MACULAR HEMORRHAGE IN
`NEOVASCULAR AGE-RELATED MACULAR
`DEGENERATION AFTER STABILIZATION
`WITH ANTIANGIOGENIC THERAPY
`
`JONATHAN P. LEVINE, MD,* INNA MARCUS, MD,†
`JOHN A. SORENSON, MD,‡§ RICHARD F. SPAIDE, MD,‡§
`MICHAEL J. COONEY, MD, MBA,‡§ K. BAILEY FREUND, MD‡§
`
`Purpose: To study patients with neovascular age-related macular degeneration (AMD)
`who experienced a macular hemorrhage after stabilization with intravitreal antivascular
`endothelial growth factor (anti-VEGF) agents to improve current treatment regimens and
`prevent disease progression.
`Methods: Retrospective chart review of six patients. The main outcome measures included
`time between last intravitreal anti-VEGF treatment and date of hemorrhage, time between last
`office visit and date of hemorrhage, and visual acuity before and after hemorrhage.
`Results: Three of 6 eyes had a macular hemorrhage within 4 weeks of a stable
`examination. One eye had optical coherence tomography (OCT) that demonstrated no fluid
`1 day before the macular hemorrhage. The average time between the date of the last
`injection and macular hemorrhage was 16.8 weeks (range, 7.3–28.9 weeks). The average
`time between the last stable examination and an event was 4.2 weeks (range, 1 day to 7.3
`weeks). Three of six patients had a persistent decline in vision after the hemorrhage.
`Among the 4 patients, who had better than 20/200 vision before the macular hemorrhage,
`2 dropped to 20/200 or worse.
`Conclusion: Sight-threatening macular hemorrhages from AMD can occur within days
`to weeks after a stable examination and absence of fluid on OCT. Regimens that treat “as
`needed” based on clinical findings and OCT may not be appropriate for certain patients.
`RETINA 29:1074 –1079, 2009
`
`Bevacizumab and ranibizumab have been wel-
`
`comed as breakthrough antiangiogenic therapies
`for the treatment of neovascular AMD.1– 6 The phase
`IIIb studies of ranibizumab demonstrated the efficacy
`of monthly intravitreal injections in improving visual
`outcomes in eyes with neovascular AMD for up to 2
`years. More recently, the PrONTO trial, a small non-
`randomized study, suggested that a variable dosing
`regimen, using OCT as a guide for retreatment, could
`
`From the *Albert Einstein College of Medicine, Bronx, New
`York; †New York University School of Medicine; ‡Vitreous-
`Retina-Macula Consultants of New York; and §The LuEsther T.
`Mertz Retinal Research Center, Manhattan Eye, Ear, and Throat
`Hospital, New York, New York.
`Supported in part by The LuEsther T. Mertz Retinal Research
`Center of the Manhattan Eye, Ear, & Throat Hospital and The
`Macula Foundation, Inc.
`Reprint requests: K. Bailey Freund, MD, Vitreous-Retina-Mac-
`ula Consultants of New York, 460 Park Avenue, 5th Floor, New
`York, NY 10022; e-mail: vrmny@aol.com
`
`also achieve good visual results with the advantage of
`fewer treatments than a monthly dosing regimen.7
`Despite the efficacy of existing regimens, sight-threat-
`ening recurrences of macular exudation remain a con-
`cern for the treating physician. Furthermore, although
`OCT has become the standard of care for monitoring
`patients receiving anti-VEGF therapy, it remains un-
`proven whether OCT will reliably detect evidence of
`recurrent neovascular activity before a visually signif-
`icant macular hemorrhage.
`We present a case series of six eyes with neovas-
`cular AMD stabilized after intravitreal bevacizumab
`or ranibizumab treatment that subsequently developed
`a sight-threatening macular hemorrhage.
