Retinocytoma/retinoma: comparative analysis of clinical features in 78 tumors and rate of transformation into retinoblastoma over 20 years.

PURPOSE: To explore clinical features and long-term outcomes in patients with retinocytoma/retinoma. METHODS: The medical records of patients with retinocytoma/retinoma over a 20-year period were reviewed retrospectively to compare patient age at presentation (<4 vs ≥4 years), tumor type, and tumor focality (unifocal vs multifocal). RESULTS: Of 2,021 patients with retinoblastoma, 62 (3%; median age, 5 years; 85% white; 58% male) had 78 tumors: 54 retinocytoma (69%) and 24 retinoma (31%). Median basal tumor diameter was 6.0 mm; mean thickness, 2.3 mm. Younger patients (<4 years) were more likely Hispanic (19% vs 2%; P = 0.04), with leukocoria (24% vs 0%; P = 0.003), and with calcification in ≤50% of the tumor (96% vs 70%; P = 0.007). Compared with retinoma, retinocytoma was more prevalent in older patients (median age, 9 vs 2 years; P < 0.001), with fewer symptoms (38% vs 69%; P = 0.04), larger median basal diameter (7.0 vs 3.0 mm; P < 0.001), greater thickness (2.5 vs 1.6 mm; P = 0.02), and less frequently with additional retinoblastoma in either eye (9% vs 71%; P < 0.0001). Compared with multifocal tumors, unifocal tumors occurred more frequently with lack of symptoms (62% vs 25%; P = 0.03), greater median basal diameter (6.0 vs 3.3; P = 0.003), and greater thickness (2.5 vs 1.5 mm; P = 0.006). Tumor transformation into retinoblastoma was found in 2.7% by 2 years, 9.2% by 5 years, 15.3% by 10-20 years. The only factor predictive of transformation was increasing thickness (P = 0.003; hazard ratio of 2.83 per 1 mm increase). CONCLUSIONS: In our study cohort, the rate of retinocytoma/retinoma transformation into retinoblastoma increased from 2 to 10-20 years of age. The only factor predictive of transformation was increasing tumor thickness.

Adenoma and Adenocarcinoma of the Retinal Pigment Epithelium: A Review of 51 Consecutive Patients.

PURPOSE: To describe the clinical and imaging characteristics, pathologic features, and management options of retinal pigment epithelium (RPE) adenoma/adenocarcinoma. DESIGN: Retrospective case series. PARTICIPANTS: Fifty-one patients with RPE adenoma/adenocarcinoma. METHODS: Treatment options for the patients included observation, partial lamellar sclerouvectomy (PLSU), enucleation, and others. MAIN OUTCOME MEASURES: Factors related to visual acuity (VA) outcomes (>2 Snellen lines loss, poor final VA [≤20/200], good final vision [≥20/40]), tumor growth, and need for enucleation. RESULTS: The mean patient age at diagnosis was 51 years, and the majority of patients were white (40/51, 78%) and female (34/51, 67%). Primary management included observation (29/51, 57%), PLSU (9/51, 18%), enucleation (4/51, 8%), or others (9/51, 18%). Outcomes revealed decreased VA (10/32, 31%), poor final VA (17/32, 53%), good final VA (11/32, 34%), tumor growth (12/25, 48%), and need for enucleation (7/51, 14%). By multivariable analysis, features predictive of decreased VA included increasing baseline tumor thickness (P = 0.01) and presence of vitreous hemorrhage (P = 0.05). Factors predictive of poor final VA included presence of exudative retinal detachment (P = 0.004), baseline VA 20/50 to 20/150 (P = 0.008), and baseline VA ≤20/200 (P = 0.01). Absence of feeding and/or draining vessel was predictive of good VA (P = 0.004). Tumor growth was associated with multiple treatments (P = 0.02). Increased tumor thickness (P = 0.03) and presence of exudative retinal detachment (P = 0.01) were predictive of enucleation. CONCLUSION: RPE adenoma/adenocarcinoma can simulate choroidal melanoma. Greater tumor thickness, vitreous hemorrhage, exudative retinal detachment, and poor baseline VA predict worse visual outcome and greater risk for enucleation.

