- Email: [email protected]
The histopathologic effects of transpupillary thermotherapy in human eyes
The Histopathologic Effects of Transpupillary Thermotherapy in Human Eyes Brian P. Connolly, MD,1 Carl D. Regillo, MD,1 Ralph C. Eagle, Jr., MD,2 Carol L. Shields, MD,3 Jerry A. Shields, MD,3 Heidi Moran, BS1 Objective: To determine the histopathologic effects of transpupillary thermotherapy (TTT) on the normal human retina. Design: Prospective, noncomparative small case series. Participants: Three patients with eyes scheduled for enucleation because of the presence of a malignant intraocular tumor. Intervention: Application of TTT to the posterior pole by using an 810-nm laser and the following laser parameters: 2-mm spot size, 60-second duration, and power settings of 430, 530, and 630 mW (low, medium, and high doses, respectively). Two or three TTT treatments at varying dose levels were performed in each eye. The eyes were then enucleated within 7 days of TTT, and light microscopy of serial sections was performed. Main Outcome Measures: Light microscopic histopathologic changes of the neurosensory retina, retinal pigment epithelium (RPE), and choroid in the areas of treatment compared with adjacent normal (control) tissue. Visual acuity, fundus appearance by slit-lamp biomicroscopy, and symptoms of pain, burning, or visual change were recorded before and after each TTT application. The degree of fundus pigmentation was also noted. Results: Eight treatment spots from three eyes were analyzed. The first eye was judged to have a lightly pigmented fundus, and no histopathologic alterations were seen on light microscopy at any of the three dose levels. The second eye had a more pigmented fundus. This eye had minimal outer retinal changes in the area corresponding to the low-dose treatment, more prominent changes in the outer and middle layers in the medium-dose treatment area, and full-thickness retinal alterations, along with changes in the RPE and choroid, where the high dose was applied. The third eye was found to have an unexpected extension of pigmented choroidal melanoma under the fovea, and full-thickness retinal changes were observed in this area after a medium-dose application. No histopathologic changes were seen at the low dose in this eye. Conclusions: TTT applications resulted in a spectrum of histopathologic effects of the retina that are related to both energy level and fundus pigmentation. Mild or no changes were observed in most low- or medium-dose applications. More extensive retinal damage occurred with applications that used energy levels higher than what have been used in the clinical setting or when the fundus was more heavily pigmented. Ophthalmology 2003; 110:415– 420 © 2003 by the American Academy of Ophthalmology.
Originally received: December 27, 2001. Accepted: May 14, 2002. Manuscript no. 211060. 1 Retina Service, Wills Eye Hospital, Philadelphia, Pennsylvania. 2 Pathology Service, Wills Eye Hospital, Philadelphia, Pennsylvania. 3 Ocular Oncology Service, Wills Eye Hospital, Philadelphia, Pennsylvania. Presented at the Association for Research in Vision and Ophthalmology meeting, Fort Lauderdale, Florida, May 2001, and at the annual meeting of the American Academy of Ophthalmology, New Orleans, Louisiana, November 2001. Supported by the Wills Eye Hospital Women’s Committee Macular Degeneration Fund, Philadelphia, PA (CDR); the Eye Tumor Research Foundation, Philadelphia, PA (CLS); the Award of Merit in Retina Research, Houston, TX (JAS); and the Noel T. and Sara L. Simmonds Endowment for Ophthalmic Pathology, Wills Eye Hospital, Philadelphia, PA (RCE). The authors have no proprietary interest in any of the instruments or techniques reported in this article. Reprint requests to Carl D. Regillo, MD, Retina Service, Wills Eye Hospital, 840 Walnut Street, Philadelphia, PA 19107. E-mail: [email protected]. © 2003 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Transpupillary thermotherapy (TTT) has been used to treat subfoveal choroidal neovascularization (CNV) in agerelated macular degeneration (AMD). A small, nonrandomized pilot study by Reichel et al1 showed short-term visual acuity improvement or stability in 12 of 16 eyes with AMD and subfoveal occult CNV treated with TTT. In a recently published clinical series by Newsom et al,2 TTT was used to treat both occult and classic subfoveal CNV in 44 eyes with AMD. In that study, TTT achieved angiographic closure of CNV after treatment in 75% to 78% of eyes, and visual acuity stabilized or improved in 67% to 71% of eyes, with a mean follow-up of approximately 7 months. No patient lost vision in the involved eye immediately after treatment in either study. A controlled clinical trial (TTT4CNV) is currently under way to determine the safety and efficacy of TTT to treat AMD-related subfoveal occult CNV. On the basis of mathematical models, the treatment parameters used in published reports and the ongoing ISSN 0161-6420/03/$–see front matter PII S0161-6420(02)01561-0
415
Ophthalmology Volume 110, Number 2, February 2003 TTT4CNV trial should result in a tissue temperature increase at the level of the retinal pigment epithelium (RPE) of approximately 10° C.3 Such a relatively small degree of hyperthermia should spare the overlying neurosensory retina of significant detrimental heat effects. Although subfoveal TTT seems to be safe in most exudative AMD cases, the exact tissue effects of TTT have yet to be formally analyzed in animal models or human eyes. This study was designed to access the histopathologic effects of TTT on the normal human retina. Different laser energy levels were used in a range that spanned 20% below and above the levels currently in clinical use for a given laser spot size, to determine the full spectrum of potential tissue effects of the treatment.