`
`Methods
`
`We reviewed the records of six patients with a
`history of neovascular AMD at one clinical center
`
`1074
`
`Regeneron Exhibit 2042
`Page 01 of 06
`
`

`

`HEMORRHAGE IN AMD AFTER ANTI-VEGF ● LEVINE ET AL
`
`1075
`
`20/400
`
`20/60
`
`20/200
`
`20/40
`
`3/400
`
`LP
`
`26.0
`
`7.3
`
`28.9
`
`10.0
`
`17.3
`
`11.0
`
`5.0
`
`7.3
`
`6.3
`
`3.7
`
`0.1
`
`2.9
`
`Follow-Up
`VAatLast
`
`andSRH
`
`LastInjection
`
`WeeksBetween
`
`andSRH
`
`Examination
`
`Last
`
`WeeksBetween
`
`arcades
`
`3.4DAwithin
`
`10/400
`
`NoSRForCME
`
`extrafoveal
`
`extrafovealPED
`CME,small
`
`2.5DA,
`
`20/50
`
`NoSRFor
`
`subfoveal
`
`20/100⬎12DA,
`center
`fromfoveal
`
`NoSRForCME
`
`1DA⬍200␮m
`
`subfoveal
`
`20/50
`
`NoSRForCME
`
`6DA,
`
`20/60
`
`NoSRForCME
`
`subfoveal
`
`5/400⬎12DA,
`
`NoSRForCME
`
`Size/Location
`Hemorrhage
`
`SRH
`Before
`
`VA
`
`BeforeSRH
`
`Findings
`
`OCT
`
`SRH
`
`SRH,subretinalhemorrhage;VA,visualacuity;F,female;M,male;OS,lefteye;OD,righteye;CNV,choroidalneovascularization;SRF,subretinalfluid;CME,cystoidmacular
`*Treatandextendregimen.
`
`edema;PED,pigmentepithelialdetachment;DA,diskareas.
`
`hemorrhage
`
`Disciformscar,no
`
`fluidorblood
`
`QuiescentCNV,no
`
`QuiescentCNV
`
`QuiescentCNV
`
`AtrophicCNV
`
`excentriclipid
`
`Disciformscarwith
`
`FindingsBefore
`
`Examination
`
`2
`
`6
`
`3
`
`3
`
`3
`
`4
`
`Clopidogrel
`
`OD
`
`Aspirin,Warfarin
`
`—
`
`—
`
`Aspirin
`
`OS
`
`OS
`
`OS
`
`OS
`
`OSWarfarin
`
`F
`
`F
`
`M
`
`M
`
`F
`
`M
`
`76
`
`95
`
`59
`
`75
`
`88
`
`87
`
`6
`
`5*
`
`1
`
`2
`
`3
`
`4
`
`BeforeSRH
`No.Injections
`
`Anticoagulant
`
`Eye
`Study
`
`CaseNo.AgeGender
`
`Table1.MacularHemorrhageinNeovascularAge-RelatedMacularDegenerationAfterStabilizationWithAntiangiogenicTherapy
`
`who were initially stabilized with intravitreal bevaci-
`zumab or ranibizumab therapy and who subsequently
`developed a new sight-threatening macular hemor-
`rhage between February 2006 and July 2007. All
`lesion types were included in this analysis, including
`those with significant areas of subretinal fibrosis. All
`patients initially received monthly injections of an
`anti-VEGF agent until stable. Stability was defined as
`complete resolution of both intraretinal and subretinal
`fluid detected on OCT and resolution of all macular
`hemorrhage when present. Persistent serous pigment
`epithelial detachment was not an exclusionary crite-
`rion. After stabilization, patients either received less
`frequent maintenance injections (“treat and extend”)
`or were observed for signs of recurrent neovascular
`activity at the discretion of the treating physician.
`We defined a “sight-threatening” macular hemor-
`rhage as a subretinal hemorrhage of any size within
`200 ␮m of the foveal center or a subretinal hemor-
`rhage of at least 2 disk areas within the temporal
`vascular arcades. All eyes had not shown any signs of
`choroidal neovascularization activity such as macular
`hemorrhage or fluid on OCT on the most recent ex-
`amination and OCT before the occurrence of macular
`hemorrhage. Information regarding the patient’s clin-
`ical history and type and dates of treatment were
`recorded along with clinical details surrounding the
`hemorrhage. The interval between the last treatment
`and a macular hemorrhage as well as the interval
`between the last stable examination and a macular
`hemorrhage were recorded. Visual acuity was re-
`corded for the last visit before an event and at the most
`recent visit on follow-up.