Choroidal Melanoma Simulating Adenoma of the Retinal Pigment Epithelium Arising at the Site of Congenital Hypertrophy of the Retinal Pigment Epithelium.

Neoplasms of the retinal pigment epithelium (RPE) are rare tumors that can simulate choroidal melanoma, but clinical and imaging characteristics often differentiate these lesions. We report a 70-year-old male with an abruptly elevated pigmented lesion that arose at the site of congenital hypertrophy of the RPE and demonstrated associated exudation, as well as feeding and draining vessels, suggestive of RPE adenoma. Optical coherence tomography showed retinal elevation with serous retinal detachment adjacent to the mass, and ultrasonography revealed an abruptly elevated, moderately echodense mass of 6.4-mm thickness. Fluorescein -angiography showed early tumor hypofluorescence, late -tumor hyperfluorescence with staining and leakage, and -retinal vessels buried under the mass, suggestive of a retinal tumor. The patient was monitored with the presumed diagnosis of RPE adenoma, but 3 months later, the growth was documented and fine-needle aspiration biopsy revealed choroidal melanoma. Management with I-125 plaque radiotherapy was performed leading to tumor regression and a thickness of 4.6 mm.

Choroidal Ischemia Sparing the Watershed Zone following Intra-Arterial Chemotherapy for Retinoblastoma.

PURPOSE: Intra-arterial chemotherapy (IAC) has become a mainstay in the management of retinoblastoma, especially in advanced or refractory disease. However, IAC is not without complications, and chemotherapy toxic effects can lead to partial or complete choroidal ischemia, often causing vision loss. METHODS: This is a case report. RESULTS: A 4-month-old girl with bilateral retinoblastoma was treated with secondary IAC (melphalan 5 mg) for recurrent tumor following intravenous chemotherapy. One month later, complete tumor control was achieved. However, she demonstrated broad choroidal ischemia in the nasal and temporal quadrants but sparing of the watershed zone superior and inferior to the optic disc and in the papillomacular region. Fluorescein angiography revealed poor perfusion of the choriocapillaris with visibility of the large choroidal vessels in the nasal and temporal areas but preserved perfusion of the watershed zone. The watershed zone remained intact on the 10-month follow-up, and the final visual acuity was fix and follow without strabismus. CONCLUSION: The pathophysiology of choroidal ischemia is not well understood, but the fortuitous watershed zone preservation in this case could represent uneven distribution of the chemotherapeutic drug, resulting in partial chemo-dilution of the medication in the watershed region, which represents the final downstream overlapping choroidal perfusion from both medial and lateral posterior ciliary arteries.

Primary Intra-Arterial Chemotherapy for Retinoblastoma in the Intravitreal Chemotherapy Era: Five Years of Experience.

PURPOSE: To report our 5-year experience with intra-arterial chemotherapy (IAC) in the intravitreal chemotherapy (IvitC) era. METHODS: Retrospective review of retinoblastoma treated with primary unilateral IAC in the IvitC era (2012-2017). RESULTS: There were 34 eyes treated with IAC alone versus 20 eyes treated with IAC plus IvitC for vitreous seeds. IAC (IAC alone vs. IAC plus IvitC) consisted of melphalan (41 vs. 10%) or melphalan plus topotecan (59 vs. 90%, p = 0.03). IvitC consisted of melphalan (60%) or melphalan plus topotecan (40%). Tumor control and globe salvage were achieved in 100% of group B and C eyes without IvitC. Despite more extensive vitreous seeds in the IvitC group (p < 0.01), comparison of IAC alone versus IAC plus IvitC revealed no difference in tumor control for group D (88 vs. 69%, p = 0.36) or group E (67 vs. 100%, p = 0.25) and no difference in globe salvage for group D (88 vs. 69%, p = 0.36) or group E (58 vs. 57%, p = 0.39). CONCLUSIONS: IAC is effective as primary therapy for unilateral group B, C, D, and E retinoblastoma. IvitC is an important adjuvant therapy to achieve comparable globe salvage rates for group D and E eyes with persistent active vitreous seeds.

