Efficacy and Safety of Monthly versus
`Quarterly Ranibizumab Treatment in
`Neovascular Age-related Macular
`Degeneration: The EXCITE Study
`
`
`
`Ursula Schmidt-Erfurth, MD,' Bora Eldem, MD,? Robyn Guymer, MD, PhD,? Jean-Francois Korobelnik, MD,*
`Reinier O. Schlingemann, MD,° Ruth Axer-Siegel, MD,° Peter Wiedemann, MD,’ Christian Simader, MD,®
`Margarita Gekkieva, MD,°? Andreas Weichselberger, PhD,° on behalf of the EXCITE Study Group*
`
`Objective: To demonstrate noninferiority of a quarterly treatment regimen to a monthly regimen ofranibi-
`zumabin patients with subfoveal choroidal neovascularization (CNV) secondary to age-related macular degen-
`eration (AMD).
`Design: A 12-month, multicenter, randomized, double-masked,active-controlled, phaseIllb study.
`Participants: Patients with primary or recurrent subfoveal CNV secondary to AMD (853 patients), with
`predominantly classic, minimally classic, or occult (no classic component)lesions.
`Intervention: Patients were randomized (1:1:1) to 0.3 mg quarterly, 0.5 mg quarterly, or 0.3 mg monthly
`doses of ranibizumab. Treatment comprised of a loading phase (3 consecutive monthly injections) followed by
`a 9-month maintenance phase(either monthly or quarterly injection).
`Main Outcome Measures: Mean changein best-corrected visual acuity (BCVA) and central retinal thickness
`(CRT) from baseline to month 12 and the incidence of adverse events (AEs).
`Results:
`In the per-protocol population (293 patients), BCVA, measured by Early Treatment Diabetic Reti-
`nopathy Study-like charts, increased from baseline to month 12 by 4.9, 3.8, and 8.3 letters in the 0.3 mg quarterly
`(104 patients), 0.5 mg quarterly (88 patients), and 0.3 mg monthly (101 patients) dosing groups, respectively.
`Similar results were observed in the intent-to-treat (ITT) population (353 patients). The mean decrease in CRT
`from baseline to month 12 in the ITT population was —96.0 um in 0.3 mg quarterly, —105.6 wm in 0.5 mg
`quarterly, and —105.3 um in 0.3 mg monthly group. The mostfrequent ocular AEs were conjunctival hemorrhage
`(17.6%, pooled quarterly groups; 10.4%, monthly group) and eye pain (15.1%, pooled quarterly groups; 20.9%,
`monthly group). There were 9 ocular serious AEs and 3 deaths; 1 death was suspected to be study related
`(cerebral hemorrhage; 0.5 mg quarterly group). The incidences of key arteriothromboembolic events were low.
`Conclusions: After3 initial monthly ranibizumab injections, both monthly (0.3 mg) and quarterly (0.3 mg/0.5
`mg) ranibizumab treatments maintained BCVAin patients with CNV secondary to AMD. At month 12, BCVA gain
`in the monthly regimen washigher than that of the quarterly regimens. The noninferiority of a quarterly regimen
`was not achieved with reference to 5.0 letters. The safety profile was similar to that reported in prior ranibizumab
`studies.
`Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references.
`Ophthalmology 2011;118:831—839 © 2011 by the American Academy of Ophthalmology.
`
`*Group members listed online in Appendix 1 (available at http://aaojournal.org).
`
`Vascular endothelial growth factor (VEGF)-A is a key
`factor involved in the pathogenesis of choroidal neovascu-
`larization (CNV).!° Ranibizumab (Lucentis; Novartis
`Pharma AG,Basel, Switzerland, and Genentech Inc., South
`San Francisco, CA) is a recombinant, fully humanized,
`affinity-matured monoclonalantigen-binding antibody frag-
`mentthat inhibits the binding of multiple biologically active
`forms of VEGF-A to their receptors.°~®
`Twopivotal Phase III trials, MARINA (Minimally clas-
`sic/occult trial of the Anti-VEGFantibody Ranibizumab /n
`the treatment of Neovascular Age-related macular degener-
`
`ation)? and ANCHOR(AMti-VEGEFantibodyforthe treat-
`ment of predominantly classic CHORoidal neovasculariza-
`tion in
`age-related macular degeneration),!°!!
`have
`previously demonstrated the efficacy of the monthly dosing
`regimensof ranibizumabin improvingvisual acuity (VA)in
`patients with subfoveal CNV secondary to age-related mac-
`ular degeneration (AMD). These studies also described the
`safety and tolerability profile of intravitreal treatment using
`ranibizumab. Based on its favorable benefit/risk ratio,
`ranibizumabreceived marketing authorization for thetreat-
`ment of CNV secondary to AMD from the US Food and
`
`© 2011 by the American Academy of Ophthalmology
`Published by Elsevier Inc.
