LONG-TERM FOLLOW-UP FOR TYPE 1
`(SUBRETINAL PIGMENT EPITHELIUM)
`NEOVASCULARIZATION USING A
`MODIFIED ‘‘TREAT AND EXTEND’’
`DOSING REGIMEN OF INTRAVITREAL
`ANTIVASCULAR ENDOTHELIAL
`GROWTH FACTOR THERAPY
`
`MICHAEL ENGELBERT, MD, PHD,*† SANDRINE A. ZWEIFEL, MD,†‡ K. BAILEY FREUND, MD*†§
`
`Purpose: The purpose of the study was to analyze long-term outcomes for the treatment
`of type 1 (subretinal pigment epithelium) neovascularization using a modified ‘‘treat and
`extend’’ antivascular endothelial growth factor dosing regimen.
`Methods: We performed a retrospective, noncomparative analysis of visual acuity,
`funduscopic, and optical coherence tomography data for 18 eyes of 16 consecutive
`patients with newly diagnosed type 1 neovascularization treated with intravitreal bevaci-
`zumab and/or ranibizumab with at least 24-month follow-up. Three monthly injections were
`followed by continued treatment at intervals increasing by 2 weeks per visit to a maximum
`of 10 weeks. The interval was shortened if clinical or optical coherence tomography
`evidence of recurrent fluid at the foveola or increased extrafoveolar fluid was detected.
`Results: Median baseline logarithm of the minimum angle of resolution visual acuity was
`0.53 (20/69 Snellen equivalent) and remained stable at 24 months (logarithm of the
`minimum angle of resolution 0.52, P = 0.84) after an average of 12 injections (range, 8–19
`injections) and at 36 months (logarithm of the minimum angle of resolution 0.52, P = 0.68)
`after an average of 20 injections (range, 18–25 injections). Although most eyes (15 of 18
`[83%]) continued to manifest extrafoveolar subretinal fluid throughout the course of
`treatment, only 1 eye developed geographic atrophy overlying the areas of choroidal
`neovascularization. During a cumulative observation period of 540 months, no eyes
`developed a sight-threatening submacular hemorrhage.
`Conclusion: A modified ‘‘treat and extend’’ dosing regimen of intravitreal antivascular
`endothelial growth factor therapy reduces the need for monthly visits and imaging and
`allows for stable long-term visual acuity in eyes with type 1 neovascularization.
`RETINA 30:1368–1375, 2010
`
`T ype 1 neovascularization occurs as a subtype of
`
`neovascular age-related macular degeneration
`(AMD) in which the abnormal vessels are located
`between Bruch membrane and the basal surface of the
`retinal pigment epithelium (RPE). Type 1 neovascu-
`larization typically exhibits an occult pattern with
`fluorescein angiography and always manifests some
`degree of RPE elevation with optical coherence
`tomography (OCT). The type 1 pattern has a different
`natural course and treatment response from the type 2
`(classic) and type 3 (retinal angiomatous proliferation)
`
`neovascular patterns. For example, eyes with type 1
`(occult) neovascularization presented with better visual
`acuity in the MARINA study1 than eyes with type 2
`(classic) neovascularization in the ANCHOR study.2
`In addition, in these trials, patients with type 1 neovascu-
`larization did not gain as many letters1 as patients with
`type 2 neovascularization,2 although they were treated
`with the same dosing regimen of continuous monthly
`intravitreal injections of ranibizumab during 24 months.
`It has been hypothesized that the type 1 neovascular
`pattern may be a compensatory form of neovascular
`
`1368
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 1
`
`Joining Petitioner: Apotex
`
`

`

`‘‘TREAT AND EXTEND’’ FOR TYPE 3 CHOROIDAL NEOVASCULARIZATION  ENGELBERT ET AL
`
`1369
`
`growth occurring in response to an ischemic outer
`retina.3 Type 1 vessels may represent a more mature
`form of neovascularization confined to the sub-RPE
`space and may be less responsive to antivascular
`endothelial growth factor (anti-VEGF therapy) than
`other neovascular patterns. In addition, by recapitulating
`the choriocapillaris, these vessels may provide nutri-
`tional support to the outer retina and could theoretically
`protect against the advent of geographic atrophy (GA).