CHOROIDAL NEVUS TRANSFORMATION INTO MELANOMA PER MILLIMETER INCREMENT IN THICKNESS USING MULTIMODAL IMAGING IN 2355 CASES: The 2019 Wendell L. Hughes Lecture.

PURPOSE: To analyze risk of nevus transformation into melanoma per millimeter increment. METHODS: Retrospective analysis of 3,806 choroidal nevi for transformation into melanoma per incremental millimeter thickness (flat [≤1.0 mm], thin [1.1-2.0 mm], thicker [2.1-3.0 mm], and thickest [>3.0 mm]) RESULTS:: The median nevus thickness was 1.4 mm, and nevi were categorized (flat, thin, thicker, and thickest) in 1,140 (30%), 2052 (54%), 555 (15%), and 59 (<1%), respectively. There were differences in tumor diameter (2.5, 4.8, 7.5, and 9.3 mm; P < 0.01), optical coherence tomography detection of overlying subretinal fluid (<1, 4, 15, and 11%; P < 0.01), overlying retinal edema (<1, 3, 14, and 25%; P < 0.01), overlying drusen (23, 49, 64, and 64%; P < 0.01), overlying retinal pigment epithelial detachment (1, 4, 4, and 9%; P < 0.01), and overlying lipofuscin hyperautofluoresence (<1, 3, 6, and 7%; P < 0.01). Choroidal nevus transformation into melanoma (n = 90/2,355 cases, 3.8%) was found by Kaplan-Meier 7-year estimates (2.2, 6.1, 31.7, and 34.5%; P < 0.0001) and by hazard ratio (HR) compared with nevus ≤1.0 mm (not available, 4.7 [P = 0.01], 35.7 [P < 0.0001], and 52.0 [P < 0.0001]). For all thicknesses, those with growth displayed increase in mean basal diameter of 2.4 mm and thickness of 1.1 mm, optical coherence tomography increase in subretinal fluid (65%), autofluorescence increase in lipofuscin (40%), and ultrasonography increase in hollowness (30%). Multivariable risk factors, recalled by the mnemonic "To Find Small Ocular Melanoma Doing IMaging" (TFSOM-DIM) representing Thickness >2 mm (ultrasonography), Fluid subretinal (optical coherence tomography), Symptom vision loss (Va), Orange pigment (autofluorescence), Melanoma hollow (ultrasonography), and DIaMeter >5 mm, revealed factors per incremental thickness category (compared with flat) including thin (Fluid overlying, HR 6.1; DIaMeter >5 mm, HR 3.3), thicker (Fluid subretinal ≤3 mm from nevus, HR 5.7; Melanoma acoustic hollowness, HR 2.7), and thickest (Orange pigment, HR 9.1). CONCLUSION: Each incremental increase in choroidal nevus thickness demonstrated risk of growth into melanoma with HR (compared with flat) 4.7 for thin, 35.7 for thicker, and 52.0 for thickest. The increase from ≤2.0 mm to >2.0 mm thickness conferred the greatest rise for transformation.

CHOROIDAL NEVUS IMAGING FEATURES IN 3,806 CASES AND RISK FACTORS FOR TRANSFORMATION INTO MELANOMA IN 2,355 CASES: The 2020 Taylor R. Smith and Victor T. Curtin Lecture.

