EFFECT OF RANIBIZUMAB
`RETREATMENT FREQUENCY ON
`NEUROSENSORY RETINAL VOLUME IN
`NEOVASCULAR AMD
`
`PEARSE A. KEANE, MRCOPHTH, MSC,* KAREN T. CHANG, BS,*
`SANDRA LIAKOPOULOS, MD,*† RENU V. JIVRAJKA, BS,*
`ALEXANDER C. WALSH, MD,* SRINIVAS R. SADDA, MD*
`
`Purpose: To determine the characteristics of patients with neovascular age-related
`macular degeneration who show initial anatomic improvements on optical coherence
`tomography in response to treatment with ranibizumab, but who subsequently regress
`toward their anatomic baseline.
`Methods: Data from 50 consecutive patients, receiving ranibizumab therapy for neo-
`vascular age-related macular degeneration, were collected. Raw StratusOCT images were
`analyzed using custom software (“OCTOR”). Changes in volume of neurosensory retina at
`months 1, 3, and 6 were calculated. Baseline demographic and morphologic characteris-
`tics were compared.
`Results: Forty-two patients (84%) showed a reduction in total retinal volume 1 month
`after initial treatment with ranibizumab. Of the patients that initially showed a reduction, 16
`(38%) maintained this reduction through month 6, whereas 26 patients (62%) demon-
`strated a subsequent increase in retinal volume. Patients who maintained a reduction in
`edema received 3.75 ⫾ 1.18 injections of ranibizumab versus 2.96 ⫾ 1.34 injections for
`patients who did not (P ⫽ 0.049). Regression of initial anatomic improvements was
`associated with worsening of visual acuity (r ⫽ 0.599, P ⫽ 0.002).
`Conclusion: Patients receiving fewer injections of ranibizumab appeared less likely to
`maintain anatomic improvements achieved following commencement of ranibizumab ther-
`apy; regression of these improvements was associated with deterioration in visual acuity.
`RETINA 29:592– 600, 2009
`
`Ranibizumab (Lucentis; Genentech, Inc., San Fran-
`
`cisco, CA), an antibody fragment that blocks all
`isoforms of vascular endothelial growth factor-A, is
`the first treatment for neovascular age-related macular
`degeneration (AMD) that has been shown to lead to
`significant improvements in visual acuity in Phase III
`clinical trials.1,2 In the MARINA and ANCHOR trials,
`
`From the *Doheny Image Reading Center, Doheny Eye Institute,
`Keck School of Medicine of the University of Southern California,
`Los Angeles, California; and †Department for Vitreoretinal Sur-
`gery, Center of Ophthalmology, University of Cologne, Germany.
`Supported in part by NIH Grant EY03040 and NEI Grant R01
`EY014375.
`Drs. Walsh and Sadda are coinventors of Doheny intellectual
`property related to optical coherence tomography that has been
`licensed by Topcon Medical Systems. However, it is not related to
`the article’s subject matter.
`Reprint requests: Srinivas R. Sadda, MD, Doheny Eye Institute-
`DEI 3623, 1450 San Pablo Street, Los Angeles, CA 90033; e-mail:
`sadda@usc.edu
`
`patients received monthly injections of ranibizumab
`over a 2-year period—a retreatment regimen that has
`both practical and economic implications, particularly
`with the availability of inexpensive alternatives, such
`as bevacizumab (Avastin; Genentech).3,4 As a result,
`much effort has focused on the development of alter-
`native treatment regimens, which would reduce the
`number of injections required, while maintaining the
`efficacy demonstrated in the MARINA and ANCHOR
`studies.5,6
`Optical coherence tomography (OCT) formed the
`basis for
`retreatment
`in the recently published
`PrONTO study,5 and OCT-guided retreatment proto-
`cols have become widely adopted in clinical practice.
`Caution is required, however, when utilizing quanti-
`tative information from OCT in a clinical setting, as
`significant errors are known to occur in data obtained
`from the widely used StratusOCT analysis software.7
`
`592
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 1
`
`Joining Petitioner: Apotex
`
`