`Although a monthly dosing regimen of intravitreal
`ranibizumab has the greatest scientific support for
`efficacy, it may be difficult to sustain in the elderly
`population with AMD. In addition, by inhibiting
`a potentially compensatory neovascular response, GA
`could be accelerated in eyes treated with more
`aggressive anti-VEGF dosing regimens. However,
`the relatively infrequent dosing in the PIER trial,4 in
`which patients received quarterly injections after an
`initial series of three monthly injections, resulted in
`inferior visual results when compared with the trials
`that were using monthly dosing.
`The PrONTO Study5 attempted to customize the
`dosing to the individual needs of the patient based on
`acuity decline, clinical findings, or OCT evidence of
`disease activity. The results from the 37 patients who
`completed this trial seemed to be favorable at both
`during 12- and 24-month period.5,6 As a result,
`PrONTO-style dosing has become popular in the
`retina community. Nonetheless,
`this strategy does
`require monthly visits, clinical examinations, and
`OCTs, and patients are uncertain if or when they will
`need treatment. In addition, there have been more
`recent concerns that patients who are no longer
`receiving regular maintenance intravitreal anti-VEGF
`injections can occasionally experience sudden sight-
`threatening macular hemorrhages within days or
`weeks after a stable clinical examination and an
`OCT showing no apparent sub- or intraretinal fluid.7–9
`The ‘‘treat and extend’’ dosing regimen is a strategy
`intended to resolve macular exudation and then maintain
`the macula in this ‘‘dry’’ state indefinitely with, when
`possible,
`fewer patient visits and treatments than
`monthly dosing.10 We recently reported encouraging
`
`From the *Edward S. Harkness Eye Institute, Columbia
`University College of Physicians and Surgeons, New York, New
`York; the †LuEsther T. Mertz Retinal Research Center, New York,
`New York;
`the ‡Department of Ophthalmology, University
`Hospital Zurich, Zurich, Switzerland; and the §Vitreous-Retina-
`Macula Consultants of New York, New York.
`Supported by The Macula Foundation, Inc.
`The authors have no conflicts of interest to disclose.
`Reprint requests: K. Bailey Freund, MD, Vitreous Retina
`Macula Consultants of New York, 460 Park Avenue, 5th Floor, New
`York, NY 10022; e-mail: kbfnyf@aol.com
`
`long-term results using this regimen in eyes with type 3
`neovascularization.11 The dosing strategy consists of an
`initial induction or ‘‘loading’’ sequence of at least three
`initial monthly injections. If stable visual acuity, an
`absence of macular hemorrhage, and a dry OCT have
`been achieved at this point, patients continue to receive
`regular maintenance injections at increasing intervals.
`At 6 weeks after the last of the 3 initial monthly
`injections, visual acuity, clinical findings, and OCT
`changes are recorded again, and patients receive an
`injection regardless of the presence or absence of disease
`activity. However, the interval to the next visit (and
`scheduled injection) is based on an observed change in
`these parameters. If there are no changes, the next visit is
`scheduled for 8 weeks. If there is a change, the patient
`returns for another scheduled injection and examination
`after 4 weeks. The observation and scheduled treatment
`interval
`is extended (hence the phrase ‘‘treat and
`extend’’). In our clinical experience, the risk of recurrent
`sight-threatening hemorrhages seems to increase be-
`cause the interval between injections of anti-VEGF
`agents is extended. Because of this concern, 10 weeks
`was chosen as the longest interval between office visits
`and treatments in this and in our previous study of eyes
`with type 3 neovascularization.7–9,11
`We now report on 18 eyes of 16 consecutive patients
`with newly diagnosed type 1 neovascularization treated
`with intravitreal bevacizumab and/or
`ranibizumab
`using a modified ‘‘treat and extend’’ dosing regimen
`and followed for at least 24 months.
`
`Materials and Methods
`
`Waiver of authorization for use of protected health
`information for the referenced research and a waiver
`of consent for this retrospective chart review were
`obtained from the Institutional Review Board Com-
`mittee of the Manhattan Eye Ear and Throat Hospital,
`New York, NY.
