Indocyanine Green-Enhanced Transpupillary Thermotherapy for Retinoblastoma: Analysis of 42 Tumors

Indocyanine Green-Enhanced Transpupillary Thermotherapy for Retinoblastoma: Analysis of 42 Tumors Murat Hasanreisoglu, MD; Jarin Saktanasate, MD; Rachel Schwendeman, NR-CMA; Jerry A. Shields, MD; Carol L. Shields, MD ABSTRACT Purpose: To evaluate the efficacy of indocyanine greenenhanced transpupillary thermotherapy (ICG-TTT) for retinoblastoma that shows suboptimal response to conventional treatments. Methods: A single center, retrospective chart review. The technique involved ICG infusion (range: 0.3 to 0.5 mg/kg) 1 minute prior to applying TTT using the indirect ophthalmoscope technique with a spot size of 1.2 mm. Results: There were 42 retinoblastomas in 30 eyes of 21 patients treated with ICG-TTT. The reasons for ICG enhancement included suboptimal response to standard TTT (n = 31, 74%), recurrence after standard TTT (n = 3, 7%), or minimally pigmented fundus with poor standard TTT uptake (n = 8, 19%). The mean patient age at treatment was 12 months (median: 11.6 months, range: 3 to 31 months). The mean tumor base was 3.5 mm (median: 3 mm), mean tumor thickness was 2.5 mm (median; 2 mm), mean distance to the foveola was 2.6 mm (median: 3 mm), and mean distance to the optic disc was 2.2 mm (median: 0.75 mm). Treatment parameters included a spot size of 1.2 mm, mean power of INTRODUCTION The primary strategy for retinoblastoma treatment is focused on patient survival, with an attempt to salvage the globe and vision. Advanced treatment options such as chemoreduction and targeted therapy using intra-arterial chemotherapy 760 mw (median: 800 mw, range: 400 to 1,200 mw), and mean duration of 4 minutes (median: 4 minutes, range: 0.5 to 14 minutes). Following a median of 2 sessions (range: 1 to 5 sessions) of ICG-TTT, 33 (79%) tumors demonstrated complete regression. The mean tumor thickness postoperatively was 1.7 mm. Two (5%) tumors showed minimal regression after ICG-TTT. During a mean follow-up of 46 months (median: 33 months), tumor recurrence after ICG-TTT developed in 7 (17%) cases at a mean interval of 7 months. Local complications of ICG-TTT included focal paraxial cataract (n = 2, 7%), iris atrophy (n = 1, 3%), and transient retinal hemorrhage (n = 2, 7%). Systemic problems included ICG allergy (n = 1, 5%). Overall, tumor control and globe salvage was achieved in all 30 (100%) eyes. There were no metastatic events. Conclusions: ICG-TTT is an effective alternative for retinoblastoma control, particularly for small tumors that show suboptimal response to standard TTT. [J Pediatr Ophthalmol Strabismus. 2015;52(6):348-354.] combined with intravitreal chemotherapy have led to remarkable tumor control. This previously deadly childhood cancer has become a potentially curable condition during the past 50 years. 1-4 Currently, there is continuing focus toward more selective, targeted therapy for permanently control- From the Ocular Oncology Service, Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. Submitted: May 12, 2015; Accepted: August 13, 2015 Supported by the Eye Tumor Research Foundation, Philadelphia, PA (CLS). The authors have no financial or proprietary interest in the materials presented herein. Correspondence: Murat Hasanreisoglu, MD, Ocular Oncology Service, Suite 1440, Wills Eye Institute, 840 Walnut Street, Philadelphia, PA 19107. E-mail: rmurat95@yahoo.com doi: 10.3928/01913913-20150929-17 348 Copyright SLACK Incorporated

