AMERICAN ACADEMY™
`
`OF OPHTHALMOLOGY
`Intravitreal Aflibercept for Diabetic
`Macular Edema
`
`®CrossMark
`
`148-Week Results from the VISTA and VIVID Studies
`
`
`
`Jeffrey S. Heier, MD,' Jean-Francois Korobelnik, MD,?** David M. Brown, MD,° Ursula Schmidt-Erfurth, MD,°
`Diana V. Do, MD,’ Edoardo Midena, MD," David S. Boyer, MD,” Hiroko Terasaki, MD, '° Peter K. Kaiser, MD,'!
`Dennis M. Marcus, MD,'? Quan D. Nguyen, MD,’ Glenn J. Jaffe, MD,'° Jason S. Slakter, MD,'*
`Christian Simader, MD,° Yuhwen Soo, PhD,'? Thomas Schmelter, PhD,'° Robert Vitti, MD,'°
`Alyson J. Berliner, MD, PhD,’ Oliver Zeitz, MD,'°'” Carola Metzig, MD,'° Frank G. Holz, MD'®
`
`Purpose: To compare efficacy and safety of intravitreal aflibercept injection (IAI) with macular laser photo-
`coagulation for diabetic macular edema (DME) over 3 years.
`Design: Twosimilarly designed phase3trials: VISTAO“® and VIVIDOME.
`
`Participants: Patients (eyes;n
`872) with central-involved DME.
`Methods: Eyes received IAI 2 mg every 4 weeks (24), IAI 2 mg every 8 weeks after 5 monthly doses (2q8), or
`laser control. From week 24, if rescue treatmentcriteria were met, IAI patients received active laser, and laser
`control patients received IAl 2q8. From week 100, laser control patients who had notreceived IAI rescue treat-
`ment received IAI as needed perretreatmentcriteria.
`Main Outcome Measures:
`Theprimary end point was the change from baseline in best-corrected visual
`acuity (BCVA) at week 52. We report the 148-weekresults.
`Results: Mean BCVA gain from baseline to week 148 with IAI 24, IAl 2q8, and laser control was 10.4, 10.5,
`and 1.4 letters (P < 0.0001) in VISTA and 10.3, 11.7, and 1.6 letters (P < 0.0001) in VIVID, respectively. The
`proportion of eyes that gained >15 letters from baseline at week 148 was 42.9%, 35.8%, and 13.6% (P < 0.0001)
`in VISTA and 41.2%, 42.2%, and 18.9% (P < 0.0001) in VIVID, respectively. Greater proportions of eyes treated
`with IAI 2q4 and IAI 2q8 versus those treated with laser control had an improvement of >2 steps in the Diabetic
`Retinopathy Severity Scale (DRSS) score in both VISTA (29.9% and 34.4% vs. 20.1% [P
`0.0350, IAI 2q4; P
`0.0052, IAI 2q8}) and VIVID (44.3% and 47.8% vs. 17.4% [P < 0.0001 for both}). In an integrated safety analysis,
`the most frequent ocular serious adverse event wascataract (3.1%, 2.1%, 0.3% for 2q4, 2q8, and control).
`Conclusions: Visual improvements observed with both IAI regimens(over laser control) at weeks 52 and 100
`were maintained at week 148, with similar overall efficacy in the IAl 2q4 and IAI 2q8 groups. Treatment with IAI
`also had positive effects on the DRSS score. Over 148 weeks, the incidence of adverse events was consistent
`with the knownsafety profile of IAl. Ophthalmology 2016;123:2376-2385 © 2016 by the American Academy of
`Ophthalmology
`
`Supplemental material is available at www.aaojournal.org.
`e
`
`functional outcomes,° '' anti-VEGF agents have rapidly
`The diabetes mellitus epidemic is growing. According to
`replaced macular laser photocoagulation as the standard of
`current predictions, by 2040, approximately 1 in every 10
`adults (642 million) worldwide will have the disease.'
`care to treat DME.
`Aflibercept, a 115-kKDA recombinant fusion protein, is
`Diabetic
`retinopathy and associated diabetic macular
`edema (DME)are serious diabetes mellitus complications
`composed of the key VEGF binding domains of human
`
`and are the leading causes of blindness and_visual VEGFreceptors 1 and 2 fused to the constant Fe domain of
`disability in working-age adults.”
`human immunoglobulin G1,'* and it binds VEGF-A with
`high affinity.'’ Unlike ranibizumab and bevacizumab,
`Current treatment options for DMEinclude macular laser
`photocoagulation,*
`corticosteroids,”
`and
`anti-vascular
`aflibercept
`also binds
`to placental growth factor.'°
`endothelial growth factor (VEGF) agents (i.e., intravitreal
`Intravitreal aflibercept injection (IAI), which is also known
`aflibercept,
`ranibizumab,
`and off-label use of bev-
`as “VEGF Trap Eye” or “IVT-AFL” in the scientific
`acizumab).° * There is a large body of evidence to support
`literature,
`is currently indicated to treat neovascular age-
`anti-VEGF use. Because of
`superior
`anatomic
`and
`related macular degeneration (AMD), macular edema
`
`2376
`
`© 2016 by the American Academy of Ophthalmology
`Published by ElsevierInc.