`The diagnosis of type 1 neovascularization was
`performed by the treating physician (K.B.F.) based on
`the clinical, fluorescein angiographic, OCT, and, in
`some cases, indocyanine green angiographic findings.
`To be considered as having the type 1 neovascular
`pattern, eyes had to have clinical and OCT evidence of
`subretinal fluid and/or hemorrhage with an associated
`elevation of the RPE. Fluorescein angiography of these
`eyes exhibited late leakage and staining in an indistinct
`or ‘‘occult’’ pattern. When available, a well-delineated
`‘‘plaque’’ found on indocyanine green angiography was
`correlated with the OCT finding of a vascularized
`pigment epithelial detachment (PED). Only patients
`with recent symptoms, hemorrhage, or evidence of
`recent disease progression were included in this
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 2
`
`Joining Petitioner: Apotex
`
`

`

`1370 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2010  VOLUME 30  NUMBER 9
`
`analysis. Patients treated previously with thermal laser,
`photodynamic therapy, or
`intravitreal pegaptanib
`(Macugen, Eyetech Pharmaceuticals Inc., Palm Beach
`Gardens, FL) or who presented with subfoveal fibrosis
`or atrophy, a history of vitrectomy, aphakia, a history of
`idiopathic or autoimmune-associated uveitis in either
`eye, or diabetic retinopathy more severe than mild
`nonproliferative stage were excluded from this study.
`Patients with preexisting cardiac or cerebrovascular
`conditions were not excluded from the study.
`Patient characteristics,
`including age, sex, and
`presence or absence of disease in the contralateral
`eye, were recorded. Treatment consisted of intravitreal
`injection of 1.25 mg/0.05 mL bevacizumab or 0.5 mg/
`0.05 mL ranibizumab. For the purpose of this analysis,
`no distinction was made between the antiangiogenic
`drugs. Before intravitreal injection, topical anesthesia
`and surface disinfection with 5% povidone-iodine were
`performed. Intravitreal injections were administered at
`the time of diagnosis and subsequently following
`a protocol we have termed ‘‘treat and extend.’’ In
`contrast to a PrONTO-style regimen, patients did not
`have to return for monthly examinations. Instead, all the
`patients received at least 3 initial monthly injections
`followed by continued examination and treatment at
`intervals increasing by 2 weeks per visit once visual
`acuity was stable and clinical examination and OCT
`showed an absence of intra- and subretinal fluid at the
`foveola, resolution of all macular hemorrhage, and no
`further
`reduction in extrafoveolar subretinal fluid.
`Because most eyes with the type 1 neovascular pattern
`continued to manifest PED and/or extrafoveolar
`subretinal fluid after the initial 3 monthly injections,
`resolution of PEDs and/or extrafoveolar subretinal fluid
`that was judged not to affect visual acuity was not
`required before the treatment intervals were lengthened.
`The treatment interval was extended to a maximum of
`10 weeks ‘‘maintenance’’ unless clinical examination
`or OCT detected new hemorrhage, persistent/recurrent
`intra- or subretinal fluid at the foveola, or an increase in
`PED size and/or extrafoveolar subretinal fluid. In those
`cases, the interval was shortened by 2 weeks and
`maintained at that duration provided this restored the
`clinical and OCT findings back to their previous level.
`In this study, the main outcome measure was visual
`acuity after
`treatment. The number of injections
`needed and change in funduscopic or tomographic
`appearance were also assessed. Specifically, presence
`of a PED, subretinal fluid, sight-threatening submacu-
`lar hemorrhage, defined as a subretinal hemorrhage of
`any size within 200 mm of the foveal center or
`a subretinal hemorrhage of at least 2 disk areas within
`the temporal vascular arcades as well as presence and
`progression of GA were recorded.
`
`Snellen visual acuity was measured by a certified
`ophthalmic technician. Snellen acuity was converted
`into logarithm of the minimum angle of resolution
`(logMAR) for statistical analysis at baseline and subse-
`quently at 1, 2, 3, 24, and 36 months after injection of an
`antiangiogenic agent. Changes in logMAR-converted
`acuities were tested with a paired Student’s t-test and
`accepted as significant if the P value was , 0.05.