ling retinoblastoma and avoiding enucleation and visual loss. Thermotherapy is a method of delivering heat to tissue to raise local temperature to cytotoxic levels, leading to tissue death. Thermotherapy can be generated with ultrasound, infrared radiation, microwave devices, and other methods. 5 Laser-induced interstitial thermotherapy is a minimally invasive treatment for tumors in the brain, prostate, liver, and eye. 6 Transpupillary thermotherapy (TTT) is a specific form of ocular thermotherapy using infrared radiation (diode laser) and has been most effective for small retinoblastomas that are less than 3 mm in diameter. 6 This technique requires pigment for uptake of the infrared wavelength and is normally accomplished with background choroidal pigment and retinal pigment epithelium. In eyes with minimal pigment (ie, those that manifest chorioretinal atrophy from previous treatment or those occurring in eyes with minimally pigmented fundus) thermotherapy is less effective. 7 In such cases, we have employed a chromophore, indocyanine green (ICG), to achieve the optimal photothermal effect. The peak emission wavelength of the infrared laser for TTT is in the near-infrared zone (810 nm), which coincides with the absorption spectrum of ICG. Theoretically, the therapeutic effect of TTT can be enhanced by the chromophore ICG. 8 We report our experience with ICG- TTT in the treatment of retinoblastoma. PATIENTS AND METHODS A review of computer-coded medical charts at the Ocular Oncology Service at Wills Eye Hospital, Philadelphia, Pennsylvania, was performed to select all of the patients with a diagnosis of retinoblastoma treated with ICG-TTT between January 1997 and January 2014. Institutional review board approval was obtained for this retrospective study. There were 21 consecutive patients with 42 retinoblastomas in 30 eyes identified for inclusion. The collected data included patient age, sex, race, hair color, iris color, and fundus color. The tumor inheritance (sporadic or familial) and laterality (unilateral or bilateral) were recorded. Each eye was assessed for visual acuity, heterochromia, and classification using the International Classification of Retinoblastoma. The treated tumor was analyzed for previous treatments, tumor diameter (mm), thickness (mm), proximity to optic nerve (mm), proximity to foveola (mm), anteroposterior location (between the vascular arcades, along the vascular arcades, vascular arcades to the equator, or equator to the ora), and quadrant location (superotemporal, inferotemporal, inferonasal, superonasal, or macula). Related features of subretinal fluid, active subretinal seeds, and active vitreous seeds were recorded. Each tumor was assessed for need of ICG-TTT. The treatment was performed under anesthesia and delivered by indirect ophthalmoscope-equipped infrared diode laser at 810 nm (Iridex, Mountain View, CA), focused with a 20-diopter lens. The ICG was delivered intravenously from 0.3 to 0.5 mg/kg 1 minute prior to laser treatment. The laser was delivered using a spot size of 1.2 mm to completely heat the tumor using a slow painting technique with care to avoid normal tissue and particularly the papillomacular region. The endpoint goal was a light gray tumor color without causing vascular spasm or rapid tumor whitening. The power was started at 200 mw and increased to or decreased by 50-mW increments until adequate slow onset uptake was observed. Fundus fluorescein angiography using RetCam (Clarity Medical Systems, Pleasanton, CA) and portable handheld spectral-domain optical coherence tomography (ivue; Optovue, Fremont, CA) were performed. Data were collected regarding foveal status, subretinal fluid, and tumor features. Each treatment session was evaluated for details on dose and timing of ICG infusion, infrared power (mw), duration (minutes), endpoint (no uptake, low uptake), and intermediate or high uptake based on color change. Follow-up examinations were performed on a monthly basis until tumor control was achieved. At the most recent visit, visual acuity, tumor status, foveal status, and optic disc status were evaluated. Specifically, the treated tumors were assessed for basal diameter (mm), thickness (mm), regression type (0 = no visible tumor scar; 1 = completely calcified tumor scar; 2 = completely non-calcified tumor scar; 3 = partially calcified tumor scar; and 4 = flat, atrophic chorioretinal atrophy with no residual tumor), tumor regression (no regression, partial regression, complete regression, or recurrence), and treatment complications. Treatment of tumor recurrence was recorded. Journal of Pediatric Ophthalmology & Strabismus Vol. 52, No. 6, 2016 349

