Contrast Sensitivity Outcomes in the ABC Trial: A Randomized Trial of Bevacizumab for Neovascular Age-Related Macular Degeneration

Abstract

Purpose.: To report the impact of intravitreous bevacizumab therapy on contrast sensitivity in patients with neovascular age-related macular degeneration (nAMD).

Methods.: This was a prospective, multicenter, double-masked, randomized, controlled trial of 131 patients with nAMD. The patients with nAMD had received intravitreal bevacizumab (n = 65) or standard therapy (n = 66) in the study eye with a 6-week cycle of assessment. The bevacizumab treatment was 1.25 mg/0.05 mL, given as three initial treatments with further retreatment as needed according to standard retreatment criteria and a 1-year (54-week) follow-up. Contrast sensitivity was determined during the study using a Pelli-Robson chart.

Results.: At the week-54 examination, bevacizumab-treated patients were more likely to gain at least 6 letters or more of contrast sensitivity than the patients receiving standard care (23 [35.4%] versus 10 [15.2%], P = 0.009). In addition the bevacizumab-treated patients were less likely to lose 6 or more letters with a better mean letter change at week 54 than the patients receiving standard care (3 [4.6%] versus 14 [21.2%], and +4.0 versus −0.7 letters; P < 0.05 for both comparisons).

Conclusions.: Consistent with the visual acuity outcomes, bevacizumab improved the chances of a clinically relevant gain in contrast sensitivity in the study population. Given the association between contrast sensitivity and visual disability, the beneficial effects of bevacizumab therapy on contrast sensitivity outcomes are expected to have a favorable impact on patients' daily activities. (www.controlled-trials.com number, ISRCTN83325075.)

The management of neovascular age-related macular degeneration has been transformed by the introduction of agents that block the action of vascular endothelial growth factor-A (VEGF). ^{1 –4} The most commonly used agents ranibizumab (Lucentis, Genentech, Inc., San Francisco, CA) and bevacizumab (Avastin, Genentech, Inc.) have been shown to improve visual acuity in patients with nAMD when administered by intravitreous injection. However, although loss of contrast sensitivity is a frequent consequence of nAMD ⁵ and can have a serious impact on quality of life and functional ability, ⁶ there have been no reports of the impact of anti-VEGF therapy on contrast sensitivity. 

Contrast sensitivity is an important additional measure of visual function in patients with subfoveal nAMD, and both measures provide complementary information about visual function. While distance visual acuity provides a measure of the ability to resolve detail at high contrast, contrast sensitivity describes the ability to see low-contrast patterns. In particular, contrast sensitivity may be a better predictor of performance in tasks requiring distance judgment of real-world targets (recognizing faces, ⁷ road signs and objects ⁸ ), night driving and mobility over conventional visual acuity. Several studies have shown only a moderate correlation between visual acuity and contrast sensitivity both in population-based studies ^9,10 and in patients with nAMD. ¹¹ Both measures of visual function are independently associated with difficulties in performing everyday activities. ⁹ Furthermore, contrast sensitivity may better relate to health-related quality of life in patients with age-related macular degeneration. ¹² It is therefore important to consider the impact of newer therapies on contrast sensitivity in addition to the effect on visual acuity. 

In a report from the TAP study, the effect of verteporfin therapy on contrast sensitivity in patients with nAMD was described, ⁵ but there have been no reports of the impact of anti-VEGF therapy on contrast sensitivity in clinical trials for nAMD. In a recent report, we presented the visual acuity results from the first double-masked, randomized, controlled trial of intravitreous bevacizumab for the treatment of nAMD. ⁴ In this article, we present the contrast sensitivity outcomes from the ABC Trial (Avastin [bevacizumab] for Choroidal Neovascularization) Trial and describe the impact of bevacizumab on contrast sensitivity in patients with subfoveal nAMD. 