`

`RANIBIZUMAB RETREATMENT FREQUENCY IN AMD ● KEANE ET AL
`
`593
`
`To address this issue, we have previously developed
`publicly available custom image-measurement soft-
`ware, entitled “OCTOR.”8,9 This software facilitates
`accurate and reproducible manual segmentation of
`OCT images and allows quantitative analysis of any
`area of interest in these images, e.g., retina, subretinal
`fluid, subretinal tissue, or pigment epithelium detach-
`ment (PED).
`Quantitative OCT data derived from accurate seg-
`mentation of the neurosensory retina may facilitate
`retreatment decisions, although its precise role re-
`mains unresolved.5,10 Using OCTOR, we have previ-
`ously reported results from a cohort of patients treated
`with ranibizumab according to an OCT-guided retreat-
`ment regimen.11 In this previous study, the effect of
`ranibizumab on the neurosensory retina appeared to be
`attenuated over time—a finding we speculated was
`secondary to treatment tachyphylaxis. In the present
`study, we aim to determine if there are differences in
`the characteristics of patients with neovascular AMD
`who show anatomic regression after initial response to
`therapy, versus those who maintain anatomic im-
`provement, as determined by OCT.
`
`using the export feature available in the StratusOCT
`version 4.0 analysis software.
`The number and type of any previous treatments for
`CNV secondary to AMD in the study eye were re-
`corded. After their initial intravitreal injection, pa-
`tients were treated at the discretion of the physician
`based on assessment of disease activity as determined
`by OCT and/or fluorescein angiography, and utilizing
`many of the criteria specified in the PrONTO study.5
`In the PrONTO study, ranibizumab retreatment was
`performed if one of the following changes was ob-
`served: a loss of five letters in conjunction with fluid
`in the macula as detected by OCT, an increase in OCT
`central retinal thickness of at least 100 ␮m, new-onset
`classic CNV, new macular hemorrhage, or persistent
`macular fluid detected by OCT at least 1 month after
`the previous injection of ranibizumab. The number
`and timing of retreatments were recorded for each
`patient. Other data collected included patient age and
`gender, as well as best-corrected Snellen visual acuity
`at the time of initial ranibizumab injection. Fluores-
`cein angiographic images taken for each patient at the
`time of initial treatment were also collected.
`
`Materials and Methods
`
`Baseline Patient Characteristics
`
`Data Collection
`
`For this retrospective study, we collected and re-
`viewed data from all patients receiving their initial
`intravitreal injections of ranibizumab at the Doheny
`Eye Institute between July 2006 and September 2007.
`Approval for data collection and analysis was ob-
`tained from the institutional review board of the Uni-
`versity of Southern California. The research adhered
`to the tenets set forth in the Declaration of Helsinki.
`For inclusion in the study, eyes were required to have
`subfoveal choroidal neovascularization (CNV) second-
`ary to AMD; StratusOCT imaging performed before
`their initial treatment with ranibizumab; and a total of
`6 months of subsequent follow-up, with StratusOCT
`imaging at each visit. Any patient switched to an
`alternative treatment for neovascular AMD, during
`this follow-up period, was excluded from the study.
`StratusOCT images were collected at baseline and
`at Months 1, 3, and 6 after initial treatment with ranibi-
`zumab. Images were obtained with a single StratusOCT
`machine using the Radial Lines protocol of six high-
`resolution B-scans. The Fast Macular Thickness pro-
`tocol was used only when photographers were unable
`to obtain adequate high-resolution images, most com-
`monly in patients with unstable fixation or poor co-
`operation. Data for each case were exported to disk
`
`Fifty consecutive patients, receiving their initial
`intravitreal injections of ranibizumab at the Doheny
`Eye Institute, and meeting the inclusion criteria were
`included in this study. Two patients were excluded
`from consideration as they were switched to an alter-
`native treatment (intravitreal bevacizumab) during the
`course of follow-up. Of the 50 patients included, 30
`(60%) were women, whereas 20 (40%) were men. The
`mean age of patients was 80 years (SD ⫽ 7), whereas
`the median age was 81 years (range, 63–92 years).
`Mean visual acuity at time of initial treatment with
`ranibizumab was 20/123. Twenty-eight (56%) patients
`had undergone prior treatment for CNV in their study
`eye, consisting of photodynamic therapy with verte-
`porfin, thermal laser photocoagulation, pegaptanib, or
`bevacizumab.
`
`Computer-Assisted Grading Software
`
`The software used for OCT analysis (entitled
`“OCTOR”) was written by Doheny Image Reading Cen-
`ter software engineers to facilitate viewing and manual
`grading. OCTOR is publicly accessible at http://www.
`diesel.la (accessed June 23rd, 2008) and has been de-
`scribed and validated in previous reports.8,9,11–13 This
`software, which effectively operates as a painting pro-
`gram and calculator, imports data exported from the
`StratusOCT machine and allows the grader to use a
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 2
`
`Joining Petitioner: Apotex
`
`