TABLE 1 Demographics and Ocular Features of ICG-TTT for 42 Retinoblastomas Features Mean age at initial presentation (m) Mean age at ICG-TTT treatment (m) Sex (n = 21 patients) No. 8 (5.5, 0.5 to 28) 12 (11.6, 3 to 31) Male 13 (62%) Female 8 (38%) Race (n = 21 patients) White 19 (90%) African-American 0 (0%) Asian 1 (5%) Hispanic 1 (5%) Family history of retinoblastoma (n = 21 patients) Tumor laterality (n = 21 patients) 4 (19%) Unilateral 1 (5%) Bilateral 20 (95%) Eye involved (n = 30 eyes) Right eye 17 (57%) Left eye 13 (43%) Iris color (n = 30 eyes) Blue 22 (73%) Brown 8 (27%) Fundus pigmentation (n = 30 eyes) Normal 17 (57%) Light 13 (43%) RESULTS The patient demographics and ocular features of treated eyes are listed in Table 1. There were 42 retinoblastomas in 30 eyes of 21 patients treated with ICG-TTT included in this study. There were 13 (62%) males and 8 (38%) females of White (n = 19, 90%), Hispanic (n = 1, 5%), or Asian (n = 1, 5%) heritage with a median age of 5.5 months (range: 0.5 to 28 months) at the time of retinoblastoma diagnosis. At the time of ICG-TTT, the median age was 11.6 months (range: 3 to 31 months). Lightly pigmented blue iris was noted in 22 (73%) eyes and minimally pigmented (blonde) choroid in 13 (43%) eyes. TABLE 2 Prior and Concomitant Treatments of ICG-TTT for 42 Retinoblastomas Treatments No. (%) TTT prior to ICG-TTT 34 (81) Reason for ICG-TTT Suboptimal tumor response to standard TTT 31 (74) Tumor recurrence after standard TTT 3 (7) Tumor in eye with minimally pigmented fundus Chemotherapy at the time of ICG-TTT (n = 21 patients) 8 (19) Intravenous chemotherapy 20 (95) Intra-arterial chemotherapy 1 (5) Prior treatments and reasons for ICG-TTT are listed in Table 2. The reasons for using ICG-TTT included suboptimal response to standard TTT (n = 31, 74%), recurrence after standard TTT (n = 3, 7%), or minimally pigmented fundus with poor standard TTT uptake (n = 8, 19%). All (100%) of the patients were receiving chemotherapy at the time of the ICG-TTT via intravenous route (n = 20, 95%) or intra-arterial route (n = 1, 5%). The tumor features at the time of ICG-TTT are shown in Table 3. The mean tumor basal diameter was 3.5 mm and mean thickness by ultrasonography was 2.5 mm. The mean tumor distance to the foveola and optic nerve was 2.6 and 2.2 mm, respectively. All of the treated tumors were posterior to the equator and 18 (43%) of 42 tumors were located in the macular region. Treatment parameters are listed in Table 4. The mean number of ICG-TTT sessions was 2.2 (median: 2 sessions, range: 1 to 5 sessions). Despite absence of previous standard TTT uptake, 24 (57%) showed intermediate to high uptake following the first ICG-TTT session. Following the second session of ICG-TTT, uptake was still intermediate to high in 19 (57%) tumors and the uptakes subsequently reduced thereafter. Final tumor thickness decreased from a mean of 2.5 to 1.7 mm. The outcomes are listed in Table 5. Tumor control with complete regression was achieved in 33 (79%) tumors, at a mean follow-up of 46 months 350 Copyright SLACK Incorporated