Method

Study Design

The design of the ABC Trial and the treatments administered have been described in detail previously. ¹³ All patients had provided informed consent to participate in the clinical trial, and the research adhered to the tenets of the Declaration of Helsinki with ethics approval in place before trial recruitment. In brief, this was a double-masked, randomized, controlled trial with two parallel treatment groups. The major inclusion criteria for the study eye were the diagnosis of previously untreated nAMD with a baseline protocol visual acuity letter score of 25 to 70 letters (1 meter equivalent) using the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at 4 m (Snellen equivalent of approximately 20/40 to 20/320) with evidence of macular thickening on time-domain OCT imaging and no permanent structural damage to the fovea. Eligible patients were randomized in a 1:1 ratio to receive either intravitreal bevacizumab or standard care (PDT [photodynamic therapy], pegaptanib intravitreal injections, or sham treatment). 

Patients and observing investigators and staff were masked to the treatment assignment at all visits. Only one eye per patient was included in the study, and it was selected before randomization. Standard care was determined before trial enrollment, at which point the patients were allocated to treatment groups by minimization, a dynamic process that reduces the imbalance between trial arms with respect to standard treatment eligibility and site. During trial recruitment, patients with well-defined (classic, no occult, or predominantly classic CNV) were randomized to either bevacizumab or PDT (standard care). PDT was performed at baseline and then at 3-month intervals as per the phase III trials of verteporfin. ¹⁴ In the absence of national funding for treatment of minimally classic or occult CNV patients with these lesion subtypes were randomized to either standard care (pegaptanib or sham treatment, based on National Health Service funding of pegaptanib therapy) or bevacizumab. Pegaptanib was given as per the product license with a dose of 0.3 mg in 0.09 mL of pegaptanib given as continuous therapy by intravitreal injection every 6 weeks for 1 year (nine injections in 54 weeks). Sham treatment was used as standard care for patients with minimally classic or occult CNV when no National Health Service funding for treatment was available. In addition, as the treatment of patients randomized to bevacizumab or pegaptanib involves the comparison of a pro re nata (prn) treatment (bevacizumab) to a treatment given continuously (nine intravitreal injections of pegaptanib at 6-week intervals), to maintain masking, sham treatments were given to patients randomized to bevacizumab not requiring intravitreal treatment at that visit (weeks 18–48). Placebo PDT was used in patients with predominantly classic CNV randomized to bevacizumab to maintain masking. A summary of the enrollment and randomization process has been published in detail. ¹³ The study was conducted before ranibizumab was available and was funded by the National Health Service in the UK. Patients returned for follow-up examination at 6-week intervals (after the initial review, 1 week after the first treatment). Three intravitreous bevacizumab treatments were given in the loading phase of treatment (at weeks 0, 6, and 12) with the potential for further injections at weeks 18, 24, 30, 36, 42, and 48 if standardized retreatment criteria were met (minimum of three and maximum of nine injections). The Patients exited the study after the 54-week (∼1 year) follow-up. 

Outcome Measures

Contrast sensitivity was measured at baseline and at weeks 1, 12, 24, 36, 48, and week 54 using a standardized protocol. Visual acuity was measured using Early Treatment Diabetic Retinopathy Study (ETDRS) charts after refraction to obtain best corrected visual acuity, as published previously. Contrast sensitivity measurement was performed after visual acuity measurements, with the Pelli-Robson chart (Clement Clarke Inc., Harlow, UK) at a distance of 1 m and chart luminance of 80 to 120 cd/m². The right eye was tested, followed by the left eye, on charts 1 and 2, respectively, with +0.75 D added to the patient's refraction. The patient was asked to name each letter on the chart starting with the high-contrast letters in the top left-hand corner and reading horizontally across the entire line. As low-contrast letters can take some time to appear, the patient was given instructions to keep looking and not to give up too soon. The examiner circled each letter read correctly and crossed out each letter read incorrectly. The test was stopped when the patient failed to correctly identify 2 or more letters correctly in a triplet. The letter-by-letter scoring advocated by Elliot et al. ¹⁵ was used to score the test in view of the better repeatability of this method of scoring. 