`
`http://dx.doi.org/10.1016/j.ophtha.2016.07.032
`ISSN 0161-6420/16
`
`IPR2023-00884
`Samsung et al. v. Regeneron
`Regeneron Pharmaceuticals,Inc.
`Exhibit 2139
`Page 1
`
`

`

`Heier et al
`
` Intravitreal Aflibercept for DME
`
`Table 1. Treatment Experience from Baseline to Week 148
`
`Laser Control
`(n ¼ 154)
`
`VISTA
`IAI 2q4
`(n ¼ 155)
`
`IAI 2q8
`(n ¼ 152)
`
`Laser Control
`(n ¼ 133)
`
`VIVID
`IAI 2q4
`(n ¼ 136)
`
`IAI 2q8
`(n ¼ 135)
`
`No. of scheduled treatments through week 148,
`mean (SD)
`Macular laser photocoagulation
`Intravitreal aflibercept
`Study eyes that received rescue treatment* from
`week 24 to week 148, n (%)
`Mean (SD) No. of rescue treatment
`y
`IAI
`Laser control eyes that received rescue or PRN
`treatment from week 24 to week 148, n (%)
`Mean (SD) number of IAI injections
`
`3.8 (2.4)
`e
`63 (40.9)*
`
`e
`29.6 (9.8)
`7 (4.5)*
`
`e
`18.1 (4.8)
`16 (10.5)*
`
`2.6 (2.0)
`e
`47 (35.3)*
`
`e
`32.0 (9.7)
`10 (7.4)*
`
`e
`18.1 (5.1)
`16 (11.9)*
`
`13.5 (3.9)
`134 (87.0)
`
`1.4 (0.8)
`e
`
`1.4 (1.1)
`e
`
`13.5 (4.3)
`109 (82.0)
`
`2.3 (1.5)
`e
`
`1.9 (1.0)
`e
`
`9.8 (5.0)
`
`e
`
`e
`
`9.3 (5.2)
`
`e
`
`e
`
`e ¼ not applicable; IAI ¼ intravitreal aflibercept injection; PRN ¼ pro re nata; SD ¼ standard deviation; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI
`every 8 weeks after 5 initial monthly doses.
`Safety analysis set.
`*Rescue treatment was 2 mg IAI every 4 weeks for 5 initial doses followed by dosing every 8 weeks in the laser control group, and active laser for the IAI 2q4
`and 2q8 groups.
`y
`Laser control patients who did not meet criteria for rescue treatment during weeks 24 to 96 received IAI 2 mg PRN per the prespecified retreatment criteria
`from week 100 to week 144. In VISTA and VIVID, respectively, 71 and 64 laser control patients received a mean (SD) of 6.53.2 and 6.03.3 PRN IAI
`injections from week 100 to week 148.
`
`secondary to retinal vein occlusion, myopic choroidal neo-
`vascularization, and DME. Intravitreal aflibercept injection
`is approved for the treatment of DME in the United States,
`the European Union, Australia, and Japan.
`The efficacy and safety of IAI in DME have been
`demonstrated over 2 years in the VISTADME and VIV-
`IDDME studies.7,14 Both trials showed that, after 52 and 100
`weeks of treatment, IAI provides significantly greater im-
`provements in both functional and anatomic outcomes when
`laser photocoagulation.7,14
`compared with macular
`In
`addition, the proportion of eyes with 2-step improvement
`in the Early Treatment Diabetic Retinopathy Study
`(ETDRS) Diabetic Retinopathy Severity Scale (DRSS)
`score was significantly greater with IAI than with laser
`control, suggesting a beneficial effect on the underlying
`diabetic retinopathy.7,14 We report the 148-week results of
`the VISTA and VIVID studies.
`
`Methods
`
`Study Design
`
`VISTA and VIVID were 2 similarly designed, double-masked,
`randomized, active-controlled, 148-week, phase 3 trials. VISTA
`(registered at www.clinicaltrials.gov; NCT01363440) was con-
`ducted across 54 sites in the United States, and VIVID (registered
`at www.clinicaltrials.gov; NCT01331681) was conducted in 73
`sites across Europe, Japan, and Australia.7,14 Each clinical site’s
`respective institutional review board or ethics committee approved
`the study. All patients provided written informed consent. Both
`VISTA and VIVID were conducted in compliance with the Inter-
`national Conference on Harmonization guidelines and the Health
`Insurance Portability and Accountability Act of 1996.15,16 Data for
`this report, which present the 148-week results, were collected
`between May 2011 and March 2015.