`
`Results
`
`Four of 6 patients had baseline vision of 20/200 or
`better, whereas 2 had vision worse than 20/400 (Table
`1). Five eyes were treated with an OCT-guided regi-
`men, whereas one was treated with a “treat and ex-
`tend” strategy. The average number of injections be-
`fore the macular hemorrhage was 3.5 (range, 2– 6).
`The average time between the date of the last injection
`and the macular hemorrhage was 16.8 weeks (range,
`7.3–28.9 weeks). The average time between the last
`stable examination and the macular hemorrhage was
`4.2 weeks (range, 1 day to 7.3 weeks). Three of 6 eyes
`had a macular hemorrhage within 4 weeks of a stable
`examination. One eye had an OCT showing no fluid
`on the day before a hemorrhage occurred. Among the
`4 patients who had better than 20/200 vision, 2
`dropped to 20/200 or worse.
`
`Regeneron Exhibit 2042
`Page 02 of 06
`
`

`

`1076 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 8
`
`Fig. 1. Case 1. Left, Color
`photograph of patient 1 show-
`ing a quiescent fibrovascular
`scar secondary to neovascular
`AMD. Right, Photograph of
`the same eye 2 weeks later
`showing a bullous retinal de-
`tachment secondary to a mas-
`sive subretinal hemorrhage.
`
`Case 1
`Patient 1 was an 87-year-old woman with a history
`of advanced neovascular AMD in the left eye for 5
`years and a longstanding disciform scar in the right
`eye. The patient had bilateral 2⫹ nuclear sclerotic
`cataracts. She was using latanoprost once daily in the
`right eye for glaucoma. The patient’s medical history
`included hypertension. She was on warfarin for ca-
`rotid stenosis. The patient began intravitreal ranibi-
`zumab (0.5 mg/0.05 mL) therapy in her left eye for
`chronic subretinal fluid and a recent small subretinal
`hemorrhage in the macula. She received a total of two
`injections. Five weeks after the second injection, vi-
`sual acuity was stable at 5/400 in the left eye. She was
`noted to have a quiescent fibrotic choroidal neovascu-
`lar membrane with no clinically apparent hemorrhage
`and no fluid detected on OCT (Figure 1, left). Twenty
`days later, the patient returned with pain and tearing in
`the left eye for 3 days. On presentation, the patient’s
`visual acuity was light perception in the left eye.
`Intraocular pressures were 19 in the right eye and 58
`in the left eye. Gonioscopy showed a closed angle for
`360° in the left eye. Funduscopic examination showed
`a massive subretinal hemorrhage in the left eye (Fig-
`ure 1, right). The patient was placed on dorzolamide/
`timolol and brimonidine drops and acetazolamide
`(500 mg) orally twice a day to lower the intraocular
`pressure. A laser iridotomy was performed the next
`day. Three days later, visual acuity in the left eye
`remained light perception and the intraocular pressure
`was 10. There was no view to the posterior pole, but
`a subretinal and vitreous hemorrhage was evident on
`B-scan ultrasonography. One month later, the vision
`was still light perception, and there was persistent
`subretinal and vitreous hemorrhage.