`

`594
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 5
`
`computer mouse to draw various boundaries in the reti-
`nal cross-sectional images.
`After the grader draws the required layers in each of
`the 6 B-scans, the software calculates the distance in
`pixels between the manually drawn boundary lines for
`each of the various defined spaces. Using the dimen-
`sions of the B-scan image, the calculated pixels are
`converted into micrometers to yield a thickness mea-
`surement at each location. The thickness at all un-
`sampled locations between the radial lines is then
`interpolated based on a polar approximation to yield a
`thickness map analogous to the StratusOCT output
`data. After interpolation, thickness values are con-
`verted into volumes (mm3) by multiplying the av-
`erage thickness measurement by the sampled area.
`The interpolation algorithm, intergrader reliability,
`and intragrader
`reproducibility have previously
`been validated.8,9
`Analogous to the StratusOCT software, OCTOR
`provides a report showing the calculated thickness/
`volume values for the nine Early Treatment Diabetic
`Retinopathy Study macular subfields. The means and
`standard deviations of the foveal center point thick-
`ness are also calculated. In contrast to the StratusOCT
`output, OCTOR provides separate maps for the vari-
`ous macular spaces of interest.
`
`Grading Procedure
`
`Optical coherence tomography scans were analyzed
`by certified OCT graders at the Doheny Image Read-
`ing Center (PAK, SL, KTC). Boundaries drawn in
`each of the six OCT B-scans included the internal
`limiting membrane, outer border of the photorecep-
`tors, borders of subretinal fluid and subretinal tissue
`(if present), inner surface of the retinal pigment epi-
`thelium and estimated normal position of the retinal
`pigment epithelium layer (in cases of retinal pigment
`epithelium elevation). All boundaries were drawn in
`accordance with the standard OCT grading protocol of
`the Doheny Image Reading Center.8 After completion
`of the grading, OCTOR was used to calculate output
`parameters for the various spaces: retina, subretinal
`fluid, subretinal tissue, and PED.
`Angiographic CNV lesion type at baseline was de-
`termined by certified FA graders at the Doheny Image
`Reading Center (PAK, KTC) using the Treatment of
`Age-Related Macular Degeneration with Photody-
`namic Therapy Study grading protocol.14 Disagree-
`ment regarding angiographic classification of CNV
`lesion type was resolved by open adjudication and
`consensus between graders, in accordance with stan-
`dard reading center practices.
`
`Data Analysis and Statistical Methods
`
`The change from baseline in total retinal volume
`was determined for each patient at Month 1. Patients
`showing an initial reduction in total retinal volume by
`Month 1 were used to formulate two groups for the
`purposes of comparison. Patients who maintained, or
`had further reduction, in total retinal volume, between
`Month 1 and the final examination at Month 6 were
`included in the first group; patients demonstrating an
`increase in retinal volume from Month 1 to Month 6
`were included in the second group (Figure 1).
`Groups were compared for each of the following
`variables: age, gender, previous treatment for neovas-
`cular AMD, angiographic CNV lesion classification at
`baseline, visual acuity at baseline, and number of
`injections of ranibizumab received. OCTOR software
`was also used to compare the volumes of neurosen-
`sory retina, subretinal fluid, PED, and subretinal tis-
`sue, between the two groups at baseline and at each
`follow-up. For each comparison between groups, a t-
`test (2-tail distribution) was used in the case of nor-
`mally distributed interval variables, whereas a chi-
`square test or Fisher’s Exact Test was used in the case
`of categorical variables. For each group the mean
`change from baseline, in total volume of the neuro-
`sensory retina, was also calculated at each follow-up
`visit, and analyzed using a paired t-test. P values
`⬍0.05 were considered statistically significant. Statis-
`tical analysis was performed using commercially
`available software (Intercooled Stata for Windows,
`Version 9, Statacorp LP).
`
`Results
`
`Baseline Morphologic Characteristics
`
`At baseline, the neovascular lesions were catego-
`rized by fluorescein angiography as predominantly
`classic (10 eyes, 20%), as minimally classic (8 eyes,
`16%) and as occult with no classic (32 eyes, 64%).
`The total area of visible classic CNV and occult CNV
`was considered the area of CNV leakage. The CNV
`lesion was defined as the area of CNV leakage plus
`any contiguous areas of thick hemorrhage, blocked
`fluorescence, staining scars or serous PED which
`could be obscuring the boundaries of the CNV. The
`mean total CNV lesion area was 5.7 ⫾ 5.1 Disk Areas,
`whereas the mean total area of CNV leakage was
`3.8 ⫾ 3.2 Disk Areas. The mean greatest linear di-
`mension (GLD) of CNV lesions was 3.9 ⫾ 1.7 mm,
`whereas the mean GLD of CNV leakage was 3.5 ⫾
`1.6 mm. The total volume of the neurosensory retina,
`measured using OCTOR, was 7.35 ⫾ 0.65 mm3 for
`the entire cohort at baseline. The total volume of
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 3
`
`Joining Petitioner: Apotex
`
`