`Qualitative assessments of retinal thickness were
`initially made using Stratus OCT (Carl Zeiss Meditec,
`Dublin, CA). Later in the study, including at last
`follow-up visits, the Topcon OCT (Topcon 3D OCT-
`1000, Topcon Medical Systems, Paramus, NJ) or
`Spectralis HRA + OCT (Heidelberg Engineering, Inc.,
`Heidelberg, Germany) was used.
`Fundus photography, fluorescein angiograms, enface
`OCT scan images, and, where available, autofluores-
`cence photography were examined for the presence and
`progression of GA overlying the areas of type 1
`neovascularization.
`
`Results
`
`Eighteen eyes of 16 consecutive patients with newly
`diagnosed type 1 neovascularization treated with
`intravitreal injections of bevacizumab and/or ranibizu-
`mab with at least 24-month follow-up were included in
`this study. Nine eyes completed 36-month follow-up.
`Median patient age was 79 years (range, 67–90
`years). Twelve of 16 patients were women. Four
`contralateral eyes had evidence of neovascular AMD
`and 2 of these eyes were treated with a ‘‘treat and
`extend’’ regimen and included in the study.
`Median logMAR visual acuity at presentation was
`0.53 (Snellen equivalent 20/69) and transiently
`improved to 0.41 (Snellen equivalent 20/51) at 1 month
`and maintained at this level during the next 2 months
`(Figure 1). The difference in logMAR-converted
`visual acuity was statistically significant at all early
`time points (paired 2-tailed t-test, P , 0.05). At 24
`months and after an average of 12 injections (range, 8–
`19), median logMAR visual acuity was 0.52, which
`was not
`statistically significantly different
`from
`baseline (P = 0.84). For 9 eyes, 36-month follow-up
`data after an average of 20 injections (range, 18–25
`injections) were available. Visual acuity remained
`stable compared with the 24-month time point with
`a logMAR of 0.52, which also was no different
`from the baseline visual acuity of 0.54 for these 9 eyes
`(P = 0.68).
`As mandated by our inclusion criteria, all eyes had
`serous or vascularized PEDs present at the initiation of
`anti-VEGF treatment. Ten of the 18 eyes had OCT
`evidence of intraretinal fluid on presentation, and 17 of
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 3
`
`Joining Petitioner: Apotex
`
`

`

`‘‘TREAT AND EXTEND’’ FOR TYPE 3 CHOROIDAL NEOVASCULARIZATION  ENGELBERT ET AL
`
`1371
`
`practices, and the healthcare system as a whole. In
`addition, safety concerns about long-term monthly
`injections argue in favor of exploring alternative
`dosing regimens.
`The only randomized, double-blind, sham-con-
`trolled trial investigating an alternative dosing scheme
`is the PIER study,4 which showed that a regimen
`consisting of three initial monthly injections followed
`by mandated quarterly dosing gives inferior visual
`results compared with a monthly dosing regimen.
`Presumably, persistent and/or
`recurrent exudation
`occurring during the extended intervals between
`treatments was related to these inferior visual out-
`comes. The PrONTO study investigated a strategy
`intended to limit macular exudation in which three
`initial monthly injections were followed by dosing on
`an as-needed basis based on changes in visual acuity,
`clinical findings, and evaluation of OCT.5,6 Although
`this open-label, nonrandomized study seemed to show
`that visual results similar to monthly dosing could be
`achieved with fewer injections, patients still required
`monthly visits, examinations, and OCTs. Furthermore,
`after the initial mandated series of three injections,
`fluid was allowed to reaccumulate at
`the foveola
`before the treatment was repeated, raising concerns
`regarding incremental long-term vision loss and the
`possibility of new hemorrhages occurring during long
`periods without VEGF inhibition.
`Type 1 (occult) neovascularization tends to have
`a variable but often less aggressive natural course
`compared with type 2 (well-defined [classic]) neo-
`vascularization and type 3 (retinal angiomatous
`proliferation) neovascularization based on the pre-
`senting acuities and long-term natural history data.