PURPOSE: To use multimodal imaging for identification of risk factors for choroidal nevus transformation into melanoma. METHODS: Retrospective chart review of 3806 consecutive choroidal nevi with imaging and 2355 choroidal nevi with additional follow up to identify factors predictive of transformation of choroidal nevus into melanoma. RESULTS: The median patient age was 62.5 years and Caucasian race in 3167 (95%). The choroidal nevus demonstrated median basal diameter of 4.0 mm and thickness of 1.4 mm. Imaging included optical coherence tomography (OCT) showing subretinal fluid (SRF) in 312 (9%), ultrasonography (US) with acoustic hollowness in 309 (9%), and hyper-autofluorescence (AF) in 100 (3%). Of those 2355 choroidal nevi with follow up, Kaplan-Meier estimates of nevus transformation into melanoma at 1, 5, and 10 years were 1.2%, 5.8%, and 13.9%, respectively. Multivariate analysis, using multimodal imaging for detection of factors predictive of nevus transformation into melanoma, included thickness >2 mm on US (hazard ratio (HR) 3.80, p < 0.0001), SRF on OCT as cap over nevus (HR 3.00, p < 0.0001) or SRF ≤3 mm from nevus margin (HR 3.56, p = 0.0003), symptomatic vision loss ≤20/50 on Snellen visual acuity (VA) (HR 2.28, p = 0.005), orange pigment (lipofuscin) hyperautofluorescence on AF (HR 3.07, p = 0.0004), acoustic hollowness on US (HR 2.10, p = 0.0020), and tumor diameter >5 mm on photography (HR 1.84, p = 0.0275). These factors can be recalled by the mnemonic "To Find Small Ocular Melanoma Doing IMaging" (TFSOM-DIM) representing Thickness >2 mm (US), Fluid subretinal (OCT), Symptoms vision loss (VA), Orange pigment (AF), Melanoma hollow (US), and DIaMeter >5mm (photography). The mean 5-year estimates of nevus growth into melanoma were 1% (HR 0.8) for those with 0 risk factor, 11% (HR 3.09) with 1 factor, 22% (HR 10.6) with 2 factors, 34% (HR 15.1) with 3 factors, 51% (HR 15.2) with 4 factors, 55% (HR 26.4) with 5 risk factors, and not-estimable with all 6 risk factors. CONCLUSION: In this analysis, multimodal imaging was capable of detecting risk factors for nevus transformation into melanoma, including thickness >2 mm (US), fluid subretinal (OCT), symptoms vision loss (Snellen acuity), orange pigment (AF), melanoma hollowness (US), and diameter >5 mm (photography). Increasing number of risk factors imparts greater risk for nevus transformation into melanoma, including thickness >2 mm (US), fluid subretinal (OCT), symptoms vision loss (Snellen acuity), orange pigment (AF), melanoma hollowness (US), and diameter >5 mm (photography). Increasing number of risk factors imparts greater risk for transformation.

OPHTHALMIC VASCULAR EVENTS AFTER INTRA-ARTERIAL CHEMOTHERAPY FOR RETINOBLASTOMA: Real-World Comparison Between Primary and Secondary Treatments.

PURPOSE: To determine whether treatment order affects ophthalmic vascular event rates after intra-arterial chemotherapy (IAC) for retinoblastoma. METHODS: Patients who received IAC as primary or secondary treatment for retinoblastoma from January 2009 to January 2018 were included. All eyes were imaged with fundus photography and fluorescein angiography. Patient characteristics and vascular event rates were compared using t-test and Fisher's exact test. RESULTS: There were 196 patients treated with 682 infusions of IAC, divided into primary (no previous therapy, 98 eyes of 98 patients, 328 infusions) and secondary (after other therapy, 105 eyes of 98 patients, 354 infusions) treatment. Overall, ophthalmic vascular events were found after 5% of infusions (17% eyes). A comparison of ophthalmic vascular events (primary vs. secondary IAC), with mean three infusions per eye (median 3, range 1-7), revealed no difference in overall percentage of eyes affected (18% vs. 15%, P = 0.57). Adverse vascular events per eye included retinal vasculature attenuation (1% vs. 0%, P = 0.99), peripheral retinal pruning (1% vs. 0%, P = 0.99), branch retinal artery occlusion (0% vs. 1%, P = 0.99), central retinal artery occlusion (0% vs. 1%, P = 0.99), macular ischemia (0% vs. 2%, P = 0.51), vitreous hemorrhage (2% vs. 3%, P = 0.92), subretinal hemorrhage (1% vs. 0%, P = 0.99), retinal pigment epithelium atrophy (6% vs. 3% P = 0.43), choroidal atrophy (4% vs. 2%, P = 0.92), optic disk pallor (1% vs. 0%, P = 0.99), and ophthalmic artery occlusion (9% vs. 6%, P = 0.35). CONCLUSION: Ophthalmic vascular events after IAC for retinoblastoma affect only 5% of eyes per infusion (17% of treated eyes). Vascular event risk per eye is similar when using IAC as primary or secondary treatment.