TABLE 3 Tumor Features Immediately Before ICG-TTT for 42 Retinoblastomas Features No. (%) Anterior-posterior tumor location Macula 18 (43) Macula to equator 24 (57) Quadrant Superior 7 (17) Inferior 7 (17) Nasal 8 (19) Temporal 2 (5) Macular 18 (43) Mean basal diameter (mm) Mean thickness (mm) Mean distance to foveola (mm) Mean distance to disc (mm) 3.5 (3, 0.2 to 8) 2.5 (2, 0.2 to 6) 2.6 (3, 0 to 9) 2.2 (0.75, 0 to 9) (median: 33 months, range: 0.9 to 198 months). Two (5%) of 42 tumors showed minimal regression and 7 (17%) of 42 had recurrence. Recurrence or non-response was treated with cryotherapy (n = 1) or plaque radiotherapy (n = 8). There was no subretinal or vitreous seeding. The complications are listed in Table 6. In 30 eyes, the most important complications included paraxial cataract (n = 2, 7%), iris atrophy (n = 1, 3%), transient retinal hemorrhage (n = 2, 7%) (Figure 1), foveal edema (n = 1, 3%) or foveal atrophy (n = 2, 7%), and optic disc traction (n = 2, 7%) or optic disc atrophy (n = 1, 3%). One (5%) patient had ICG allergy without permanent side effect. Overall, ICG-TTT provided tumor control in 79% and globe salvage in 100% of eyes. No eyes required enucleation and no patient experienced metastasis or death. DISCUSSION Retinoblastoma is a malignancy of early childhood that can lead to death if untreated. The treatment of retinoblastoma involves a critical balance of the child s life with globe salvage and ultimate TABLE 4 Treatment Parameters of ICG-TTT for 42 Retinoblastomas Total Treatment Sessions (n = 93 Sessions) Session #5 (n = 2 Tumors) Session #4 (n = 4 Tumors) Session #3 (n = 12 Tumors) Session #2 (n = 33 Tumors) Session #1 (n = 42 Tumors) Features 760 (800, 400 to 1,200) 750 (750, 700 to 800) 800 (800, 800 to 800) 790 (800, 700 to 800) 760 (800, 400 to 1,200) 745 (800, 400 to 1,200) Mean power (mw) 4, (4, 0.5 to 14) 4 (4, 4 to 4) 3.25 (3, 3 to 4) 3.8 (3.5, 1 to 10) 4.3 (4, 1 to 14) 3.9 (3.00, 0.5 to 10) Mean duration (min) 184 (144, 24 to 756) 180 (180, 168 to 192) 156 (144, 144 to 192) 181 (144, 42 to 480) 200 (144, 45 to 756) 176 (144, 24 to 486) Mean total energy (J) ICG-TTT uptake High 14 (33%) 7 (21%) 2 (17%) 0 (0%) 0 (0%) 23 (25%) Intermediate 10 (24%) 12 (36%) 2 (17%) 0 (0%) 0 (0%) 24 (26%) Low 13 (31%) 9 (27%) 6 (50%) 1 (25%) 2 (100%) 31 (33%) No uptake 5 (12%) 5 (15%) 2 (17%) 3 (75%) 0 (0%) 15 (16%) ; High = white appearance after ICG-TTT; Intermediate = gray appearance after ICG-TTT; Low = light gray appearance after ICG-TTT; No uptake = no color change after ICG-TTT Journal of Pediatric Ophthalmology & Strabismus Vol. 52, No. 6, 2016 351

TABLE 5 Outcomes of ICG-TTT for 42 Retinoblastomas a Feature No. (%) Tumor control (n = 42 tumors) Complete regression 33 (79) Minimal regression 2 (5) Recurrence 7 (17) Treatment of recurrence/ no response (n = 9 tumors) Cryotherapy 1 (11) Plaque radiotherapy 8 (89) Visual acuity (n = 30 eyes) Fix and follow 29 (97) No fix and follow 1 (3) Enucleation (n = 30 eyes) 0 (0) Metastasis (n = 21 patients) 0 (0) Death (n = 21 patients) 0 (0) a Data are given as value (percentage) unless otherwise indicated, some percentage do not sum to 100 because of rounding. TABLE 6 Treatment of Complications of ICG-TTT for 42 Retinoblastomas a Complication Focal findings No. (%) (n = 30 Eyes) Cataract, paraxial 2 (7) Iris atrophy 1 (3) Retinal hemorrhage 2 (7) Foveal edema 1 (3) Foveal atrophy 2 (7) Retinal pigment epithelium abnormalities 1 (3) Optic disc traction 2 (7) Optic disc atrophy 1 (3) Systemic findings (n = 21 patients) ICG allergy 1 (5) a No patients had complications of corneal abnormality, central cataract, vitreous hemorrhage, retinal detachment, foveal epiretinal membrane, foveal subretinal fluid, choroidal neovascularization, optic disc edema, glaucoma, hypotony, or phthisis bulbi. visual potential. Several major treatment modalities have been used to control this condition, including enucleation, radiotherapy (external beam radiation therapy and brachytherapy), and chemotherapy using various delivery routes and chemotherapy protocols. Focal treatments such as cryotherapy, TTT, and laser photocoagulation are considered viable adjuvant treatments to potentiate globe survival and maximize visual prognosis. 