Materials and Methods The patients eligible to participate in this study were those who were diagnosed with uveal melanoma for which an enucleation was recommended. Enrollment took place between May 2000 and September 2000. The institutional review board of Wills Eye Hospital approved the protocol and the consent form for this study before enrollment of any patients. Before any treatment, all patients underwent a complete ocular examination in the Oncology Service at Wills Eye Hospital. Bestcorrected Snellen visual acuity, intraocular pressure, and all anterior and posterior segment findings were recorded. A detailed fundus drawing and fundus photography were also performed. To be eligible to participate, the patient’s involved eye needed to have clear media and an area of ophthalmoscopically normal-appearing retina along the horizontal meridian on either side of the optic disc (preferably the macula), to accommodate three 2-mm laser spots placed one spot-width apart. Within 1 week of scheduled enucleation, TTT was performed. Up to three applications with different energy levels were placed along the horizontal meridian in the posterior pole, with an attempt to treat the foveal center when possible (depending on the location of the tumor). As indicated previously, a 2-mm laser spot size was used, and each application was separated by approximately one spot-width to have a standard area of intervening untreated retina to serve as normal control tissue. The ophthalmoscopic appearance of the treatment area was monitored during the entire course of each laser application (60second duration) and after its completion. Any retinal changes were recorded. If visible retinal alterations occurred during the laser application, the treatment at that energy level was immediately stopped and the time noted. After each TTT application, the patients were asked whether they had any symptoms of pain or burning. Visual acuity determination and repeat retinal examination of the involved eye were performed in all patients on the day of the enucleation. TTT was delivered through a Haag–Streit slit lamp by using an 810-nm infrared diode laser (Oculight SLX; Iridex Corp., Mountain View, CA). A laser beam width of 2 mm was used, as indicated previously. A Goldmann 903 (Bern, Switzerland) contact lens was used, with methylcellulose as a coupling agent and 0.5% topical proparacaine for topical anesthesia. (Because of the optics of the Goldmann lens, the actual area treated was predicted to be 2.15 mm with these settings; however, we will refer to this as a 2-mm spot for the purpose of this article.) With the same laser spot size and duration, the power was set at 430, 530, or 630 mW to achieve three different energy-level applications (low, medium, and high doses, respectively.) The
416
TTT medium dose (2-mm spot, 60-second duration, and 530-mW setting) has approximately the same power–spot diameter ratio as TTT with a 3-mm spot size and 800 mW setting (i.e., 265–267 mW/mm) and, theoretically, should result in the same desired degree of hyperthermia (8°–10° C).3 The low and high doses represent energy levels of approximately 20% less and more, respectively, than the medium dose. When possible, each eye received all three doses, with the fovea being treated with the medium dose. If a color change in the retina was observed at a given dose level (e.g., 530 mW), only a dose lower than the one that produced the color change (e.g., 430 mW) was used in that eye thereafter. Immediately after enucleation and before fixation, the eyes were sectioned meridionally through the pars plana. The posterior segment was then carefully and expeditiously submerged in 10% neutral buffered formalin to ensure optimal fixation and prevent artifactual retinal detachment. After fixation, the posterior segments were carefully inspected and photographed with a dissection microscope by using detailed clinical fundus drawings as a guide. Blocks of tissue containing TTT application sites were prepared under the dissecting microscope, embedded in paraffin, and sectioned serially for light microscopy. Sections were stained with hematoxylin– eosin and periodic acid–Schiff. Additional tissue segments containing the tumor and the anterior segment also were submitted for routine diagnostic purposes. Histopathologic analysis and photomicrography were performed by an experienced ophthalmic pathologist (RCE). During analysis, measurements performed with an optical micrometer helped to confirm the location of TTT application sites. Segments of normal retina and choroid adjacent to application sites served as internal control tissue for the purposes of the study.
Results Eight treatment spots from three eyes were analyzed. Six additional treatment spots from two eyes were excluded because artifactual retinal detachment occurred during fixation of the specimen, as indicated previously. All eyes were enucleated because they contained large choroidal malignant melanomas. Details of the findings from the three eyes are described below.
Case 1 The first eye was from an 83-year-old white man who had a large superotemporal ciliochoroidal melanoma in the left eye. The tumor measured 20 ⫻ 20 ⫻ 11.5 mm. It was associated with subretinal fluid that extended into the temporal macula area but spared the fovea. A moderate amount of nuclear sclerosis was present. The visual acuity was 6/30 before the TTT applications. Clinically, the fundus was judged to have a light to medium degree of pigmentation. Three TTT applications were oriented along the horizontal meridian, on either side of the optic nerve. The first application (530 mW) was placed just nasal to, but included, the foveal center. The second application (430 mW) was placed one spot width (2 mm) nasal to the optic nerve, and the final application (630 mW) was placed one spot width nasal to the second spot. No ophthalmoscopically visible retinal changes occurred at any dose level during or immediately after the TTT applications. Furthermore, the patient did not experience any subjective discomfort (pain or burning) during or after any of the three applications. Enucleation was performed the next day. Immediately before the enucleation, the visual acuity was 6/30, and no retinal changes were observed ophthalmoscopically. Gross inspection of the posterior segment after fixation disclosed no TTT-related retinal
Connolly et al 䡠 Histopathologic Effects of Transpupillary Thermotherapy changes (Fig 1). Light microscopic evaluation of serial sections that included TTT applications revealed no abnormalities of the neurosensory retina, RPE, or choroid at any dose.
stroma contained a mild infiltrate of polymorphonuclear leukocytes.
Case 3 Case 2 The second case was a 41-year-old white man. A choroidal melanoma was present inferotemporally in the left eye and measured 13 ⫻ 13 ⫻ 14 mm. The melanoma extended through Bruch’s membrane and had a typical mushroom-shaped configuration. It was associated with a small area of subretinal fluid that extended just inferotemporal to, but did not involve, the foveal center. Before the TTT administration, the visual acuity was 6/12. A mild to moderate amount of nuclear sclerosis was present. The degree of fundus pigmentation was judged clinically to be moderate to heavy. Three TTT applications were also performed in this eye. The first TTT application (530 mW) was placed just nasal to the fovea in attached retina. Because the tumor overhung the temporal half of the macula, full foveal TTT application could not be confirmed. The second TTT application (430 mW) was placed just nasal to the nerve, and the final application (630 mW) was placed a spot-width more nasally, similar to the first case. The patient reported mild discomfort with the high (630 mW) TTT dose starting at 25 seconds into the treatment and continuing for the remainder of the 60-second laser application. No ophthalmoscopically evident