`
`ISSN 0161-6420/11/$-see front matter
`doi:10.1016/j.ophtha.2010.09.004
`
`831
`
`U.S. Pat. 9,669,069, Exhibit 2092
`
`Mylan v. Regeneron, IPR2021-00880
`
`Exhibit 2092
`Page 01 of 09
`
`

`

`Ophthalmology Volume 118, Number 5, May 2011
`
`Drug Administration, the European Medicines Evaluation
`Agency, and many other national health authorities around
`the world since 2006.
`Although the monthly regimen of ranibizumab provides
`the best known treatment outcome as indicated by cumula-
`tive clinical evidence,10,11 there was a need to evaluate
`whether a less frequent
`treatment regimen can also be
`effective, while decreasing the treatment burden caused by
`monthly intravitreal injections. In this context, the PIER (A
`Phase IIIb, Multicenter, Randomized, Double Masked,
`Sham Injection Controlled Study of the Efficacy and Safety
`of Ranibizumab in Subjects with Subfoveal Choroidal Neo-
`vascularization [CNV] with or without Classic CNV Sec-
`ondary to Age- Related Macular Degeneration) study of the
`12-month efficacy of quarterly dosing of ranibizumab after
`3 consecutive monthly injections (6 doses per year instead
`of 12 for the first treatment year) was the first to test an
`alternative maintenance regimen.12 The 12-month efficacy
`result of PIER showed that both 0.3 mg and 0.5 mg ranibi-
`zumab injections provided statistically significant superior-
`ity in VA improvement as compared with sham treatment,
`with corresponding treatment differences of ⱖ3 lines. How-
`ever, mean changes in best-corrected VA (BCVA) from
`baseline to month 12 in the quarterly ranibizumab dosing
`groups (⫺1.6 letters for 0.3 mg and ⫺0.2 letters for 0.5
`mg) was lower than that observed with the monthly
`dosing regimens of 0.3 mg and 0.5 mg ranibizumab in the
`MARINA (⫹7.2 letters) and ANCHOR studies (⫹11.3 let-
`ters). Importantly, because these studies did not directly
`compare the monthly and quarterly dosing regimens, an
`appropriate inference of the clinical benefits of the different
`maintenance treatment regimens is limited.
`The first prospective trial designed to directly compare
`monthly and quarterly ranibizumab dosing regimens,
`EXCITE evaluated patients with subfoveal CNV secondary
`to AMD. This 1-year study had an active control arm of
`continuous monthly injections (0.3 mg) versus the less
`frequent dosing schedules of 3 initial monthly injections of
`0.3 mg or 0.5 mg ranibizumab followed by quarterly injec-
`tions of the respective doses. The primary objective of this
`study was to investigate whether a maintenance strategy
`using a quarterly dosing regimen (0.3 and 0.5 mg) was
`noninferior to a monthly dosing regimen as determined by
`the mean change in BCVA from baseline to month 12 in the
`study population. The key secondary objectives were to
`assess possible differences in the proportion of patients with
`loss or gain of BCVA of ⱖ15 letters, loss of BCVA of ⱖ30
`letters, mean change in central retinal thickness (CRT) from
`baseline, overall safety, and tolerability.
`
`Methods
`
`Study Design
`The EXCITE study was a 1-year, randomized, double-masked,
`active-controlled, multicenter, Phase IIIb study in patients with
`subfoveal CNV secondary to AMD, comparing the efficacy and
`safety of quarterly dosing regimens of ranibizumab with a monthly
`dosing regimen during the maintenance phase, that is, from month
`3 onward.
`Eligible patients were randomly assigned in a 1:1:1 ratio to any
`of the following 3 double-masked treatment arms (Fig 1): loading
`doses of 3 initial monthly intravitreal injections of 0.3 mg (arm A)
`or 0.5 mg (arm B) ranibizumab followed by quarterly injections of
`the respective doses at months 5, 8, and 11 (i.e., a total of 6
`injections) or 0.3 mg ranibizumab administered monthly from
`baseline to month 11 (arm C, active control) (i.e., a total of 12
`injections). Primary end point analysis was at month 12. To
`maintain masking, patients in treatment arms A and B were ad-
`ministered a sham injection during the monthly visits for which no
`intravitreal injection was scheduled.
`This study was conducted in a total of 59 study centers in 16
`European countries, Australia, Brazil, Israel, and Turkey in accor-
`dance with the declaration of Helsinki and International Confer-
`ence on Harmonization Good Clinical Practice guidelines. Ap-
`proval was obtained from the independent Ethics Committee or
`Institutional Review board at each participating center. All patients
`provided signed informed consent before participating in the study.
`The trial is registered at clinicaltrials.gov (NCT00275821).