`

`RANIBIZUMAB RETREATMENT FREQUENCY IN AMD ● KEANE ET AL
`
`595
`
`Fig. 1. Optical coherence to-
`mography B-scans (each at
`90°) from a subject demon-
`strating initial anatomic im-
`provements, with subsequent
`regression and associated de-
`cline in visual acuity. Base-
`line, Patient is newly diagnosed
`with neovascular age-related
`macular degeneration. Visual
`acuity is 20/40 and the patient
`is
`treated with intravitreal
`ranibizumab. Month 1, Treat-
`ment with ranibizumab leads
`to a significant reduction in to-
`tal retinal volume. Visual acu-
`ity is 20/30, and patient is ob-
`served. Month 3, Regression of
`anatomic changes is clearly ev-
`ident. Visual acuity is 20/80
`and the patient is treated with a
`second intravitreal injection of
`ranibizumab. Month 6, Total
`retinal volume remains ele-
`vated. Visual acuity is 20/100,
`and patient receives third injec-
`tion of ranibizumab.
`
`subretinal fluid at baseline was 0.34 ⫾ 0.58 mm3,
`whereas that of subretinal tissue was 0.29 ⫾ 0.49
`mm3. The total volume of PED was 1.04 ⫾ 1.79 mm3
`for the entire cohort at baseline. The total volumes of
`subretinal fluid, PED, and subretinal tissue, at baseline
`and at each follow-up, are summarized in Table 1.
`
`Morphologic Outcomes—Baseline to Month 1
`
`Forty-two patients (84%) showed a reduction in
`total retinal volume 1 month after their initial treat-
`ment with ranibizumab—“Initial Responders.” Eight
`patients (16%) showed an increase in total retinal
`
`volume at Month 1, despite treatment with intravitreal
`ranibizumab—“Initial Non-Responders” (Figure 2).
`Patients who demonstrated an initial anatomic re-
`sponse had a greater total retinal volume at baseline
`than those who failed to respond (7.45 ⫾ 0.61 mm3
`versus 6.84 ⫾ 0.68 mm3, P ⫽ 0.035).
`No statistically significant difference was found be-
`tween “Initial Responders” and “Initial Non-Respond-
`ers” in terms of age (P ⫽ 0.781), baseline visual
`acuity (P ⫽ 0.115), previous treatment for neovascu-
`lar AMD (P ⫽ 0.277), total CNV lesion area (P ⫽
`0.914), total CNV leakage area (P ⫽ 0.541), GLD of
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 4
`
`Joining Petitioner: Apotex
`
`