`
`Patient eligibility for the VISTA and VIVID studies has been
`described.14 Briefly, adult patients with type 1 or 2 diabetes
`mellitus who presented with central-involved DME (defined as
`retinal
`thickening involving the central 1-mm subfield [central
`subfield thickness {CST}] as determined by spectral domain op-
`tical coherence tomography [SD OCT]) were eligible for enroll-
`ment if best-corrected visual acuity (BCVA) was between 73 and
`24 letters (20/40 to 20/320 Snellen equivalent) in the study eye.
`Only 1 eye per patient was enrolled in the study. Eyes were ran-
`domized in a 1:1:1 ratio to 3 groups to receive 1 of the following
`treatments (a) 2 mg IAI every 4 weeks (2q4), (b) 2 mg IAI every 8
`weeks after 5 initial monthly doses (2q8), and (c) macular laser
`photocoagulation at baseline. Treatments continued through
`week 148.
`Beginning at week 12, study eyes in all treatment groups were
`assessed for laser retreatment. If any ETDRS-defined, clinically
`significant macular edema was present (defined as thickening of the
`retina or hard exudates at 500 mm of center of the macula, or at
`least 1 zone of retinal thickening 1 disc area or larger, any part of
`which was within 1 disc diameter of center of the macula), study
`eyes in the IAI 2q4 and IAI 2q8 groups received sham laser and
`those in the laser group received active laser, but no more
`frequently than every 12 weeks.
`Beginning at week 24, study eyes in all treatment groups also
`could receive additional (rescue) treatment if DME worsened, as
`defined by a 10-letter loss at 2 consecutive visits or 15-letter
`loss at 1 visit from the best previous measurement, when BCVA
`was not better than baseline. When these criteria were met, study
`eyes in the IAI 2q4 and IAI 2q8 groups could receive active laser
`(rather than sham laser) from week 24 onward and continued with
`the existing IAI regimen; study eyes in the laser control group
`received 5 doses of 2 mg IAI every 4 weeks followed by dosing
`every 8 weeks until the end of the study (rather than sham in-
`jections), in addition to laser, when the laser retreatment criteria
`were met. Patients could receive both laser and IAI, when appli-
`cable, at the same visit.
`Beginning at week 100, patients in the laser control group who
`did not meet criteria for rescue treatment during weeks 24 to 96
`
`2377
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 2
`
`

`

`Ophthalmology Volume 123, Number 11, November 2016
`
`Figure 1. Mean (standard deviation) change in best-corrected visual acuity (BCVA) from baseline through week 148 in VISTA (A) and VIVID (B).
`Primary analysis method (LOCF): last observation carried forward, censoring measurements after rescue treatment was given; measurements after as needed
`(PRN) treatment was given were not censored. Ancillary analysis method (aLOCF): last observation carried forward, including measurements after
`additional or PRN treatment was given. Full analysis set. In VISTA, n ¼ 154 for laser control, n ¼ 154 for intravitreal aflibercept injection (IAI) 2q4, and
`n ¼ 151 for IAI 2q8. In VIVID, n ¼ 132 for laser control, n ¼ 136 for IAI 2q4, and n ¼ 135 for IAI 2q8. aP < 0.0001, bP ¼ 0.0002, cP ¼ 0.0345, and dP ¼
`0.0021 versus laser control from the analysis of covariance. aLOCF ¼ last observation carried forward, including measurements after additional or pro re nata
`(PRN) treatment was given; LOCF ¼ last observation carried forward, censoring measurements after rescue treatment was given; measurements after PRN
`treatment was given were not censored; SD ¼ standard deviation; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses.
`
`received 2 mg IAI as needed (pro re nata [PRN]) when any 1 of the
`following criteria was met: a >50 mm increase in CST compared
`with the lowest previous measurement; (b) new or persistent cystic
`retinal changes or subretinal fluid on optical coherence tomography
`(OCT), or persistent diffuse edema in the central subfield on OCT;
`(c) a loss of 5 letters in BCVA from the best previous mea-
`surement in conjunction with any increase in CST; or (d) an in-
`crease of 5 letters in BCVA between the current and the most
`recent visit.
`
`Outcome Measures
`
`The primary efficacy end point, change from baseline BCVA in
`ETDRS letters at week 52, and the prespecified secondary and
`exploratory efficacy end points at week 52 and week 100 have been
`reported.7,14 We report the 148-week results of the VISTA and
`
`VIVID studies. Prespecified efficacy end points at week 148 were
`exploratory and included the change from baseline in BCVA,
`proportion of eyes that gained or lost 10 and 15 letters from
`baseline, proportion of eyes with a 2-step improvement from
`baseline in the DRSS score,17 and change from baseline in CST as
`determined by SD OCT.