`
`Inclusion and Exclusion Criteria
`Patients aged ⱖ50 years and suffering from primary or recurrent
`subfoveal CNV secondary to AMD, with predominantly classic,
`minimally classic, or occult (with no classic component) lesions
`were included in the study. The reading center (DARC) required
`active CNV for confirmation of the patient inclusion. Other inclu-
`sion criteria, based on study eye characteristics were as follows:
`total area of CNV (including classic and occult components)
`ⱖ50% of the total lesion area; the total lesion area ⱕ12 disc areas
`for minimally classic or occult with no classic component or ⱕ9
`disc areas (5400 ␮m) for predominately classic lesions; and BCVA
`score between 73 and 24 letters (approximately 20/40 to 20/320
`Snellen equivalent).
`Exclusion criteria were as follows: BCVA score of ⬍34 letters
`in both eyes; previous treatment or participation in a clinical trial
`(for either eye) with antiangiogenic drugs; use of any other inves-
`tigational drugs at the time of screening, or within 30 days or 5
`half-lives of screening; prior treatment in the study eye with
`verteporfin, external-beam radiation therapy, subfoveal focal laser
`photocoagulation, vitrectomy, or transpupillary thermotherapy;
`operative intervention for AMD in the past in the study eye; laser
`photocoagulation in the study eye within 1 month preceding base-
`line; angioid streaks or precursors of CNV in either eye due to
`
`Figure 1. Dosing schedule of ranibizumab regimen in the EXCITE study.
`
`832
`
`Exhibit 2092
`Page 02 of 09
`
`

`

`Schmidt-Erfurth et al
`
`䡠 Monthly versus Quarterly Ranibizumab Dosing in EXCITE Study
`
`other causes; clinically significant subretinal hemorrhage in the
`study eye that involved the foveal center; or any other significant
`clinical condition detrimental to the study outcome.
`Patients’ eligibility was confirmed by an independent masked
`Central Reading Center, DARC, at screening by fundus photog-
`raphy and fluorescein angiography. The DARC also classified the
`lesion types and assessed lesion area, area of CNV, and leakage
`activity based on fluorescein angiography at months 6 and 12. A
`separate independent masked Central Reading Center (Vienna
`Reading Center) reviewed optical coherence tomography (OCT)
`images to provide an objective assessment of retinal thickness for
`each monthly assessment of all patients.
`
`Study Assessments
`
`Efficacy. Visual acuity was assessed in both eyes at each study
`visit using Early Treatment Diabetic Retinopathy Study-like charts
`at an initial testing distance of 4 m. The change in BCVA from
`baseline to each visit was assessed. The mean change in BCVA
`from baseline to month 12 was the primary end point. In addition,
`change in BCVA was assessed as the proportion of patients with
`⬍15 letters loss, ⱖ30 letters loss, ⱖ0 letters gain, and ⱖ15 letters
`gain in BCVA from baseline to month 12. The CRT was measured
`in both eyes by time domain OCT at screening, and at each
`monthly visit until month 12. Baseline BCVA and OCT were
`performed before treatment. Fluorescein angiograms were used to
`evaluate CNV lesions at screening, month 6, and month 12. In 84%
`of patients (296 out of 353 patients), visual function contrast
`sensitivity was assessed in both eyes at baseline, month 6, and
`month 12 using Pelli-Robson charts.
`Safety. Adverse events (AEs), serious AEs, and changes in
`vital signs were assessed monthly during the study. Biochemical
`values were measured at screening and at the end of the study visit
`(month 12), and hematology, blood chemistry, and urine were
`regularly monitored. Intraocular pressure measurement (before and
`after each administration by tonometry) and standard ophthalmic
`examination were also performed monthly.
`
`Statistical Analysis
`A population size of 350 randomized patients was planned to reach
`a sample size of 101 per protocol (PP) patients per treatment arm,
`assuming a dropout and protocol deviation rate of 13%. The
`dropout rate and protocol deviation calculations were based on
`results of the MARINA clinical study data. The PP population was
`
`chosen as the primary analysis population to assess the primary
`end point and to evaluate the null hypothesis of noninferiority of
`quarterly treatment regimen to monthly treatment regimen in terms
`of change in BCVA from baseline to month 12. Assuming that
`there is no difference between quarterly and monthly treatment
`regimens, there was a power of ⱖ83% to reject this null hypothesis
`and therefore conclude that quarterly treatment is noninferior to
`monthly treatment using 6.8 letters as the noninferiority margin.