1,2 Thermotherapy is a technique in which focused heat is applied to tissues at a subphotocoagulation level to promote cell destruction. This method is used in several subspecialties for oncologic care. 6 Lagendijk is considered the first to use whole eye thermotherapy as an adjuvant treatment for retinoblastoma in combination with radiotherapy. 9 In the early 1990s, he found that thermotherapy enhanced retinoblastoma radiosensitivity so that a lower effective dose of radiotherapy was necessary and radiotherapeutic complications were minimized. 5 In the mid to late 1990s, improved heat delivery systems allowed precise thermotherapy delivery directly to tiny retinoblastomas using a diode laser system delivered through a transpupillary approach. Initially, the laser was delivered through an operating microscope, then through the indirect ophthalmoscope, and later still through a transscleral applicator. The TTT technique became most useful and popular. In 1999, Shields et al. 7 evaluated retinoblastoma control with thermotherapy coupled with chemotherapy in 188 tumors. They found thermotherapy for small tumors less than 3 mm in basal diameter overall provided tumor control in 86% of cases. They identified that complications of laser-induced focal iris atrophy and focal paraxial lens opacity (24%) were found mostly with the operating microscope system in poorly dilated eyes, so they switched to the indirect ophthalmoscope thermotherapy system with fewer complications and equivalent control. 7 Thermotherapy is also important in tumor consolidation for malignant melanoma of the skin and choroid. This method is most effective for heavily pigmented tumors. Amelanotic melanoma, with little pigment, received enhancement with intravenous ICG administration immediately preceding thermotherapy to improve heat uptake. 10,11 The application of ICG dye for thermotherapy enhancement was based on previous observations demonstrating that ICG has slow diffusion out of fenestrated small choroidal vessels and displays an absorption peak of 352 Copyright SLACK Incorporated

Figure 1. Unilateral multifocal retinoblastoma in the left eye of a boy with minimally pigmented fundus. (A) Before treatment, the viable retinoblastomas are noted. (B) Following 5 sessions of intravenous chemotherapy, remnant noncalcified tumor is seen nasal to the optic disc and in the superior foveal region. (C) Following completion of 6 sessions of intravenous chemotherapy and 1 session of indocyanine green-enhanced thermotherapy, the tumor nasal to the disc appeared unchanged. (D) Following chemotherapy and 3 indocyanine green-enhanced thermotherapy sessions, all of the tumors appear regressed, particularly the nasal juxtapapillary lesion and the macular tumor. Figure 2. Bilateral sporadic retinoblastoma in a boy with minimally pigmented fundus. Note the peripheral laser for retinopathy of prematurity. (A) Following 2 sessions of intravenous chemotherapy, tumor along the superotemporal arcade showed limited response to systemic chemotherapy. Note the minimally pigmented fundus appearance. (B) Following 3 sessions of intravenous chemotherapy and 1 session of indocyanine green-enhanced transpupillary thermotherapy (ICG- TTT), good response was observed with regression of the noted tumor. (C) Following 4 sessions of intravenous chemotherapy and 2 sessions of ICG-TTT, further regression of the tumor was noted. (D) Six months after the completion of 6 sessions of intravenous chemotherapy and 4 sessions of ICG-TTT, complete regression of tumor was noted with normal fovea and optic nerve. 805 nm, near the 810 nm diode laser. 10,11 The use of ICG enhancement was then applied to retinoblastoma. We first used this coupling technique in 1997 and found greater uptake in tumor absorption. 