`
`Case 2
`Patient 2 was an 87-year-old woman with neovas-
`cular AMD in both eyes. The right eye had a long-
`
`standing disciform scar. The left eye received 3
`monthly intravitreal injections of bevacizumab (1.25
`mg/0.5 mL) for subretinal hemorrhage associated with
`poorly defined subfoveal choroidal neovasculariza-
`tion. She was monitored for 6 months without addi-
`tional treatment but later developed a recurrence of
`subretinal hemorrhage and received 2 additional in-
`jections of intravitreal ranibizumab. Four months after
`her second ranibizumab injection, she was seen for a
`routine follow-up examination. Visual acuity was
`counting fingers in the right eye and 20/60 in the left
`eye. Clinical examination revealed a stable disciform
`scar in the right eye and pigmentary changes in the left
`eye without hemorrhage or fluid detected on OCT
`(Figure 2, left). One day later, she presented with
`acute loss of vision in the left eye. On examination,
`visual acuity in the left eye was 20/400. The patient
`was noted to have a new subfoveal hemorrhage (Fig-
`ure 2, right). The patient received 4 more intravitreal
`ranibizumab injections over the next 6 months, but
`visual acuity remained 3/400. The most recent exam-
`ination of the left eye revealed a stable fibrotic scar
`with no hemorrhage or fluid seen clinically and no
`fluid on OCT.
`
`Case 3
`
`Patient 3 was a 68-year-old man with a history of
`neovascular AMD in the right eye. His left eye was
`treated with verteporfin photodynamic therapy fol-
`lowed by 3 monthly injections of intravitreal ranibi-
`zumab (0.5 mg/0.5 mL). On examination, 4 weeks
`after the third injection, visual acuity was 20/800 in
`the right eye and 20/40 in the left eye. Clinical exam-
`ination showed a stable disciform scar in the right eye
`and a small stable area of subretinal fibrosis in the left
`eye. Fluorescein angiography in the left eye showed
`no active leakage in the left eye, and OCT in the left
`eye showed no retinal fluid. The patient returned 5
`weeks later with decreased vision in the left eye with
`
`Regeneron Exhibit 2042
`Page 03 of 06
`
`

`

`HEMORRHAGE IN AMD AFTER ANTI-VEGF ● LEVINE ET AL
`
`1077
`
`Fig. 2. Case 2. Top and bottom left, Color photograph and OCT of patient 3 demonstrating an absence of subretinal hemorrhage and fluid. Top right,
`Color photograph of the same eye 1 day later demonstrating subfoveal hemorrhage. Bottom right, OCT demonstrates new subretinal and subretinal
`pigment epithelium fluid.
`
`a central gray scotoma. Visual acuity was 20/800 in
`the right eye and 20/100 in the left eye. Clinical
`examination showed new subretinal blood in the left
`eye with new retinal fluid on OCT. The hemorrhage
`and fluid resolved after
`two monthly intravitreal
`ranibizumab injections. During a follow-up visit 6
`weeks later, the patient’s visual acuity was 20/800 in
`the right eye and 20/50 in the left eye with no clinical
`evidence of macular hemorrhage or fluid on OCT
`(Figure 3, left). Based on these findings, no further
`treatment was given. On examination 4 weeks later,
`the vision had dropped to 20/125 in the left eye and
`subretinal hemorrhage and fluid were observed on
`clinical examination in the left eye (Figure 3, right).
`The patient was then placed on a maintenance regimen
`with intravitreal injections of ranibizumab given at
`intervals of every 5 to 6 weeks. He received 5 addi-
`tional intravitreal injections of ranibizumab over the
`next 7 months. At last follow-up, visual acuity had
`improved to 20/40 in the left eye with no recurrence of
`macular hemorrhage or fluid on OCT.