`For both alternative dosing treatment arms (0.3 and 0.5 mg
`quarterly),
`the noninferiority to the reference arm (0.3 mg
`monthly) was tested using 1-sided testing procedures (or equiva-
`lent, using 1-sided confidence intervals [CIs]), while keeping an
`overall type I error level of 0.025. The Hochberg procedure was
`used to control for multiplicity; that is, the null hypothesis was
`rejected if either or both comparisons were statistically significant
`at a 0.025 level or ⱖ1 comparison was statistically significant at a
`0.0125 level. For both quarterly dosing arms (0.3 and 0.5 mg), the
`⫺ um
`ⱕ – 6.8 and the alternative hypothesis
`null hypothesis H0: uq
`⫺ um
`⬎ – 6.8 were tested, where uq and um were the mean
`Ha: uq
`changes in BCVA from baseline/month 3 to month 12 in the
`quarterly dosing treatment arms (q) and the monthly reference arm
`(m), respectively, with a noninferiority limit of – 6.8. The nonin-
`feriority limit was based on the results of a previous study in which
`the value of 6.8 was approximately one half of the minimum
`estimated difference (13.6; lower limit of a 2-sided 95% CI) in the
`mean change in BCVA from baseline to month 12, with testing
`distance of 4 m between the ranibizumab 0.3 mg and sham
`injection groups.9 Noninferiority of 0.5 mg quarterly to 0.3 mg
`monthly was assessed based on the change from baseline to month
`12, and noninferiority analysis of 0.3 mg quarterly versus 0.3 mg
`monthly could be based on the change from month 3 to month 12
`because any differences at month 3 between the 0.3 mg groups
`could be attributed to chance (up to the month 3 assessment there
`was no difference in the corresponding treatment regimen).
`The mean change in BCVA from baseline to month 12 was
`analyzed by using an analysis of variance with treatment, baseline
`BCVA (ⱕ52 vs ⱖ53 letters), and lesion type as factors.
`The primary end point was analyzed for both PP and intent-to-
`treat (ITT) populations. The PP population was a subset of the ITT
`population and included patients who had an assessment for
`BCVA at month 12 and with no major study protocol deviation.
`The ITT population comprised all randomized patients. The last
`observation carried forward method was used to impute missing
`values for the ITT population for all efficacy measures. All the
`
`Table 1. Summary of the EXCITE Patient Disposition
`
`0.3 mg
`Quarterly, n (%)
`
`0.5 mg
`Quarterly, n (%)
`
`0.3 mg
`Monthly, n (%) Total, n (%)
`
`Enrolled
`Randomized
`Completed*
`Early discontinued from study
`Adverse event(s)
`Administrative problems
`Patient withdrew consent
`Lost to follow-up
`Death
`Abnormal test procedure result(s)
`Unsatisfactory therapeutic effect
`Protocol deviation
`
`—
`120
`106 (88.3)
`14 (11.7)
`4 (3.3)
`3 (2.5)
`0
`0
`0
`0
`2 (1.7)
`5 (4.2)
`
`—
`118
`95 (80.5)
`23 (19.5)
`12 (10.2)
`4 (3.4)
`2 (1.7)
`1 (0.8)
`2 (1.7)
`0
`1 (0.8)
`1 (0.8)
`
`—
`115
`103 (89.6)
`12 (10.4)
`5 (4.3)
`4 (3.5)
`1 (0.9)
`1 (0.9)
`1 (0.9)
`0
`0
`0
`
`482
`353
`304 (86.1)
`49 (13.9)
`21 (5.9)
`11 (3.1)
`3 (0.8)
`2 (0.6)
`3 (0.8)
`0
`3 (0.8)
`6 (1.7)
`
`*Completed the study and underwent visual acuity assessment at month 12.
`
`833
`
`Exhibit 2092
`Page 03 of 09
`
`

`

`Ophthalmology Volume 118, Number 5, May 2011
`
`safety parameters were calculated for the safety (i.e., ITT, in this
`study) population.
`
`10.2% in 0.5 mg quarterly; 4.3% in 0.3 mg monthly). Details of
`patient disposition are given in Table 1.
`
`Results
`
`Patients
`A total of 482 patients were screened and 353 patients were
`randomized for treatment with the study medication. As per the
`study design, patients received ranibizumab 0.3 mg quarterly (120
`patients), ranibizumab 0.5 mg quarterly (118 patients), or ranibi-
`zumab 0.3 mg monthly dosing (115 patients). The PP population
`included 104 patients (86.7%) from the 0.3 mg quarterly, 88
`(74.6%) from the 0.5 mg quarterly, and 101 (87.8%) from the 0.3
`mg monthly dosing groups. The study was completed by 106
`patients (88.3%) in the ranibizumab 0.3 mg quarterly group, 95
`(80.5%) in the ranibizumab 0.5 mg quarterly group, and 103
`(89.6%) in the ranibizumab 0.3 mg monthly treatment group. In all
`3 treatment groups, the most frequently reported reason for early
`discontinuation from study was AEs (3.3% in 0.3 mg quarterly;
`
`Baseline Characteristics and Treatment Exposure
`Baseline demographic and ocular disease characteristics of pa-
`tients (ITT population) in the EXCITE study are summarized in
`Table 2. The treatment groups were balanced with respect to
`baseline BCVA, CRT, and fluorescein angiography of the study
`eye. Approximately 20% of patients had predominantly classic
`lesion, 40% patients had minimally classic lesion, and 40% pa-
`tients had occult (with no classic component) lesion.