Data Analysis

All efficacy analyses were performed with one of two commercial programs (Prism ver. 5.00 for Windows; (GraphPad Software, San Diego, CA, and SPSS ver. 16.0 for Windows; SPSS Inc., Chicago, IL) with a strict intent-to-treat definition that included all data for all randomized patients. The last observation was carried forward for missing efficacy values, for all but the time-to-event analyses. In keeping with other trials, the proportion of patients who gained or lost 6 letters or more in each group at 54 weeks was calculated with 95% confidence intervals (CIs) computed by the exact binomial method. A Fisher's exact test was conducted to assess whether the observed differences in proportions were statistically significant. A 6-letter loss of contrast sensitivity translates into a doubling of the amount of contrast necessary to see the test letters and has approximately the same impact on self-reported visual disability as a 15-letter loss of visual acuity (doubling of visual angle). ¹⁶ For example, a 6-letter loss of contrast sensitivity increases the odds of reporting difficulty with everyday visual tasks by a factor of three to five. In previous work we confirmed that a 6-letter change is consistent with clinical change rather that test–retest variability in patients with AMD. ¹⁷ Similarly, the proportion of patients gaining or losing any letters or 15 letters or more of contrast sensitivity were also reported with a Fisher's exact test where appropriate. A 15-letter loss of contrast sensitivity translates into a 5.5-fold increase in the contrast required to see the test letters, and has approximately the same impact on visual disability as a 30-letter loss of visual acuity (quadrupling of visual angle). ¹⁶  

Mean changes in contrast sensitivity from baseline were compared using a two-sample t-test. The time-to-event analyses of the times to a gain or loss of 6 or more letters of contrast sensitivity were performed using the Kaplan-Meier method and the log-rank test. A loss or gain of 6 or more letters at a visit was counted as an event only if it was confirmed at the last visit. Additional prespecified exploratory analyses were conducted to report the change in contrast sensitivity and the proportion of eyes that gained or lost at least 6 letters of contrast sensitivity in patients categorized into subgroups by lesion composition (predominantly classic, minimally classic, or occult). We also examined the relationship between contrast sensitivity and visual acuity at baseline and the relationship between change in CS and change in VA with therapy, using the Pearson correlation coefficient. 

Results

Baseline Characteristics

Between August 2006 and November 2007, 131 patients were enrolled in the present study and randomly allocated to two intervention groups (65 eyes of 65 patients to bevacizumab and 66 of 66 patients to standard care). The baseline characteristics of patients are described in detail elsewhere. ⁴ In brief, the characteristics of the bevacizumab-treated and the patients treated with standard care were similar (Table 1). The study follow-up for both groups was high (bevacizumab group 98% and standard care group 94% at week 54). There was a weak correlation between contrast sensitivity and visual acuity at baseline (r = 0.2). 

Table 1.

 

Baseline Characteristics

Characteristic	Standard Therapy (n = 66)*	Bevacizumab (n = 65)
Sex, % (n)
Male	37.9 (25)	40.0 (26)
Female	62.1 (41)	60.0 (39)
Mean age, y	81	79
Contrast sensitivity in study eye (n letters)
Mean	19.4	20.2
Median (IQR)	21 (16, 24)	21 (16, 24)
ETDRS visual acuity score in study eye (n letters)
Mean	52.8	50.9
Median (IQR)	53 (47, 60)	50 (43, 61)
Type of choroidal neovascularization, % (n)
Minimally classic/occult	75.8 (50)	75.4 (49)
Predominantly classic	24.2 (16)	24.6 (16)
OCT central 1-mm macular thickness
Median (IQR)	330 (256, 359)	328 (271, 376)

 Some data were taken from a previous report. ⁴ IQR, interquartile range.

*  Pegabtanib, n = 38; verteporfin, n = 16; sham, n = 12.

Contrast Sensitivity Outcomes

In the total study population, patients who received bevacizumab intravitreous injections gained significantly more letters of contrast sensitivity on the Pelli-Robson chart compared with patients receiving standard care (Fig. 1) with no fall-off in gain over the variable retreatment phase of the study (weeks 18–48). 