`Patients with type 1 neovascularization who were
`enrolled in the MARINA study presented with a mean
`Early Treatment Diabetic Retinopathy Study letter
`score of 53 (Snellen equivalent 20/802),1 whereas
`patients with type 2 neovascularization enrolled in the
`ANCHOR study presented with only 45 letters
`(Snellen equivalent of 20/125).2 It is well known that
`some patients who have evidence of type 1 (occult)
`neovascularization may never experience vision loss
`(often despite continued growth of the lesion),12 or
`they may experience only a mild gradual visual
`decline. This benign natural course may relate to the
`theory that, in some eyes, the type 1 neovascular
`growth pattern may develop as a compensatory form
`of neovascularization providing nutritional support to
`an ischemic outer retina by recapitulating the normal
`choriocapillaris.3 Our finding that only 1 of 18 eyes
`(6%) developed GA overlying the areas of type 1
`neovascularization may support
`this hypothesis.
`However, in some eyes, the type 1 neovascular pattern
`
`Fig. 1. Mean Snellen visual acuity of patients with type 1 choroidal
`neovascularization on a ‘‘treat and extend’’ regimen was assessed at
`baseline and 1, 2, 3, 24, and 36 months and plotted after conversion to
`logMAR. Horizontal lines represent the median.
`
`18 eyes had subretinal fluid. In all these latter cases, the
`subretinal fluid involved the foveola. At the last follow-
`up examination, PEDs had resolved in only 3 of 18 eyes
`(17%). Most eyes (15 of 18 [83%]) continued to
`manifest extrafoveolar subretinal fluid throughout the
`course of treatment. Of the 10 eyes with intraretinal
`fluid on presentation, complete resolution of this fluid
`occurred in only 3 eyes. In the remaining seven eyes,
`the intraretinal fluid resolved at
`the foveola but
`persisted elsewhere within the macula. Three eyes that
`initially presented without intraretinal fluid developed
`extrafoveolar intraretinal fluid during the follow-up
`period. One eye developed a new PED while on the
`anti-VEGF treatment regimen.
`Only 1 of 18 eyes developed GA overlying the areas
`of type 1 neovascularization. Accordingly, visual acuity
`decreased from 20/200 to 20/400 in this patient. During
`a cumulative observation period of 540 months, no eyes
`developed a sight-threatening submacular hemorrhage.
`Figure 2 shows a representative case of a 75-year-
`old woman treated with the modified ‘‘treat and
`extend’’ dosing regimen (31 ranibizumab injections)
`during 36 months. The patient’s visual acuity
`remained stable despite persistent extrafoveolar sub-
`retinal and intraretinal fluid.
`
`Discussion
`
`Although intravitreal anti-VEGF therapy for neo-
`vascular AMD has produced visual outcomes superior
`to previous therapies, the optimal dosing regimen for
`these agents remains uncertain. Similarly, whether
`different neovascular patterns respond differently or
`require different dosing regimens remains unclear.
`Although monthly dosing of anti-VEGF agents gives
`visual outcomes superior to previous treatments,1,2 it
`can place a tremendous burden on patients, retinal
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 4
`
`Joining Petitioner: Apotex
`
`

`

`1372 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2010  VOLUME 30  NUMBER 9
`
`Fig. 2. Representative case. A–D. Left eye of a 75-year-old woman with 20/200 visual acuity at baseline. A. Color fundus photograph showing a large
`area of subretinal hemorrhage temporal to a vascularized PED. B. Fluorescein angiogram shows type 1 neovascularization nasal to blocked
`fluorescence resulting from sub-RPE and subretinal hemorrhage. C. Late indocyanine green angiogram shows a plaque of choroidal neovascularization
`representing type 1 neovascularization. D. Spectral domain OCT scan (Topcon 3D) shows a vascularized PED and subretinal fluid. E–H. Images at 36-
`month follow-up dosed according to the ‘‘treat and extend’’ protocol. Visual acuity is stable. E. Color fundus photograph. F. Red-free photograph
`shows resolution of the hemorrhage. G. Late indocyanine green angiogram shows a persistent plaque representing type 1 neovascularization. H.