retinal changes were seen during or immediately after any of the three TTT applications. The eye was enucleated the next day. On the morning of the enucleation, the visual acuity remained 6/12. Ophthalmoscopically, there was gray discoloration with edema corresponding to sites of the medium- and high-dose TTT applications. It was also evident at this time that the medium-dose application was nasal to (and did not involve) the foveal center. Gross inspection of the enucleated globe confirmed the retinal color changes that were evident on pre-enucleation examination (Fig 2). Light microscopy of serial sections through the three treatment sites revealed increasingly severe tissue alterations that seemed to have a dose–response relationship. The changes in the area treated with the low-dose (430-mW) application were mild and confined to the outer retina (Fig 3). Compared with adjacent normal retina (Fig 4), there was distortion and mild edema of the photoreceptors layer that particularly affected cone inner segments. The cone inner segments were rounded and swollen, and several had detached from their cell bodies and migrated toward the RPE. The photoreceptor nuclei comprising the outer nuclear layer appeared mildly pyknotic. The RPE and choroid were normal. Retinal tissue alterations in the area treated with the mediumdose (530-mW) application were more extensive (Figs 5 and 6). Compared with adjacent untreated retina, the photoreceptor layer appeared thickened, and the photoreceptors were edematous and distorted (Fig 5). The RPE was also mildly thickened, but the choroid appeared normal. There was extensive vacuolization of the inner nuclear layer (Fig 6). Fewer vacuoles were present in the outer part of the outer plexiform layer and the inner part of the photoreceptor layer bordering the external limiting membrane. The outer nuclear layer was irregular, and its nuclei were smaller and pyknotic. The apical surface of some of the RPE cells had an abnormal domed configuration. In the area treated with the high dose (630 mW), vacuolization involved the entire thickness of the retina, which appeared diffusely pale and edematous (Fig 7). The photoreceptors were in disarray, and the RPE was thickened and focally depigmented. The architecture of the inner choroid appeared intact, but its inner
The third eye was from a 40-year-old white man who had a juxtapapillary choroidal melanoma in the left eye. The tumor was located primarily nasal to the optic disc and overhanging it. The tumor measured 7 ⫻ 6 ⫻ 5.6 mm. There was no associated subretinal fluid. The patient’s visual acuity was 6/7.5. This eye was judged to have a moderate degree of fundus pigmentation. A medium TTT dose (530 mW) was applied to the fovea. A mild graying change to the retina was observed at the end of the application. The patient did not experience any subjective discomfort during the procedure but did report a “purplish” discoloration of the vision after this treatment. The second TTT application (430 mW) was placed one spot-width temporal to the first. No retinal changes or symptoms occurred after this application. Because there was a visible change present at 530 mW power, a high-dose (630-mW) application was not used. Therefore, this eye received only two TTT applications. The following morning (the day of the enucleation), the visual acuity had decreased to 6/60. Ophthalmoscopically, there was whitening and edema of the fovea where the medium TTT dose (530 mW) was applied, but there were no changes at the site of the low-dose treatment. This finding was also observed on gross inspection of the enucleated globe. Light microscopy of the medium-dose (530-mW) treatment site revealed extensive damage that involved the retina, RPE, and choroid (Fig 8). The tissue alterations were similar to, but much more severe than, those caused by the high-dose application in case 2. Prominent vacuolization involved all layers of the retina and parts of the inner choroid. The RPE was necrotic and focally detached. Beneath the treatment area, the choroid was largely replaced by a tongue of moderately pigmented uveal melanoma approximately 300 m in thickness, which was not evident clinically. The choroidal tissues were partially necrotic and infiltrated by polymorphonuclear leukocytes. No histopathologic changes were found in the area corresponding to the low-dose (430-mW) application.
Discussion The management of exudative AMD is in a state of evolution. The Treatment of Age-Related Macular Degeneration with Photodynamic Therapy trial demonstrated that photodynamic therapy (PDT) with verteporfin was beneficial for patients with predominantly classic, subfoveal CNV.4 This therapy is now generally considered to be the new gold standard for treating selected eyes that present with predominantly classic, subfoveal CNV associated with AMD. More recently, in a different randomized trial, the Verteporfin in Photodynamic Therapy Study Group demonstrated a modest benefit with the use of verteporfin PDT compared with the control group in eyes with AMD and purely occult CNV that were either small (four disc areas or fewer) or associated with lower levels of visual acuity (approximately 20/50 or less).5 However, before the publication of the 2-year verteporfin PDT results in treating occult CNV, investigators had already started to explore alternative treatments. One such treatment technique was TTT. The first published report on
417
Ophthalmology Volume 110, Number 2, February 2003
418
Connolly et al 䡠 Histopathologic Effects of Transpupillary Thermotherapy Figure 1. Case 1. Macroscopic view of posterior segment of equatorially sectioned eye after fixation. No transpupillary thermotherapy–related changes are evident. Figure 2. Case 2. Horizontally oriented tissue block of the left posterior segment containing the optic disc and fovea. Two transpupillary thermotherapy applications are evident grossly. The 530-mW treatment site nearly abuts the superior nasal border of the fovea. The site of the 630-mW application is evident as a depigmented focus approximately 4 mm nasal to the optic disc. The 430-mW application is not visible. Figure 3. Case 2. The 430-mW transpupillary thermotherapy application. The photoreceptor layer contains scattered vacuoles. Photoreceptors appear mildly swollen and distorted compared with the adjacent normal control retina seen in Figure 4. Cone inner segments are rounded and swollen, and several have detached and migrated toward the retinal pigment epithelium, which appears normal. There is mild pyknosis of photoreceptor nuclei (stain, hematoxylin– eosin; original magnification, ⫻250). Figure 4. Case 2. Normal control retina adjacent to the 430-mW transpupillary thermotherapy application (stain, hematoxylin– eosin; original magnification, ⫻250). Figure 5. Case 2. The 530-mW transpupillary thermotherapy application. The treated area shows prominent vacuolization of the inner nuclear layer and scattered vacuoles external to external limiting membrane (XLM). The photoreceptor layer appears thickened and in mild disarray compared with the adjacent untreated retina. The retinal pigment epithelium overlaying the treatment site is mildly thickened (stain, hematoxylin– eosin; original magnification, ⫻50). Figure 6. Case 2. The 530-mW transpupillary thermotherapy application. Vacuolization involves the inner nuclear layer and, to a lesser extent, the outer part of the outer plexiform layer and the photoreceptor layer bordering XLM. The outer nuclear layer appears somewhat disorderly, with nuclear pyknosis. Some of the retinal pigment epithelium cells have abnormal dome-shaped apical surfaces (stain, hematoxylin– eosin; original magnification, ⫻100). Figure 7. Case 2. The 630-mW transpupillary thermotherapy application. Vacuolization involves the full-thickness retina, which appears pale and edematous. The photoreceptors are in disarray, and the retinal pigment epithelium is thickened and focally depigmented. The inner choroid contains a mild infiltrate of polymorphonuclear leukocytes (stain, hematoxylin– eosin; original magnification, ⫻100). Figure 8. Case 3. The 530-mW transpupillary thermotherapy application overlaying the tongue of pigmented choroidal melanoma. Severe vacuolization involves all layers of the retina and inner choroid. There is focal detachment of necrotic retinal pigment epithelium. Polymorphonuclear leukocytes infiltrate the choroid, which is partially necrotic (stain, hematoxylin– eosin; original magnification, ⫻25). Š