Figure 1.

 

Mean changes in contrast sensitivity from baseline in the total study population. The last-observation-carried-forward method was used if there were missing values.

The patients who received bevacizumab were more likely to gain at least 6 or 15 letters of contrast sensitivity than were the patients who received standard care over 54 weeks (Table 2). In the total study population, 23 bevacizumab-treated patients (35.4%) gained six or more letters at week 54 (95% CI, 24%–48%), compared with 10 patients (15.2%) given standard care (95% CI, 7.5%–26%; P = 0.009). Similarly at week 54, 3 (4.6%) bevacizumab-treated patients (95% CI, 1%–13%) and 14 (21.2%) patients given standard care lost 6 or more letters (95% CI, 12%–33%; P = 0.008). 

Table 2.

 

Summary of Contrast Sensitivity Outcomes

Outcome	Bevacizumab (n = 65)	Standard Care (n = 66)	P
Patients who gained ≥15 letters	4 (6.2)	2 (3.0)	0.44
Patients who gained ≥6 letters	23 (35.4)	10 (15.2)	<0.01
Patients who gained any letters	47 (72.3)	23 (34.8)	<0.01
Patients who lost any letters	12 (18.5)	33 (50.0)	<0.01
Patients who lost ≥6 letters	3 (4.6)	14 (21.2)	<0.01
Patients who lost ≥15 letters	2 (3.0)	2 (3.0)	—

 Data are expressed as n (%). The last-observation-carried-forward method was used if there were missing values.

The mean change in contrast sensitivity letter score at week 54 compared with baseline was +4.0 letters for the bevacizumab group (95% CI, +2.4 to +5.7 letters, P < 0.0001) and −0.7 letters for the standard care group (95% CI, −2.5 to +1.0 letters, P = 0.41). 

The time-to-event analysis of the time to a gain or loss of at least 6 letters of contrast sensitivity also showed a significant treatment benefit for patients who received bevacizumab therapy (Figs. 2, 3). These treatment benefits were seen as early as week 12 and were sustained after the loading phase of treatment over the variable dosage phase of the study. These outcomes were achieved with a mean of seven intravitreous bevacizumab injections (median 7; range, 3–9). For the standard care group, for the patients receiving pegaptanib the mean number of pegaptanib injections was 8.9 (median 9; range, 6–9). Including the required administration at day 0, active verteporfin PDT was administered a mean of 3.2 times (median, 3; range, 2–5) in the verteporfin standard care group and placebo verteporfin administered a mean of 2.5 times (median. 3; range, 1–5) in the bevacizumab group. The mean number of sham treatments was 7.3 (median, 9; range, 3–9) in the 12 patients randomized to sham intravitreal injections in the standard care group. 

Figure 2.

 

Kaplan-Meier estimate of the time until a gain of at least 6 letters of contrast sensitivity relative to baseline. *Log-rank test.

Figure 3.

 

Kaplan-Meier estimate of the time until a loss of at least 6 letters of contrast sensitivity relative to baseline. *Log-rank test.

There was a moderate correlation between the change in contrast sensitivity and the change in visual acuity with therapy at week 54 (r = 0.4). 

Prespecified Subgroup Analyses

Contrast sensitivity outcomes were better in the bevacizumab-treated patients both in the subgroup with predominantly classic lesions and in the subgroup with minimally classic or occult lesions at baseline, compared with standard care. At the week-54 examination, the patients with predominantly classic lesions at baseline who received standard care (n = 16) gained a mean of 0.4 letters (median, 0 letters) of contrast sensitivity, whereas the bevacizumab-treated patients (n = 16) gained 5.9 letters (median gain, 4.0 letters). In addition, the patients with predominantly classic lesions who received bevacizumab were more likely to have gained at least 6 letters and were less likely to have lost at least 6 letters of contrast sensitivity at week 54. In this subgroup, five (31.3%) of the bevacizumab-treated patients had gained at least 6 letters of contrast sensitivity at week 54, compared with two (12.5%) of the standard-care patients, with no bevacizumab-treated patients losing 6 letters of contrast sensitivity or more at week 54, compared with three (18.8%) of the standard-care patients. 

