 (31.4)38 (54.2)10 (14.2) Hyperreflectivity of RPE1 (17.0)3 (50.0)2 (33.0)0.54†Chi square test. Pigment epithelial detachment11 (19.0)27 (47.6)20 (32.4)0.26†Chi square test.Treatment type Ranibizumab/bevacizumab combination16 (23.5)36 (52.9)16 (23.5)0.74†Chi square test. Ranibizumab34 (28.6)60 (50.4)25 (21.0)No. of injections at 12 months, median (IQR)6 (4–8)5 (4–7)4 (3–5)0.01‡Analysis of variance test.Residual fluid on OCT at 3 months, n (%) No residual fluid/dry37 (31.4)59 (50.0)22 (18.6) IRF with or without SRF9 (18.4)22 (44.9)18 (36.7)0.15†Chi square test. SRF4 (20.0)9 (45.0)7 (35.0)Pigment epithelial detachment on OCT at 3 months, n (%) Yes10 (25.8)17 (45.2)11 (29.0)0.85†Chi square test. No40 (26.7)73 (49.2)36 (24.2)RPE hyperreflectance on OCT at 3 months, n (%) Hyperreflectivity of RPE5 (17.2)10 (34.5)14 (48.3) No hyperreflectivity45 (28.7)82 (51.6)31 (19.7)0.006†Chi square test.BCVA = best-corrected LogMAR visual acuity; CNV = choroidal neovascularization; DA = disc areas; OCT = optical coherence tomography; IQR = interquartile range; IRF = intraretinal fluid; RPE = retinal pigment epithelium; SRF = subretinal fluid. Includes minimally classic, occult, retinal angiomatous proliferation, fibrovascular RPE detachment.† Chi square test.‡ Analysis of variance test. Open table in a new tab Table 3Results of Generalized Estimating Equations Modeling of the Effects of Patient and Treatment Characteristics on Visual Acuity Outcomes at 12 MonthsPredictorImproved vs Stable BCVA⁎Reference group is improved VA.Improved vs Decreased BCVA⁎Reference group is improved VA.OR95% CIP ValueOR95% CIP ValueOCT appearance No residual fluid (dry)1.00Reference1.00Reference IRF with or without SRF1.930.56–6.300.276.081.53–24.140.01 SRF1.620.36–7.300.534.310.40–46.920.23Male gender1.860.72–4.800.200.520.16–1.720.28No. of injections0.870.70–1.070.190.670.50–0.880.004Delay in treatment0.980.96–1.000.150.970.95–1.020.06BCVA = best-corrected LogMAR visual acuity; CI = confidence interval; IRF = intraretinal fluid; OCT = optical coherence tomography; OR = odds ratio; SRF = subretinal fluid. Reference group is improved VA. Open table in a new tab Table 4Results of Generalized Estimating Equations Modeling of the Effects of Patient and Treatment Characteristics on Visual Acuity Outcome at 12 MonthsPredictorImproved vs. Stable BCVA⁎Reference group is improved VA.Improved vs. Decreased BCVA⁎Reference group is improved VA.OR95% CIP-ValueOR95% CIP-ValueOCT appearance No hyperreflectivity of RPE1.00Reference1.00Reference Hyperreflectivity of RPE1.320.26–6.660.744.481.11–16.410.03Male gender1.600.62–4.130.330.440.14–1.390.16Number of injections0.850.69–1.040.120.740.56–0.980.03Delay in treatment0.990.96–1.010.200.970.95–1.020.06BCVA = best-corrected logMAR visual acuity; OR = odds ratio; CI = confidence interval; SRF = subretinal fluid; IRF = intraretinal fluid. Reference group is improved VA. Open table in a new tab Figure 2A, Change in mean visual acuity over time, depending on location of fluid within the macula on optical coherence tomography. B, Change in mean visual acuity over time, depending on presence and location of fluid within the macula on optical coherence tomography after 3 antivascular growth factor treatments.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table 5Clinical Features of Patients with a Band of Hyperreflectivity at the Level of the Retinal Pigment Epithelium (RPE) on Optical Coherence Tomography (OCT) after 3 Antivascular Endothelial Growth Factor TreatmentsOCT AppearancesClinical AppearanceFibrosisAtrophyChronic PEDSubretinal HemorrhageHyperreflectivity with residual fluid (n = 14)8105Hyperreflectivity with no residual fluid (n = 15)4812atrophy = window defect in early phase of angiogram with loss of retinal pigment epithelium clinically; fibrosis = >25% of lesion is fibrotic scar; PED = retinal pigment epithelial detachment; subretinal hemorrhage = large subfoveal hemorrhage with significant masking of details on angiography. Open table in a new tab BCVA = best-corrected logMAR visual acuity; CNV = choroidal neovascularization; DA = disc areas; IQR = interquartile range; IRF = intraretinal fluid; OCT = optical coherence tomography; RPE = retinal pigment epithelium; SRF = subretinal fluid. BCVA = best-corrected LogMAR visual acuity; CNV = choroidal neovascularization; DA = disc areas; OCT = optical coherence tomography; IQR = interquartile range; IRF = intraretinal fluid; RPE = retinal pigment epithelium; SRF = subretinal fluid. BCVA = best-corrected LogMAR visual acuity; CI = confidence interval; IRF = intraretinal fluid; OCT = optical coherence tomography; OR = odds ratio; SRF = subretinal fluid. BCVA = best-corrected logMAR visual acuity; OR = odds ratio; CI = confidence interval; SRF = subretinal fluid; IRF = intraretinal fluid. atrophy = window defect in early phase of angiogram with loss of retinal pigment epithelium clinically; fibrosis = >25% of lesion is fibrotic scar; PED = retinal pigment epithelial detachment; subretinal hemorrhage = large subfoveal hemorrhage with significant masking of details on angiography. Safety and Efficacy of a Flexible Dosing Regimen of Ranibizumab in Neovascular Age-Related Macular Degeneration: The SUSTAIN StudyOphthalmologyVol. 118Issue 4PreviewTo evaluate the safety and efficacy of individualized ranibizumab treatment in patients with neovascular age-related macular degeneration. Full-Text PDF