10 In the current study of 42 consecutive retinoblastomas that were treated with ICG-TTT, 79% (n = 33) showed complete regression at the 4-year followup (Figure 2). There were several reasons for using ICG enhancement, including recurrence after standard TTT (n = 3, 7%), poor initial response to standard TTT (n = 31, 74%), and recognition of light fundus pigment at risk for poor uptake of standard TTT (n = 8, 19%). Of the 9 tumors that showed recurrence, all of them were salvaged with cryotherapy or plaque radiotherapy. No eyes required enucleation and no patient developed metastasis. Most of the patients in our study were treated before intra-arterial chemotherapy was available. We realize that intra-arterial chemotherapy can provide solid tumor control for recurrent retinoblastoma in Journal of Pediatric Ophthalmology & Strabismus Vol. 52, No. 6, 2016 353

up to 93% of cases and we currently use this modality in our practice. 3 However, there are centers where intra-arterial chemotherapy is not available or situations where there is a desire to avoid the invasiveness and potential complications of intra-arterial chemotherapy; therefore, ICG-TTT remains an option. Intravitreal chemotherapy is another recently popularized modality for control of vitreous retinoblastoma seeds and rarely for subretinal seeds, but with limited documented effect on solid intraretinal tumors, as found in our study. Francis et al. 8 studied 16 eyes with retinoblastoma treated with ICG-TTT in combination with intra-arterial chemotherapy, and found tumor control in all of the cases. We did not find the need for TTT consolidation in eyes treated with intraarterial chemotherapy because the chemotherapy was sufficiently potent to control tumor without the need for additional TTT consolidation. 3 Hence, we used ICG-TTT mostly for retinoblastomas treated with intravenous chemotherapy but rarely for those treated with intra-arterial chemotherapy. ICG-enhanced TTT is an effective method to improve thermotherapy uptake in eyes with nonpigmented retinoblastoma. This approach is most useful in those eyes with blonde fundus and little background pigment. REFERENCES 1. Shields CL, Lally SE, Leahey AM, et al. Targeted retinoblastoma management: when to use intravenous, intra-arterial, periocular, and intravitreal chemotherapy. Curr Opin Ophthalmol. 2014;25:374-385. 2. Shields CL, Fulco EM, Arias JD, et al. Retinoblastoma frontiers with intravenous, intra-arterial, periocular, and intravitreal chemotherapy. Eye (Lond). 2013;27:253-264. 3. Shields CL, Manjandavida FP, Lally SE, et al. Intra-arterial chemotherapy for retinoblastoma in 70 eyes: outcomes based on the International Classification of Retinoblastoma. Ophthalmology. 2014;121:1453-1460. 4. Ghassemi F, Shields CL, Ghadimi H, Khodabandeh A, Roohipoor R. Combined intravitreal melphalan and topotecan for refractory or recurrent vitreous seeding from retinoblastoma. JAMA Ophthalmol. 2014;132:936-941. 5. Legendijk JJW. A microwave heating technique for the hyperthermic treatment of tumors in the eye, especially retinoblastoma. Phys Med Biol. 1995;27:1313-1324. 6. Palanker D, Blumenkranz MS. Retinal laser therapy: biophysical basis and applications. In: Ryan SJ, Schachat AP, Wilkinson CP, Hinton DR, Sadda SR, Wiedemann P, eds. Retina, 5th ed. St. Louis: Mosby.; 2012:746-760. 7. Shields CL, Santos MC, Diniz W, et al. Thermotherapy for retinoblastoma. Arch Ophthalmol. 1999;117:885-893. 8. Francis JH, Abramson DH, Brodie, SE, Marr BP. Indocyanine green enhanced transpupillary thermotherapy in combination with ophthalmic artery chemosurgery for retinoblastoma. Br J Ophthalmol. 2013;97:164-168. 9. Lagendijk JJ. A microwave heating technique for the hyperthermic treatment of tumours in the eye, especially retinoblastoma. Phys Med Biol. 1982;27:1313-1324. 10. Shields CL, Shields JA, Cater J, et al. Transpupillary thermotherapy for choroidal melanoma: tumour control and visual results in 100 consecutive cases. Ophthalmology. 1998;105:581-590. 11. Chong LP, Ozler SA, de Queiroz J, Liggett PE. Indocyanine green-enhanced diode laser treatment of melanoma in a rabbit model. Retina. 1993;13:251-259. 354 Copyright SLACK Incorporated