the use of TTT to treat exudative AMD was by Reichel et al in 1999.1 The investigators used an 810-nm infrared diode laser, with spot sizes ranging from 1.2 to 3 mm, a 60-second duration laser application, and power settings ranging from 360 to 1000 mW, depending on the spot size used. Treatment was performed in such a way as to have no visible or only a “barely detectable light-gray appearance” to the CNV lesion at the end of the laser application. Sixteen eyes of 15 patients with acute exudative AMD and subfoveal occult CNV were treated. The study demonstrated decreased exudation after 1 treatment in 15 (94%) of 16 eyes and stabilization or improvement in visual acuity at a final mean follow-up of 13 months in 12 (75%) of 16 eyes. No deleterious effects from the treatment were observed. This favorable initial pilot experience, along with other similarly favorable but uncontrolled data, such as those that were later published by Newsom et al,2 ultimately led to the formation of the TTT4CNV multicenter controlled clinical trial, which is currently under way. The laser settings in the TTT4CNV trial were derived from those used in the pilot studies, which were in agreement with mathematical models of retinal temperature increases in eyes illuminated with various light sources. Both clinical experiences and theoretical mathematical models indicated a constant power– diameter relationship (ratio) for the purpose of inducing and sustaining a thermal increase of approximately 8° to 10° C. A tissue temperature increase of this degree at the level of the RPE theoretically should be safe for the overlying neurosensory retina, and yet it may still be sufficient to induce favorable local tissue changes that could lead to a reduction of exudation and ultimately involution of CNV.3 In comparison, standard threshold retinal laser photocoagulation causes a temperature increase of 20° to 40° C at the level of the retina and RPE and, predictably, results in full-thickness damage to the retina
and other histopathologic alterations that extend deep into the choroid.6 In contrast, PDT should be expected to result in a very minimal degree of temperature increase of approximately 2° to 3° C.3 In the TTT4CNV trial, TTT is being applied with an 810-nm laser and treatment settings of 3-mm spot size, 60-second duration, and 800 mW. As indicated in Materials and Methods, we used laser settings that delivered power levels consistent with the power– diameter ratio currently used in the TTT4CNV trial for the medium dose, 20% less for the low dose, and 20% more for the high dose. In doing so, a spectrum of histopathologic changes was observed and a dose response demonstrated. In our study, retinal changes were more likely to be found (and to a greater degree) histopathologically when higher dose settings were used. However, the energy level of TTT applications alone did not fully account for the occurrence of histopathologic changes in this study. Not surprisingly, we also demonstrated that retinal tissue alterations were more likely to be found in eyes with greater fundus pigmentation, whether it be otherwise normal general pigmentation of the choroid and RPE (such as in case 2) or pathologic increased pigmentation in the choroid from choroidal melanoma (such as in case 3). It is likely that a greater degree of laser energy absorption and a corresponding greater increase in local tissue temperature took place in the area where TTT was applied. Five of eight TTT applications in our study produced either no detectable tissue effects or very limited degree changes confined to the outer retina, as determined by light microscopic analysis. It is highly likely that retinal function was preserved in these areas, especially if no histopathologic changes occurred. This conclusion is supported by the lack of visual acuity change after a medium-dose TTT application to the fovea in the first case, in which the fovea
419
Ophthalmology Volume 110, Number 2, February 2003 appeared normal by light microscopy. Retinal alterations confined to the outer layers, as seen with the low-dose application in case 2, are probably also compatible with preserved (or recoverable) visual function. Similar histopathologic changes were frequently observed in animal studies with verteporfin PDT, and that treatment very rarely causes permanent visual loss in clinical practice.4,7 In three of eight TTT applications, histopathologic evaluation disclosed somewhat greater retinal alterations. Extrapolating from human laser photocoagulation experiments,7 the full-thickness retinal changes that developed at the high TTT dose in case 2 and the medium TTT dose in case 3 are incompatible with preserved retinal function. The loss of visual acuity with foveal TTT in case 3 of our study supports this conclusion. Whether or not the partial-thickness retinal changes observed in the (nonfoveal) area of medium TTT dose application in case 2 are compatible with preserved (or recoverable) retinal function cannot be determined on the basis of our data. The data in this study suggest that significant, full-thickness retinal alterations are not likely to occur in otherwise normal human eyes if the power–laser spot diameter ratio is kept at or below the level used in the TTT4CNV trial (267 mW/mm) and if there is no more than a moderate degree of fundus pigmentation. The likelihood of producing an undesirable degree of neurosensory retinal damage in a normal human model seems to be high if the energy level is increased by more than approximately 20% or if the fundus pigmentation is relatively prominent (either naturally or pathologically). However, extrapolating these data and the laser parameter guidelines above to eyes with exudative AMD has significant limitations. In such eyes, there is subretinal fluid (and other exudative material such as lipid and blood) in the subretinal (or sub-RPE) space. Of course, there is also abnormal fibrovascular tissue (i.e., CNV) at the level of the RPE. It is highly likely that such pathologic features would influence the effects of the local temperature increase induced by TTT. Subretinal fluid, for example, probably helps to minimize any excessive heat-damaging effects of TTT to the neurosensory retina. Theoretically, otherwise normal foveal retina may respond differently to a given TTT energy level than nonfoveal retina because of regional differences in both anatomy (e.g., RPE and xanthophyll pigmentation) and physiology (e.g., choroidal blood flow). Our study design does not allow us to determine whether the degree of hyperthermia
420
induced by a given TTT application differs from one part of the retina to another. However, it is conceivable that a clinically subthreshold TTT application outside the fovea may not be subthreshold when applied to the fovea, and vice versa. In summary, our data show that variable histopathologic retinal changes occur within 24 hours of TTT. The degree of tissue alterations relates to both the energy level and the degree of fundus pigmentation. Unless there is a relatively high degree of (normal or abnormal) fundus pigmentation, or an energy level is used that is greater than what is currently being used in a formal clinical trial, minimal or no histopathologic change by light microscopy occurs in otherwise normal human retina.