`Spectral domain OCT scan (Heidelberg) shows persistent sub- and intraretinal fluid.
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 5
`
`Joining Petitioner: Apotex
`
`

`

`‘‘TREAT AND EXTEND’’ FOR TYPE 3 CHOROIDAL NEOVASCULARIZATION  ENGELBERT ET AL
`
`1373
`
`may follow a more aggressive course similar to type 2
`vessels, or these vessels may erode through the RPE
`becoming type 2 neovascularization within 1 year
`leading to more rapid vision loss.13–15
`Given a more variable natural course, a rigid
`monthly, ‘‘one-size-fits-all’’ dosing regimen may be
`less suitable for eyes with type 1 vessels than for eyes
`with other neovascular patterns and (in theory) could
`inhibit an important compensatory mechanism aimed
`at preventing loss of overlying neurosensory elements.
`We believe that aggressive strategies aimed at
`eliminating type 1 neovascularization could ultimately
`prove to be detrimental by accelerating GA in some
`patients. This concern is limited not only to a fixed-
`schedule continuous monthly regimen, but also to
`combination strategies using verteporfin photody-
`namic therapy or radiation that aim to completely
`occlude the neovascular lesion. It is well known that
`photodynamic therapy monotherapy affects patho-
`logic neovascularization and vessels perfusing the
`normal choriocapillaris.16–18 Combination strategies
`that add an antiangiogenic agent19 or an anti-inflam-
`matory agent seem to enhance and prolong choroidal
`hypoperfusion, occasionally resulting in profound
`visual loss.20,21 It is unknown what effects this hypo-
`perfusion may have on RPE and the outer retina, but
`obvious concerns are GA and photoreceptor damage.
`The ‘‘treat and extend’’ dosing regimen is a tailored
`maintenance regimen intended to achieve optimal
`visual results with two additional goals.10 One goal is
`to reduce the treatment burden by reducing the number
`of patient visits and the number of imaging studies
`performed by eliminating the need for the monthly
`visits necessitated by alternative dosing strategies.
`We recently reported success in achieving this goal in
`a small cohort of eyes with newly diagnosed type 3
`neovascularization.11 In this previous report, patients
`experienced a sustained visual improvement of ;2
`Snellen lines with nearly half the number of office
`visits and injections compared with a monthly dosing
`regimen.
`A second goal of the ‘‘treat and extend’’ dosing
`regimen is to reduce the risk of new sight-threatening
`submacular hemorrhages. We recently showed a sta-
`tistically significant increase in macular hemorrhages
`when patients in the PIER trial were switched from
`a monthly to quarterly dosing regimen.8 Unfortu-
`nately, large and potentially devastating submacular
`hemorrhages may occur almost
`immediately after
`a high-quality OCT examination showing an absence
`of fluid.7,9 Theoretically, eyes treated with an OCT-
`guided as-needed regimen in which patients may
`go long intervals without VEGF suppression could
`be at greater risk for sight-threatening submacular
`
`hemorrhages compared with eyes receiving more
`frequent and regular anti-VEGF treatments. Because
`of our concern regarding the risk of new hemorrhages
`with long intervals between treatments, we limited the
`interval between anti-VEGF injections to no longer
`than 10 weeks.
`In applying the ‘‘treat and extend’’ strategy to eyes
`with type 1 neovascularization, we elected to modify
`the regimen used in our previous report in which the
`dosing interval was extended only in the absence of
`intraretinal fluid, subretinal fluid, and PED.11
`In
`contrast to eyes with type 2 and 3 vessels, eyes with
`type 1 neovascularization often continue to manifest
`extrafoveolar subretinal fluid (83% in this series)
`and/or PED (83% in this series) after a loading
`sequence of 3 monthly intravitreal injections of an
`anti-VEGF agent. The type 1 neovascular lesions are
`typically larger and may represent a more mature
`neovascular phenotype that is less responsive to anti-
`VEGF treatment. In this study, after three monthly
`treatments, we extended the dosing interval even in the
`presence of a persistent PED and provided any
`remaining fluid spared the foveola and was judged
`not to be affecting visual acuity.