`The mean (standard deviation) number of active treatment
`injections received over the study treatment period from baseline
`to month 11 were 5.7 (0.80), 5.5 (1.05), and 11.4 (1.69) in the 0.3
`mg quarterly, 0.5 mg quarterly, and 0.3 mg monthly groups,
`respectively.
`
`Efficacy
`The mean change in BCVA in the study eye from baseline over
`time (PP population) is shown in Figure 2A. In the PP population,
`
`Table 2. Demographics and Baseline Characteristics of the Study Eye of Patients who Entered Treatment in the EXCITE Study
`(Intent-to-Treat Population)
`
`Characteristic
`
`Gender, n (%)
`Women
`Men
`Race, n (%)
`Caucasian
`Asian
`Other
`Age (yrs)
`Mean (SD)
`Age group, n (%)
`50–64
`65–74
`75–84
`ⱖ85
`History
`Years since first diagnosis, mean (SD)
`BCVA (letters)*
`Mean (SD)
`ⱕ52
`ⱖ53
`BCVA (Snellen equivalent)*
`ⱕ20/200
`⬎20/200 and ⬍20/40
`ⱕ20/40
`CNV classification
`Predominantly classic
`Minimally classic
`Occult (no classic)
`Retinal thickness at central point (␮m)§
`n
`Mean (SD)
`Retinal thickness at central subfield (␮m)
`Mean (SD)
`
`0.3 mg
`Quarterly (n ⴝ 120)
`
`0.5 mg
`Quarterly (n ⴝ 118)
`
`0.3 mg
`Monthly (n ⴝ 115)
`
`Total (n ⴝ 353)
`
`70 (58.3)
`50 (41.7)
`
`118 (98.3)
`1 (0.8)
`1 (0.8)
`
`75.1 (7.45)
`
`13 (10.8)
`37 (30.8)
`61 (50.8)
`9 (7.5)
`
`0.57 (1.424)
`
`55.8 (11.81)
`46 (38.3)
`74 (61.7)
`
`5 (4.2)
`94 (78.3)
`21 (17.5)
`
`25 (20.8)
`50 (41.7)
`45 (37.5)
`
`100
`313.6 (85.05)
`
`321.4 (86.80)
`
`73 (61.9)
`45 (38.1)
`
`117 (99.2)
`0
`1 (0.8)
`
`75.8 (6.96)
`
`12 (10.2)
`28 (23.7)
`72 (61.0)
`6 (5.1)
`
`0.52 (1.14)
`
`57.7 (13.06)
`33 (28.0)
`85 (72.0)
`
`8 (6.8)
`84 (71.2)
`26 (22.0)
`
`27 (22.9)
`46 (39.0)
`45 (38.1)
`
`66 (57.4)
`49 (42.6)
`
`113 (98.3)
`0
`2 (1.7)
`
`75 (8.26)
`
`10 (8.7)
`45 (39.1)
`46 (40.0)
`14 (12.2)
`
`0.56 (2.177)
`
`56.5 (12.19)
`36 (31.3)
`79 (68.7)
`
`6 (5.2)
`86 (74.8)
`23 (20.0)
`
`21 (18.3)
`46 (40.0)
`48 (41.7)
`
`209 (59.2)
`144 (40.8)
`
`348 (98.6)
`1 (0.3)
`4 (1.1)
`
`75.3 (7.56)
`
`35 (9.9)
`110 (31.2)
`179 (50.7)
`29 (8.2)
`
`0.55 (1.629)
`
`56.7 (12.4)
`115 (32.6)
`238 (67.4)
`
`19 (5.4)
`264 (74.8)
`70 (19.8)
`
`73 (20.7)
`142 (40.2)
`138 (39.1)
`
`100
`324.5 (115.94)
`
`95
`320.6 (118.55)
`
`295
`319.5 (107.13)
`
`331.9 (105.74)
`
`326.6 (99.44)
`
`326.6 (97.38)
`
`AMD ⫽ age-related macular degeneration; BCVA ⫽ best-corrected visual acuity; CNV ⫽ choroidal neovascularization; ETDRS ⫽ Early Treatment
`Diabetic Retinopathy Study; SD ⫽ standard deviation.
`*Measured using ETDRS-like charts at a distance of 4 m.
`§Measured using optical coherence tomography.
`
`834
`
`Exhibit 2092
`Page 04 of 09
`
`

`

`Schmidt-Erfurth et al
`
`䡠 Monthly versus Quarterly Ranibizumab Dosing in EXCITE Study
`
`baseline to month 12.’ However, given that the 97.5% CI barely
`includes 0, it can be interpreted that 0.3 mg quarterly treatment is
`numerically inferior to the 0.3 mg monthly treatment regimen.