References 1. Reichel E, Berrocal AM, Ip M, et al. Transpupillary thermotherapy of occult subfoveal choroidal neovascularization in patients with age-related macular degeneration. Ophthalmology 1999;106:1908 –14. 2. Newsom RSB, McAlister JC, Saeed M, McHugh JDA. Transpupillary thermotherapy (TTT) for the treatment of choroidal neovascularisation [published erratum appears in Br J Ophthalmol 2001;85:505]. Br J Ophthalmol 2001;85:173– 8. 3. Mainster MA, Reichel E. Transpupillary thermotherapy for age-related macular degeneration: long-pulse photocoagulation, apoptosis, and heat shock proteins. Ophthalmic Surg Lasers 2000;31:359 –73. 4. Treatment of Age-related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Photodynamic therapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin. One-year results of 2 randomized clinical trials—TAP report 1 [published erratum appears in Arch Ophthalmol 2000;118:488]. Arch Ophthalmol 1999;117: 1329 – 45. 5. Verteporfin in Photodynamic Therapy Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in agerelated macular degeneration: two-year results of a randomized clinical trial including lesions with occult with no classic choroidal neovascularization—verteporfin in photodynamic therapy report 2. Am J Ophthalmol 2001;131:541– 60. 6. Brooks HL Jr, Eagle RC Jr, Schroeder RP, et al. Clinicopathologic study of organic dye. Laser in the human fundus. Ophthalmology 1989;96:822–34. 7. Kramer M, Miller JW, Michaud N, et al. Liposomal benzoporphyrin derivative verteporfin photodynamic therapy: selective treatment of choroidal neovascularization in monkeys. Ophthalmology 1996;103:427–38.
Originally received: December 27, 2001. Accepted: May 14, 2002. Manuscript no. 211060. 1 Retina Service, Wills Eye Hospital, Philadelphia, Pennsylvania. 2 Pathology Service, Wills Eye Hospital, Philadelphia, Pennsylvania. 3 Ocular Oncology Service, Wills Eye Hospital, Philadelphia, Pennsylvania. Presented at the Association for Research in Vision and Ophthalmology meeting, Fort Lauderdale, Florida, May 2001, and at the annual meeting of the American Academy of Ophthalmology, New Orleans, Louisiana, November 2001. Supported by the Wills Eye Hospital Women’s Committee Macular Degeneration Fund, Philadelphia, PA (CDR); the Eye Tumor Research Foundation, Philadelphia, PA (CLS); the Award of Merit in Retina Research, Houston, TX (JAS); and the Noel T. and Sara L. Simmonds Endowment for Ophthalmic Pathology, Wills Eye Hospital, Philadelphia, PA (RCE). The authors have no proprietary interest in any of the instruments or techniques reported in this article. Reprint requests to Carl D. Regillo, MD, Retina Service, Wills Eye Hospital, 840 Walnut Street, Philadelphia, PA 19107. E-mail: [email protected]. © 2003 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Transpupillary thermotherapy (TTT) has been used to treat subfoveal choroidal neovascularization (CNV) in agerelated macular degeneration (AMD). A small, nonrandomized pilot study by Reichel et al1 showed short-term visual acuity improvement or stability in 12 of 16 eyes with AMD and subfoveal occult CNV treated with TTT. In a recently published clinical series by Newsom et al,2 TTT was used to treat both occult and classic subfoveal CNV in 44 eyes with AMD. In that study, TTT achieved angiographic closure of CNV after treatment in 75% to 78% of eyes, and visual acuity stabilized or improved in 67% to 71% of eyes, with a mean follow-up of approximately 7 months. No patient lost vision in the involved eye immediately after treatment in either study. A controlled clinical trial (TTT4CNV) is currently under way to determine the safety and efficacy of TTT to treat AMD-related subfoveal occult CNV. On the basis of mathematical models, the treatment parameters used in published reports and the ongoing ISSN 0161-6420/03/$–see front matter PII S0161-6420(02)01561-0
415
Ophthalmology Volume 110, Number 2, February 2003 TTT4CNV trial should result in a tissue temperature increase at the level of the retinal pigment epithelium (RPE) of approximately 10° C.3 Such a relatively small degree of hyperthermia should spare the overlying neurosensory retina of significant detrimental heat effects. Although subfoveal TTT seems to be safe in most exudative AMD cases, the exact tissue effects of TTT have yet to be formally analyzed in animal models or human eyes. This study was designed to access the histopathologic effects of TTT on the normal human retina. Different laser energy levels were used in a range that spanned 20% below and above the levels currently in clinical use for a given laser spot size, to determine the full spectrum of potential tissue effects of the treatment.
Materials and Methods The patients eligible to participate in this study were those who were diagnosed with uveal melanoma for which an enucleation was recommended. Enrollment took place between May 2000 and September 2000. The institutional review board of Wills Eye Hospital approved the protocol and the consent form for this study before enrollment of any patients. Before any treatment, all patients underwent a complete ocular examination in the Oncology Service at Wills Eye Hospital. Bestcorrected Snellen visual acuity, intraocular pressure, and all anterior and posterior segment findings were recorded. A detailed fundus drawing and fundus photography were also performed. To be eligible to participate, the patient’s involved eye needed to have clear media and an area of ophthalmoscopically normal-appearing retina along the horizontal meridian on either side of the optic disc (preferably the macula), to accommodate three 2-mm laser spots placed one spot-width apart. Within 1 week of scheduled enucleation, TTT was performed. Up to three applications with different energy levels were placed along the horizontal meridian in the posterior pole, with an attempt to treat the foveal center when possible (depending on the location of the tumor). As indicated previously, a 2-mm laser spot size was used, and each application was separated by approximately one spot-width to have a standard area of intervening untreated retina to serve as normal control tissue. The ophthalmoscopic appearance of the treatment area was monitored during the entire course of each laser application (60second duration) and after its completion. Any retinal changes were recorded. If visible retinal alterations occurred during the laser application, the treatment at that energy level was immediately stopped and the time noted. After each TTT application, the patients were asked whether they had any symptoms of pain or burning. Visual acuity determination and repeat retinal examination of the involved eye were performed in all patients on the day of the enucleation. TTT was delivered through a Haag–Streit slit lamp by using an 810-nm infrared diode laser (Oculight SLX; Iridex Corp., Mountain View, CA). A laser beam width of 2 mm was used, as indicated previously. A Goldmann 903 (Bern, Switzerland) contact lens was used, with methylcellulose as a coupling agent and 0.5% topical proparacaine for topical anesthesia. (Because of the optics of the Goldmann lens, the actual area treated was predicted to be 2.15 mm with these settings; however, we will refer to this as a 2-mm spot for the purpose of this article.) With the same laser spot size and duration, the power was set at 430, 530, or 630 mW to achieve three different energy-level applications (low, medium, and high doses, respectively.) The
416
TTT medium dose (2-mm spot, 60-second duration, and 530-mW setting) has approximately the same power–spot diameter ratio as TTT with a 3-mm spot size and 800 mW setting (i.e., 265–267 mW/mm) and, theoretically, should result in the same desired degree of hyperthermia (8°–10° C).3 The low and high doses represent energy levels of approximately 20% less and more, respectively, than the medium dose. When possible, each eye received all three doses, with the fovea being treated with the medium dose. If a color change in the retina was observed at a given dose level (e.g., 530 mW), only a dose lower than the one that produced the color change (e.g., 430 mW) was used in that eye thereafter. Immediately after enucleation and before fixation, the eyes were sectioned meridionally through the pars plana. The posterior segment was then carefully and expeditiously submerged in 10% neutral buffered formalin to ensure optimal fixation and prevent artifactual retinal detachment. After fixation, the posterior segments were carefully inspected and photographed with a dissection microscope by using detailed clinical fundus drawings as a guide. Blocks of tissue containing TTT application sites were prepared under the dissecting microscope, embedded in paraffin, and sectioned serially for light microscopy. Sections were stained with hematoxylin– eosin and periodic acid–Schiff. Additional tissue segments containing the tumor and the anterior segment also were submitted for routine diagnostic purposes. Histopathologic analysis and photomicrography were performed by an experienced ophthalmic pathologist (RCE). During analysis, measurements performed with an optical micrometer helped to confirm the location of TTT application sites. Segments of normal retina and choroid adjacent to application sites served as internal control tissue for the purposes of the study.