`
`Discussion
`
`ranibizumab and bevacizumab have
`Intravitreal
`transformed the prognosis for patients with neovascu-
`lar AMD.8 Although the optimal dosing regimen of
`these agents remains uncertain, current treatment al-
`
`gorithms are largely based on the phase IIIb MARINA
`and ANCHOR trials of ranibizumab in which patients
`received continuous monthly injections for 2 years.1–5
`Because monthly visits and injections are costly to the
`healthcare system and difficult to maintain in this
`elderly patient population, alternative dosing strate-
`gies continue to be explored. In the PIER trial of
`ranibizumab, an initial gain in visual acuity with three
`monthly injections was lost when patients were
`switched from monthly injections to quarterly injec-
`tions as was mandated by the study protocol.9 This
`decline in visual acuity was presumably the result of
`recurrent neovascular activity and associated exuda-
`tion occurring between injections. More recently, the
`PrONTO study, using an as-needed dosing regimen
`guided by monthly eye examinations and OCT, dem-
`onstrated visual outcome data similar to monthly dos-
`ing. In this small nonrandomized trial, the total num-
`ber of patient visits remained the same, but
`the
`number of injections was reduced by approximately
`half. The PrONTO strategy is based on the assumption
`that fluid in the macula will occur before sight-threat-
`ening macular hemorrhages and that treating after
`fluid recurs, rather than before, will give visual results
`similar to monthly maintenance injections.5,7
`A recent analysis of the ANCHOR, MARINA, and
`PIER data demonstrated that monthly intravitreal
`
`Regeneron Exhibit 2042
`Page 04 of 06
`
`

`

`1078 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2009 ● VOLUME 29 ● NUMBER 8
`
`Fig. 3. Case 3. Top and bottom left, Color photograph and OCT of patient 4 demonstrating a quiescent subfoveal fibrovascular scar. Top right,
`Photograph of the same eye 1 month later showing new subretinal hemorrhage. Bottom right, OCT showing increased retinal thickness temporally.
`
`ranibizumab dosing significantly reduced the fre-
`quency of macular hemorrhages compared with the
`sham controls or photodynamic therapy-treated pa-
`tients regardless of lesion type. The effect was lost
`when patients were switched from monthly to quar-
`terly dosing in the PIER study.10 Reducing the fre-
`quency of injections should, therefore, be done with
`caution.
`In our case series, 3 of 6 eyes on intravitreal anti-
`VEGF therapy developed a sight-threatening macular
`hemorrhage within 4 weeks of a stable clinical exam-
`ination and OCT showing an absence of intra- or
`subretinal fluid. We based our definition of a sight-
`threatening macular hemorrhage on its size and prox-
`imity to the fovea rather than on vision loss per se,
`although three of our patients had a drop in vision
`from the hemorrhage. We felt it appropriate to include
`“near-miss” hemorrhages even if they were not sub-
`foveal or resulted in vision loss. The visual signifi-
`cance of hemorrhagic events is likely influenced by
`multiple factors such as their size, thickness, proxim-
`ity to the fovea, and the manner in which they are
`managed.11 Hemorrhage size and proximity to the
`fovea seem to correlate with worse visual outcome in
`our series (Table 1).
`For some patients, a monthly examination schedule
`similar to the PrONTO strategy may be sufficient to
`detect early recurrence and allow for timely treatment
`as needed with fewer treatments than a monthly dos-
`
`ing regimen. However, our findings related to the
`timing and severity of macular hemorrhages in three
`of our patients challenge the strategy of treating all
`patients in this manner. A maintenance regimen may
`be more appropriate for eyes identified as high risk, in
`particular eyes with preserved foveal function and
`patients with poor vision in the fellow eye.
`Tilanus et al12 identified warfarin use as a risk
`factor for massive intraocular hemorrhage in AMD
`and noted a possible association between massive
`hemorrhage and antiplatelet therapy. In our study, 4 of
`6 patients were on anticoagulants, 2 were on Couma-
`din, one was on aspirin, and one was taking clopi-
`dogrel. The significance of these agents is uncertain
`because we do not know the prevalence of anticoag-
`ulation use in our general AMD population.
`In our study, one patient underwent the “treat and
`extend” regimen, whereas the other five were treated
`as needed, based on examination and OCT results.
`Data on “treat and extend” are limited and primarily
`based on anecdotal evidence. Furthermore, it is diffi-
`cult to determine in advance how far one can safely
`extend a patient’s treatment interval without risking a
`macular hemorrhage. For our patients, the interval
`between the last injection and an event ranged from
`7.3 to 28.9 weeks. A fluorescein angiography may be
`useful when monitoring patients who are 8 to 10
`weeks past their last treatment, especially if consider-
`ing increasing their interval of retreatment.