Combination of multimodal imaging features predictive of choroidal nevus transformation into melanoma.

AIM: To characterise combinations of multimodal imaging risk factors and predictive value for choroidal nevus transformation into melanoma. METHODS: This is a retrospective review of multimodal imaging features for 3806 choroidal nevi from 1 January 2007 through 1 January 2017. Kaplan-Meier estimates and Cox regression analyses were used to calculate 5-year percentages of growth to melanoma and HR. RESULTS: Using multimodal imaging, six risk factors predictive of choroidal nevus transformation into melanoma were identified, namely tumour thickness >2 mm, subretinal fluid, symptoms of visual acuity loss to 20/50 or worse, orange pigment, hollow acoustic density and tumour largest basal diameter >5 mm. Kaplan-Meier 5-year estimated tumour growth was found in 1% of nevi with no risk factors, 11% (range 9%-37%) with one factor, 22% (12%-68%) with two factors, 34% (21%-100%) with three factors, 51% (0%-100%) with four factors and 55% (0%-100%) with five factors. HR for growth was 0.1 with no factor, 2.1-7.8 with one factor, 1.8-12.1 with two factors, 4.0-24.4 with three factors, 4.6-170.0 with four factors and 12.0-595.0 with five factors. The highest HR with each combination of two, three, four or five risk factors always included symptoms of visual acuity loss and orange pigment. CONCLUSION: Six risk factors for choroidal nevus transformation into melanoma by multimodal imaging have been identified. Risk for transformation into melanoma is 1% when no factors are present, and approaches 100% with specific combinations of three or more risk factors. Understanding how combinations of factors influence risk of transformation into melanoma can guide counselling and treatment decisions.

Ophthalmic Vascular Events after Primary Unilateral Intra-arterial Chemotherapy for Retinoblastoma in Early and Recent Eras.

PURPOSE: To assess risk factors for ophthalmic vascular events after intra-arterial chemotherapy (IAC) for retinoblastoma. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients who received unilateral IAC as primary treatment for retinoblastoma from January 1, 2009, to November 30, 2017, at a single center. METHODS: Records were reviewed for patient demographics, tumor features, IAC parameters, and treatment-related vascular events in the early IAC era (2009-2011) compared with the recent era (2012-2017) using the t test and Fisher exact test. Change in event rates over time was assessed using Poisson regression analysis, with Spearman's rho used to test correlation. MAIN OUTCOME MEASURES: Rate of IAC-induced ophthalmic vascular events. RESULTS: There were 243 chemotherapy infusions in 76 eyes of 76 patients, divided into early (22 eyes, 57 infusions) and recent (54 eyes, 186 infusions) eras. Intra-arterial chemotherapy consisted of melphalan (243 infusions), topotecan (124 infusions), and carboplatin (9 infusions). A comparison (early vs. recent era) revealed fewer mean number of infusions (2.6 vs. 3.4, P = 0.02) with similar mean patient age and presenting tumor features. Event rates decreased over time (P < 0.01), with fewer ophthalmic vascular events (early era vs. recent era) in the recent era (59% vs. 9% per eye, 23% vs. 3% per infusion, P < 0.01), including peripheral retinal nonperfusion (5% vs. 2% per eye, P = 0.50), vitreous hemorrhage (9% vs. 2%, P = 0.20), subretinal hemorrhage (0% vs. 2%, P = 0.99), branch retinal vein occlusion (5% vs. 0%, P = 0.29), choroidal ischemia (14% vs. 4%, P = 0.14), and ophthalmic artery spasm/occlusion (27% vs. 0%, P < 0.01). Events did not correlate to patient age (P = 0.75), tumor diameter (P = 0.32), tumor thickness (P = 0.59), or cumulative dosage of melphalan (P = 0.13) or topotecan (P = 0.59). There were no IAC-induced vascular events in 72 infusions of 21 consecutively treated eyes in 2016 to 2017. CONCLUSIONS: Ophthalmic vascular events after IAC have decreased from the early era (2009-2011) through the current era (2012-2017) at this center. Experience performing this highly specialized procedure could be an important factor predicting IAC-related vascular events.