Results Eight treatment spots from three eyes were analyzed. Six additional treatment spots from two eyes were excluded because artifactual retinal detachment occurred during fixation of the specimen, as indicated previously. All eyes were enucleated because they contained large choroidal malignant melanomas. Details of the findings from the three eyes are described below.
Case 1 The first eye was from an 83-year-old white man who had a large superotemporal ciliochoroidal melanoma in the left eye. The tumor measured 20 ⫻ 20 ⫻ 11.5 mm. It was associated with subretinal fluid that extended into the temporal macula area but spared the fovea. A moderate amount of nuclear sclerosis was present. The visual acuity was 6/30 before the TTT applications. Clinically, the fundus was judged to have a light to medium degree of pigmentation. Three TTT applications were oriented along the horizontal meridian, on either side of the optic nerve. The first application (530 mW) was placed just nasal to, but included, the foveal center. The second application (430 mW) was placed one spot width (2 mm) nasal to the optic nerve, and the final application (630 mW) was placed one spot width nasal to the second spot. No ophthalmoscopically visible retinal changes occurred at any dose level during or immediately after the TTT applications. Furthermore, the patient did not experience any subjective discomfort (pain or burning) during or after any of the three applications. Enucleation was performed the next day. Immediately before the enucleation, the visual acuity was 6/30, and no retinal changes were observed ophthalmoscopically. Gross inspection of the posterior segment after fixation disclosed no TTT-related retinal
Connolly et al 䡠 Histopathologic Effects of Transpupillary Thermotherapy changes (Fig 1). Light microscopic evaluation of serial sections that included TTT applications revealed no abnormalities of the neurosensory retina, RPE, or choroid at any dose.
stroma contained a mild infiltrate of polymorphonuclear leukocytes.
Case 3 Case 2 The second case was a 41-year-old white man. A choroidal melanoma was present inferotemporally in the left eye and measured 13 ⫻ 13 ⫻ 14 mm. The melanoma extended through Bruch’s membrane and had a typical mushroom-shaped configuration. It was associated with a small area of subretinal fluid that extended just inferotemporal to, but did not involve, the foveal center. Before the TTT administration, the visual acuity was 6/12. A mild to moderate amount of nuclear sclerosis was present. The degree of fundus pigmentation was judged clinically to be moderate to heavy. Three TTT applications were also performed in this eye. The first TTT application (530 mW) was placed just nasal to the fovea in attached retina. Because the tumor overhung the temporal half of the macula, full foveal TTT application could not be confirmed. The second TTT application (430 mW) was placed just nasal to the nerve, and the final application (630 mW) was placed a spot-width more nasally, similar to the first case. The patient reported mild discomfort with the high (630 mW) TTT dose starting at 25 seconds into the treatment and continuing for the remainder of the 60-second laser application. No ophthalmoscopically evident retinal changes were seen during or immediately after any of the three TTT applications. The eye was enucleated the next day. On the morning of the enucleation, the visual acuity remained 6/12. Ophthalmoscopically, there was gray discoloration with edema corresponding to sites of the medium- and high-dose TTT applications. It was also evident at this time that the medium-dose application was nasal to (and did not involve) the foveal center. Gross inspection of the enucleated globe confirmed the retinal color changes that were evident on pre-enucleation examination (Fig 2). Light microscopy of serial sections through the three treatment sites revealed increasingly severe tissue alterations that seemed to have a dose–response relationship. The changes in the area treated with the low-dose (430-mW) application were mild and confined to the outer retina (Fig 3). Compared with adjacent normal retina (Fig 4), there was distortion and mild edema of the photoreceptors layer that particularly affected cone inner segments. The cone inner segments were rounded and swollen, and several had detached from their cell bodies and migrated toward the RPE. The photoreceptor nuclei comprising the outer nuclear layer appeared mildly pyknotic. The RPE and choroid were normal. Retinal tissue alterations in the area treated with the mediumdose (530-mW) application were more extensive (Figs 5 and 6). Compared with adjacent untreated retina, the photoreceptor layer appeared thickened, and the photoreceptors were edematous and distorted (Fig 5). The RPE was also mildly thickened, but the choroid appeared normal. There was extensive vacuolization of the inner nuclear layer (Fig 6). Fewer vacuoles were present in the outer part of the outer plexiform layer and the inner part of the photoreceptor layer bordering the external limiting membrane. The outer nuclear layer was irregular, and its nuclei were smaller and pyknotic. The apical surface of some of the RPE cells had an abnormal domed configuration. In the area treated with the high dose (630 mW), vacuolization involved the entire thickness of the retina, which appeared diffusely pale and edematous (Fig 7). The photoreceptors were in disarray, and the RPE was thickened and focally depigmented. The architecture of the inner choroid appeared intact, but its inner
The third eye was from a 40-year-old white man who had a juxtapapillary choroidal melanoma in the left eye. The tumor was located primarily nasal to the optic disc and overhanging it. The tumor measured 7 ⫻ 6 ⫻ 5.6 mm. There was no associated subretinal fluid. The patient’s visual acuity was 6/7.5. This eye was judged to have a moderate degree of fundus pigmentation. A medium TTT dose (530 mW) was applied to the fovea. A mild graying change to the retina was observed at the end of the application. The patient did not experience any subjective discomfort during the procedure but did report a “purplish” discoloration of the vision after this treatment. The second TTT application (430 mW) was placed one spot-width temporal to the first. No retinal changes or symptoms occurred after this application. Because there was a visible change present at 530 mW power, a high-dose (630-mW) application was not used. Therefore, this eye received only two TTT applications. The following morning (the day of the enucleation), the visual acuity had decreased to 6/60. Ophthalmoscopically, there was whitening and edema of the fovea where the medium TTT dose (530 mW) was applied, but there were no changes at the site of the low-dose treatment. This finding was also observed on gross inspection of the enucleated globe. Light microscopy of the medium-dose (530-mW) treatment site revealed extensive damage that involved the retina, RPE, and choroid (Fig 8). The tissue alterations were similar to, but much more severe than, those caused by the high-dose application in case 2. Prominent vacuolization involved all layers of the retina and parts of the inner choroid. The RPE was necrotic and focally detached. Beneath the treatment area, the choroid was largely replaced by a tongue of moderately pigmented uveal melanoma approximately 300 m in thickness, which was not evident clinically. The choroidal tissues were partially necrotic and infiltrated by polymorphonuclear leukocytes. No histopathologic changes were found in the area corresponding to the low-dose (430-mW) application.