`In our series, patients following this modified ‘‘treat
`and extend’’ regimen were seen on average 12 times
`(range, 8–19) during the first 24 months reflecting the
`variable course of these eyes and a reduced need for
`retreatment. Although patients on a ‘‘treat and extend’’
`dosing regimen receive a mandated injection at each
`visit, their eyes received a similar number of injections
`as received by the 37 patients who completed 24-
`month follow-up in the PrONTO study (average, 9.9;
`range, 3–25) during the first 24 months.6 The
`significant reduction in patient visits of nearly 50%
`without an increase in the number of treatments could
`potentially decrease the burden on patients, practi-
`tioners, and the healthcare system as a whole.
`None of the 18 eyes in our series experienced
`a sight-threatening hemorrhage during a cumulative
`observation period of 540 months. This finding
`seemed to support our hypothesis that more frequent
`and consistent dosing of anti-VEGF treatment may
`help reduce
`the occurrence of new macular
`hemorrhages.
`In our goal of evaluating the long-term results of the
`‘‘treat and extend’’ dosing regimen, we wanted to
`investigate whether visual acuity could be improved
`and maintained with this anti-VEGF treatment
`regimen and how any such effect observed would
`compare historically with outcomes reported for the
`large randomized MARINA1 trial, which examined
`almost exclusively type 1 neovascular lesions. Our
`patients’ baseline visual acuity of 20/69 was somewhat
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 6
`
`Joining Petitioner: Apotex
`
`

`

`1374 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES  2010  VOLUME 30  NUMBER 9
`
`than the mean baseline Early Treatment
`better
`Diabetic Retinopathy Study letter
`score in the
`MARINA study, which was 53.6 letters (Snellen
`equivalent of 20/802). Although, unlike in the
`MARINA trial, there was not a statistically significant
`visual improvement at 24 months, the mean visual
`acuity of 20/70 at this time point was similar to that of
`the MARINA trial. In addition, the reported results of
`the MARINA trial apply only to those patients who
`were willing and able to complete 2 years of monthly
`visits and injections. In the MARINA trial, 14.1% of
`enrolled patients did not complete the 24-month
`follow-up visit. Similarly, in our experience, it is often
`difficult to sustain monthly visits in the population
`with neovascular AMD.
`It is problematic to compare our visual acuity data
`with the PrONTO study5,6 in which patients gained an
`average of 11.3 letters because the lesion compositions
`in the 2 studies were dissimilar. In the PrONTO study,
`75% of eyes had lesions with at least some classic
`choroidal neovascularization (type 2 neovasculariza-
`tion) and 25% had retinal angiomatous proliferation
`(type 3 neovasularization). These more aggressive
`lesions tend to present with more active exudation and
`worse visual acuity (20/80+ in the PrONTO study)
`compared with type 1 lesions. This may offer such
`patients a greater chance for visual improvement after
`resolution of exudation. We have recently reported on
`type 3 lesions treated according to the ‘‘treat and
`extend’’ regimen in which an average of ;2 lines of
`improvement was found after 24 months.11 This is
`similar to the ANCHOR study in which patients with
`exclusively classic lesions gained 11.3 letters. The
`superior visual results in eyes with type 3 neo-
`vascularization may relate to a smaller size and a more
`robust response to treatment than are typical for type 1
`lesions. In addition, in our previous study, we used
`a more aggressive treatment regimen in which an
`absence of PED and fluid both centrally and in the
`extrafoveolar macula was required before the interval
`between injections was extended.
`Although historical comparisons between different
`studies may be relevant, they are hampered by the
`impossibility of statistical analysis. Conclusions about
`the superiority of one treatment protocol or the other
`remain speculative at best, particularly when the
`numbers in the studies being compared are low.
`Unlike the MARINA and PrONTO studies, which
`are limited to 24-month follow-up, 9 of our 17 patients
`completed 36-month follow-up with a mean of 20
`injections. Visual acuity remained stable at the 24-
`month level, suggesting true long-term stabilization.