`The BCVA using the ETDRS protocol4 and CST using SD
`OCT were assessed every 4 weeks. Color fundus photography
`was performed at baseline and weeks 24, 52, 72, 100, 124, and
`148. Masked readers at
`independent central
`reading centers
`evaluated OCT images for CST (Duke Reading Center, Durham,
`NC, for VISTA, and Vienna Reading Center, Vienna, Austria,
`for VIVID) and fundus images including assessment of
`the
`DRSS score (Digital Angiography Reading Center, Great Neck,
`NY, for VISTA, and Vienna Reading Center, Vienna, Austria,
`for VIVID).
`
`2378
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 3
`
`

`

`Heier et al
`
` Intravitreal Aflibercept for DME
`
`Table 2. Eyes with Vision Gains and Losses from Baseline at Week 148 in VISTA
`
`Laser Control
`(n ¼ 154)
`
`LOCF
`IAI 2q4
`(n ¼ 154)
`
`IAI 2q8
`(n ¼ 151)
`
`P
`Value
`
`Laser Control
`(n ¼ 154)
`
`aLOCF
`IAI 2q4
`(n ¼ 154)
`
`IAI 2q8
`(n ¼ 151)
`
`Vision gain, n (%)
`15 letters
`10 letters
`
`Vision loss, n (%)
`10 letters
`15 letters
`
`21 (13.6)
`
`66 (42.9)
`
`54 (35.8)
`
`<0.0001*
`
`37 (24.0)
`
`68 (44.2)
`
`58 (38.4)
`
`48 (31.2)
`
`90 (58.4)
`
`89 (58.9)
`
`<0.0001*
`
`74 (48.1)
`
`92 (59.7)
`
`93 (61.6)
`
`30 (19.5)
`
`15 (9.7)
`
`9 (5.8)
`
`6 (3.9)
`
`5 (3.3)
`
`4 (2.6)
`
`y
`0.0004
`z
`<0.0001
`y
`0.0386
`z
`0.0107
`
`8 (5.2)
`
`7 (4.5)
`
`10 (6.5)
`
`7 (4.5)
`
`4 (2.6)
`
`4 (2.6)
`
`P
`Value
`
`y
`0.0002
`z
`0.0069
`y
`0.0291
`z
`0.0177
`y
`0.6032
`z
`0.2531
`y
`0.9884
`z
`0.3753
`
`aLOCF ¼ ancillary last observation carried forward, including measurements after additional or as needed (PRN) treatment was given; IAI ¼ intravitreal
`aflibercept injection; LOCF ¼ last observation carried forward, censoring measurements after rescue treatment was given; measurements after PRN treatment
`was given were not censored; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses.
`Full analysis set.
`*For both IAI 2q4 and 2q8 compared with laser control.
`y
`For IAI 2q4 compared with laser control.
`z
`For IAI 2q8 compared with laser control.
`
`Statistical Analyses
`
`All outcome measures at week 148 were analyzed in an explor-
`atory manner, and P values reported are considered nominal (not
`prespecified). Efficacy end points were evaluated at a 2-sided
`significance level of 2.5% in the full analysis sets from each in-
`dividual study. The full analysis sets included eyes that received
`study treatment and had a baseline and at least 1 postbaseline
`BCVA assessment. Continuous variables were analyzed with an
`analysis of covariance with the baseline value as covariate and
`treatment group and geographic region (VIVID only) or medical
`history of myocardial
`infarction or cerebrovascular accident
`(VISTA only) as fixed factors. Proportions were analyzed using a
`
`CochraneManteleHaenszel test stratified by geographic region
`(VIVID) and history of myocardial infarction or cerebrovascular
`accident (VISTA). Missing values were imputed using the last
`observation carried forward method, and for eyes that received
`rescue treatment, the last value before rescue treatment was used
`for analyses, censoring measurements after rescue treatment was
`given (primary analysis method; LOCF). Measurements obtained
`after PRN IAI treatment in the laser group were not censored.
`Prespecified sensitivity analyses were also performed to include
`values after rescue treatment was given (ancillary analysis method;
`aLOCF). Safety was assessed on the integrated safety set from
`VISTA and VIVID,
`including all
`randomized patients who
`received any study treatment.