`The BCVA time course in the ITT population (last observation
`carried forward method) was consistent with that of the PP pop-
`ulation, with a mean change in BCVA from baseline to month 12
`of 4.0, 2.8, and 8.0 letters for the ranibizumab 0.3 mg quarterly, 0.5
`mg quarterly, and 0.3 mg monthly groups, respectively (P ⫽
`0.0751 [95% CI, ⫺7.7 to ⫺0.9] for the 0.3 mg quarterly and P ⫽
`0.1678 [95% CI, ⫺8.6 to ⫺1.7] for the 0.5 mg quarterly, both
`compared with the 0.3 mg monthly group). In the monthly treat-
`ment regimen, the initially gained mean BCVA remained stable
`during the treatment period, whereas it gradually decreased in the
`quarterly treatment regimens with a pattern reflecting the impact
`of the quarterly injections (Fig 2B). The BCVA values at baseline
`and the change from baseline at month 12 are given in Table 3 for
`both the PP and the ITT populations.
`The proportion of patients who lost ⬍15 letters from baseline
`to month 12 was similar across the treatment groups (ITT popu-
`lation) with 93.3%, 91.5%, and 94.8% in the 0.3 mg quarterly, 0.5
`mg quarterly, and 0.3 mg monthly ranibizumab groups, respec-
`tively (Fig 3A). The proportion of patients who had a VA gain of
`ⱖ15 letters from baseline to month 12 was 14.2% in the ranibi-
`zumab 0.3 mg quarterly group, 17.8% in the ranibizumab 0.5 mg
`quarterly group, and 28.7% in the ranibizumab 0.3 mg monthly
`group (Fig 3B). The proportion of patients with a gain of ⱖ0 letters
`of VA were 71.7% (86/120; 0.3 mg quarterly), 66.9% (79/118; 0.5
`mg quarterly), and 82.6% (95/115; 0.3 mg monthly) at month 12.
`The percentage of patients, at month 12, with a VA Snellen
`equivalent of ⱕ20/200 (BCVA ⫽ 34 letters) was greater in the
`quarterly dosing regimen (7.5% for 0.3 mg and 6.8% for 0.5 mg)
`compared with the 0.3 mg monthly dosing regimen (2.6%). Severe
`vision loss (ⱖ30 letters) at the end of this study was observed in
`2 patients (1.7%) of each of the quarterly treatment groups and in
`none of the 0.3 mg monthly treatment group.
`Anatomically, the overall reduction in CRT of the study eye
`from baseline to month 3 and to month 12 was similar between the
`3 treatment groups in the ITT population. However, although the
`mean CRT decreased similarly from baseline to month 3 in all 3
`treatment groups, thereafter it remained more or less stable at the
`monthly dosing regimen but was variable in the quarterly dosing
`groups (mean CRT decrease 1 month after each treatment and
`increase thereafter until next treatment visit at months 5, 8, and 11;
`Fig 4). The mean change in CRT from baseline to month 12 was
`similar between the 0.5 mg quarterly group (⫺105.6 ␮m) and the
`0.3 mg monthly group (⫺105.3 ␮m). For the 0.3 mg quarterly
`group, the mean CRT change was ⫺96.0 ␮m. The overall retinal
`thickness at the central subfield of the study eye at baseline and
`months 3 and 12 was also similar between the treatment groups.
`On the basis of angiographic data, the mean decrease in CNV
`lesion area from baseline to month 12 was numerically higher in
`the 0.5 mg quarterly treatment group compared with the other
`treatment groups; however, this difference was not significant
`(⫺2.28 mm2 in the 0.3 mg quarterly, ⫺3.49 mm2 in the 0.5 mg
`quarterly, and ⫺2.63 mm2 in the 0.3 mg monthly dosing regimen;
`Table 4). The mean change (decrease) from baseline to month 12 in
`the total area of leakage and total lesion area are shown in Table 4.
`Contrast sensitivity analysis (ITT population; last observation
`carried forward method) at month 6 showed a mean change of
`0.071 log units from baseline in the 0.3 mg quarterly group (100
`patients), 0.107 log units in the 0.5 mg group quarterly group (98
`patients), and 0.123 log units in the 0.3 mg monthly treatment
`group (98 patients). In the 0.3 mg quarterly, 0.5 mg quarterly, and
`0.3 mg monthly treatment groups, the mean change from baseline
`to month 12 showed an overall improvement by 0.085, 0.081, and
`0.131 log units, respectively.
`
`835
`
`Figure 2. Mean change in best-corrected visual acuity score from baseline
`over time in the (A) per-protocol population (study visit) and (B) intent-
`to-treat population (last observation carried forward [LOCF]) of EXCITE.
`Vertical bars represent standard error of the mean.