`We are not aware of any other dosing regimen of anti-
`VEGF therapy, which has showed stable visual acuity
`
`.3 years. Furthermore, the absence of sight-threat-
`ening macular hemorrhages and rare progression of
`GA overlying the neovascular lesions suggest addi-
`tional long-term benefits of the ‘‘treat and extend’’
`dosing regimen.
`Our study is limited by its retrospective, noncom-
`parative nature and relatively few patients. Another
`limitation is a possible ascertainment bias as a result of
`a methodology that excluded patients with a follow-up
`of ,24 months and those who were noncompliant with
`the ‘‘treat and extend’’ dosing regimen. It is possible
`that patients who did not have 24-month follow-up
`may have discontinued the treatment regimen as a
`result of poor outcomes. However, few such cases
`could be recalled.
`Despite these limitations, our study supports that the
`modified ‘‘treat and extend’’ dosing regimen used in
`this study for type 1 neovascularization may be a safe
`and effective way to reduce the number of follow-up
`visits and injections required in these eyes for up to
`36 months. Despite persistent PEDs and extrafoveolar
`fluid in most eyes, visual acuity remained stable with
`no eyes experiencing sight-threatening submacular
`hemorrhages. Given these encouraging long-term
`results of a treatment strategy designed to control
`but not eliminate a potentially protective compensa-
`tory form of neovascularization, we suggest that future
`studies aimed at improving visual outcomes in eyes
`harboring type 1 neovascularization explore similar
`nondestructive treatment modalities.
`
`Key words: type 1 neovascularization, bevacizumab,
`ranibizumab, Lucentis, Avastin, treat and extend.
`
`References
`
`1. Rosenfeld PJ, Brown DM, Heier JS, et al; MARINA Study
`Group. Ranibizumab for neovascular age-related macular
`degeneration. N Engl J Med 2006;355:1419–1431.
`2. Brown DM, Kaiser PM, Michels M, et al; ANCHOR Study
`Group. Ranibizumab versus verteporforin for neovascular
`age-related macular degeneration. N Engl J Med 2006;355:
`1432–1444.
`3. Grossniklaus HE, Green WR. Choroidal neovascularization.
`Am J Ophthalmol 2004;137:496–503.
`4. Regillo CD, Brown DM, Abraham P, et al. Randomized,
`double-masked, sham-controlled trial of
`ranibizumab for
`neovascular age-related macular degeneration: PIER Study
`year 1. Am J Ophthalmol 2008;145:239–248.
`5. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical coherence
`tomography-guided, variable dosing regimen with intra-vitreal
`ranibizumab (Lucentis) for neovascular age-related macular
`degeneration. Am J Ophthalmol 2007;143:566–583.
`6. Lalwani GA, Rosenfeld PJ, Fung AE, et al. A variable-dosing
`regimen with intravitreal ranibizumab for neovascular age-
`related macular degeneration: year 2 of the PrONTO Study.
`Am J Ophthalmol 2009;148:43–58.el.
`7. Levine JP, Marcus I, Sorenson JA, Spaide RF, Cooney MJ,
`Freund KB. Macular hemorrhage in neovascular age-related
`
`Mylan Exhibit 1025
`Mylan v. Regeneron, IPR2021-00880
`Page 7
`
`Joining Petitioner: Apotex
`
`

`

`‘‘TREAT AND EXTEND’’ FOR TYPE 3 CHOROIDAL NEOVASCULARIZATION  ENGELBERT ET AL
`
`1375
`
`macular degeneration after stabilization with antiangiogenic
`therapy. Retina 2009;29:1074–1079.
`8. Barbazetto I, Saroj N, Freund KB. Dosing regimen and the
`frequency of macular hemorrhages in neovascular age-related
`macular degeneration treated with ribizumab. Presented at the
`Annual Meeting of the Retina Society, New York, NY, September
`2008.
`9. Margolis R, Freund KB. Hemorrhagic recurrence of neo-
`vascular age-related macular degeneration not predicted by
`spectral domain optical coherence tomography. Retin Cases
`Brief Rep In press.
`10. Spaide R. Ranibizumab according to need: a treatment for age-
`relat