`
`Table 3. Eyes With Vision Gains and Losses from Baseline at Week 148 in VIVID
`
`Laser Control
`(n ¼ 132)
`
`LOCF
`IAI 2q4
`(n ¼ 136)
`
`IAI 2q8
`(n ¼ 135)
`
`P
`Value
`
`Laser Control
`(n ¼ 132)
`
`aLOCF
`IAI 2q4
`(n ¼ 136)
`
`IAI 2q8
`(n ¼ 135)
`
`P
`Value
`
`Vision gain, n (%)
`15 letters
`10 letters
`
`Vision loss, n (%)
`10 letters
`15 letters
`
`25 (18.9)
`
`56 (41.2)
`
`57 (42.2)
`
`<0.0001*
`
`30 (22.7)
`
`63 (46.3)
`
`60 (44.4)
`
`39 (29.5)
`
`76 (55.9)
`
`76 (56.3)
`
`<0.0001*
`
`55 (41.7)
`
`83 (61.0)
`
`83 (61.5)
`
`26 (19.7)
`
`18 (13.6)
`
`5 (3.7)
`
`4 (2.9)
`
`3 (2.2)
`
`0 (0)
`
`<0.0001*
`y
`0.0013
`z
`<0.0001
`
`8 (6.1)
`
`6 (4.5)
`
`5 (3.7)
`
`4 (2.9)
`
`2 (1.5)
`
`1 (0.7)
`
`y
`<0.0001
`z
`0.0001
`y
`0.0013
`z
`0.0010
`y
`0.3651
`z
`0.0498
`y
`0.4900
`z
`0.0530
`
`aLOCF ¼ ancillary last observation carried forward, including measurements after additional or PRN treatment was given; IAI ¼ intravitreal aflibercept
`injection; LOCF ¼ last observation carried forward, censoring measurements after rescue treatment was given; measurements after PRN treatment was given
`were not censored; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses.
`Full analysis set.
`*For both IAI 2q4 and 2q8 compared with laser control.
`y
`For IAI 2q4 compared with laser control.
`z
`For IAI 2q8 compared with laser control.
`
`2379
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 4
`
`

`

`Ophthalmology Volume 123, Number 11, November 2016
`
`Figure 2. Proportion of eyes with a 2-step improvement in Diabetic Retinopathy Severity Scale (DRSS) score from baseline at week 148 in VISTA (A)
`and VIVID (B). Primary analysis method (LOCF): last observation carried forward, censoring measurements after rescue treatment was given; measurements
`after as needed (PRN) treatment was given were not censored. Ancillary analysis method (aLOCF): last observation carried forward, including measure-
`ments after additional or PRN treatment was given. In VISTA, analyses were performed using the full analysis set. In VIVID, analyses included only
`evaluable patients defined as those with a gradable baseline DRSS and a post-baseline DRSS score. In VISTA, n ¼ 154 for laser control, n ¼ 154 for
`intravitreal aflibercept injection (IAI) 2q4, and n ¼ 151 for IAI 2q8. In VIVID, LOCF: n ¼ 86 for laser control, n ¼ 88 for IAI 2q4, and n ¼ 92 for IAI 2q8;
`aLOCF: n ¼ 89 for laser control, n ¼ 89 for IAI 2q4, and n ¼ 93 for IAI 2q8. aP ¼ 0.0350, bP ¼ 0.0052, cP < 0.0001, dP < 0.0016, and eP < 0.0022 versus
`laser control. aLOCF ¼ last observation carried forward, including measurements after additional or PRN treatment was given; LOCF ¼ last observation
`carried forward, censoring measurements after rescue treatment was given; measurements after PRN treatment was given were not censored; 2q4 ¼ 2 mg IAI
`every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses.
`
`Results
`
`Patient Disposition and Treatment Experience
`
`VISTA treated 461 eyes, and VIVID treated 404 eyes (Appendix 1,
`available at www.aaojournal.org). Demographics and baseline
`characteristics of patients were reported by Korobelnik et al.14
`Overall, 76.6% of eyes in VISTA and 74.4% of eyes in VIVID
`completed the study through week 148 (Appendix 1, available
`at www.aaojournal.org). The most
`common
`reason
`for
`discontinuation during year 3 was withdrawal by patient in both
`VISTA and VIVID, with other common reasons being death and
`adverse events (Appendix 1, available at www.aaojournal.org).
`From baseline to week 148, study eyes in the IAI 2q4 and IAI
`2q8 groups received a mean of 29.6 and 18.1 injections in
`VISTA and 32.0 and 18.1 injections in VIVID, respectively
`(Table 1). Eyes in the laser control group received an average of
`3.8 and 2.6 laser treatments in VISTA and VIVID, respectively.
`
`in VISTA was
`From week 24 to week 148, rescue treatment
`given to 4.5% and 10.5% of eyes in the IAI 2q4 and IAI 2q8
`groups compared with 40.9% of eyes in the laser control group,
`and in VIVID to 7.4% and 11.9% of eyes in the IAI 2q4 and
`IAI 2q8 groups compared with 35.3% of eyes in the laser control
`group, respectively (Table 1). Considering PRN IAI treatment
`given from week 100 to week 148, 87.0% of laser control eyes
`in VISTA and 82.0% of laser control eyes in VIVID received
`IAI treatment (rescue or PRN) from week 24 to week 148
`(Table 1).