Discussion The management of exudative AMD is in a state of evolution. The Treatment of Age-Related Macular Degeneration with Photodynamic Therapy trial demonstrated that photodynamic therapy (PDT) with verteporfin was beneficial for patients with predominantly classic, subfoveal CNV.4 This therapy is now generally considered to be the new gold standard for treating selected eyes that present with predominantly classic, subfoveal CNV associated with AMD. More recently, in a different randomized trial, the Verteporfin in Photodynamic Therapy Study Group demonstrated a modest benefit with the use of verteporfin PDT compared with the control group in eyes with AMD and purely occult CNV that were either small (four disc areas or fewer) or associated with lower levels of visual acuity (approximately 20/50 or less).5 However, before the publication of the 2-year verteporfin PDT results in treating occult CNV, investigators had already started to explore alternative treatments. One such treatment technique was TTT. The first published report on
417
Ophthalmology Volume 110, Number 2, February 2003
418
Connolly et al 䡠 Histopathologic Effects of Transpupillary Thermotherapy Figure 1. Case 1. Macroscopic view of posterior segment of equatorially sectioned eye after fixation. No transpupillary thermotherapy–related changes are evident. Figure 2. Case 2. Horizontally oriented tissue block of the left posterior segment containing the optic disc and fovea. Two transpupillary thermotherapy applications are evident grossly. The 530-mW treatment site nearly abuts the superior nasal border of the fovea. The site of the 630-mW application is evident as a depigmented focus approximately 4 mm nasal to the optic disc. The 430-mW application is not visible. Figure 3. Case 2. The 430-mW transpupillary thermotherapy application. The photoreceptor layer contains scattered vacuoles. Photoreceptors appear mildly swollen and distorted compared with the adjacent normal control retina seen in Figure 4. Cone inner segments are rounded and swollen, and several have detached and migrated toward the retinal pigment epithelium, which appears normal. There is mild pyknosis of photoreceptor nuclei (stain, hematoxylin– eosin; original magnification, ⫻250). Figure 4. Case 2. Normal control retina adjacent to the 430-mW transpupillary thermotherapy application (stain, hematoxylin– eosin; original magnification, ⫻250). Figure 5. Case 2. The 530-mW transpupillary thermotherapy application. The treated area shows prominent vacuolization of the inner nuclear layer and scattered vacuoles external to external limiting membrane (XLM). The photoreceptor layer appears thickened and in mild disarray compared with the adjacent untreated retina. The retinal pigment epithelium overlaying the treatment site is mildly thickened (stain, hematoxylin– eosin; original magnification, ⫻50). Figure 6. Case 2. The 530-mW transpupillary thermotherapy application. Vacuolization involves the inner nuclear layer and, to a lesser extent, the outer part of the outer plexiform layer and the photoreceptor layer bordering XLM. The outer nuclear layer appears somewhat disorderly, with nuclear pyknosis. Some of the retinal pigment epithelium cells have abnormal dome-shaped apical surfaces (stain, hematoxylin– eosin; original magnification, ⫻100). Figure 7. Case 2. The 630-mW transpupillary thermotherapy application. Vacuolization involves the full-thickness retina, which appears pale and edematous. The photoreceptors are in disarray, and the retinal pigment epithelium is thickened and focally depigmented. The inner choroid contains a mild infiltrate of polymorphonuclear leukocytes (stain, hematoxylin– eosin; original magnification, ⫻100). Figure 8. Case 3. The 530-mW transpupillary thermotherapy application overlaying the tongue of pigmented choroidal melanoma. Severe vacuolization involves all layers of the retina and inner choroid. There is focal detachment of necrotic retinal pigment epithelium. Polymorphonuclear leukocytes infiltrate the choroid, which is partially necrotic (stain, hematoxylin– eosin; original magnification, ⫻25). Š
the use of TTT to treat exudative AMD was by Reichel et al in 1999.1 The investigators used an 810-nm infrared diode laser, with spot sizes ranging from 1.2 to 3 mm, a 60-second duration laser application, and power settings ranging from 360 to 1000 mW, depending on the spot size used. Treatment was performed in such a way as to have no visible or only a “barely detectable light-gray appearance” to the CNV lesion at the end of the laser application. Sixteen eyes of 15 patients with acute exudative AMD and subfoveal occult CNV were treated. The study demonstrated decreased exudation after 1 treatment in 15 (94%) of 16 eyes and stabilization or improvement in visual acuity at a final mean follow-up of 13 months in 12 (75%) of 16 eyes. No deleterious effects from the treatment were observed. This favorable initial pilot experience, along with other similarly favorable but uncontrolled data, such as those that were later published by Newsom et al,2 ultimately led to the formation of the TTT4CNV multicenter controlled clinical trial, which is currently under way. The laser settings in the TTT4CNV trial were derived from those used in the pilot studies, which were in agreement with mathematical models of retinal temperature increases in eyes illuminated with various light sources. Both clinical experiences and theoretical mathematical models indicated a constant power– diameter relationship (ratio) for the purpose of inducing and sustaining a thermal increase of approximately 8° to 10° C. A tissue temperature increase of this degree at the level of the RPE theoretically should be safe for the overlying neurosensory retina, and yet it may still be sufficient to induce favorable local tissue changes that could lead to a reduction of exudation and ultimately involution of CNV.3 In comparison, standard threshold retinal laser photocoagulation causes a temperature increase of 20° to 40° C at the level of the retina and RPE and, predictably, results in full-thickness damage to the retina