`
`the mean BCVA increase from baseline to month 12 (primary end
`point) was 4.9, 3.8, and 8.3 letters in the 0.3 mg quarterly, 0.5 mg
`quarterly, and 0.3 mg monthly groups, respectively. In all the 3
`treatment arms, the mean BCVA increased from baseline to month
`3 (monthly dosing phase for all treatment arms) by 6.8, 6.6, and 7.5
`letters, in the 0.3 mg quarterly, 0.5 mg quarterly, and 0.3 mg
`monthly groups, respectively. However, between months 3 and 12
`(maintenance phase), patients in the quarterly treatment groups lost
`1.8 (0.3 mg quarterly) and 2.8 (0.5 mg quarterly) letters, whereas
`patients in the monthly treatment group gained 0.8 letters on
`average. Up to month 3, there was no notable difference between
`the treatment arms. The first notable difference was observed at
`month 4, that is, 2 months after the last loading dose (Fig 2A).
`Although this study was designed to test noninferiority of the
`quarterly treatment regimen versus monthly treatment regimen,
`this was not achieved for the 0.5 mg quarterly regimen, as evi-
`denced by the lower CI limits for the corresponding treatment
`difference being below the noninferiority threshold of ⫺6.8 letters
`(95% CI, ⫺7.9 to ⫺0.7; 97.5% CI, ⫺8.4 to ⫺0.2; P ⫽ 0.0867).
`For the comparison of 0.3 mg quarterly versus 0.3 mg monthly
`treatment groups (from months 3 to 12), the lower CI limit (97.5%
`CI, ⫺5.6 to 0.22; P ⫽ 0.0008), however, indicates a theoretical
`noninferiority, also driven by the smaller variability in the end
`point ‘change from months 3 to 12’ compared with ‘change from
`
`Exhibit 2092
`Page 05 of 09
`
`

`

`Ophthalmology Volume 118, Number 5, May 2011
`
`Table 3. Best-Corrected Visual Acuity at Baseline and Mean Change from the Baseline in the Study
`Eye at Month 12
`
`0.3 mg Quarterly
`
`0.5 mg Quarterly
`
`0.3 mg Monthly
`
`PP population (observed)
`n
`Baseline mean (SD)
`Month 12 mean (SD)
`Change from baseline, mean (SD)
`Comparison vs monthly dosing
`Mean difference (SE)
`95% CI
`97.5% CI§
`P-value*§
`ITT population (LOCF)
`n
`Baseline mean (SD)
`Month 12 mean
`Change from baseline, mean (SD)
`Comparison vs monthly dosing
`Mean difference (SE)
`95% CI
`97.5% CI
`P-value*
`
`104
`55.3 (12.11)
`60.2 (16.01)
`4.9 (13.13)
`
`⫺3.3 (1.76)
`⫺7.1, ⫺0.2
`⫺7.6, 0.3
`0.0365
`
`120
`55.8 (11.81)
`59.8 (17.20)
`4.0 (14.88)
`
`⫺3.9 (1.75)
`⫺7.7, ⫺0.9
`⫺8.2, ⫺0.4
`0.0751
`
`88
`57.5 (13.07)
`61.3 (16.32)
`3.8 (13.33)
`
`⫺4.5 (1.84)
`⫺7.9, ⫺0.7
`⫺8.4, ⫺0.2
`0.0867
`
`118
`57.7 (13.06)
`60.5 (16.50)
`2.8 (13.78)
`
`⫺5.2 (1.76)
`⫺8.6, ⫺1.7
`⫺9.1, ⫺1.2
`0.1678
`
`101
`56.2 (12.33)
`64.5 (16.27)
`8.3 (11.31)
`
`115
`56.5 (912.19)
`64.5 (15.85)
`8.0 (11.27)
`
`CI ⫽ confidence interval; ITT ⫽ intent to treat; LOCF ⫽ last observation carried forward; PP ⫽ per protocol;
`SD ⫽ standard deviation; SE ⫽ standard error.
`*One-sided test of H0: mean difference(test-reference) ⱕ ⫺6.8.
`§The CI for the difference between months 3 and 12 (0.3 mg quarterly group) is –5.6, 0.22 (P ⫽ 0.0008).
`
`Safety
`The AEs (ⱖ3% in any group) are summarized in Table 5 (avail-
`able online at http://aaojournal.org). The most frequently reported
`ocular AEs were eye pain (18.3%, 11.9%, and 20.9% in the 0.3 mg
`quarterly, 0.5 mg quarterly, and 0.3 mg monthly groups, respec-
`tively), conjunctival hemorrhage (19.2 %, 16.1%, and 10.4% in the
`0.3 mg quarterly, 0.5 mg quarterly, and 0.3 mg monthly groups,
`respectively), reduced VA (13.3%, 16.1%, and 7.8% in the 0.3 mg
`quarterly, 0.5 mg monthly, and 0.3 mg monthly groups, respec-
`tively), and increased intraocular pressure of ⬎10 mmHg (5.0%,
`5.9%, and 14.8% in the 0.3 mg quarterly, 0.5 mg quarterly, and 0.3
`mg monthly groups, respectively). Among the nonocular AEs
`reported, the incidence of nasopharyngitis was the highest (9.2%,
`3.4%, and 7.0% in the 0.3 mg quarterly, 0.5 mg quarterly, and 0.3
`mg monthly groups, respectively), followed by hypertension (8.3%
`for 0.3 mg quarterly, 5.1% for 0.5 mg quarterly, and 7.0% for 0.3
`mg monthly). There was no apparent trend of a dose or treatment
`frequency–related change in AE incidences, although differences
`between groups were observed with respect to individual AEs.