`
`Regeneron Exhibit 2042
`Page 05 of 06
`
`

`

`HEMORRHAGE IN AMD AFTER ANTI-VEGF ● LEVINE ET AL
`
`1079
`
`Because this is a retrospective case series without a
`control group, any conclusions should be appropri-
`ately restricted. Given the small size of this series, we
`were unable to draw meaningful conclusions regard-
`ing specific lesion characteristics such as neovascular
`subtype, lesion size, or presence of subretinal fibrosis
`that would predispose to these hemorrhagic events.
`Cross-sectional study comparison is also limited be-
`cause, unlike the phase IIIb ranibizumab trials, we
`included eyes with vision worse than 20/400, lesions
`greater than 12 disk areas, and eyes with significant
`areas of subretinal fibrosis. We also included patients
`treated with either intravitreal bevacizumab or ranibi-
`zumab. Despite these limitations, we feel these cases
`represent a significant cohort encountered in clinical
`practice and provide useful information that can be
`applied to a broader AMD population.
`As the PrONTO study has demonstrated, an as-
`needed treatment regimen based on the clinical sta-
`tus of the individual patient can be an effective
`strategy with fewer treatments than a monthly dos-
`ing regimen. For some patients, however, a sight-
`threatening macular hemorrhage can occur without
`the presence of preceding fluid on OCT performed
`as recently as 1 day before the event. Prophylactic
`maintenance therapy may be appropriate in some
`cases.
`Key words: age-related macular degeneration, be-
`vacizumab,
`choroidal neovascularization,
`ranibi-
`zumab, subretinal hemorrhage.
`
`References
`
`1. Brown DM, Kaiser PK, Michels M, et al; ANCHOR Study
`Group. Ranibizumab versus verteporfin for neovascular age-
`
`related macular degeneration. N Engl J Med 2006;355:1432–
`1444.
`2. Rosenfeld PJ, Brown DM, Heier JS, et al; MARINA Study
`Group. Ranibizumab for neovascular age-related macular de-
`generation. N Engl J Med 2006;355:1419 –1431.
`3. Stone EM. A very effective treatment for neovascular macular
`degeneration. N Engl J Med 2006;355:1493–1495.
`4. Rosenfeld PJ, Rich RM, Lalwani GA. Ranibizumab: phase III clin-
`ical trial results. Ophthalmol Clin North Am 2006;19:361–372.
`5. Spaide R. Ranibizumab according to need: a treatment for
`age-related macular degeneration. Am J Ophthalmol 2007;
`143:679 – 680.
`6. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA, Nasir
`MA, Giust MJ. Intravitreal bevacizumab (Avastin) for neovas-
`cular age-related macular degeneration. Ophthalmology 2006;
`113:363–372.e5.
`7. Fung AE, Lalwani GA, Rosenfeld PJ, et al. An optical
`coherence tomography-guided, variable dosing regimen
`with intravitreal ranibizumab (Lucentis) for neovascular
`age-related macular degeneration. Am J Ophthalmol 2007;
`143:566 –583.
`8. Steinbrook R. The price of sight—ranibizumab, bevacizumab,
`and the treatment of macular degeneration. N Engl J Med
`2006;355:1409 –1412.
`9. Regillo CD, Brown DM, Abraham P, et al. Randomized,
`double-masked, sham-controlled trial of ranibizumab for neo-
`vascular age-related macular degeneration: PIER Study year 1.
`Am J Ophthalmol 2008;145:239 –248.
`10. Barbazetto et al. Dosing regimen and the frequency of macular
`hemorrhages in neovascular age-related macular degeneration
`treated with ranibizumab. Presented at the Annual Meeting of
`the Retina Society; September 2008.
`11. Scupola A, Coscas G, Soubrane G, Balestrazzi E. Natural
`history of macular subretinal hemorrhage in age-related mac-
`ular degeneration. Ophthalmologica 1999;213:97–102.
`12. Tilanus MA, Vaandrager W, Cuypers MH, Verbeek AM,
`Hoyng CB. Relationship between anticoagulant medication
`and massive intraocular hemorrhage in age-related macular
`degeneration. Graefes Arch Clin Exp Ophthalmol 2000;238:
`482– 485.
`
`Regeneron Exhibit 2042
`Page 06 of 06
`
`