At the week-54 examination, the patients with minimally classic or occult lesions at baseline treated with standard care (n = 50) lost a mean of 1.1 letters (median, 1 letter loss) of contrast sensitivity, whereas the bevacizumab-treated patients (n = 49) gained 3.8 letters (median gain, 4.0 letters). The patients with minimally classic or occult lesions who received bevacizumab were more likely to gain at least 6 letters and were less likely to lose at least 6 letters of contrast sensitivity at week 54. In this subgroup, 18 (30.5%) of the bevacizumab-treated patients gained at least 6 letters of contrast sensitivity at week 54, compared with 7 (14%) of the standard-care patients, and 3 (6.1%) of the bevacizumab-treated patients and 3 standard-care patients (6.0%) lost 6 letters of contrast sensitivity or more at week 54. Contrast sensitivity change by standard care treatment is presented as plots of mean contrast sensitivity change (± SE) for bevacizumab against PDT or pegaptanib (Supplementary Fig. S1). 

Discussion

The results from the ABC Trial show that bevacizumab intravitreous injections, given as part of a variable-dose strategy, led to a greater chance of improved contrast sensitivity, a reduced risk of loss of contrast sensitivity, and improved mean contrast sensitivity at 1 year in patients with nAMD compared with standard care. In addition, visual acuity in more than 35% of the bevacizumab-treated patients improved 6 or more letters, a threshold that exceeds the variability of the measurement of contrast sensitivity. ¹⁷ The gain in contrast sensitivity was maintained throughout the variable-dosage phase of the study (after three fixed bevacizumab injections) up to week 54. 

The results from the TAP study with verteporfin showed a mean loss of 0.4 letters of contrast sensitivity at 12 months in patients with predominantly classic CNV. The mean change in the 16 patients with predominantly classic CNV randomized to PDT was not dissimilar to the outcome at 12 months in the TAP study, with a mean gain of 0.4 letters. In comparison, this study shows that bevacizumab improved contrast sensitivity with a mean gain of 4 letters at week 54 in the cohort overall, compared with baseline, with a mean change of +4.9 letters in the 16 patients with predominantly classic CNV randomized to bevacizumab. 

The overall gain in contrast sensitivity with bevacizumab in this study is consistent with the visual acuity outcomes, with no loss of vision function during the variable-dosage phase of the study (from week 18 onward) using the ABC Trial retreatment criteria (Patel PJ, Tufail A, manuscript submitted) with OCT guided retreatment. Although exploratory analyses must be interpreted with caution, even when prespecified, it is reassuring to see that all subtypes of CNV included in this study respond to bevacizumab treatment with a gain in contrast sensitivity. The patients randomized to pegaptanib showed loss of contrast sensitivity. 

The relatively modest correlation between contrast sensitivity and visual acuity in this report from the ABC Trial both at baseline and when evaluating change in these vision function measures with therapy, strengthens the rationale for including contrast sensitivity as an outcome measure in clinical trials for nAMD. 

Limitations of this study include the mix of treatments in the comparator arm, making it difficult to compare the effect of bevacizumab against a single comparator. However, the lack of a single treatment in the comparator arm reflects the treatment of patients in the United Kingdom at the time of the study, and none of the treatments in the standard therapy arm have been shown to improve contrast sensitivity in randomized controlled trials. The lack of a vision related quality of life questionnaire in this study is a further limitation, as it is difficult to draw definitive conclusions regarding the impact of treatment benefit on quality of life. Previous work however, suggests that contrast sensitivity relates better to health related quality of life than high contrast, distance visual acuity in age-related macular degeneration. ¹²  

Strengths of the study include the double-masked, multicenter, randomized design from which an unbiased estimate of the effect of bevacizumab on contrast sensitivity has been obtained. In addition, unlike most published phase IIII clinical trials, the ABC Trial involved noncontinuous dosing rather than a continuous regimen. This is a strength of the study, as it is an attempt to make the results more translatable to the real world by delivering individualized therapy using OCT-guided retreatment. Using 6 weekly study visits reduces the number of hospital visits by one third over conventional 4-week cycles of assessment. This consideration is important in elderly population with developing nAMD. 