`A total of 12 patients (10.0%) in the 0.3 mg quarterly group, 10
`patients (8.5%) in the 0.5 mg quarterly group, and 13 patients
`(11.3%) in the monthly treatment group experienced AEs that
`could be potentially related to systemic VEGF inhibition (Table 6;
`available online at http://aaojournal.org). Arteriothromboembolic
`events reported in this study showed no increased risk of stroke in
`the monthly dosing regimen as compared with that of the quarterly
`dosing regimens (Table 6). There were 3 incidences of angina
`pectoris (1 in each of the groups) and 2 incidences of myocardial
`infarction (1 each in 0.3 mg quarterly and monthly groups). Other
`incidences of arteriothromboembolic events were cerebrovascular
`accident (1 in 0.3 mg monthly group) and pulmonary embolism (1
`in the 0.3 mg monthly group). Nonocular hemorrhage was reported
`in the 0.5 mg quarterly group (4 patients; 3.4%) and in the 0.3 mg
`monthly group (1 patient; 0.9%).
`
`836
`
`Incidences of serious AEs were reported in 15 patients (12.5%)
`in the ranibizumab 0.3 mg quarterly group, 23 patients (19.5%) in
`the 0.5 mg quarterly group, and 20 patients (17.4%) in the 0.3
`mg monthly treatment group (Table 7; available online at http://
`aaojournal.org). The incidence of ocular serious AEs in the study
`eye was low: 2.5% in the 0.3 mg quarterly group, 4.2% in the 0.5
`mg quarterly group, and 0.9% in the 0.3 mg monthly group. Three
`deaths occurred during this study (Table 7), which were due to
`cardiorespiratory arrest and cerebral hemorrhage (both in the 0.5
`mg quarterly group) and lung infection (1 in the 0.3 mg monthly
`group). Of the 3 deaths, 1 was suspected to be related to the study
`medication. This patient (male, 73 years) received treatment of 0.5
`mg ranibizumab quarterly and had an active medical condition,
`including diabetes mellitus, hypertension, chronic renal failure,
`drug hypersensitivity, cataract in both eyes, dementia Alzheimer’s
`type, gastritis, vitamin B complex deficiency, hypercholesterol-
`emia, and hyperuricemia, and was under multiple concomitant
`medications. The patient died owing to cerebral hemorrhage 41
`days after the previous ranibizumab administration.
`Study discontinuation owing to AEs was higher in the 0.5 mg
`quarterly group (13 patients, 11%), compared with the 0.3 mg
`quarterly (4 patients, 3.3%) or the 0.3 mg monthly treatment group
`(6 patients, 5.2%). Ocular AEs of the study eye that led to treat-
`ment discontinuation were reported in 3 patients (2.5%) in the 0.3
`mg quarterly group, 5 patients (4.2%) in the 0.5 mg quarterly
`group, and 1 patient (0.9%) in the 0.3 mg monthly treatment group.
`Two patients (1.7%) from the 0.5 mg quarterly group discontinued
`because of AEs of the fellow eye.
`
`Discussion
`
`The EXCITE trial is the first study directly comparing
`visual outcomes between monthly and quarterly dosing
`
`Exhibit 2092
`Page 06 of 09
`
`

`

`Schmidt-Erfurth et al
`
`䡠 Monthly versus Quarterly Ranibizumab Dosing in EXCITE Study
`
`0.3 mg monthly dosing regimen (8.0 letters) is in line with
`that of the ANCHOR (8.5 letters, 0.3 mg; 11.3 letters, 0.5
`mg)10 and MARINA (6.5 letters, 0.3 mg; 7.2, 0.5 mg)9
`trials. Also, the other VA outcomes in this group, such as
`the proportion of patients with loss or gain of BCVA (15
`letters) and BCVA equaling ⱕ34 letters, are similar to
`those of the ANCHOR study and better than those of the
`MARINA study. A comparison of the quarterly results from
`this study to the pivotal PIER study12 revealed numerically
`better BCVA improvement in EXCITE patients. The PIER
`study compared the efficacy of quarterly dosing of ranibi-
`zumab with that of sham treatment. The mean BCVA in-
`creased from baseline to month 12 for the quarterly dosing
`groups in EXCITE by 4.0 letters in the 0.3 mg quarterly
`group and 2.8 letters in the 0.5 mg group, wher