`

`596
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 5
`
`Table 1. Additional Morphologic Outcomes Derived From “OCTOR” Analysis (Mean ⫾ Standard Deviation)
`
`Total (n ⫽ 50)
`
`Sustained Responders
`(n ⫽ 16)
`
`Regressors (n ⫽ 26)
`
`P*
`
`Subretinal fluid
`Baseline—total vol. (mm3)
`Month 1—total vol. (mm3)
`Month 3—total vol. (mm3)
`Month 6—total vol. (mm3)
`Pigment epithelium detachment
`Baseline—total vol. (mm3)
`Month 1—total vol. (mm3)
`Month 3—total vol. (mm3)
`Month 6—total vol. (mm3)
`Subretinal tissue
`Baseline—total vol. (mm3)
`Month 1—total vol. (mm3)
`Month 3—total vol. (mm3)
`Month 6—total vol. (mm3)
`
`*Ranksum test.
`
`0.36 ⫾ 0.60
`0.09 ⫾ 0.26
`0.11 ⫾ 0.24
`0.09 ⫾ 0.21
`
`1.01 ⫾ 1.76
`0.73 ⫾ 1.51
`0.66 ⫾ 1.02
`0.52 ⫾ 0.79
`
`0.30 ⫾ 0.52
`0.23 ⫾ 0.34
`0.24 ⫾ 0.40
`0.25 ⫾ 0.33
`
`0.39 ⫾ 0.71
`0.10 ⫾ 0.21
`0.16 ⫾ 0.32
`0.06 ⫾ 0.16
`
`0.52 ⫾ 0.42
`0.61 ⫾ 1.26
`0.93 ⫾ 1.40
`0.52 ⫾ 0.88
`
`0.43 ⫾ 0.78
`0.21 ⫾ 0.39
`0.26 ⫾ 0.45
`0.21 ⫾ 0.29
`
`0.34 ⫾ 0.53
`0.09 ⫾ 0.29
`0.08 ⫾ 0.15
`0.12 ⫾ 0.24
`
`1.31 ⫾ 2.17
`0.80 ⫾ 1.67
`0.48 ⫾ 0.67
`0.52 ⫾ 0.75
`
`0.23 ⫾ 0.25
`0.24 ⫾ 0.31
`0.22 ⫾ 0.37
`0.27 ⫾ 0.35
`
`0.506
`0.202
`0.956
`0.171
`
`0.490
`0.804
`0.546
`0.659
`
`0.141
`0.310
`0.671
`0.611
`
`lesion (P ⫽ 0.524), GLD of leakage (P ⫽ 0.541), or
`angiographic classification of CNV at baseline (P ⫽
`0.312). Additional analysis with OCTOR software
`found no statistically significant difference between
`the groups in terms of baseline subretinal fluid volume
`(P ⫽ 0.665), subretinal tissue volume (P ⫽ 0.207), or
`PED volume (P ⫽ 0.761).
`
`Morphologic Outcomes—Month 1 to Month 6
`
`Of the patients who initially responded, 16 (38%)
`maintained or had further reduction through to the
`final examination at Month 6 —“Sustained Respond-
`ers” (Figure 3). Of the 42 patients who initially
`showed a reduction in total retinal volume by Month 1,
`
`*
`
`1.2
`
`1
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`0
`
`-0.2
`
`-0.4
`
`-0.6
`
`-0.8
`
`Retinal Volume (mm3)
`
`0
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`Month
`Fig. 2. Mean change from baseline in total volume of neurosensory retina—patients who failed to respond anatomically to initial treatment with
`intravitreal ranibizumab (“Initial Non-Responders”). Vertical lines, 1 standard error of the mean. *P ⬍ 0.05.
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 5
`
`Joining Petitioner: Apotex
`
`

`

`RANIBIZUMAB RETREATMENT FREQUENCY IN AMD ● KEANE ET AL
`
`597
`
`**
`
`**
`
`**
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`0.2
`
`0
`
`-0.2
`
`-0.4
`
`-0.6
`
`Retinal Volume (mm3)
`
`-0.8
`
`-1
`
`0
`
`Month
`Fig. 3. Mean change from baseline in total volume of neurosensory retina—patients who responded initially, and who continued to respond by month
`6 (“Sustained Responders”). Vertical lines, 1 standard error of the mean. **P ⬍ 0.01.
`
`26 patients (62%) had regressed toward baseline ret-
`inal volume measurements between Month 1 and
`Month 6 —“Regressors” (Figure 4). The patients who
`maintained the Month 1 reduction received 3.75 ⫾
`1.18 injections of ranibizumab versus 2.96 ⫾ 1.34
`injections for the patients who did not (P ⫽ 0.049).
`No statistically significant difference was found be-
`tween the groups in terms of age (P ⫽ 0.196), baseline
`visual acuity (P ⫽ 0.321), previous treatment for
`neovascular AMD (P ⫽ 0.907), baseline neurosensory
`retinal volume (P ⫽ 0.579), total CNV lesion area
`(P ⫽ 0.303), total CNV leakage area (P ⫽ 0.493),
`GLD of lesion (P ⫽ 0.847), GLD of leakage (P ⫽
`0.471), or angiographic classification of CNV at base-
`line (P ⫽ 0.180). Additional analysis with OCTOR
`software found no statistically significant difference
`between the groups in terms of subretinal fluid vol-
`ume, subretinal tissue volume, or PED volume, at
`baseline, or at any of the follow-up visits (Table 1).
`
`Visual Outcomes
`
`The mean changes from baseline in visual acuity for
`each of the subgroups are illustrated in Figure 5. The
`change in visual acuity between baseline and Month 1
`was not correlated with the change in total retinal
`volume over the same time period for any of the
`
`subgroups: “Initial Responders” (P ⫽ 0.609), “Initial
`Non-Responders” (P ⫽ 0.201), or entire cohort (P ⫽
`0.638). In addition, the change in visual acuity be-
`tween baseline and Month 6 was not correlated with
`the change in total retinal volume between baseline
`and Month 6 (P ⫽ 0.221). However, between Month
`1 and Month 6, a statistically significant correlation
`was detected between the change in visual acuity and
`the change in total retinal volume (r ⫽ 0.359, P ⫽
`0.014). This correlation was also seen for “Initial
`Responders” (r ⫽ 0.400, P ⫽ 0.011), “Initial Non-
`Responders” (r ⫽ 0.7617, P ⫽ 0.078), and for “Regres-
`sors” (r ⫽ 0.599, P ⫽ 0.002). For the same time period,
`no correlation was detected between the change in visual
`acuity and the change in total retinal volume for the
`“Sustained Responders” (r ⫽ 0.235, P ⫽ 0.399).
`
`Discussion
`
`In this study, we analyzed changes in neurosensory
`retinal volume occurring in response to treatment with
`ranibizumab for neovascular AMD to determine the
`characteristics of patients who initially respond to
`treatment, but who subsequently regress toward their
`anatomic baseline.
`OCTOR analysis of the neurosensory retina re-
`vealed that cases showing initial anatomic improve-
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 6
`
`Joining Petitioner: Apotex
`
`