Although ranibizumab is the licensed drug for the treatment for nAMD, bevacizumab is widely used worldwide as an off-label treatment. It is therefore important to understand the effect of this treatment on visual function. This report is the first to describe the effect of anti-VEGF agents on contrast sensitivity, an important measure of visual function that provides complementary information to high-contrast distance visual acuity measurement and is better correlated with difficulty in practical daily tasks. The results show that bevacizumab therapy, given as part of the ABC Trial OCT-guided, variable treatment strategy leads to improvement in contrast sensitivity in patients with nAMD. 

Supplementary Materials

Text s01, TIF - Text s01, TIF 

Footnotes

 Supported by The Special Trustees of Moorfields Eye Hospital. This research has received a proportion of its funding from the Department of Health's NIHR Biomedical Research Centre for Ophthalmology at Moorfields Eye Hospital and UCL Institute of Ophthalmology. The views expressed in the publication are those of the authors and not necessarily those of the Department of Health.

Footnotes

 Disclosure: P.J. Patel, None; F.K. Chen, None; L. Da Cruz, None; G.S. Rubin, None; A. Tufail, Novartis (C), GSK (C), Allergan (C), Pfizer (C)

Footnotes

 No reprints will be available.

Appendix

ABC Trial Study Group

Chief Investigator: Adnan Tufail 

Trial Manager: Laura Henderson and Ola Segun-Odumosu 

NIHR Biomedical Research Centre for Ophthalmology, Moorfields Eye Hospital and UCL Institute of Ophthalmology, London UK: Adnan Tufail, consultant ophthalmologist, medical retina department, Praveen Patel, principal investigator and locum consultant ophthalmologists, and Catherine Egan, Philip Hykin, Lyndon da Cruz, Zdenek Gregor, and Jonathan Dowler, consultant ophthalmologists 

Bristol Eye Hospital, Bristol, UK: Mohammed Majid, principal investigator and consultant ophthalmologist; Clare Bailey, consultant ophthalmologist 

Gloucestershire Eye Department, Cheltenham General Hospital, Cheltenham, UK: Quresh Mohamed, principal investigator and consultant ophthalmologist; Robert Johnston, consultant ophthalmologist 

Data management: Catey Bunce, senior statistician, Wen Xing, statistician, and Richard Seebaran, data support 

Pharmacy/drug management: Jagdev Bains and Moorfields Pharmaceuticals 

Trial staff in the Clinical Trials Unit: Kerry Waller, Felicia Ikeji, Matthew Richardson, and Kanom Bibi 

Optometry: Dan Ehrlich, Catherine Grigg, Sonal Rughani, Jen Smith, Shima Shah, Mina Devani, Graham Brown, Rebecca Black 

Treating doctors: Sobha Sivaprasad, Andrew Browning, Yvonne D'Souza, Nachi Acharya, Andrew Kaines, Sam Fraser-Bell, Maria Niskopolou, Noel Horgan, Fred K. Chen, Waheeda Rahman, Rajen Gupta, Richard Hanson, Tariq Aslam, Mohammed Musadiq, Tryfon Rotsos, Gayatri Banerjee 

Image acquisition: Felicia Ikeji, Matthew Richardson, Kulwant Sehmi, and Richard Poynter 

Central angiographic resource facility, Queen's University of Belfast: Alison Murphy, Nicola Duff, and Liam Patton 

Data and Safety Monitoring Committee: Marion Campbell, chair, Frank Holz, and Robyn Guymer 

Steering Committee: Astrid Fletcher, chair, Catey Bunce, Praveen Patel, Adnan Tufail, and Richard Wormald