`
`Efficacy Outcomes
`
`In both VISTA and VIVID, eyes with DME treated with IAI 2q4
`and IAI 2q8 demonstrated sustained visual acuity gains through
`week 148. With the primary analysis method (LOCF), which
`censored measurements after rescue treatment was given, but
`included measurements after PRN treatment, the mean  standard
`
`2380
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 5
`
`

`

`Heier et al
`
` Intravitreal Aflibercept for DME
`
`Figure 3. Mean change in central subfield thickness (CST) from baseline to week 148 in VISTA (A) and VIVID (B). Primary analysis method (LOCF):
`last observation carried forward, censoring measurements after rescue treatment was given; measurements after as needed (PRN) treatment was given were
`not censored. Ancillary analysis method (aLOCF): last observation carried forward, including measurements after additional or PRN treatment was given.
`Full analysis set. In VISTA, n ¼ 154 for laser control, n ¼ 154 for intravitreal aflibercept injection (IAI) 2q4, and n ¼ 151 for IAI 2q8. In VIVID, n ¼ 132
`for laser control, n ¼ 136 for IAI 2q4, and n ¼ 135 for IAI 2q8. aP < 0.0001, bP ¼ 0.0001, cP ¼ 0.0003, dP < 0.0033, and eP ¼ 0.0002 versus laser control.
`aLOCF ¼ last observation carried forward, including measurements after additional or PRN treatment was given; DRSS ¼ Diabetic Retinopathy Severity
`Scale; LOCF ¼ last observation carried forward, censoring measurements after rescue treatment was given; measurements after PRN treatment was given
`were not censored; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses; IAI ¼ intravitreal aflibercept injection.
`
`deviation change from baseline BCVA in the IAI 2q4 and IAI 2q8
`groups at week 148 was þ10.414.2 and þ10.512.7 letters
`versus þ1.414.5 letters in the laser control group (P < 0.0001 for
`both IAI groups compared with laser control)
`in VISTA,
`and þ10.312.5 and þ11.710.1 letters versus þ1.612.7 letters
`(P < 0.0001 for both IAI groups compared with laser control) in
`VIVID (Fig 1), respectively. With the ancillary analysis method
`(aLOCF), which included values after all IAI treatments (rescue
`or PRN), the between-group differences narrowed, but remained
`numerically (VISTA) and statistically (VIVID) in favor of the IAI
`groups (Fig 1).
`In both VISTA and VIVID, more eyes treated with IAI gained
`10 and 15 letters, whereas more eyes treated with laser control
`lost 10 and 15 letters from baseline at week 148 using the
`primary analysis method (LOCF) (Tables 2 and 3). The between-
`group differences in the proportions of patients who gained 15
`letters narrowed, but remained in favor of the IAI groups when
`
`measurements after all IAI treatments (rescue or PRN) were
`included in the analyses (aLOCF, Tables 2 and 3).
`With the primary analysis method (LOCF), higher propor-
`tions of eyes treated with IAI 2q4 and IAI 2q8 compared with
`those treated with laser control had at least a 2-step improvement
`in the DRSS score in both VISTA and VIVID (Fig 2). The
`between-group differences narrowed, but remained numerically
`in favor of the IAI groups when measurements after all IAI
`treatments (rescue or PRN) were included in the analyses
`(aLOCF) (Fig 2).
`With the primary analysis method (LOCF), the improvements
`from baseline CST in the IAI 2q4 and IAI 2q8 groups were robust
`throughout the study and significantly greater than those seen in the
`laser control group at week 148 in both VISTA and VIVID (Fig 3).
`The between-group differences at week 148 disappeared when
`measurements after all IAI treatments (rescue or PRN) were
`included in the analyses (aLOCF) (Fig 3).