`

`598
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 5
`
`**
`
`**
`
`0.1
`
`0
`
`-0.1
`
`-0.2
`
`-0.3
`
`-0.4
`
`-0.5
`
`-0.6
`
`-0.7
`
`-0.8
`
`Retinal Volume (mm3)
`
`0
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`Month
`Fig. 4. Mean change from baseline in total volume of neurosensory retina—patients who responded initially, but who subsequently regressed toward
`their anatomic baseline by month 6 (“Regressors”). Vertical lines, 1 standard error of the mean. **P ⬍ 0.01.
`
`ment, but subsequent regression, received fewer
`injections of intravitreal ranibizumab than cases demon-
`strating a sustained anatomic response. Although our
`study did not use a fixed retreatment schedule, our
`findings appear consistent with anatomic results from
`the recent phase IIIb PIER Study, the aim of which
`was to assess the efficacy of a less frequent ranibi-
`zumab-dosing regimen (monthly for 3 months and
`then quarterly).6 In the PIER Study, the transition to
`less frequent therapy was associated with increased
`central retinal thickness on OCT, as well as increased
`CNV leakage on fluorescein angiography.6,15 Al-
`though quarterly ranibizumab therapy in the PIER
`Study maintained baseline VA, it was associated with
`a 4.5-letter decline in the maximal VA benefit
`achieved during the initial “induction” period of ther-
`apy. It is possible that the return of retinal edema
`contributed, at least in part, to the decline in the
`maximal VA benefit achieved. Our study provides
`additional evidence that some patients with neovascu-
`lar AMD may require more frequent ranibizumab
`injections to control CNV leakage and OCT features
`of exudation.
`The alternative dosage regimen used in the PIER
`Study was selected on the basis of evidence from
`Phase I/II studies indicating that the pharmacody-
`namic activity of ranibizumab administered monthly
`
`for three doses might persist for a subsequent 90
`days.16,17 Despite this evidence, it appears from our
`study, and the PIER trial results, that any persistence
`in the activity of ranibizumab may not be sufficient to
`maintain initial anatomic benefits over an extended
`period. Alternatively, a rebound effect may explain
`the anatomic changes observed. Experience with other
`pathophysiological systems has shown that rebound
`hyperfunction commonly develops on withdrawal of a
`drug.18 Clinical evidence of rebound macular edema
`has previously been described in a small case series of
`patients receiving another antivascular endothelial
`growth factor therapy (bevacizumab), in the context of
`retinal vein occlusion.19 If a rebound phenomenon
`occurs, the underlying mechanism may involve tran-
`sient up-regulation of vascular endothelial growth fac-
`tor production within the eye secondary to adminis-
`tration of ranibizumab, and a more frequent treatment
`regimen (as in the MARINA and ANCHOR trials)
`may be required to overcome this factor.
`Quantitative OCT subanalysis using OCTOR re-
`vealed that cases showing initial anatomic improve-
`ment, but subsequent regression, did not differ from
`cases demonstrating a sustained anatomic response in
`terms of baseline subretinal fluid volume, baseline
`subretinal tissue volume, or baseline PED volume.
`Furthermore, there was no statistically significant dif-
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 7
`
`Joining Petitioner: Apotex
`
`