and other histopathologic alterations that extend deep into the choroid.6 In contrast, PDT should be expected to result in a very minimal degree of temperature increase of approximately 2° to 3° C.3 In the TTT4CNV trial, TTT is being applied with an 810-nm laser and treatment settings of 3-mm spot size, 60-second duration, and 800 mW. As indicated in Materials and Methods, we used laser settings that delivered power levels consistent with the power– diameter ratio currently used in the TTT4CNV trial for the medium dose, 20% less for the low dose, and 20% more for the high dose. In doing so, a spectrum of histopathologic changes was observed and a dose response demonstrated. In our study, retinal changes were more likely to be found (and to a greater degree) histopathologically when higher dose settings were used. However, the energy level of TTT applications alone did not fully account for the occurrence of histopathologic changes in this study. Not surprisingly, we also demonstrated that retinal tissue alterations were more likely to be found in eyes with greater fundus pigmentation, whether it be otherwise normal general pigmentation of the choroid and RPE (such as in case 2) or pathologic increased pigmentation in the choroid from choroidal melanoma (such as in case 3). It is likely that a greater degree of laser energy absorption and a corresponding greater increase in local tissue temperature took place in the area where TTT was applied. Five of eight TTT applications in our study produced either no detectable tissue effects or very limited degree changes confined to the outer retina, as determined by light microscopic analysis. It is highly likely that retinal function was preserved in these areas, especially if no histopathologic changes occurred. This conclusion is supported by the lack of visual acuity change after a medium-dose TTT application to the fovea in the first case, in which the fovea
419
Ophthalmology Volume 110, Number 2, February 2003 appeared normal by light microscopy. Retinal alterations confined to the outer layers, as seen with the low-dose application in case 2, are probably also compatible with preserved (or recoverable) visual function. Similar histopathologic changes were frequently observed in animal studies with verteporfin PDT, and that treatment very rarely causes permanent visual loss in clinical practice.4,7 In three of eight TTT applications, histopathologic evaluation disclosed somewhat greater retinal alterations. Extrapolating from human laser photocoagulation experiments,7 the full-thickness retinal changes that developed at the high TTT dose in case 2 and the medium TTT dose in case 3 are incompatible with preserved retinal function. The loss of visual acuity with foveal TTT in case 3 of our study supports this conclusion. Whether or not the partial-thickness retinal changes observed in the (nonfoveal) area of medium TTT dose application in case 2 are compatible with preserved (or recoverable) retinal function cannot be determined on the basis of our data. The data in this study suggest that significant, full-thickness retinal alterations are not likely to occur in otherwise normal human eyes if the power–laser spot diameter ratio is kept at or below the level used in the TTT4CNV trial (267 mW/mm) and if there is no more than a moderate degree of fundus pigmentation. The likelihood of producing an undesirable degree of neurosensory retinal damage in a normal human model seems to be high if the energy level is increased by more than approximately 20% or if the fundus pigmentation is relatively prominent (either naturally or pathologically). However, extrapolating these data and the laser parameter guidelines above to eyes with exudative AMD has significant limitations. In such eyes, there is subretinal fluid (and other exudative material such as lipid and blood) in the subretinal (or sub-RPE) space. Of course, there is also abnormal fibrovascular tissue (i.e., CNV) at the level of the RPE. It is highly likely that such pathologic features would influence the effects of the local temperature increase induced by TTT. Subretinal fluid, for example, probably helps to minimize any excessive heat-damaging effects of TTT to the neurosensory retina. Theoretically, otherwise normal foveal retina may respond differently to a given TTT energy level than nonfoveal retina because of regional differences in both anatomy (e.g., RPE and xanthophyll pigmentation) and physiology (e.g., choroidal blood flow). Our study design does not allow us to determine whether the degree of hyperthermia
420
induced by a given TTT application differs from one part of the retina to another. However, it is conceivable that a clinically subthreshold TTT application outside the fovea may not be subthreshold when applied to the fovea, and vice versa. In summary, our data show that variable histopathologic retinal changes occur within 24 hours of TTT. The degree of tissue alterations relates to both the energy level and the degree of fundus pigmentation. Unless there is a relatively high degree of (normal or abnormal) fundus pigmentation, or an energy level is used that is greater than what is currently being used in a formal clinical trial, minimal or no histopathologic change by light microscopy occurs in otherwise normal human retina.
References 1. Reichel E, Berrocal AM, Ip M, et al. Transpupillary thermotherapy of occult subfoveal choroidal neovascularization in patients with age-related macular degeneration. Ophthalmology 1999;106:1908 –14. 2. Newsom RSB, McAlister JC, Saeed M, McHugh JDA. Transpupillary thermotherapy (TTT) for the treatment of choroidal neovascularisation [published erratum appears in Br J Ophthalmol 2001;85:505]. Br J Ophthalmol 2001;85:173– 8. 3. Mainster MA, Reichel E. Transpupillary thermotherapy for age-related macular degeneration: long-pulse photocoagulation, apoptosis, and heat shock proteins. Ophthalmic Surg Lasers 2000;31:359 –73. 4. Treatment of Age-related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Photodynamic therapy of subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin. One-year results of 2 randomized clinical trials—TAP report 1 [published erratum appears in Arch Ophthalmol 2000;118:488]. Arch Ophthalmol 1999;117: 1329 – 45. 5. Verteporfin in Photodynamic Therapy Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in agerelated macular degeneration: two-year results of a randomized clinical trial including lesions with occult with no classic choroidal neovascularization—verteporfin in photodynamic therapy report 2. Am J Ophthalmol 2001;131:541– 60. 6. Brooks HL Jr, Eagle RC Jr, Schroeder RP, et al. Clinicopathologic study of organic dye. Laser in the human fundus. Ophthalmology 1989;96:822–34. 7. Kramer M, Miller JW, Michaud N, et al. Liposomal benzoporphyrin derivative verteporfin photodynamic therapy: selective treatment of choroidal neovascularization in monkeys. Ophthalmology 1996;103:427–38.