`
`2381
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 6
`
`

`

`Ophthalmology Volume 123, Number 11, November 2016
`
`Table 4. Ocular Serious Adverse Events from Baseline to Week 148
`
`Control (n [ 287)
`
`IAI 2q4 (n [ 291)
`
`IAI 2q8 (n [ 287)
`
`All IAI (n [ 578)
`
`Any ocular SAEs in study eye, n (%)
`Cataract
`Cataract subcapsular
`Corneal epithelium defect
`Diabetic retinal edema
`Diabetic retinopathy
`Hyphema
`Lens dislocation
`Macular degeneration
`Punctate keratitis
`Retinal artery occlusion
`Retinal detachment
`Retinal exudates
`Retinal hemorrhage
`Retinal ischemia
`Retinal neovascularization
`Retinal vascular disorder
`Visual acuity reduced
`Vitreous hemorrhage
`Injection site injury
`Endophthalmitis
`Intraocular pressure increased
`Visual acuity tests abnormal
`Visual field defect
`Cataract operation
`
`18 (6.3)
`1 (0.3)
`1 (0.3)
`1 (0.3)
`0
`4 (1.4)
`0
`0
`1 (0.3)
`0
`0
`0
`1 (0.3)
`1 (0.3)
`0
`3 (1.0)
`1 (0.3)
`0
`5 (1.7)
`0
`0
`0
`1 (0.3)
`0
`2 (0.7)
`
`25 (8.6)
`9 (3.1)
`0
`0
`1 (0.3)
`0
`1 (0.3)
`1 (0.3)
`0
`1 (0.3)
`1 (0.3)
`3 (1.0)
`0
`0
`1 (0.3)
`0
`2 (0.7)
`1 (0.3)
`4 (1.4)
`1 (0.3)
`2 (0.7)
`0
`0
`0
`2 (0.7)
`
`18 (6.3)
`6 (2.1)
`2 (0.7)
`0
`0
`0
`0
`0
`0
`0
`1 (0.3)
`2 (0.7)
`0
`0
`0
`0
`0
`1 (0.3)
`3 (1.0)
`0
`1 (0.3)
`1 (0.3)
`0
`1 (0.3)
`1 (0.3)
`
`43 (7.4)
`15 (2.6)
`2 (0.3)
`0
`1 (0.2)
`0
`1 (0.2)
`1 (0.2)
`0
`1 (0.2)
`2 (0.3)
`5 (0.9)
`0
`0
`1 (0.2)
`0
`2 (0.3)
`2 (0.3)
`7 (1.2)
`1 (0.2)
`3 (0.5)
`1 (0.2)
`0
`1 (0.2)
`3 (0.5)
`
`IAI ¼ intravitreal aflibercept injection; SAE ¼ serious adverse event; 2q4 ¼ 2 mg IAI every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly
`doses.
`Integrated safety analysis set.
`
`Adverse Events
`
`There were no clinically relevant differences in pattern or fre-
`quency of ocular serious adverse events (SAEs) between the
`treatment groups (Table 4). Study eye endophthalmitis was
`reported in 3 eyes in total: 2 eyes (0.7%) in the IAI 2q4 group
`and 1 eye (0.3%) in the IAI 2q8 group. The incidence of
`nonocular SAEs was slightly higher for some events in the
`combined IAI group (e.g., anemia, coronary artery disease,
`and cerebrovascular accident) and for others in the control
`group (e.g., acute myocardial
`infarction and hyperkalemia)
`(Appendix 2, available at www.aaojournal.org). The overall
`incidences of arterial
`thromboembolic events defined by the
`Anti-Platelet Trialists’ Collaboration criteria were 10.7%,
`7.3%, and 7.7% in the IAI 2q4, IAI 2q8, and control groups,
`respectively (Table 5). The incidences of death in the IAI 2q4,
`IAI 2q8, and control groups were 7.7%, 3.9%, and 3.2% in
`VISTA, 5.1%, 5.2%, and 2.3% in VIVID, and 6.5%, 4.5%,
`and 2.8% in the integrated dataset, respectively. A listing of
`
`shown in Appendix 2 (available at
`
`causes of death is
`www.aaojournal.org).
`
`Discussion
`
`In the VISTA and VIVID trials, eyes treated with IAI 2q4
`and IAI 2q8 achieved significantly greater improvements in
`functional and anatomic outcomes when compared with
`laser control at both weeks 52 and 100. The improvements
`observed at 100 weeks7 were maintained over 148 weeks of
`treatment. In both studies, based on the primary analysis
`method LOCF (i.e., censoring measurements after rescue
`treatment was given and including measurements in the
`laser control group after IAI PRN treatment), mean BCVA
`changes from baseline to week 148 were significantly
`greater
`in eyes
`treated with IAI 2q4 and IAI 2q8
`compared with laser control. In addition, a significantly
`
`Table 5. Anti-Platelet Trialists’ CollaborationeDefined Arterial Thromboembolic Events from Baseline to Week 148
`
`Control (n [ 287)
`
`IAI 2q4 (n [ 291)
`
`IAI 2q8 (n [ 287)
`
`All IAI (n [ 578)
`
`Any APTC-ATEs,* n (%)
`Nonfatal myocardial infarction
`Nonfatal stroke
`Vascular death
`
`22 (7.7)
`9 (3.1)
`10 (3.5)
`4 (1.4)
`
`31 (10.7)
`10 (3.4)
`11 (3.8)
`11 (3.8)
`
`21 (7.3)
`9 (3.1)
`7 (2.4)
`6 (2.1)
`
`52 (9.0)
`19 (3.3)
`18 (3.1)
`17 (2.9)
`
`APTC-ATEs ¼ Anti-Platelet Trialists’ Collaborationedefined arterial thromboembolic events; IAI ¼ intravitreal aflibercept injection; 2q4 ¼ 2 mg IAI
`every 4 weeks; 2q8 ¼ 2 mg IAI every 8 weeks after 5 initial monthly doses.