`

`RANIBIZUMAB RETREATMENT FREQUENCY IN AMD ● KEANE ET AL
`
`599
`
`Initial Non Responders
`Continued Responders
`Regressors
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`0.1
`
`0.08
`
`0.06
`
`0.04
`
`0.02
`
`0
`
`-0.02
`
`-0.04
`
`-0.06
`
`-0.08
`
`Visual Acuity (LogMAR)
`
`-0.1
`
`0
`
`Month
`Fig. 5. Mean change from baseline in visual acuity (Snellen visual acuities converted to LogMAR units for the purposes of statistical analysis).
`
`ference between the groups in terms of age, previous
`treatment for neovascular AMD, CNV lesion area,
`CNV leakage area, CNV classification, or baseline
`visual acuity—findings consistent with subgroup anal-
`yses from the MARINA, ANCHOR, and PIER tri-
`als.6,20,21 From these trials, it appears that CNV lesion
`classification and size have limited prognostic value
`for ranibizumab therapy and, consequently, fluores-
`cein angiography may play a less prominent role in
`ranibizumab therapy than it has for photodynamic
`therapy with verteporfin.
`OCTOR analysis also demonstrated that the initial
`treatment with ranibizumab led to reductions in total
`retinal volume in the majority of eyes, but that this
`decrease was not correlated with an improvement in
`visual acuity, either at Month 1 or at Month 6. How-
`ever, the change in visual acuity between Month 1 and
`Month 6 was found to correlate with the change in
`total retinal volume over this time period, and this
`correlation was particularly evident in the “Regres-
`sors” subgroup. In other words, increases in total
`retinal volume, after initial decreases, were strongly
`associated with deterioration in visual acuity. Many
`studies have failed to find a consistent correlation
`between changes in central retinal thickness on OCT,
`
`in patients with neovascular
`and visual acuity,
`AMD.4,15,22 As a result, our findings may be of con-
`siderable interest if they can be replicated in larger
`prospective studies utilizing Early Treatment Diabetic
`Retinopathy Study visual acuities obtained with pro-
`tocol refractions.
`Our study has a number of limitations. The break-
`down of patients into subgroups for comparison, al-
`though hypothesis-driven, was an arbitrary construct.
`In addition, the patient cohort examined was small and
`collected retrospectively;
`therefore, all conclusions
`must be considered preliminary. Although patients in
`our study were treated utilizing many of the criteria
`specified in the PrONTO study, the purpose of our
`study was not specifically to address the merits of
`OCT-guided retreatment. The PrONTO study has pro-
`vided evidence that an OCT-guided variable dosing
`regimen could result in visual acuity outcomes similar
`to the Phase III clinical studies, but require fewer
`injections.5 Despite these excellent results, the issue of
`ranibizumab retreatment remains controversial, and a
`number of alternative retreatment regimens have been
`suggested, e.g., “inject and extend.”10 Large phase III
`clinical trials (EXCITE trial, SUSTAIN trial), evalu-
`ating retreatment regimens based on disease activity,
`
`Mylan Exhibit 1051
`Mylan v. Regeneron, IPR2021-00881
`Page 8
`
`Joining Petitioner: Apotex
`
`