`Integrated safety analysis set.
`*Adjudicated by a masked committee.
`
`2382
`
`Samsung et al. v. Regeneron IPR2023-00884
`Regeneron Pharmaceuticals, Inc. Exhibit 2139 Page 7
`
`

`

`In conclusion, visual acuity improvements observed with
`both IAI regimens (over laser control) at weeks 52 and 100
`were maintained at week 148. Because the IAI 2q4 and IAI
`2q8 regimens had similar efficacy, the IAI 2q8 regimen may
`potentially reduce the treatment burden of anti-VEGF ther-
`apy to manage most patients with DME.
`
`Acknowledgments. Assistance with the study design and
`conduct and data analysis was provided by Karen Chu, MS, and
`Xiaoping Zhu, PhD, Regeneron Pharmaceuticals, Inc (VISTA),
`and Jana Sachsinger, PhD, and Christiane Norenberg, MS, Bayer
`HealthCare (VIVID and integrated safety analysis of VISTA and
`VIVID). Editorial and administrative assistance to the authors was
`provided by Hadi Moini, PhD, Regeneron Pharmaceuticals, Inc.
`Additional editorial assistance was provided by Leigh Prevost,
`PAREXEL, and funded by Bayer HealthCare.
`
`References
`
` Intravitreal Aflibercept for DME
`Heier et al
`greater proportion of IAI-treated eyes gained 10 and 15
`letters at week 148.
`We found that a significantly greater proportion of eyes
`treated with IAI 2q4 and IAI 2q8 had an improvement of 2
`steps in the DRSS score compared with eyes treated with
`laser control. These data support previous observations that
`IAI may promote regression of the underlying diabetic
`retinopathy beyond the macular area.7,14
`Improvements from baseline CST also were significantly
`greater in the IAI 2q4 and IAI 2q8 groups compared with
`the laser control group at week 148 in both studies, and
`there was no evidence that the observed fluctuations in CST
`in the IAI 2q8 group were associated with any limitation in
`visual acuity benefit over the 148 weeks of follow-up.
`On average, the IAI 2q8 regimen provided visual and
`anatomic improvements similar to those achieved by the IAI
`2q4 regimen through week 148 with substantially fewer
`injections. This ability to achieve equivalent visual and
`anatomic benefits with less frequent dosing supports the
`potential for reduced treatment burden,
`including fewer
`clinic visits, for these patients. However, because of the
`spectrum of disease severity, there may be patients with
`DME who would benefit from more frequent treatment, as it
`has been shown for a subpopulation of patients with
`AMD.18 Further analysis is warranted to identify these
`patients.
`(i.e.,
`With the ancillary analysis method (aLOCF)
`including values after all IAI treatments [rescue or PRN]),
`between-group differences narrowed, but remained in favor of
`IAI 2q4 and IAI 2q8 for visual end points when compared
`with laser control. However, between-group differences in
`CST completely disappeared by week 148. These findings
`were expected because a large number of laser control patients
`(>80%) received IAI treatment as rescue beginning at week
`24 or as PRN starting at week 100. Of note, the catch-up with
`visual and anatomic improvements after delayed treatment
`with IAI was more pronounced than those seen in prior trials
`with ranibizumab.19 The remaining between-group differ-
`ences in visual outcomes between the laser and IAI groups are
`not reflective of anatomic improvements (because these were
`similar across treatment groups at week 148), indicating there
`may be irreversible vision losses because of delayed treatment
`with IAI in the laser control group.
`The presence of underlying systemic comorbidities in
`patients with DME can put them at a potentially high risk
`for adverse events, and repeated intravitreal injections over a
`prolonged period of time may increase their cumulative risk
`of complications.20 In both VISTA and VIVID, the overall
`incidences of ocular and nonocular SAEs were similar
`across treatment groups through 148 weeks of treatment.
`The overall
`rates of Anti-Platelet Trialists’ Collabo-
`rationedefined arterial
`thromboembolic events were low
`and comparable across the treatment groups. Although
`deaths were more frequent in the IAI groups, the overall rate
`was low and causes of death were consistent with the un-
`derlying disease and comorbidities present in this patient
`population. In general, even with a longer treatment period,
`no new safety signals were observed, and the adverse event
`profile of IAI over the 148 weeks of the study was consistent
`with the known safety profile of IAI.21
`
`1. International Diabetes Foundation. IDF Diabetes Atlas. http://
`www.diabetesatlas.org/key-messages.html. Accessed January
`2016.
`2. Antonetti DA, Klein R, Gardner TW. Diabetic retinopathy.
`N Engl J Med 2012;366:1227 39.
`3. Boyer