`

`600
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 5
`
`are currently underway, and may provide more defin-
`itive guidance regarding ranibizumab dosing.23,24
`In summary, patients receiving fewer injections of
`ranibizumab appeared less likely to maintain anatomic
`improvements achieved following commencement of
`ranibizumab therapy. Furthermore, regression of ini-
`tial anatomic improvement was strongly associated
`with worsening of visual acuity. These phenomena, if
`confirmed in other studies, may have implications for
`the future design of ranibizumab retreatment regimens
`in neovascular AMD.
`Key words: age-related macular degeneration, op-
`tical coherence tomography, ranibizumab.
`
`References
`
`1. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus
`verteporfin for neovascular age-related macular degeneration.
`N Engl J Med 2006;355:1432–1444.
`2. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for
`neovascular age-related macular degeneration. N Engl J Med
`2006;355:1419 –1431.
`3. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA, Nasir
`MA, Giust MJ. Intravitreal bevacizumab (Avastin) for neovas-
`cular age-related macular degeneration. Ophthalmology 2006;
`113:363–372.
`4. Spaide RF, Laud K, Fine HF, et al. Intravitreal bevacizumab
`treatment of choroidal neovascularization secondary to age-
`related macular degeneration. Retina 2006;26:383–390.
`5. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical coher-
`ence tomography-guided, variable dosing regimen with intra-
`vitreal ranibizumab (Lucentis) for neovascular age-related
`macular degeneration. Am J Ophthalmol 2007;143:566 –583.
`6. Regillo CD, Brown DM, Abraham P, et al. Randomized,
`double-masked, sham-controlled trial of ranibizumab for neo-
`vascular age-related macular degeneration: PIER study year 1.
`Am J Ophthalmol 2008;145:239 –248.
`7. Sadda SR, Wu Z, Walsh AC, et al. Errors in retinal thickness
`measurements obtained by optical coherence tomography.
`Ophthalmology 2006;113:285–293.
`8. Joeres S, Tsong JW, Updike PG, et al. Reproducibility of
`quantitative optical coherence tomography subanalysis in neo-
`vascular age-related macular degeneration. Invest Ophthalmol
`Vis Sci 2007;48:4300 – 4307.
`9. Sadda SR, Joeres S, Wu Z, et al. Error correction and quanti-
`tative subanalysis of optical coherence tomography data using
`computer-assisted grading. Invest Ophthalmol Vis Sci 2007;
`48:839 – 848.
`10. Spaide R. Ranibizumab according to need: a treatment for
`age-related macular degeneration. Am J Ophthalmol 2007;
`143:679 – 680.
`11. Keane PA, Liakopoulos S, Ongchin SC, et al. Quantitative
`
`subanalysis of optical coherence tomography after treatment
`with ranibizumab for neovascular age-related macular degen-
`eration. Invest Ophthalmol Vis Sci 2008;49:3115–3120.
`12. Joeres S, Kaplowitz K, Brubaker JW, et al. Quantitative com-
`parison of optical coherence tomography after pegaptanib or
`bevacizumab in neovascular age-related macular degeneration.
`Ophthalmology 2008;115:347–354.
`13. Keane PA, Liakopoulos S, Chang KT, et al. Comparison of the
`optical coherence tomographic features of choroid neovascular
`membranes in pathological myopia versus age-related macular
`degeneration, using quantitative subanalysis. Br J Ophthalmol
`2008;92:1081–1085.
`14. Barbazetto I, Burdan A, Bressler NM, et al. Photodynamic
`therapy of subfoveal choroidal neovascularization with verte-
`porfin: fluorescein angiographic guidelines for evaluation and
`treatment—TAP and VIP report No. 2. Arch Ophthalmol
`2003;121:1253–1268.
`15. Benz MS, Brown DM, Shapiro H, Tuomi L. Ranibizumab for
`neovascular age-related macular degeneration (AMD): optical
`coherence tomography (OCT) vs. visual acuity (VA) changes
`in the PIER study. Invest Ophthalmol Vis Sci 2008;49:ARVO
`E-Abstract 347.
`16. Heier JS, Antoszyk AN, Pavan PR, et al. Ranibizumab for
`treatment of neovascular age-related macular degeneration: a
`phase I/II multicenter, controlled, multidose study. Ophthal-
`mology 2006;113:633;e1– e4.
`17. Rosenfeld PJ, Heier JS, Hantsbarger G, Shams N. Tolerability
`and efficacy of multiple escalating doses of ranibizumab (Lu-
`centis) for neovascular age-related macular degeneration. Oph-
`thalmology 2006;113:623– 632;e1.
`18. Gillen D, McColl KE. Problems related to acid rebound and
`tachyphylaxis. Best Pract Res Clin Gastroenterol 2001;15:
`487– 495.
`19. Matsumoto Y, Freund KB, Peiretti E, Cooney MJ, Ferrara DC,
`Yannuzzi LA. Rebound macular edema following bevaci-
`zumab (Avastin) therapy for retinal venous occlusive disease.
`Retina 2007;27:426 – 431.
`20. Boyer DS, Antoszyk AN, Awh CC, Bhisitkul RB, Shapiro H,
`Acharya NR; MARINA Study Group. Subgroup analysis of
`the MARINA study of ranibizumab in neovascular age-related
`macular degeneration. Ophthalmology 2007;114:246 –252.
`21. Kaiser PK, Brown DM, Zhang K, et al. Ranibizumab for
`predominantly classic neovascular age-related macular degen-
`erati